HDR Version Finale COMPLET

Habilitation à diriger des recherches
CARACTERISATION ET FACTEURS DE DEGRADATION DE LA QUALITE DE

L’EAU. RECHERCHE D’INDICATEURS DE RISQUES ENVIRONNEMENTAUX
ET SANITAIRES
Présentée par
Aude‐Valérie JUNG
Enseignante‐Chercheure Ecole des métiers de l’Environnement (EME, Bruz)
Chercheure associée LERES (EHESP‐IRSET U1085 INSERM, Rennes)
Soutenue le 12 décembre 2016 devant la commission d’examen :
Dr. Hélène Budzinski, Université Bordeaux 1, EPOC ‐ UMR 5805 CNRS

Pr. Geneviève Feuillade, Université Limoges, GRESE ‐ EA 4330
Pr. Montserrat Filella, Université de Genève, Institut F.‐A. Forel
Dr. Stéphane Mounier, Université de Toulon, PROTEE ‐ EA 3819
Pr. Barbara Le Bot, Université Rennes 1, LERES/EHESP, IRSET ‐ UMR 1085
Dr. Gérard Gruau, Université Rennes 1, Géosciences UMR 6118
ECOLE DES METIERS DE L’ENVIRONNEMENT

Etablissement privé d’enseignement supérieur - Reconnu par l’Etat
Habilité à recevoir la taxe d’apprentissage
Campus de Ker Lann – Tél : +33(0)2 99 05 88 00 Association loi 1901

Rennes contact@ecole-eme.fr SIRET : 392 925 129 000 26
Avenue Robert Schuman www.ecole-eme.fr APE : 8542 Z
35170 BRUZ - FRANCE
1
2
« Il n’y a pas de hasard, il n’y a que des forces en marche, à toi de les créer et le hasard
suivra »
ANTOINE DE SAINT‐ EXUPERY
« La vie, ce n'est pas les molécules, mais les liens qui existent entre elles ».
Pr LINUS PAULING, Prix Nobel de chimie et Prix Nobel de la paix
3
4
Remerciements
J’adresse mes remerciements à Jean‐Luc Bersillon, ancien Professeur au Laboratoire

Environnement et Minéralurgie (LEM) à Vandoeuvre‐lès‐Nancy qui fut mon directeur de thèse
ainsi qu’à Céline Frochot du Laboratoire des Réactions et du Génie des Procédés (LRGP‐Nancy)
et Bruno Lartiges du Laboratoire Géosciences Environnement de Toulouse pour leurs conseils et
co‐encadrements lors de mes « premiers pas » en recherche en tant que jeune chercheure.
Merci à Olivier Thomas, Professeur émérite au Laboratoire d’Etudes et de Recherche en

Environnement Santé (LERES‐EHESP Rennes) de m’avoir accueillie dans son laboratoire en
2009 en tant que chercheure associée, et ainsi de m’avoir permis de travailler à ses côtés sur des
projets nationaux et européens. La spectroscopie UV a encore (il me semble) de beaux jours à
venir en terme de développement et je le remercie de m’avoir initiée à cet outil !
Merci à Philippe Quénel, directeur actuel du LERES, de continuer à me faire confiance dans la
collaboration scientifique engagée. Je nous souhaite un partenariat fructueux à l’interface santé‐
environnement au sein de l’Institut de Recherche‐Santé‐Environnement‐Travail (IRSET, UMR
INSERM 1085).
Je remercie vivement Geneviève Feuillade, Professeure à l’Ecole Nationale Supérieure

d’Ingénieurs de Limoges, Monserrat Filella, Chercheure et Professeure à l’Institut Forel de
l’Université de Genève, et enfin Hélène Budzinski, Directrice de recherche du Laboratoire de
Physico‐Toxicochimie de l’Environnement (LPTC) à l’Université de Bordeaux, pour l’honneur
qu’elles me font en acceptant d’être rapporteurs de ce travail. Un grand merci à Stéphane
Mounier, Directeur de recherche au Laboratoire Processus de Transferts et d’Échanges dans
l’Environnement (PROTEE) à l’Université de Toulon, Barbara Le Bot, Professeure à l’Université
de Rennes 1 et chercheure au LERES – EHESP et à Gérard Gruau, Directeur de Recherche de
l’Equipe « Transferts » à l’Université de Rennes 1 (Géosciences Rennes, UMR 6118) pour le
plaisir qu’ils me font de participer au jury de la soutenance orale de ce travail.
Je tiens aussi à remercier l’ensemble de mes collègues de l’Ecole des Métiers de l’Environnement
(EME) et du LERES pour leur sympathie et leur bonne humeur, en particulier Marie‐Florence
Thomas pour son soutien et ses conseils précieux pendant les quelques années passées à ses
côtés en tant que responsable recherche à l’EME puis aujourd’hui en tant que collègue au LERES.
J’exprime également ma reconnaissance à mes anciennes collègues de l’EME Laurence Legal et
Valérie Nouvel pour avoir initié les projets Nacre, DCNS et Record présentés dans ce rapport.
Je remercie aussi l’ensemble des ingénieurs d’étude et de recherche ayant participé aux résultats
expérimentaux, et notamment Jaafar Ghanbaja (Institut Jean Lamour, Université de Lorraine)
pour la mise en œuvre de la microscopie électronique à transmission, Mustapha Abdelmoula,
chercheur au LCPME de Nancy pour la spectroscopie Mossbauer, Delphine Pellé, responsable du
Pôle Analyse du LERES et toute l’équipe du pôle de chimie pour les nombreuses analyses
réalisées notamment lors des thèses de Ianis Delpla et de Jean Causse.
Enfin, je ne peux terminer sans un immense merci à tous les étudiants ayant travaillé à mes côtés
sans qui tout cela n’aurait jamais existé, que ce soient les étudiants ingénieurs en dernière année
de l’EME au travers des nombreux travaux d’ingéniérie encadrés, ou plus particulièrement
encore Ianis Delpla et Jean Causse lors de leurs travaux de thèse : vous m’avez permis de
découvrir à vos côtés le cocktail très vivant des doutes et de questionnements, des moments de
suspens et de rebondissements … mais surtout de petites et grandes joies qui font le quotidien
d’un encadrement de doctorats, particulièrement quand il s’agit d’études environnementales de
terrain !
5
6
Présentation chronologique
Après des classes préparatoires scientifiques aux grandes écoles (PCSI, PC) au
lycée A. Schweizer de Mulhouse, j’ai intégré l’école de chimie de Mulhouse (ENSCMu) en
1998 (diplôme d’ingénieur obtenu en 2001). Pendant cette scolarité, j’ai réalisé ma
dernière année en suivant des cours complémentaires me permettant de valider un DEA
de chimie‐physique de l’université de Haute‐Alsace. J’ai ensuite intégré le Laboratoire
Environnement et Minéralurgie de Nancy (LEM, UMR 7569)1 afin de réaliser un doctorat
de géosciences. L’objectif de ce travail de thèse (co‐financé CNRS‐Région) était de
réaliser une étude à l’échelle d’une zone atelier (Zone Atelier Moselle) afin de
caractériser la matière organique naturelle et ses interactions avec les micropolluants,
dans l’objectif final de mieux comprendre les phénomènes de transport des
xénobiotiques à l’échelle d’un bassin versant. Un volet « traitement de l’eau » a
également été réalisé lors de cette thèse pour comprendre les mécanismes d’échange
entre les polluants, la matière organique et le coagulant utilisé. De 2006 à 2007, j’ai
intégré un service de santé au travail (ALSMT, Nancy) en tant qu’ingénieur prévention.
Ce service assure le suivi médico‐professionnel de plus de 140 000 salariés répartis sur
plus de 12 000 entreprises adhérentes dans le département de Meurthe et Moselle. Dans
le contexte de la réforme de la médecine du travail (démarrée en 2004), j’ai donc rejoint
l’équipe pluridisciplinaire afin de réaliser des évaluations de risques chimiques au sein
des entreprises adhérentes et conseiller les médecins du travail sur les bases de données
de toxicologie existantes. Ce fut l’opportunité de mettre en œuvre sur le terrain, et pour
des secteurs d’activités variés, des méthodes d’évaluation des risques professionnels en
particulier chimiques, et également d’acquérir des compétences en métrologie
atmosphérique. J’ai à ce titre obtenu une habilitation en tant qu’intervenante en
prévention des risques professionnels (IPRP) en 2006, délivrée par le Collège inter‐
régional CARSAT Alsace‐Moselle. En 2007, en suivant mon souhait initial de travailler
dans l’enseignement et la recherche, je suis devenue enseignante‐chercheure à l’Ecole
des Métiers de l’Environnement (EME) à Bruz. J’y enseigne la chimie générale et
organique en classe préparatoire, ainsi que la caractérisation des eaux en cycle
ingénieur. Grâce à mon expérience professionnelle en tant qu’IPRP, je suis également
enseignante dans les modules de sécurité/santé au travail et des installations classées
pour la protection de l’environnement. Ces deux derniers modules sont également
dispensés à un autre niveau aux étudiants Bachelor « Coordinateurs en Environnement »
de l’EME, et en anglais dans le Master of Science « Sustainable Management and Eco‐
Innovation » en partenariat avec l’Ecole Supérieure de Commerce de Rennes. A partir de
2007, j’ai mené différents contrats d’études en tant que consultante pour l’EME,
notamment une revue bibliographique sur la mise en place de tests de génotoxicité pour
les salariés du déchet, via le Réseau Coopératif de Recherche sur les Déchets (RECORD).
Ce travail a été réalisé en partenariat avec Olivier Fardel et Laurent Verhnet (INSERM
U620‐SeRAIC) de 2008 à 2009. J’ai également été consultante pour la DCNS de Brest
quant à la problématique de substitution de leurs produits cancérigènes, mutagènes et
reprotoxiques (2008‐2009). J’ai également été responsable scientifique pour l’EME du
projet NACRE (NAvire Conduit pour le Respect de l’Environnement), labellisé par le Pôle
Mer Bretagne (2007‐2013). Pour ce projet, j’ai pu acquérir des compétences en
évaluation des impacts environnementaux dans le domaine maritime, et supervisé
1
Aujourd’hui rattaché au Laboratoire Interdisciplinaire des environnements continentaux (LIEC, UMR
7360)
7
plusieurs projets d’ingéniérie réalisés par les étudiants ingénieurs de dernière année de
l’EME.
En 2009, je retourne à la recherche plus académique et je deviens chercheure associée

du LERES (EHESP‐IRSET) pour appuyer le volet « caractérisation de la matière
organique » pour le projet européen ERA‐NET « Environment change and rising DOC
trends ». Avec ce projet, j’ai pu appréhender l’impact du changement climatique sur la
qualité des eaux de surface, en me concentrant particulièrement sur le carbone
organique dissous et découvrir les nombreuses possibilités d’exploitation de la
spectrophotométrie UV, d’après les méthodes de travail développées par Olivier
Thomas, ancien directeur du LERES. J’ai également pu participer à l’encadrement de la
thèse de Ianis Delpla dans le cadre de ce projet européen. En 2012, le projet Oracle, en
partenariat avec Frédéric Pitois (SARL Limnologie, Rennes) fut l’opportunité de
m’intéresser à la problématique des cyanobactéries dans les lacs bretons et ainsi
acquérir une expérience complémentaire concernant la qualité des eaux naturelles, en
termes d’évaluation des risques sanitaires (eaux de baignade). J’ai pu participer ainsi au
suivi de 2 étudiantes de master 1 de l’EME ayant réalisé leurs stages dans le cadre de ce
projet. En 2012, j’ai débuté le co‐encadrement de thèse de Jean Causse, pour un projet
initié par Coop de France Ouest, regroupement d’entreprises agricoles, concernant les
questions de temporalité des transferts de nutriments dans les bassins versants à
algues vertes. Ce projet m’a permis d’élargir la problématique de la contamination
organique des eaux à d’autres flux de nutriments (formes azotées et phosphorées) et
d’appréhender l’importance du rapport C/N dans l’environnement. La valorisation de ce
travail, aujourd’hui toujours en cours, permettra je l’espère de développer de nouveaux
indicateurs de vulnérabilité des zones environnementales à risque, notamment
soumises à l’eutrophisation.
8
CURRICULUM VITAE
Informations personnelles
Née le 11/10/1978
1 place du vert buisson, 35 170 BRUZ, Tel 06.33.80.31.96
Situation actuelle :
Enseignant‐Chercheur en chimie/sciences de l’environnement depuis 2007 à
l’École des Métiers de l’Environnement (EME) de Rennes – Chercheure associée au
Laboratoire d’Etudes en et de Recherche en Santé Environnement (LERES) Ecole
des Hautes Etudes en Santé Publique de Rennes (35) Institut de Recherche Santé
Environnement Travail (UMR U 1085) depuis 2009
Expérience professionnelle :
2006 – 2007 Ingénieur prévention « risques chimiques », Association Lorraine
de Santé en Milieu de Travail, 12 200 entreprises, 147 000 salariés, Nancy (54)
 Création de bases de données, de fiches de données simplifiées, fiches
de postes des produits chimiques utilisés par les entreprises adhérentes
 Evaluation des risques chimiques en entreprise: méthodologie INRS,
protocoles de métrologie développés en coopération avec des laboratoires
d’analyse
 Conseils aux médecins du travail et aux entreprises (hygiène, sécurité,
environnement, toxicologie)
 Veille technologique et législative (Reach, Plans Santé‐Environnement et
Santé au Travail)
 Encadrement de stagiaires (Master I toxicologie environnementale de
Metz), formation dispensée aux médecins du travail, aux assistantes
techniques, conférences (chef de projet de rédaction de plaquettes grand
public)
Diplômes et formation :
Diplômes
2004 Docteur en Géosciences (Interactions de la matière organique avec les
micro polluants et devenir lors des opérations de traitement de l’eau
par coagulation), Laboratoire Environnement et Minéralurgie, UMR CNRS
7569, Institut National Polytechnique de Lorraine (INPL)
2001 Ingénieur chimiste (spécialité « hygiène, sécurité et
environnement »), Ecole Nationale Supérieure de Chimie de Mulhouse
(ENSCMu) et Diplôme d’Etudes Approfondies de Physique‐Chimie,
Université de Haute Alsace
9
Stages professionnels et autres formations
2015 Comprendre la santé publique et le système de santé, MOOC EHESP
2014 IED: quelles obligations pour les ICPE ? CCI, Paris

Eléments de santé au travail pour les ingénieurs et managers, MOOC
FUN, CNAM
2013 Open‐innovation, identifier les réseaux d’experts internationaux sur vos
projets R et D, Techniques des Sciences de l’Ingénieur, Paris
Matinales Socotec, REACH et risques chimiques, Socotec, Cesson‐Sévigné
Certificat basique « gestion de projet », 1er MOOC de France, Ecole
Centrale Lille
2012 Piloter et suivre l’évolution de ses installations classées (ICPE),
Techniques des Sciences de l’Ingénieur, Paris
2010 Nouvel étiquetage CLP/SGH des produits chimiques, Techniques des
Sciences de l’Ingénieur, Paris
2007 Réaliser une campagne de prélèvements d’atmosphère, Institut National
de Recherche en sécurité (INRS), Nancy
2007 Le risque biologique, Institut National de Recherche en sécurité (INRS),
Paris
2006 Habilitation technique professionnelle en tant qu’Intervenant en
Prévention des Risques Professionnels (IPRP), Collège inter‐régional
CARSAT Alsace‐Moselle
Qualifications aux fonctions de Maître de Conférences

Section 31 CNU depuis 2005
Activités d’enseignement et ingénierie pédagogique

Enseignement
Activités d’enseignement en cours
Chimie générale, 1ère année cycle ingénieur, 14h cours, 10h TD
Chimie organique, 2ème année cycle ingénieur, 16h cours, 10h TD
Caractérisation des eaux potables et usées, 3ème année cycle ingénieur, 20h TP
Santé sécurité au travail, 4ème année cycle ingénieur, 10h cours, 6h TD
Installations classées pour la protection de l’environnement (ICPE), 4ème année cycle
ingénieur, 6h TD
Prélèvements d’eaux de surface, 5ème année cycle ingénieur, 4h TP
Risque chimique, 2ème année Bachelor Coordinateurs en environnement, 4h de cours, 6h
de TD
Santé sécurité au travail, 3ème année Bachelor Coordinateurs en environnement, 10h
cours, 6h TD
Installations classées pour la protection de l’environnement (ICPE), 3ème année Bachelor
Coordinateurs en environnement, 2h cours, 6h TD
ICPE/REACH : 11h d’enseignement en anglais, module « International comparative
sustainable development laws », Master of Science in Eco‐Design and sustainable
management, Ecole Supérieure de Commerce de Rennes
Conférences 1ère, 2ème, 5ème année (6h): Santé environnement, REACH, Caractérisation de
la matière organique par spectroscopies
10
Activités d’enseignements antérieures :
Toxicologie, Diplôme d’étude professionnelle spécialisée (DEPS), 4h cours, 6h TD (2007
à 2012)
Toxicologie appliquée au bâtiment, Diplôme d’étude professionnelle spécialisée (DEPS,
4h cours (2010)
Chimie des solutions (PH métrie‐complexométrie‐oxydo‐réduction), 1ère année cycle
préparatoire intégré, 16h cours, 10h TD (2008)
Caractérisation des eaux potables et usées, Diplôme d’étude professionnelle spécialisée
(DEPS, (2007‐2010),4h cours, 4h TD, 8h TP
Ecotoxicologie, 4ème année cycle ingénieur, 6h cours, 4h TD (2007‐2009)
Génie des procédés/techniques membranaires et distillation, 4ème année cycle ingénieur,
6h cours, 6h TD (2007‐2009)
REACH, 5ème année cycle ingénieur et étudiants de l’école supérieure de commerce
(enseignement commun dans le parcours Eco‐management) (2010‐2013)
Activités d’enseignements antérieures données dans d’autres établissements:

Santé sécurité au travail/ICPE, 4ème année Ecole Supérieure de Logistique Industrielle
(ESLI, Redon), 14h cours, 7h TD (2008‐2010)
Vacations TP pilote de filtration sur sable, 4ème année Ecole Nationale Supérieure de
Géologie (ENSG, Nancy), 4h TP (2003‐2004)
Caractérisation de la matière organique naturelle, 4ème année Ecole Nationale Supérieure
de Géologie (ENSG, Nancy), 3h cours (2003‐2004)
Ingénierie
Création des modules « sécurité/prévention en entreprise » et ICPE 4ème année cycle
ingénieur, intégrant les nouvelles technologies d’informations et de communication
pour l’enseignement (TICE)
Responsabilité pédagogique
Responsable 1ère année cycle ingénieur (2008‐2009) : organisation générale du parcours
pédagogique et des projets, stages d’observation, conseils de classe, relation avec les
parents d’élèves
Responsable des travaux d’ingénierie, 5ème année cycle ingénieur (depuis 2011):
prospection de sujets d’étude auprès d’entreprises et de collectivités et gestion des
conventions entre les entreprises et les laboratoires de recherche, suivi des prestations
réalisées par les étudiants pour les partenaires, organisation de journée de restitution
ouverte au public avec membres de jury externes à l’EME (entreprises, conseils
généraux, chambre de commerce et d’industrie, bureaux d’étude)
Activités d’encadrement
> Co‐encadrement de thèse et participation à l’encadrement de thèse (directeur :
Thomas O., LERES‐EHESP) Doctorats (Ecole doctorale Vie‐Agro‐Santé, Université
Rennes 1) :
Causse J., (2012‐15) projet Algues vertes, collaboration Coop de France Ouest.
Jury de thèse 30 juin 2015.
Delpla I., (2008‐11) Contribution à l’évaluation de l’impact des changements
climatiques sur la qualité des eaux destinées à la consommation humaine, dans le cadre
11
du projet européen « ERA‐Environment and Health ». Jury de thèse le 16 décembre
2011
> Encadrement de projets d’ingéniérie (5ème année cycle ingénieur EME) :
2009 : Élaboration d’une méthode d’étude d’impact appliquée aux rejets des navires,
projet NACRE (Navire conduit dans le respect de l’environnement)
2010 : Evaluation des impacts environnementaux des rejets des navires, projet NACRE
Mise en place d’un système de sécurité au travail sur le bâtiment de l’EME
2011 : Méthodes d’analyses environnementales comparatives de rejets de navires, projet
NACRE
2012 : Gain environnemental, résultats finaux projet NACRE
2013 : Identification des situations à risque de contamination microbienne, projet ISIS
Audit énergétique du bâtiment de l’ECAM
2014 :Caractérisation de la matière organique naturelle et agricole par
spectrophotométrie UV
Valorisation des eaux dans la ville (projet Valocity avec le cluster Eco‐origin,
LERES/EHESP, Hydrocity)
2015 : Identification des substances dangereuses dans les effluents hospitaliers (RSDE),
recherche des sources et élaboration d’un plan d’actions
> Encadrement de projets bibliographiques, étudiants 3ème année cycle ingénieur

EME
> Encadrements de stage d’étudiants (4ème et 5ème année cycle ingénieur EME) et
de projets d’étudiants ERASMUS (Masters)
Activités de Recherche
Participation à des projets de recherche ou de partenariats
1. Mise en place de test de génotoxicité pour les salariés du déchet, Réseau Coopératif
de Recherche sur les Déchets (RECORD), partenaire avec Olivier Fardel et Laurent
Verhnet (INSERM U620‐SeRAIC), 2008‐2009, RECORD
2. Environment change and rising DOC trends: implications for public health (CCWQ).
Programme ERANET Environment and Health (7ème PCRDT). Partenaires U. Bangor
(UK), U. Ultrecht (NL), CEH (UK), Anses, 2009‐2012, chercheure associée
3. Coop de France Ouest, Impact de facteurs émergents sur le transfert de nutriments à
l’échelle d’un bassin versant pilote, contrat CIFRE, 2012‐2015, co‐encadrante de
thèse
4. Oracle, Objectiver le Risque lié à l’Apparition de Cyanobactéries et toxines dans Les
ressources en Eau, en partenariat avec Frédéric Pitois (SARL Limnologie, Rennes),
2012 – 2015, APR EST 2011 ANSES, chercheure associée, co‐encadrante 2 stagiaires
M1
12
Coordination de projets de recherche ou de partenariats
 NACRE (NAvire Conduit pour le Respect de l’Environnement), Pôle Mer Bretagne,

2007‐2013, responsable scientifique du projet pour l’EME
 Substitution des produits cancérigènes, mutagènes et reprotoxiques, DCNS Brest,
consultante, expertise, 2008‐2009
Formation de personnel hautement qualifié
Co‐encadrante thèse Jean Causse (2012‐2015) Impact de facteurs émergents sur

le transfert de nutrients à l’échelle d’un bassin versant pilote, contrat CIFRE, sous la
direction d’Olivier Thomas.
Participation à l’encadrement de thèse de Ianis Depla (2008‐2011) Contribution à
l’évaluation de l’impact des changements climatiques sur la qualité des eaux destinées à
la consommation humaine, soutenue le 16 décembre 2011, sous la direction d’Olivier
Thomas.
Co‐encadrement d’étudiants ingénieurs 4ème année dans le cadre de projets de
recherche :
o Mathilde Paque, 2014, Etude de la relation carbone‐azote par
spectrophotométrie UV
o Aurélie Fuchez, 2013, Objectiver le risque lié à l’apparition de
cyanobactéries : mise en place d’un diagnostic rapide
o Caroline Farenc, 2012, Objectiver le risque lié à l’apparition de
cyanobactéries : étude des données statistiques (projet ORACLE)
Production scientifique2
ACL
JUNG A.‐V., FROCHOT C., BERSILLON J.‐L., 2015, Fluorescence spectroscopy as a specific
tool for the interaction study of two surfacants with natural and synthetic compounds, Colloids
and Surfaces A: Physicochemical and Engineering Aspects, 481, 567‐576.
DELPLA, I., MONTEITH, D., FREEMAN, C., HAFTKA, J., JONES, T., BAURÈS, E., JUNG, A.‐V.,
THOMAS O., 2015, Heavy rainfall impacts on THM formation in contrasting north‐west European
potable waters", Journal of Environmental Quality, 44, 1241‐1251.
CAUSSE J., BAURES E., MERY Y., JUNG A.V., THOMAS O., 2015, N sources and variability of
N exportation in water from plot to watershed in rural area: a review, Critical Review in
Environmental Science and Technology, 45, 2245‐2281.
JUNG A.‐V., LE CANN P., ROIG B., THOMAS O., BAURES E., THOMAS M.‐F., 2014, Microbial
Contamination Detection in Water Resources: Interest of Current Optical Methods, Trends and
Needs in the Context of Climate Change, International Journal of Environmental Research and
Public Health, 11, 4292‐4310; doi:10.3390/ijerph110404292.
THOMAS O., JUNG A.‐V., CAUSSE J., LOUYER M.‐V., PIEL S., BAURES, E., THOMAS M.‐F.,
2014, Revealing organic carbon–nitrate linear relationship from UV spectra of freshwaters in
agricultural environment, Chemosphere, 107, 115–120.
2
Les références surlignées sont celles citées dans le rapport de synthèse joint
13
DELPLA, I., MONTEITH, D., FREEMAN, C., HAFTKA, J., JONES, T., BAURÈS, E., JUNG, A.‐V.,
THOMAS O., 2014, A Decision Support System for Drinking water production integrating Health
Risk Assessment. International Journal of Environmental Research and Public Health, 11, 7354‐
7375.
BAURES E., DELPLA I., MEREL S., JUNG A.‐V.., PETAVY F., CLEMENT M., THOMAS O.,
2013a, Variation of organic carbon ‐ nitrate relationship with hydrologic regime of river, Journal
of Hydrology, 6, 477, 86–93.
JUNG A.‐V., CHANUDET V., LARTIGES B.S., GHANBAJA J., ABDELMOULA M., BERSILLON J.‐
L., 2012. Association of iron oligomeric species with natural organic matter: a combined EELS
and Mössbauer investigations. Aquatic Sciences, 74, 769‐779.
ROIG B., DELPLA I., BAURES E., JUNG A.‐V., THOMAS O., 2011, Analytical issues for
drinking water contamination monitoring related to short term intense rainy events. Trends in
Analytical Chemistry, 30, 8, 1243‐1251.
DELPLA I., BAURES E., JUNG A.‐V., CLEMENT M., THOMAS O., 2011a, Issues of drinking
water quality of small scale water services towards climate change, Water Science and
Technology, 63, 2, 227‐232.
DELPLA I., BAURES E., JUNG A.‐V., THOMAS O., 2011b, Impacts of rainfall events on
runoff water quality in an agricultural environment in temperate areas, Science of the Total
Environment, 409, 1683‐1688.
JUNG A.‐V., FROCHOT C., VILLIERAS F., LARTIGES B.S., PARANT S., VIRIOT M.‐L.,
BERSILLON J.‐L., 2010, Interaction of pyrene fluoroprobe with natural and synthetic humic
substances: Examining the local molecular organization from photophysical and interfacial
processes, Chemosphere, 80, 228‐234.
DELPLA I., JUNG A.‐V., BAURES E., CLEMENT M., THOMAS O., 2009, Impacts of climate
changes on surface water quality in relation to drinking water production, Environment
International, 35, 8, 1225‐1233.
JUNG A.‐V., FROCHOT C., PARANT S., LARTIGES B.S., VIRIOT M.‐L., SELVE C., BERSILLON
J.‐L., 2005a, Synthesis of amino‐phenolic humic‐like substances and comparison with natural
river‐extracted humic acids: a multi‐analytical techniques approach, Organic Geochemistry, 36,
9, 1252‐1271.
JUNG A.‐V., CHANUDET V., GHANBAJA J., LARTIGES B.S., BERSILLON J.‐L., 2005b,
Coagulation of humic substances and dissolved organic matter with a ferric salt: an Electron
Energy Loss Spectroscopy investigation, Water Research, 39, 16, 3849‐3862.
ACLN
BAURES E., PITOIS F., JUNG A.V., THOMAS O., 2013b, Oracle: objectivizing cyanobacteria‐
associated risks in recreational waters, Water and Society, WIT Transactions on Ecology and The
Environment, WIT Press, 178, 133‐143.
PETAVY F., JUNG A.‐V., DJELAL H., 2009, Traitement des effluents de navires, Récents
progrès en génie des procédés n°98, SFGP Ed., Paris, 6pp.
FARDEL O., VERNHET L., JUNG A.‐V., NOUVEL V., LEGRAND‐LORANS A., 2009, Utilisation
des tests de génotoxicité pour la surveillance de l’exposition des travailleurs dans l’industrie du
traitement et recyclage des déchets, Revue RECORD (Réseau Coopératif de Recherche sur les
Déchets) n°07‐0667/1A, 163 pp, disponible sur http://www. www.record‐net.org/
14
OS
ROIG B., BAURES E., JUNG A.‐V., DELPLA I., THOMAS O., 2012, Rainfall and water quality.
In MARTIN, Olga E., ROBERTS, Tricia M. (dir.), Rainfall: Behavior, Forecasting and Distribution.
Nova Science Publishers. ISBN : 978‐1620815519.
C‐ACTI
CAUSSE J, THOMAS O., MERY Y., JUNG A.‐V., BAURES E., 2014, Impact of heavy rainfalls
on global nutrients export during the fertilization period in a coastal agricultural watershed,
21st century watershed technology conference, November 3rd‐6th, Hamilton, Nouvelle‐Zélande.
BAURES E., PITOIS F., JUNG A.‐V., THOMAS O., 2013, ORACLE‐ To objective
cyanobacteria‐associated risks in recreational waters, 7th International Conference on
Sustainable Water Resources Management, May 21st‐23rd, New Forest, United Kingdom.
DELPLA I., MEREL S., BAURES E., JUNG A.‐V., PETAVY F., CLEMENT M., THOMAS O., 2011,
Surface water quality and climate change issues: impact of hydrology and temperature on total
organic carbon and nitrate in small scale water services, SETAC Europe, May 15th‐19th, Milan,
Italy.
BAURES E, DELPLA I, JUNG A.‐V., CLEMENT M, THOMAS O., 2010, Issues of drinking
water quality of small scale water services towards climate change, IWA World Water Congress
and Exhibition, September 19th‐24th, Montréal, Canada.
DELPLA I., BAURES E., JUNG A.‐V., THOMAS O., 2010, Water quality in Small Scale Water
Services and climate change issues, SETAC Europe, May 23rd‐27th, Séville, Spain.
J.‐L., 2003, Interactions de micropolluants anthropiques avec des modèles de substances
humiques, 5ème colloque sur la matière organique naturelle/groupe français de l'IHSS
(International Humic Substances Society), March 26th‐28th, Aubière, France.
J.‐L., 2003, Relations entre matière organique naturelle et micropolluants, Colloque Le fil de
l'Eau, March 10th‐12th, 2003, Nancy, France.
C‐AFF
JUNG A.V., CAUSSE J., THOMAS M.F., DUFOUR V., IAHNS A., JULIEN L., GAUTRONNEAU M.,
ROCQUET A., BAURES E., THOMAS O., 2015, Caractérisation de la matière organique naturelle et
agricole par spectrophotométrie UV, Carrefour des Gestions de l’Eau, Université Européenne de
Bretagne (UEB)/ Eco‐origin, 28 janvier, Rennes, France.
THOMAS M.F., AUGERO V., DESHORS J., FAQUET V., JEANNEY C., YIN T., BAURES E.
BEAUREPAIRE J.F., CARFANTAN G., BLANC‐MARTEAU N., JANVIER R., JUNG A.V., 2015, Valocity :
Valorisation des eaux dans la ville, Carrefour des Gestions de l’Eau, Université Européenne de
Bretagne (UEB)/ Eco‐origin, 28 janvier, Rennes, France.
PITOIS F., THORAVAL I., JUNG A.‐V., THOMAS O., BAURES E., 2014, ORACLE – Objectiver
le risque lié à l’apparition de cyanobactéries et toxines dans les ressources en eau, Journées
informations eaux de l’ESIP, 3‐6 novembre, Poitiers, France.
15
CAUSSE J., THOMAS O., JUNG A.‐V., BAURES E., 2014, Lutte contre les marées vertes:
améliorer le diagnostic, Les Rencontres de la Transition Écologique, Carrefour des Gestions de
l’Eau, Université Européenne de Bretagne (UEB)/ Eco‐origin, 29 janvier, Rennes, France.
JUNG A.‐V., FAURE E., LIEBAUT M., PAVILLA M., ROTH L., SERDOBBEL V., THOMAS M.‐F.,
CAUSSE J., THOMAS O., BAURES E., 2014, Qualité des eaux et activités anthropiques : étude de la
relation carbone‐azote par spectrophotométrie UV, Colloque Ferdinand Bonn: gestion de l’eau et
risques pour la santé: action à mettre en place pour réduire les risques, 82ème Congrès de
l’ACFAS, 12‐16 mai, Université Concordia, Québec.
JUNG A.‐V., DELPLA I., THOMAS O., BAURES E., 2011, Changement climatique et qualité
de l’eau de boisson: de la ressource aux sous‐produits de désinfection. Congrès national Santé
Environnement – 14‐15 décembre, Paris, France.
DELPLA I., BAURES E., JUNG A.‐V., THOMAS O., 2011, A new approach for assessment of
chlorination by products formation potential from surface water, Environmental health
conference, February 6th‐9th, Salvador, Brazil.
DELPLA I., BAURES E., JUNG A.V., THOMAS O., 2010, Impacts of typical rainfall events on
runoff water quality in an agricultural area, European Doctoral College on Environment and
Health (EHESP), June 7th‐11th, Rennes, France.
DELPLA I., BAURES E., JUNG A.‐V., THOMAS O., 2010, Impacts of climate change on
coastal areas: expected trends on surface and littoral water quality, Ist International Conference
on Environmental Pollution, Restoration, and Management SETAC Asia Pacific Joint Center,
March 1st‐5th, Ho Chi Minh City, Vietnam.
FROCHOT C., JUNG A.‐V., NOUVEL C., VIRIOT M.‐L., 2006, Fluorescence spectroscopy as a
characterization tool to study systems organization in solutions or in biological media, IVth
International conference on tracers and tracing methods, October 3rd‐5th, Austrans/Grenoble,
France.
JUNG A.‐V., CHANUDET V., GHANBAJA J., LARTIGES B.S., MONTARGES‐PELLETIER E.,
BERSILLON J.‐L., 2003, Coagulation of Natural Organic Matter with ferric salts: a TEM‐EDX‐EELS
investigation, European Science Foundation (ESF), October 9th, Acquafredda di Maretea, Naples,
Italy.
JUNG A.‐V., FROCHOT C., KAZPARD V., LARTIGES B.S., PELLETIER‐MONTARGES E.,
VIRIOT M.‐L., BERSILLON J.‐L., 2002, Removal of synthetic humic macromolecules by
aluminium‐based coagulant species: a pyrene fluorescence investigation, XIXth IUPAC
Symposium on photochemistry, July 14th‐19th, Budapest, Hungary.
C‐COM
CAUSSE J., JUNG A.V., MERY Y., BAURES E., THOMAS O., 2015, La coopération entre les
acteurs du territoire et les scientifiques au service de la qualité de l’eau : améliorer la lutte
contre les marées vertes, Carrefour des Gestions de l’Eau, Université Européenne de Bretagne
(UEB)/ Eco‐origin, 29 janvier , Parc des expositions, Rennes, France.
CAUSSE J., BAURES E., JUNG A.‐.V, THOMAS O.., 2014, Lutte contre les marées vertes :
améliorer le diagnostic, Les Rencontres de la Transition Écologique, Carrefour des Gestions de
l’Eau, Université Européenne de Bretagne (UEB)/ Eco‐origin, 29 janvier, Parc des expositions,
Rennes, France.
16
CAUSSE J., BAURES E., MERY Y., JUNG A.‐V., THOMAS O., 2014, Les nitrates dans les eaux
en 2014, Colloque de l’ASEES, Faculté de pharmacie, 22 et 23 mai, Lyon, France.
DELPLA I., BAURES E., JUNG A.‐V., MONTEITH D., FREEMAN C., THOMAS O., 2011,
Environmental changes and rising DOC trends: Impacts on drinking water quality, ACQWA
workshop on science and data gaps in EU climate change and water‐related projects, Riederalp,
Switzerland.
JUNG A.‐V., FARDEL O., 2010, Mise en place de tests de génotoxicité pour les salariés du
déchet, Congrès annuel RECORD (avec comité et diffusion restreinte des actes), EDF/GDF, Paris,
France.
JUNG A.‐V., Congrès national de la Caisse Primaire d’Assurance Maladie (CPAM), 2006,
Risque CMR et évaluation des risques chimiques en entreprise, 13 avril, Nancy, France.
JUNG A.‐V., FROCHOT C., LARTIGES B.S., VIRIOT M.‐L., BERSILLON J.‐L., 2002,
Purification de l’eau et élimination de la matière organique par coagulation au sel de fer, Salon
Pollutec, Lyon, France.
AP
JUNG A.‐V., 2014, émission radio wikiradio Université Européenne de Bretagne (UEB)
« Les échos de l’EME : les travaux d’ingéniérie », 17 avril 2014 à 17h.
JUNG A.‐V., 2014, témoignage « la femme ingénieure au travers de la formation à l’EME »,

1ère journée de la femme ingénieure en Bretagne, 27 février, INSA, Rennes, France.
CAUSSE J, THOMAS O., MERY Y., JUNG A.‐V., BAURES E., 2014, Impact of Heavy Rainfalls
on Global Nutrients export During the Fertilization Period in a Coastal Agricultural Watershed,
3th day of the Young Researchers of Irset, December 16th, Rennes, France.
CAUSSE J., BAURES E., JUNG A.‐.V, THOMAS O.., 2014, Impact de facteurs secondaires sur
l’export de nutriments dans les bassins versants à algues vertes, Rencontres‐ateliers du
CRESEB, 14 juin, Pleumeur Bodou, France.
CAUSSE J., BAURES E., JUNG A.‐.V, THOMAS O., 2014, Green algaes: a public health
concern, Rencontres du réseau doctoral, 16 avri,l Paris, France.
DELPLA I., BAURES E., JUNG A.‐V., THOMAS O., 2011, Impact of climatic events on surface
and drinking water quality in rural area: health risk assessment and adaptation tool for water
supply, Journées de l’Hôtel Dieu, 12‐13 mai, EHESP, Paris, France.
JUNG A.‐V., 2011, La réglementation REACH face aux enjeux des industries
agroalimentaires, Journée technique CTCPA (Institut de recherche agroalimentaire) Nettoyage
en IAA : réduire les coûts et les impacts environnementaux, 3 novembre, Ecole des métiers de
l’environnement, Bruz, France.
DELPLA I., BAURES E., JUNG A.V., THOMAS O., 2010, Impacts of typical rainfall events on
runoff water quality in an agricultural area, Journées de l’Hôtel Dieu, 20 janvier, EHESP, Rennes,
France.
JUNG A.‐V., 2008, Réglementation et méthodologie d’évaluation des risques chimiques,

Journée Technique RECORD, Ecole des Métiers de l’Environnement, Bruz, France.
JUNG A.‐V., MARCHAND C., 2006, Produits chimiques dangereux, Séminaire Médef, 12
octobre, Nancy, France.
17
JUNG A.‐V., 2004, Interactions de la matière organique avec les micropolluants
anthropiques et devenir lors des opérations de coagulation, Séminaire Ecole Doctorale RP2E, 15
janvier, Nancy, France.
J.‐L., Interactions de la matière organique avec les micropolluants anthropiques et devenir lors
des opérations de coagulation, 2004, Séminaire Zone Atelier Moselle, 3 février, Metz, France.
JUNG A.‐V., 2002, Interactions de la matière organique avec les micropolluants

anthropiques, Séminaire Ecole Doctorale de Lorraine (Doc’Lor), 12‐17 novembre, La Bresse,
France.
J.‐L., 2002, Interactions et transport des micropolluants anthropiques du bassin mosellan par la
matière organique naturelle : Etude sur substances modèles dans le compartiment dissous,
Séminaire Zone Atelier Moselle, 10 octobre, Mirecourt, France.
Rapports de contrats de recherche, contrats avec le monde socio‐économique et

subventions :
ORACLE –N°11‐ 115, 2013, Projet financé dans le cadre du PNR EST, objectiver le risque
lié à l’apparition de cyanobactéries et toxines dans les ressources en eau, Rapport intermédiaire,
8 pp.
ERA‐ENVHEALTH CSA, 2012, Coordination Action Grant agreement number 219337, P7‐
ENV‐2007‐CSA‐1.2.3‐01, ANSES call, Final report, Environmental Change and Rising DOC Trends,
18 pp.
NACRE‐ JUNG A.‐V., THOMAS M.‐F., 2012, Rapport final d’activité et financier, Nacre,
PROJET labellisé par le pôle de compétitivité VALORIAL, Région Bretagne, 15 pp.
NACRE‐ JUNG A.‐V., THOMAS M.‐F., 2012, Rapport final d’exécution, Projet Nacre, Pôle
Mer Bretagne/Fond Unique Interministériel, 34 pp.
ENV‐2007‐CSA‐1.2.3‐01, ANSES call, Second intermediate report, Environmental Change and
Rising DOC Trends, 14 pp.
ENV‐2007‐CSA‐1.2.3‐01, ANSES call, First Intermediate report, Environmental Change and
Rising DOC Trends, 23 pp.
NACRE‐ PETAVY F., JUNG A.‐V., DJELAL H., 2009, Les déchets à bord des navires, Rapport
intermédiaire, Pôle Mer Bretagne, DCNS Brest, 10 pp.
DCNS‐ JUNG A.‐V., 2009, Substitution des produits cancérigènes mutagènes et toxiques
(CMR) pour la reproduction pour la DCNS de Brest, Rapport final d’étude EME/DCNS, 231 pp.
RECORD (Réseau Coopératif de Recherche sur les Déchets)‐ FARDEL O., VERNHET L.,
JUNG A.‐V., NOUVEL V., LEGRAND‐LORANS A., 2008, Utilisation des tests de génotoxicité pour la
surveillance de l’exposition des travailleurs dans l’industrie du traitement et recyclage des
déchets, rapport intermédiaire RECORD n°07‐0667/1A, 151 pp.
18
Communications orales lors de meetings nationaux et internationaux devant les
comités de pilotage pour des contrats de recherche :
JUNG A.‐V., DELPLA I., BAURES E., THOMAS O., 2013, Final results concerning DOC
th
characterization with UV/Visible spectroscopy, Final Meeting ERA‐ENVHEALTH, January 9-10 ,
London, United Kingdom.
JUNG A.‐V., DELPLA I., BAURES E., THOMAS O., 2012, UV/Visible spectroscopy indexes to
th
characterize organic matter, Intermediate Meeting ERA‐ENVHEALTH, February 7-8 ,
Windermere, United Kingdom.
JUNG A.‐V., DELPLA I., BAURES E., THOMAS O., 2011, Climat change and NOM
characterization by UV/Visible spectroscopy, Intermediate Meeting ERA‐ENVHEALTH, May,
Saint-Malo, France.
JUNG A.‐V., DELPLA I., BAURES E., THOMAS O., 2010, NOM characterization by
UV/Visible spectroscopy, Intermediate Meeting ERA‐ENVHEALTH, September, Ultrecht
University, Netherland.
JUNG A.‐V., DELPLA I., BAURES E., THOMAS O., 2010, Impacts of heavy rainfalls events
on runoff water in an agricultural area: application of UV/Visible Spectroscopy, Intermediate
Meeting ERA‐ENVHEALTH, February, Bangor University, United Kingdom.
traitement et recyclage des déchets, Présentation du rapport final RECORD, 8 janvier, ADEME,
Paris, France.
traitement et recyclage des déchets, Présentation du rapport intermédiaire RECORD, 12 juin,
ADEME, Paris, France.
Organisation de séminaires
Co‐organisatrice et participante à la Journée Technique du Réseau Coopératif de

Recherche sur les Déchets (RECORD), 2008, Ecole des Métiers de l’Environnement,
Utilisation des tests de génotoxicité pour la surveillance de l’exposition des travailleurs
dans l’industrie du traitement et recyclage des déchets
Organisatrice pour l’EME des « Rencontres de la Vilaine » (LERES‐EME‐SAUR‐
Institut d’Aménagement de la Vilaine) depuis 2010, usine de traitement de l’eau du
Drézet, Férel (56)
19
Autres activités
Depuis 2007 : participation à des réunions du Collège des Hautes Études du

développement durable (CHEDD) formant des cadres dirigeants et des responsables au
développement durable
2007‐2009 : contribution au Réseau Coopératif de Recherche sur les Déchets (RECORD)
dans le cadre d’une étude technique, d’organisation de journée technique et de
participation aux journées de restitution annuelle du réseau
2007‐2013 : participation au Pôle de compétitivité Pôle Mer Bretagne dans le cadre
d’une étude technique (NAvire Conçu dans le Respect de l’Environnement, projet Nacre)
et de participation aux assemblées générales du Pôle
Depuis 2013 : ‐ groupe de travail eau/métrologie réseau Eco‐origin (cluster des
éco‐activités au service des entreprises bretonnes pour relever le défi de l’économie
verte), encadrement d’étudiants pour la réalisation de posters (carrefour de l’eau à
Rennes)
‐ ateliers de réflexion « enseigner la santé sécurité au travail dans les
établissements d’enseignement supérieur » avec la CARSAT Bretagne
‐ participation aux matinales de SOCOTEC (thématiques santé au
travail – risques professionnels)
‐ membre de l’ASTEE (Association Scientifique et Technique pour
l'Eau et l'Environnement)
‐ journées de sensibilisation au Développement Durable de l’EME en
partenariat avec l’ADEME
2013 : participation atelier recherche journée des 5èmes rencontres nationales du
développement durable de la conférence des grandes écoles (CGE)
2014 : groupe de travail de l’UEB (Université Européenne de Bretagne) : référentiel
compétence de l’enseignant‐chercheur pour la mise en œuvre des technologies de
l'information et de la communication pour l'enseignement (TICE)
2015 : participation commission scientifique et technique du CRESEB (Centre de
Ressources et d’Expertise Scientifique sur l’eau de Bretagne) "outils et méthodes de
mesure de la qualité de l’eau développés par la recherche : quelle opérationnalité pour
l’évaluation et pour l’action ?"
Reviewer de plusieurs articles dans des journaux internationaux (ex : Chemical
Engineering Journal, Global Environnemental Change, Environmental Science and
Technology…) et de projets de recherche internationaux (applications for allocations
from the Research Grants Council (RGC) of Hong Kong)
Autres responsabilités
Responsabilités administratives
‐ Membre du conseil de laboratoire du LERES

‐ Membre du conseil scientifique de l’EME
‐ Participation aux conseils de perfectionnement et conseils de vie de l’EME
‐ Responsable pédagogique des travaux d’ingénierie (5ème année cycle ingénieur
EME)
20
‐ Jury « Validation des Acquis de l’Expérience » pour les candidats externes au
titre d’ingénieur en génie industriel de l’environnement de l’EME
Autres activités
‐ Animation scientifique journée de la Vilaine (Thèmatique 2014 : chlorure de

vinyl monomère et trihalométhanes dans les réseaux)
‐ Participation aux entretiens de recrutement des étudiants du cycle
préparatoire, ingénieur, Bachelor, sélection sur concours d’entrée aux grandes
écoles d’ingénieur (E3A)
‐ Représentation de l’EME sur des salons étudiants et journées portes ouvertes
‐ Participation atelier « audit énergétique » organisé par des bureaux d’étude
(CERUR, Audiar) sur le bâtiment de l’EME
‐ Animation stand « traitement de l’eau » sur le village de la Fête de la Science
(Rennes)
‐ Participation aux ateliers « Bretagne Mobilité Augmentée » BMA sur le Campus
de Ker Lann, projet initié par Novincie (CCI de Rennes‐Bretagne), auprès de 40
Entreprises et Collectivités bretonnes et accompagné par l’ADEME dans le
cadre du programme « véhicule du futur »
‐ Participation aux cafés TICE, nouvelles technologies d’enseignement avec
l’université de Rennes 1.
Réponses à appels à projet
‐ ANR Eco‐TS, 2013 : Projet ISIS Identification des Situations à rISque de

contamination microbienne dans les eaux (partenariat SAUR, LERES, IAV, EME) –
noté A, non financé par le conseil scientifique
‐ Appel à projet ANSES, 2012 : Projet ISIS Identification des Situations à rISque
de contamination microbienne dans les eaux (partenariat LERES, EME) – noté A,
non financé par le conseil scientifique
‐ ANR JCJC, 2011 : Projet Turb’eau Effets des crues turbides sur la qualité
physico‐chimique des eaux de surface utilisées pour la production d’eau potable :
étude des relations colloïdes‐micropolluants (partenariat LERES, Ecole des mines
d’Alès, EME) – noté B
‐ Appel à projet RECORD, 2009 : Mélange de polluants et toxicité (partenariat
INSERM U620, EME) – non financé
21
Synthèse des projets3
Nom du projet/ année Enjeux Travaux étudiants

encadrés
Substituer des produits
DCNS/ 2008‐2009
cancérigènes,
mutagènes te
reprotoxiques
Etudier la possibilité de
RECORD/2007‐2009 mise en œuvre de tests
de génotoxicité pour la
surveillance des salariés
des déchets
NACRE 2 projets bibliographiques
Donner une
(2008 et 2009), ingénieurs
méthodologie
1ère année
(Pôle Mer Bretagne)/2007‐ d’évaluation
2013 d’empreinte
4 travaux d’ingéniérie EME
environnementale des
(2009 à 2012), ingénieurs
navires
5ème année
ERA Net Env Health Comprendre l’impact du
changement climatique Participation à
(FP 7)/2009‐2013 sur la qualité des eaux l’encadrement de thèse
destinées à la I. Delpla (2008‐2011)
consommation humaine
Co‐encadrement de thèse
J. Causse (2012‐2015)
Algues vertes
Expliciter les transferts
Stage ERAMUS M. Paque
rapides de nutriments
(CDFO)/ 2012‐2015 (2014), M1
dans les basins versants
à algues vertes
2 travaux d’ingénierie EME
(2013 et 2014), ingénieurs
5ème année
Objectiver le risque lié à Tuteur EME stages
ORACLE
l’apparition de C. Farenc (2012) et A.
cyanobactéries et Fuchez (2013), ingénieurs
(ANSES)/2012‐2015 toxines 4ème année
ISIS Identifier les situations à 1 travail d’ingéniérie,
risques de contamination Expérimentations
(déposé à l’ANR Ecotechnologies et bactérienne dans les préliminaires au projet
Ecoservices en 2013, retenu eaux dans les ressources (2013), ingénieurs 5ème
scientifiquement mais non financé) en eau (lacs) année
TURB’EAU
Expliciter la turbidité en
lien avec les phénomènes
(déposé à l’ANR jeunes chercheurs
d’adsorption/
en 2011, retenu scientifiquement
relargage/ des
mais non financé)
micropolluants dans
l’eau
3
En grisé : projets déposés non financés, en blanc : projets réalisés
22
Dossier de synthèse4
Introduction
Le défi de la qualité de l’eau et des milieux aquatiques est une préoccupation

européenne forte et partagée par l’ensemble des acteurs, renforcée notamment par la
directive cadre sur l’eau du 23 octobre 2000, révisée le 16 décembre 2008 (Directive
2008/105/CE). La capacité de l’ensemble des parties prenantes à relever ce défi
déterminera à l’avenir la capacité des territoires à concilier développement économique
et territorial, préservation des ressources et acceptabilité des milieux.
Parmi les enjeux posés, l’étude et le suivi de la pollution organique des milieux
aquatiques est primordiale. En effet, de grandes quantités de matières organiques dans
l’eau peuvent par exemple entraîner des perturbations de l’équilibre écologique d’un
bassin hydrologique. Elles interviennent en tant que polluants principalement à deux
niveaux différents. Outre la toxicité apportée directement par certaines substances
spécifiques (pesticides, herbicides, composés aromatiques polycycliques, etc.), la
dégradation des matières organiques entraîne la désoxygénation du milieu aquatique
ainsi que la libération des micropolluants associés. Il a en effet été très largement établi
que l’ensemble des matières organiques (dissoutes, colloïdales ou particulaires) avaient
de fortes capacités de complexation pour les espèces du milieu, et en particulier les
polluants. La compréhension des mécanismes mis en jeu lors du transport des
micropolluants, la chimie des colloïdes, les interactions eau/particules est indispensable
pour mettre au point une typologie des matières organiques aquatiques. Outre une
meilleure connaissance des milieux aquatiques, une telle approche permet de mieux
gérer les ressources en eau et d’augmenter les performances des procédés de
traitement. En effet, la potabilisation d’une eau à partir d’eaux de surface est en réalité
un ensemble de procédés particulièrement vulnérables aux variations à court terme des
apports de matière organique affectant la qualité de l'eau brute traitée. La
caractérisation de la qualité des eaux de surface de façon générale, et de la matière
organique en particulier, est donc fondamentale pour respecter les enjeux globaux.
La qualité de l’eau est variable car soumise aux différentes contraintes liées à son
environnement (pressions agricoles, naturelles, industrielles ou urbaines). Les
changements dans les pratiques d'utilisation des terres (notamment agricoles) ont été
identifiés comme des moteurs potentiels de l’augmentation du carbone organique
dissous (COD), un des indicateurs de la matière organique. De surcroît, le dérèglement
climatique (augmentation de l’intensité des précipitations et de la fréquence des
événements de sécheresse et de fortes pluies), prévu par le groupement d’experts
intergouvernemental sur l’évolution du climat (GIEC), va fortement affecter la qualité de
l'eau et par voie de conséquence les opérations de traitement. De tels changements dans
les conditions hydrométéorologiques comme les fortes pluies pourraient, par exemple,
causer une augmentation rapide de matières en suspension, des taux de COD et du
carbone organique total (COT), des nutriments, de certains micropolluants ainsi que des
paramètres microbiologiques, tels que des bactéries, des protozoaires ou des virus.
4
Sauf pour celles expressément mentionnées dans le texte, la majorité des références
bibliographiques citées sont données dans la liste des communications (p.11/12) et surlignées
en gris dans cette liste
23
Inversement, les épisodes de sécheresse et leurs conséquences peuvent conduire à la
prolifération des algues et des augmentations associées en cyanotoxines ou encore à des
diminutions des teneurs en oxygène dissous et des capacités de dilution des nutriments
(ammonium, orthophosphates, etc). Il est donc également nécessaire de pouvoir mieux
appréhender ces différents facteurs de dégradation de la qualité de l’eau.
L’action des autorités sanitaires, ainsi que celle des collectivités ou de leurs délégataires
font que l’eau du robinet est globalement de bonne qualité en France grâce aux
exigences de qualité, aux progrès techniques au niveau du traitement et des analyses des
eaux. Toutefois, des épidémies peuvent encore survenir sans qu’aucun indicateur de
contamination classique ne soit positif (comme par exemple l’épidémie à crytosoridium
aux Etats‐Unis dans les années 1990). Il ainsi important non seulement de surveiller la
qualité de l’eau, mais également de minimiser les risques de contamination en disposant
d’indicateurs environnementaux et sanitaires adaptés. La sécurité sanitaire est en effet
une des questions centrales dans le champ de la santé publique, questionnant d’une part
les dispositifs de veille sur leur capacité à détecter et à alerter « en temps voulu » sur les
menaces de santé publique, et d’autre part les dispositifs de réponse à contrôler et
prévenir de manière « efficace, proportionnée et efficiente » les risques sanitaires liés à
ces menaces.
Mes travaux de recherche s’inscrivent dans une problématique de

santé/environnement, avec une préoccupation d’utilité sociale et d’aide à la décision
pour les gestionnaires et l’ensemble des acteurs du domaine de l’eau. Il s’agit
d’améliorer les connaissances de la contamination des milieux aquatiques dans une
optique d’évaluation des risques environnementaux et sanitaires. Elle se définit en 3
axes en lien avec les équipes de recherche de l’EME et du LERES/EHESP (figure 1) :
- Etude de l’interaction « matière organique – micropolluants » : les travaux

de cet axe ont été développés pendant le doctorat et valorisés pendant et au‐delà
de cette période. Ils visent à caractériser par des approches multi‐techniques un
des constituants majeur de la matière organique : les substances humiques,
obtenues soit par extraction de composés naturels soit par synthèse chimique
(composés modèles). Les interactions avec différents représentants de polluants
chimiques (hydrocarbures aromatiques polycycliques (HAP), herbicides et
tensio‐actifs) sont abordés par des techniques à l’échelle locale
(spectrofluorimétrie notamment) et interfaciale. Les questions du devenir des
complexes formés lors du traitement de l’eau (pour l’étape de coagulation) sont
également étudiés sur les différents types de composés organiques considérés
- Caractérisation et facteurs de dégradation de la qualité de l’eau/recherche

d’indicateurs : les travaux menés dans le cadre de cet axe ont pour objectif de
contribuer à une meilleure compréhension des phénomènes dynamiques qui
conditionnent l’état des masses d’eau, incluant les facteurs climatiques et les
activités humaines (rejets industriels/urbains/agricoles). Les projets menés
permettent d’étudier le comportement de la matière organique en fonction de
situations à risques (événements climatiques de types fortes pluies, apports
anthropiques lors des épandages agricoles, de l’échelle de la parcelle à celle du
24
bassin versant). Ils contribuent également au développement de méthodes et de
systèmes (basés notamment sur des méthodes optiques) pour caractériser la
matière organique bioassimilable ou suivre les flux de nutriments de l’échelle de
l’exutoire sur un événement court‐terme, à l’échelle du bassin‐versant pour un
événement intra‐annuel.
- Evaluation environnementale et prévention des risques sanitaires :

l’évaluation des risques sanitaires est abordée initialement sous l’angle « santé au
travail » suite à l’expérience professionnelle acquise dans un service de santé au
travail (ALSMT, Nancy) et au travers des activités d’ingéniérie réalisées à l’EME
pour la mise en place d’indicateurs de suivi d’exposition professionnelle des
salariés du déchet (projet RECORD) et pour la mise en pratique de la substitution
d’agents chimiques cancérigènes, mutagènes et reprotoxiques (projet DCNS).
Cette évaluation sanitaire est doublée de travaux d’évaluation d’impacts
environnementaux pour un projet de réalisation d’un navire conduit dans le
respect de l’environnement (projet NACRE, Pôle Mer Bretagne). Cet axe permet
également l’élaboration d’outils d’aide à la décision pour l’amélioration de la
gestion de la qualité de l’eau distribuée par les petites unités de traitement et
l’adaptation des méthodes d’analyse du risque à la production d’eau potable
(projet ERA‐net Environment and Health).
Figure 1 : synoptique et synergie entre les 3 axes de recherche développés
25
1. Interaction de la matière organique avec les micropolluants
anthropiques et devenir lors de la coagulation
Cette thématique a été développée lors de la thèse de doctorat (2001‐2004),

réalisée au Laboratoire Environnement et Minéralurgie (UMR 7569) rattaché au
Laboratoire Interdisciplinaire des Environnements continentaux (UMR 7360) depuis le
1er janvier 2013. Le travail de thèse s’est inscrit dans le cadre du projet Zone Atelier
Moselle qui s’attelle à sécuriser l’approvisionnement en eau du Bassin Mosellan. Réalisé
en collaboration avec seize équipes lorraines de recherche (CEGUM, CRF‐INRA, CRPG,
G2R, EBSE, IDEDE, ILGU, LAEGO, LCPME‐LSE, LHRSP, LEM, LTAC, LSE‐INRA, SERES,
SRSMC, SSAD‐INRA), ce projet, soutenu par le Contrat Plan Etat‐Région Lorraine, a porté
sur l’étendue actuelle des connaissances sur la nature, le transport et la biodégradation
des matières organiques naturelles et xénobiotiques.
Parmi l’ensemble des molécules naturelles du milieu aquatique, il existe une

famille de composants majoritaire du milieu, considérée comme un des véhicules
potentiels de transport ou d’immobilisation de micropolluants et autres produits
phytosanitaires avec risque de désorption lors de modifications physico‐chimiques du
milieu : les substances humiques. L’étude s’est principalement axée sur cette famille de
molécules aux propriétés physico‐chimiques hétérogènes et polydisperses, encore
aujourd’hui mal définies.
Le travail de thèse a pris son origine dans les interrogations suivantes : pouvons‐
nous faciliter la caractérisation du comportement des substances humiques naturelles
grâce à la synthèse de molécules « modèles » ? Quel est le comportement des substances
humiques en solution, à savoir quelles sont les modifications de conformations
éventuelles de ces macromolécules pouvant donner un éclairage sur leur capacité de
transport et de piégeage des polluants ? Quels sont les polluants susceptibles d’interagir
avec ces substances ? Comment aborder au laboratoire, par une approche modélisée, les
processus d’interaction avec les polluants retenus ? Et enfin, comment se comportent les
complexes formés entre la matière organique naturelle et les polluants lors des
opérations de traitement de l’eau (étape de coagulation/flocculation) ?
Pour répondre à ces questions, une approche utilisant différentes techniques

analytiques a été séquencée en 4 principales étapes développées dans les paragraphes
ci‐après.
1.1. Synthèse et caractérisation de composés humiques modèles –

comparaison avec des extraits naturels de matières en suspension
(approche multi‐techniques).
Des acides humiques de synthèse de différentes tailles moléculaires, obtenus par

oxydation de catéchol et de glycine ou de triglycine, ont été caractérisés et comparés à
des extraits naturels d’acides humiques associés aux matières aquatiques en suspension
(rivière Moselle). Le suivi cinétique de leur synthèse, par spectrofluorimétrie et
26
spectroscopie infra‐rouge, a permis d’appréhender le rôle des différents groupements
fonctionnels dans l’organisation structurale finale des macromolécules. Il a été montré
que la nature des groupements azotés joue un rôle non négligeable dans leur
comportement en fluorescence et à l’échelle de l’interface eau/air. En effet,
l’incorporation d’atome d’azote favorise la production de fluorophores émettant à
grande longueur d’onde. Les modèles obtenus par oxydation à l’air en milieu légèrement
basique de catéchol et de glycine ou triglycine présentaient des similitudes intéressantes
avec les composés humiques naturels, extraits des matière en suspension prélevées à
Richardménil (réserve d’eau de l’agglomération nancéenne), même s’ils sont beaucoup
moins absorbants et émissifs que les composés humiques naturels (figure 2).
Figure 2: Représentation des matrices de fluorescence (excitation‐émission‐intensité de

fluorescence) pour les matières organiques naturelles (eau de rivière Moselle, substances de
type humiques naturelles extraites de matières en suspension sous des conditions
hydrométéorologiques différentes (AHN1 ou N1 extrait pendant une période de canicule
sécheresse (été 2003) et AHN2 ou N2 extrait pendant une période de temps de pluie (octobre
2003)), et molécules modèles synthétisées au laboratoire à partir de glycine (AHCG ou CG,
>8kDa) ou de triglycine (AHCT1 ou CT1 1‐3,5kDa et AHCT2 ou CT2 > 8kDa). Les composantes
visibles sont alpha (excitation : 350nm, émission : 430nm) généralement assignée d’après la
littérature aux substances humiques d’origine aquatique autochtone ou d’origine terrigène peu
transformée, alpha’ (excitation : 270nm, émission : 450nm) généralement assignée aux
substances humiques d’origine plutôt terrigène assez transformée et delta (excitation : 270nm,
émission : 350nm), généralement assignée à la composante de type « tryptophane »
(Jung et al., 2005a)
Cette étude, par un couplage multi‐techniques (analyses élémentaires, pyrolyse

chromatographie gaz, chromatographie liquide haute‐performance, UV/Visible,
27
spectroscopies de fluorescence et infra‐rouge, titration des acidités) a permis de décrire
ces pseudo substances humiques comme contenant des structures de type protéiques
pouvant être décrites par des concepts d’association supramoléculaire définis dans la
littérature pour ces substances. Du fait des différents rapports C/N pour les composés
modèles synthétisés et avec un rapprochement des données de la littérature pour
l’assignation des signaux de fluorescence ainsi que du recoupement avec les données
physico‐chimiques, les fractions CT1, CT2 et CG ont respectivement été prises pour
modèles de substances humiques (i) d’origines plutôt « aquatiques autochtones » (ou
d’origines terrigènes mais peu transformées) (CT1 et CT2) (ii) d’origines plutôt
terrigènes ayant subi de plus importantes transformations dans l’environnement
(polymérisation/condensation) (CG).
La voie de synthèse utilisée des substances « modèles » est donc intéressante et

contrôlable, car elle permet d’obtenir des substances de réactivité proche de celles des
substances naturelles tenant compte de différences au niveau de leurs origines dans le
milieu.
1.2. Etude de l’organisation des substances humiques en solution aqueuse à

l’échelle moléculaire et macroscopique (spectrofluorimétrie et tension de
surface).
L’utilisation de sondes de fluorescence a essayé de répondre à la question du

comportement des acides humiques en solution. En effet, les sondes de fluorescence
donnent des renseignements sur le comportement de molécules en milieu aqueux, et
plus particulièrement sur leur aptitude à former des microdomaines hydrophobes, leur
capacité à incorporer des substances hydrophobes au sein des microdomaines et leur
propriétés tensioactives. Ainsi, en incorporant une sonde sensible aux modifications de
son environnement (polarité, viscosité…) au sein de systèmes susceptibles de donner
lieu à des microdomaines hydrophobes en solution aqueuse, il est possible de mettre en
évidence l’existence de tels microdomaines.
Différentes sondes de polarité et de viscosité ont été utilisées à cet effet: les
rotors de type « cinnamylidène » et « coumarine » (respectivement fonctionnalisés 2‐CN
et –CNCOOEt) déjà étudiés au Département de Chimie Physique des Réactions (DCPR)
de l’Ecole Nationale Supérieure de Chimie des Industries Chimiques (ENSIC), la sonde D
7760 commercialisée par la société Molecular Probes Europe (Leiden, The Netherlands)
pour sa chaîne latérale aliphatique positivement chargée (ammonium) de structure
apparentée aux tensio‐actifs cationiques susceptibles d’être des micro‐polluants du
milieu naturel aquatique et enfin une nouvelle sonde rotor de type « coumarine »
CouC12 possédant une chaîne aliphatique hydrocarbonée (chaîne dodécyl) non chargée
correspondant à celle du tensio‐actif SDS qui spécialement été synthétisée pour ce
travail. Seules des études sur la fraction synthétique d’acide humique catéchol/glycine
AHCG ont été réalisées à pH = 8 contrôlé.
Cette étude nous a permis d’appréhender les difficultés de l’utilisation de

certaines sondes de fluorescence avec des macromolécules aussi complexes que les
substances humiques. Une interprétation globale de ces résultats est délicate. Nous
avons cependant pu émettre deux principales hypothèses par rapport aux sondes
testées :
28
 soit les sondes ne « voient » des variations de viscosité locale que sur un
certain domaine de concentrations en substances humiques, preuve qu’elles présentent
un encombrement stérique peu favorable pour permettre l’enregistrement de variations
de polarité ou de viscosité aux plus faibles concentrations. Il leur est donc nécessaire
d’être en présence d’une quantité suffisante de molécules humiques pour détecter des
variations. A des concentrations en acide humique trop importantes, sous l’effet d’une
reconformation éventuelle des acides humiques, elles peuvent être expulsées des micro‐
domaines sous l’effet d’une contraction de structure tridimensionnelle du réseau dans
lequel elles étaient piégées.
 soit les sondes sont bien appropriées à l’étude des micro‐domaines de viscosités
différentes, mais l’existence de groupements n’absorbant pas de lumière aux longueurs
d’onde d’absorption (comme les chaînes aliphatiques par exemple), en se réorganisant
lors de la reconformation (ou agrégation) des substances humiques en solution, vont
provoquer la formation de nouvelles familles de groupements chromophores (et donc
modifier le diagramme des énergies électroniques du système initial) au sein des
macromolécules. Ces effets, propres aux substances humiques, nous ont ainsi empêché
d’avoir des variations monodales en polarité et/ou viscosité comme il est généralement
constaté pour des systèmes polymèreres mieux définis.
La question de la pertinence de l’utilisation éventuelle de sondes avec les

molécules humiques nous a fait par la suite aborder l’étude sous l’angle de la mesure de
la fluorescence intrinsèque aux substances humiques, ce qui est traitée notamment dans
la partie 1.3.2.
1.3. Etude des interactions entre la matière organique et les polluants

(techniques analytiques globales et moléculaires)
La mise en évidence d’interactions avec plusieurs familles de polluants types

(HAP, tensio‐actifs, herbicides) a été réalisée par une approche multi‐techniques
(fluorescence, tension de surface, suivi de paramètres physico‐chimiques). Différents
représentants de polluants anthropiques ont été choisis afin de déterminer leur capacité
d’interaction avec la matière organique. La méthode des ajouts dosés a été utilisée avec
des outils analytiques spécifiques tels que la spectroscopie UV/Visible, la fluorimétrie et
des mesures de pH, la conductivité et la turbidité des mélanges. Les herbicides atrazine
et sulcotrione, ainsi que les tensio‐actifs de type Nonyl‐Phénols Ethoxylés (NPE) n’ont
pas présenté d’interaction quantifiables par fluorescence avec les molécules organiques.
1.3.1 Interaction matière organique ‐ pyrène
Tout d’abord, l’interaction entre la matière organique et le pyrène a été regardée

à l’échelle locale grâce à la spectrofluorimétrie (technique du quenching) et à l’échelle
interfaciale par tensiométrie (méthode de Wilhelmy).
29
L’équation de Stern‐Volmer utilisée pour le quenching (ou inhibition) de fluorescence
[1] permet de préciser la constante d’association KSV entre la matière organique et le
pyrène d' après la formule suivante :
I0 [1]
 1  K sv [quencheur ]
I
Avec I0 et I les intensités de fluorescence du pyrène (excitation 332nm, émission

377nm) respectivement sans et avec matière organique. La concentration en quencheur
(inhibiteur de fluorescence) est déterminée par la concentration en matière organique.
Les mesures simultanées de tension de surface ont permis de déterminer les

concentrations micellaires critiques, si existantes, pour les différents composés
organiques (figure 3).
Figure 3 : mesures des tension de surface γ des substances humiques modèles (CTHA1, CTHA2,
CGHA) et naturelles extraites de matières en suspension de rivière (NHA1, NHA2) en fonction de
la concentration en substance organique (Jung et al., 2010)
Cette étude a montré l’existence de 2 principaux domaines de concentration en matière

organique permettant d’expliquer des mécanismes d’interaction différents entre les
substances humiques et le pyrène :
 Pour les faibles concentrations de 0‐30 mg/L en substance humiques : des

interactions faibles de type adsorption se produisent entre le fluorophore
(pyrène) et les sites les plus hydrophobes des composés humiques. Cette région
permet de quantifier les constantes d’interaction par l’analyse de Stern‐Volmer
30
 Pour des concentrations supérieures à 250mg/L, une pseudo phase micellaire
s’organise en solution. L’interface eau/air se structure. Grâce à l’analyse
quantitative (Tableau 1), nous avons pu déterminer que la concentration de
pyrène peut atteindre 2g/kg dans les micelles. Comme le point précédent a
montré que le pyrène pouvait être relargué lors des opérations de l’eau, ce
résultat est capital en terme de risque sanitaire par ingestion d’eau potable.
Tableau 1 : Constants d’association pyrène‐matière organique (Ksv) et concentrations

micellaires critiques (CMC) pour les différents types de substances organiques modèles (CGHA,
CTHA1 et CTHA2) et naturelles (NHA1 et NHA2) considérés (Jung et al., 2010)
1.3.2 Interaction matière organique – tensio‐actifs
Quatre techniques différentes ont été utilisées pour étudier l’interaction entre le
bromure de cétyltriméthylammonium (CTAB), surfactant cationique et le
nonylphenylethoxyl (NPE) surfactant non‐ionique, avec les 3 agrégats modèles
(composés organiques de synthèse CT1, CT2 et CG) et 2 extraits humiques naturels
extraits de matières en suspension (N1 et N2). Aucune interaction significative entre le
tensioactif NPE et un quelconque des composés étudiés n’a été trouvée. Concernant le
CTAB; les mesures du pH, de la conductivité et de la turbidité des solutions, ainsi que la
spectroscopie de fluorescence ont été combinées pour fournir une meilleure
compréhension des interactions entre les agrégats organiques et l'agent tensioactif. Les
données spectroscopiques montrent que le composé modèle CG de type « terrigène
hautement transformé » interagit d'une autre manière avec l'agent tensioactif cationique
que les composés de type « aquatique autochtone » (CT1) ou d’origine « terrigène mais
moins transformé » (CT2). En effet, en présence de CTAB mis à concentration croissante
dans le milieu, les mesures de rendements quantiques de fluorescence (définis comme le
rapport du nombre de photons émis au nombre de photons absorbés pendant la même
période de temps), ont montré des comportements différents entre les fluorophores α
(signature spectrale des composés d’origine plutôt « aquagénique autochtone ») et α’
(signature spectrale des composés aquatiques d’origine « terrigène » plus transformé).
Le centre α « voit » une exaltation de sa fluorescence alors que le centre α’ voit une
inhibition pour des concentrations croissantes de CTAB (Figure 4), avec différents
domaines correspondant à différentes étapes d’interaction.
31
Figure 4 : Exemple de suivis spécifiques d’interaction entre le CTAB et une des molécules
modèles par mesure du rendement quantique de fluorescence (Φf/Φ°) des fluorophores α
(ronds noirs) et α’ (carrés blancs) en fonction de la concentration en tensio‐actif
(Jung et al., 2015)
Cette étude a montré que dans des conditions similaires, des modifications
spectrales spécifiques du signal de fluorescence peuvent être corrélées aux variations
des paramètres physico‐chimiques (pH, turbidité, conductivité) qui sont quant à elles
des mesures non‐spécifiques. Le mécanisme de complexation a été proposé comme
essentiellement régi par la neutralisation de la charge, pouvant donc être « monitoré »
par la fluorescence.
1.4. Devenir du coagulant, de la matière organique (MO) et du complexe MO‐polluant

lors du traitement de l’eau
Après avoir caractérisé les interactions MO‐polluants, la dernière partie du

travail de doctorat posait la question du devenir de ces complexes lors de certaines
opérations du traitement de l’eau (étape de coagulation/floculation).
1.4.1. Coagulation de la matière organique (MO) par un sel de fer – spéciation du fer
Avant d’étudier le devenir des complexes MO‐polluant, il a déjà fallu caractériser

le statut du coagulant lors de son interaction avec la matière organique. Le sel de fer est
un des coagulants les plus employés dans les traitements physico‐chimiques pour la
potabilisation des eaux de surface. Cependant, comme en témoignent les nombreux
mécanismes de déstabilisation proposés encore actuellement dans la littérature pour ce
type de coagulant (neutralisation/complexation, hétéroagrégation, ennoiement dans un
précipité insoluble …), la nature de l'interaction entre les composés d’oxyhydroxyde de
fer et de matière organique naturelle (MON) peut prendre diverses formes et est encore
un sujet de débat. C’est un domaine important de recherche, en particulier pour les
applications de traitement de l'eau et de la connaissance de transport du fer dans
32
l'environnement. Lors du travail de doctorat, la nature de l'association entre espèces
oligomères de fer et de la matière organique naturelle a été étudiée en utilisant la
spectroscopie Mössbauer et la spectroscopie de perte d’énergie électronique (EELS) au
niveau du seuil Fe‐L3. Cette caractérisation a été réalisée sur les sédiments obtenus
après décantation et lyophilisation obtenus par procédure de type jar‐test (figure 5).
Figure 5 : Procédure de jar‐test mise en place pour l’étude du statut du fer et des flocs
formés par coagulation de différents types de matières organiques avec du nitrate ferrique
(Jung, 2004, thèse de doctorat)
En comparant les résultats entre du nitrate ferrique et différents types de

matières organiques (MON issue d'eau brute de la rivière Moselle, substances humiques
naturelles (NHS) extraites de la rivière (matières en suspension), et un des modèles de
composé synthétique (SHS)), les signaux obtenus par spectroscopie Mössbauer ont
montré que le fer est présent dans un environnement de coordination octaédrique. Les
zones sous les pics (au seuil du fer Fe L3) et les formes des pics de spectres EELS varient
en fonction de l'origine de la matière organique (figure 6), ce a laissé penser que
différents types de populations de Fe pouvaient être distingués en utilisant la technique
EELS. La combinaison de la sélectivité des deux spectroscopies Mössbauer (pour
identifier des états d'oxydation du Fe) et de la EELS (pour être en mesure de caractériser
la population de Fe), est apparue prometteuse pour caractériser le Fe dans les systèmes
contenant de la MON.
33
Figure 6 : Déconvolution (65% gaussienne‐35% lorentzienne) au seuil Fe‐L3 du
signal obtenu par spectroscopie de perte d’énergie EELS pour évaluer la chimie
locale du fer (Jung et al., 2012)
1.4.2. Coagulation de la matière organique (MO) ‐ spéciation des composés

organiques (techniques associées à la microscopie électronique à
transmission)
Comme nous l’avons vu précédemment, la spectroscopie EELS peut être

utilisée comme une première étape – technique peu consommatrice d’échantillon ‐
pour enquêter sur la réactivité des coagulants envers les colloïdes organiques à
l'échelle nanométrique. En particulier, la haute résolution de la méthode (<100 nm)
34
permet de caractériser des matrices naturelles d’une grande hétérogénéité. Le
principal avantage de la technique reste la possibilité de couplage d’analyses
réalisées au seuil du fer mais également au seuil du carbone pour fournir des aperçus
sur la structure des complexes Fe‐MO à l’échelle d'agrégats (figure 7).
Figure 7 : exemples d’agrégats de matières organiques (polysaccharide en haut, matrice organo‐

minérale en bas, photos par microscopie électronique à transmission et couplage avec la
spectroscopie de perte d’énergie au seuil C‐K du carbone et de dispersion d’énergie X)
(Jung et al., 2005b)
La spectroscopie de perte d’énergie a donc également permis de vérifier

l’interaction forte entre les groupements carboxyliques et les espèces hydrolysées de Fe
lors du processus de coagulation. En effet, le pic à 288 eV (correspondant à la transition
1s‐> π* de la liaison C=O) décroît rapidement lorsque le dosage en coagulant augmente
(figure 8 et 9).
35
Figure 8 : Spectre typique obtenu par spectroscopie EELS (pour une substance humique modèle
synthétisée au laboratoire coagulée avec du fer, pH 8 avec [Fe] 8.0. 10‐4 mol/L)
(Jung et al., 2005b)
36
Figure 9: Suivi par jar‐tests de la coagulation de substance humique modèle et de matière
organique naturelle avec spectroscopie de perte d’énergie au seuil C‐K du carbone
associée (Jung et al., 2005b)
Pour les différentes catégories de matière organique, la morphologie des agrégats

organo‐minéraux obtenus et les variations de la stœchiométrie N/C, Fe/C, Na/C et Ca/C
ont été comparées. Le recours aux substances humiques synthétiques a permis de
vérifier que la demande en agent coagulant augmentait avec la concentration en
carboxyles fortement dissociés. La nature chimique des matières organiques naturelles a
donc été montrée prépondérante pour l’efficacité de la coagulation, le pH ayant
également un rôle déterminant. Des mécanismes d’échange ioniques Ca/Na/Fe couplés à
d’éventuels phénomènes de sorption (aux pH plus élevés) ont été proposés pour décrire
les résultats obtenus. L’étude a en particulier montré que le rapport atomique Ca/C des
agrégats diminue avec la concentration en coagulant à pH 8. Ces résultats sont corrélés
avec des études publiées par ailleurs montrant des améliorations de rendements
d’élimination de certains composés de la matière organique avec l’ajout de sels
minéraux par exemple.
37
De plus, une élimination séquentielle de la matrice organique en fonction de la
concentration en coagulant a également été démontrée (figure 10 a). Comme la
spectroscopie de perte d’énergie des électrons permet de calculer un rapport C/N pour
les colloïdes organiques, nous avons pu vérifier que les substances
polyhydroxyaromatiques de l’eau de Moselle coagulée avec du nitrate ferrique sont
incluses dans les premiers agrégats décantables (Jung et et al., 2005b).
(a)
Figure 10 : Evolution des rapports atomiques de Fe/C*1/10 (⟤ , N/C ∎ ,

Na/C*1/10 et Ca/C pour les substances humiques extraites de matières en
suspension de la rivière Moselle a NHS pH8, optimum de coagulation : 6,4.10‐4 mol/L
et l’eau de rivière Moselle b NOM pH 7.4, optimum de coagulation : 1,7.10‐4 mol/L
d’après Jung et al. 2005b
Enfin, cette étude a également permis d’étudier les phénomènes de reconformation

de la matière organique. En effet, les substances humiques sont généralement
représentées dans la littérature sous la forme de macromolécules plus ou moins
ramifiées, dont la conformation change en fonction du pH. Une représentation
concurrente suggère qu’elles correspondent à des associations supramoléculaires de
petites molécules hétérogènes. Des effets conformationnels ont été démontrés avec
l’analyse des flocs formés par EDXS. En effet, pour expliquer que le rapport molaire Fe/C
des substances humiques agrégées à pH 8 avec des espèces hydrolysées de Fe décroit
fortement au voisinage de la concentration optimale (figure 10b), il faut en faire appel à
la dynamique de reconformation du colloïde. A pH 8, les substances humiques sont
généralement représentées sous forme « dépliée » : lorsque le nombre d’espèces
coagulantes ajoutées dans la suspension est petit, l’intervalle de temps entre chaque
collision de colloïdes déstabilisés reste suffisamment grand pour permettre à ceux‐ci de
se réorganiser, et le rapport Fe/C est par conséquent élevé. En revanche, à concentration
plus importante, la fréquence de collision entre espèces coagulantes et substances
humiques augmente, et ne laisse plus le temps pour un éventuel réarrangement de
celles‐ci. Dans ce cas, le rapport Fe/C qui est mesuré à distance d’analyse constante, va
diminuer. A pH 6, Fe/C ne varie pratiquement pas dans le domaine de concentration
étudié. Ce résultat peut être expliqué en invoquant, d’une part l’incorporation de
nanocolloïdes organiques non destabilisés dans les agrégats, d’autre part leur
volatilisation sous le faisceau d’électrons au cours de l’analyse (Jung et al., 2005b).
38
1.4.3. Coagulation de la matière organique (MO)‐ étude des mécanismes de
coagulation à deux pH de travail et devenir du complexe pyrène/MO (jar‐
tests, dosage par spectrofluorimétrie).
La dernière partie du travail de doctorat a permis de donner un aperçu des

mécanismes intervenants lors du traitement par coagulation d’eau artificiellement
polluée avec le polluant « pyrène » (choisi comme représentant des HAP). L’élimination
du pyrène lors de la coagulation a été étudiée sur toutes les substances synthétiques et
naturelles, ainsi que sur deux échantillons d’eau de Moselle. Les variations des rapports
de fluorescence (permettant d’estimer la concentration en pyrène) ont été mesurées
parallèlement à celles de la turbidité, après chaque ajout d’agent coagulant. Les
substances humiques ont eu sensiblement toutes les mêmes comportements. Si
l’élimination du pyrène est parfois presque totale, son optimum coïncide rarement avec
celui de la matière organique. Un résultat capital a donc été que cette élimination n’est
pas le fait exclusif du caractère hydrophobe des substances humiques lorsque celles‐ci
s’insolubilisent par neutralisation des charges (en supposant que les interactions
hydrophobes soient conservées). Dans tous les cas, il semble exister un stade optimal de
piégeage du pyrène dans les habitats hydrophobes des substances humiques, favorisé
par le traitement coagulant, suivi d’un stade de relargage en cas de surtraitement (figure
11).
Figure 11 : Schéma conceptuel de l’interaction entre le pyrène et la matière organique lors de la

coagulation par le sel de fer (Jung, 2004, thèse de doctorat)
39
2. Caractérisation et facteurs de dégradation de la qualité de l’eau,
recherche d’indicateurs
Lors de mon intégration en tant que chercheure associée au LERES‐EHESP/IRSET en

2009, j’aborde la problématique de la qualité de l’eau de façon plus globale, avec une
optique sanitaire plus affirmée. Cette nouvelle expérience professionnelle va m’amener
dans un premier temps, au travers de différents projets (ERA‐Net Environment and
Health, ORACLE et Algues vertes/CDFO), à l’étude d’indicateurs environnementaux de
pollution « classiques » : le carbone organique dissous (COD), lui‐même indicateur de la
matière organique matière organique naturelle précédemment étudiée pour
comprendre les mécanismes d’interaction à l’échelle des agrégats humiques (partie 1) ;
les nitrates et formes du phosphore, principaux nutriments considérés au niveau de la
pollution agricole (notamment associés aux développement des algues vertes) ou encore
la densité cellulaire de cyanobactéries utilisée pour le contrôle sanitaire des eaux de
baignade. Ces études ont toutes montré les limites de l’utilisation de ces indicateurs bien
connus, notamment pour des raisons de spécificité des indices ou par manque de prise
en compte des facteurs de dégradation (comme les effets du changement climatique par
exemple). Certains indicateurs classiques ont d’ailleurs été remis en question par
rapport à leur significativité en terme sanitaire. Dans cette optique, une partie de mes
recherches a consisté en l’investigation de nouveaux modèles, de nouvelles méthodes ou
d’indicateurs complémentaires qui seraient des descripteurs plus complets de l’état ou
du niveau de pression des milieux.
Au niveau de mes travaux, j’ai travaillé aussi bien avec des indicateurs ayant un rôle
d’ « outils de diagnostic ou de suivi » (synthétisant les processus en jeu, matérialisant les
enjeux, décrivant les actions) que « d’outils d’aide à la décision » (permettant d’agir en
toute connaissance). A la base de cette étude d’indicateurs, une analyse de données a
toujours été considérée, quelque soit le projet envisagé. De plus, pour le projet ERA‐Net,
un indicateur composite obtenu par agrégation de données ou d’indicateurs a été
développé. Cet indice, donnant une approche résumée et simplifiée de systèmes
complexes et défini par un score, permet également une communication orientée vers
un public beaucoup plus large (parties prenantes, décideurs mais aussi acteurs du
terrain). Cette partie d’étude, élaborée dans le cadre de la création d’un outil d’aide à la
décision est décrite en partie 3, car elle apparait plus comme le stade final du
déploiement d’une évaluation de risques (Figure 12).
40
Figure 12 : Fil conducteur de la méthode de recherche autour de l’identification et de la
mise en œuvre des indicateurs
Les indicateurs ont dû également être mis en œuvre à différentes échelles

spatiales et temporelles en fonction de l’action menée et des objectifs. Au niveau spatial,
nous avons travaillé (i) à l’échelle de la parcelle, car il s’agit du premier niveau
d’enregistrement des pratiques, c’est l’unité environnementale élémentaire d’émission
diffuse des polluants d’origine agricole (iii) à l’échelle de la rivière, pour appréhender la
question des risques sanitaires liés à la formation de produits de dégradation pour des
opérations de traitement d’eau prélevée à partir d’eau de surface (ii) à l’échelle du
territoire (bassin‐versant), échelle du pilotage de l’action collective concernant plusieurs
interlocuteurs. Les indicateurs sont également influencés par les évènements survenus
au cours de la période considérée. Les facteurs climatiques ainsi que l’occupation des
sols et les pratiques culturales ont donc par exemple été considérés. La méthode de suivi
déployée a également nécessité un ajustement au niveau du pas de temps considéré. Une
revue critique concernant l’adaptation des méthodes de suivi pour des événements
hydrométéorologiques rapides, en particulier lors des fortes pluies a été écrite en ce
sens et donne les besoins et solutions actuelles en terme d’échantillonnage et d’analyses
ainsi que les tendances en recherche (Roig et al., 2011).
La présentation des projets de recherche conduits est donc organisée en fonction

de ces différentes étapes et contraintes : analyse des données existantes, conception
d’expérimentation spécifique en fonction des facteurs de dégradation supposés à
différentes échelles (parcelles, rivière, bassin versant), discussion des limites des
indicateurs de suivi en place et recherche de nouveaux indicateurs.
41
2.1. Exploitation de données historiques et analyse de relations entre paramètres
descripteurs des milieux
2.1.1. Pour l’indicateur « COD »
La recherche est partie du constat que les concentrations en COD ont, dans certaines
ressources en eau potable, plus que doublé en deux décennies et continuent encore
aujourd’hui d’augmenter (Delpla et al., 2009). La réduction des dépôts acides, couplée
avec une augmentation des températures et de la fréquence des inondations ainsi que
des changements dans la distribution saisonnière des précipitations, sont responsables
de l’augmentation des concentrations en COD dans les ressources en eau potable.
L’augmentation des niveaux de COD a des impacts potentiellement importants sur la
qualité de l’eau et la santé humaine, ce qui a attiré l’attention des agences européennes
de l’environnement et des industries de l’eau. Le COD est un paramètre qui pèse sur les
procédés traditionnels de traitement de l’eau : il représente un substrat pour le
développement microbien, entraîne la formation de Sous‐Produits de Désinfection (SPD)
potentiellement toxiques et peut être associé à la présence de micropolluants (métaux
lourds, pesticides, pharmaceutiques, hormones, composés perfluorés, …).
Le projet ERA‐Net Environment and Health (FP7, 2009‐2013) avait pour objectif de
permettre à l’échelon européen5 le suivi du COD, de comprendre l’interaction avec les
micropolluants du milieu et de modéliser l’impact sanitaire des complexes matières
organiques – micropolluants formés lors des opérations de traitement de l’eau (figure
xx). Outre les impacts du changement climatique sur la disponibilité de l'eau et des
risques hydrologiques, les conséquences sur la qualité de l’eau commencent à peine à
être étudiées. Une revue de la littérature a permis de montrer le rôle des principaux
facteurs de dégradation (réchauffement et les conséquences des événements extrêmes
telles que les sécheresses et les fortes pluies) expliquant les effets du changement
climatique sur la qualité de l'eau (Delpla et al., 2009). La principale conclusion qui a pu
être tirée de la bibliographie est que la tendance de dégradation de la qualité de l'eau
potable dans le contexte du changement climatique conduit à une augmentation des
situations à risque liées à l'impact potentiel pour la santé.
En zoomant en particulier sur la problématique de la matière organique, les questions à

la recherche étaient donc les suivantes: quels sont les principaux facteurs hydro‐
climatiques permettant d’expliquer l’évolution des concentrations de COD ? Quelles sont
les conséquences de scénarios de changement climatique et de la baisse des dépôts
acides sur les flux et les concentrations de COD futures entrant dans un système de
traitement de l'eau? Quelles sont les impacts attendus pour la formation de sous‐
produits de désinfection dans le processus de traitement de l'eau ? Quelles sont les
conséquences sur la santé des changements prévus dans la formation des sous‐produits
de désinfection et d'autres contaminants chimiques dans l'eau potable ? (figure 13)
5
Partenariat : Royaume‐Uni (Université Bangor, Centre Environnement et Hydrologie de
Lancaster) et Pays‐Bas (université d’Ultrecht)
42
Temperature
Warming
Emerging
Environment substances &
pathogens fate
Water
Rising DBPs
resources production
Natural Urban
Micropollutants
Nutrients
DOM
Biological
Drinking waters
Agricultural
Intensity/volume variations
Human Health Risks ?
Rainfalls
Mitigation options?
Adapted from [Delpla et al., Env. Int., 2009]
Figure 13 : Schéma de la problématique de recherche initiée au LERES pour le projet ERA‐Net

(adapté de Delpla et al., 2009)
Le thème de recherche initial a donc été la vulnérabilité sanitaire des impacts du

changement climatique global, à savoir mieux caractériser les milieux aquatiques,
d’avantage prendre en compte les pressions anthropiques et enfin mieux considérer les
changements climatiques et l’influence du territoire sur la qualité des eaux.
Pour l'équipe française, les principaux résultats du projet ont été:
 L'exploitation des données historiques de qualité de l'eau permettant de définir

les facteurs hydro‐climatiques « à risque » (débit, température de l’eau)
 La conception de protocoles expérimentaux de suivi innovants autour
d'événements climatiques spécifiques à l’échelle de la parcelle agricole afin
d’appréhender les impacts des fortes précipitations sur la qualité des eaux de
ruissellement et de l'eau de la rivière (2.2.1)
 La mise en œuvre d'expériences de laboratoire pour comprendre le rôle des
événements de précipitations en lien avec la formation de sous‐produits de
désinfection dangereux (2.2.2)
 La définition d'une architecture générale pour le système aide à la décision
(2.2.3)
La dernière question à la recherche posée portant sur l’évaluation des risques sanitaires,
utilisant des méthodes de travail spécifiques (outil d’aide à la décision), est traitée dans
la partie 3.
43
Ainsi, l'impact du climat sur la qualité de l'eau a été évalué en Bretagne pour 3 petits
bassins hydrographiques agricoles confrontés à des pressions élevées et utilisés pour la
production d'eau potable pour de petites unités de distribution d’eau, particulièrement
vulnérables aux changements climatiques. Le débit et la température de l'eau ont été
considérés comme des variables climatiques majeures et les teneurs en COT et en nitrate
comme variables de qualité de l'eau. Les données de qualité de l'eau ont été fournies par
le programme de contrôle sanitaire et ont couvert une période de 27 ans (1983‐2009)
pour les ressources et les eaux traitées. L'impact combiné de la température et le débit
sur la qualité de l'eau ont été considérés (figure 14).
a) b)
20 30
TOC
TOC < 75% of limit
75% ‐ 100% of limit
> 100% of limit
Water tem perature (°C)

15
20
TOC (mg/L)
10
T mean
10
5
0 0
100 30
< 75% of limit
Nitrate
Nitrate
75% ‐ 100% of limit
80 > 100% of limit
Water temperature (°C)
Nitrate (mg/L)
20
60
40 T mean
10
20
0 0
-2 -1 0 1 -2 -1 0 1
Log (Q/Qm) Log (Q/Qm )
Figure 14. a) Impact du débit sur le COT et les nitrates dans la ressource de la rivière Meu (ligne
pointillée indique la limite de qualité pour les ressources en eau potable); b) Vulnérabilité des
ressources pour les eaux potables obtenus à partir d’eaux de surface par rapport aux
concentrations de COT et de nitrates selon l'hydrologie et la température de l'eau par
comparaison avec les valeurs limites réglementaires françaises
(rapport intermédiaire ERA‐Net, 2011 et compléments dans Delpla et al., 2011a).
Les résultats ont montré que les variations de débit ont un impact majeur sur la
qualité de l'eau. Comme exemple représentatif, les résultats sur la rivière Meu ont
montré que la concentration en COT diminue avec l'augmentation du débit (jusqu’à des
débits moyens), mais des augmentations pour les débits élevés ont été trouvées,
reflétant un apport plus élevé des sols vers les cours d'eau (effet d’entrainement). La
tendance inverse a été constatée pour les nitrates (figure 14a) : pour les plus hauts
débits, la dilution prédomine. Les résultats obtenus sur trois rivières a également permis
d'identifier les conditions hydroclimatiques pour lesquels la probabilité de
dépassements des limites de la qualité de l'eau est la plus élevée (figure 14b).
44
La variation du rapport C/N (COT/N‐NO3) avec le débit de la rivière a été regardée
de plus près par rapport au niveau des conditions hydrométéorologiques « à risque ». Il
a été montré le rapport augmente de façon significative pour les conditions extrêmes
(pour les très bas débits, inférieurs à 10% du débit annuel moyen et pour les très hauts
débits, supérieurs à au moins 500% du débit annuel moyen Qm) (Figure 15). Pour les bas
débits (apparaissant majoritairement en périodes d’étiages en été), les nitrates sont
assimilés par la biomasse et le relargage des sols est également limité en saison sèche, la
croissance de biomasse explique également les augmentations de COT pour ces
conditions. Pour les périodes de hauts débits (période hivernale), le rapport C/N serait
plutôt contrôlé par la dilution et le ruissellement de surface. Les valeurs de C/N trouvées
sont d’ailleurs proches de celles de la littérature pour des sols, reflétant un fort apport
de constituants organiques exogènes dans le milieu aquatique.
Figure 15 : Impacts du débit sur la variation du rapport C/N pour 3 sites

(Baurès et al., 2013a)
Cette partie d’étude a donc permis de montrer (i) des tendances historiques
contraires entre le carbone et l’azote (ii) de confirmer que la matière organique est un
paramètre de qualité de l'eau particulièrement sensible aux conditions
hydroclimatiques exceptionnelles (sécheresses et inondations) (iii) d’évaluer les
probabilités de risques sanitaires à partir de ces 2 indicateurs en fonction de données
hydroclimatiques ; les concentrations en nitrates dans les eaux de surface ont tendance
à diminuer à la fois en périodes de sécheresse et de crue, alors que les concentrations en
COT ont tendance à augmenter pendant les crues (Delpla et al., 2011a).
45
2.2.2. Pour l’indicateur « densité cellulaire »
L’étude ORACLE (2012‐2015), financée par l’ANSES (PNRSET 2012) posait les
questions à la recherche suivantes : Quelle est la place des cyanobactéries en tant que
composante de la flore planctonique ? Comment prendre en compte l’influence des
caractéristiques physiques du milieu (morphologie des sites, facteurs météorologiques
et évolution climatique actuelle) ? Comment évaluer la pertinence des méthodes de suivi
et de gestion du risque ? Peut‐on développer des outils et des protocoles d’alerte
complémentaires ?
L’exploitation de données de 26 lacs récréatifs de Bretagne , surveillés chaque année sur

une période de 7 ans de 2004 à 2011 a permis de mettre en évidence les variations
spatiales et interannuelles du développement de cyanobactéries à l’échelle régionale. La
première phase d’exploitation concernait les paramètres définissant les risques
sanitaires et notamment la densité cellulaire liée aux seuils de dépassement de l’OMS.
Figure 16 : paramètres d’évolution des crises de cyanobactéries à l’échelle régionale

(Baurès et al., 2013)
Ces résultats ont permis de montrer (i) une augmentation interannuelle de la fréquence
de dépassement des seuils fixés par l’OMS, de niveau 2 (20000 cellules/mL) et 3
(100 000 cellules/mL) pendant la durée des suivis (Figure 16 2a) (ii) une relation
annuelle positive entre la durée d’exposition et le maximum de densités cellulaire
(Figure 16 2b) (iii) une relation positive entre la récurrence annuelle et la durée
annuelle d’exposition (iv) une relation inverse entre la fréquence de détection de
microcystines et la densité cellulaire des cyanobactéries.
46
2.2. Conceptions de protocoles autour d’événements spécifiques
2.2.1. Expérimentations à l’échelle de la parcelle
Comme la majorité du COD provenant des sols et emmené à la rivière est exporté durant
des événements pluvieux, une meilleure compréhension du transport de la matière
organique, adaptée à l’échelle de l’événement (fortes pluies) est apparue comme
nécessaire.
Une étude expérimentale a été menée sur des parcelles agricoles gérées par l'Institut
national français de la recherche agronomique (INRA Rennes) dans le cadre de la thèse
de I. Delpla (projet ERA‐Net). Deux parcelles fertilisées selon les pratiques d’épandage
locales (lisier de porc (PS) et fumier de bovins (CM)) plus une parcelle de contrôle ont
été sélectionnées. Des campagnes d'échantillonnage des eaux de ruissellement par
temps de pluie ont été menées pour plusieurs événements pluvieux (douche et
tempête). L’événement de type « douche » a été défini pour une intensité maximale
(Imax)> 20mm/h et une courte durée, et la tempête a été définie pour une intensité
maximum inférieure (<10 mm/h) et une plus longue durée (Roig et al., 2012). Les
données hydrologiques et météorologiques ont été mesurées sur le site. L'eau de
ruissellement a été échantillonnée en utilisant des échantillonneurs automatiques et
analysée immédiatement pour le COT, COD et les spectres UV (Figure 17).
Figure 17 : Description schématique du dispositif de collecte et de prélèvement des eaux de

ruissellement
Grâce à un équipement dédié pour la surveillance des eaux de ruissellement et

l'échantillonnage, il a été possible de suivre avec un court intervalle de temps l'impact de
la pluie sur la variation de la principaux paramètres de qualité de l'eau tels que le
carbone organique (total, dissous et en suspension) et de l'azote Kjeldahl pendant toute
la durée de l’événement pluvieux.
A notre connaissance, ce travail a été un des premiers traitant de la variation de la

qualité de l'eau de ruissellement lors des événements pluvieux à l’échelle de parcelles
agricoles fertilisées, avec un système de surveillance à haute fréquence.
La principale conclusion pour l'événement de type « douche » a été un impact significatif

de la pluie pic sur le COT (Figure 18). Le principal résultat pour l'événement de
« tempête » en hiver (après l'épandage) est une augmentation de COT pour les eaux de
47
ruissellement avec un maximum correspondant au maximum de l'intensité du débit. Le
COD ne suit pas la même tendance avec une concentration maximale mesurée durant la
première vague de l'événement. Les spectres UV ont donné des informations en
conformité avec les concentrations de carbone. En considérant les flux de carbone,
l'exportation de COD lors d'événements pluvieux intenses est plus élevée que lors des
pluies de faible intensité. Cette étude nous a aussi permis de montrer l’influence notable
de la fertilisation sur les phénomènes d’export de la matière organique.
48
Rainfall intensity (mm.h ‐1 )
1 2 3
10 4
8 3
Absorbance (a.u.)
6 2
4 1
2 0
200 250 300 350 400
0 Wavelength (nm)
17:00 19:00 21:00 23:00 01:00 03:00 1
120
4
Absorbance (a.u.)
100
2
D O C and TO C (m g/L)
23:00 00:00 01:00 02:00
80
0
200 250 300 350 400
Wavelength (nm)
60 2
40 4
Absorbance (a.u.)
10 2
0 0
200 250 300 350 400
17:00 19:00 21:00 23:00 01:00 03:00 Wavelength (nm)
3
DOC TOC
Rainfall intensity (mm.h ‐1)
1 2 3 4
30 0.6
25
Absorbance (a.u.)
0.4
20 0.2
15 0.0
200 250 300 350 400
10
Wavelength (nm)
1
5
0 0.6
8:00 10:00 12:00 14:00 16:00

Absorbance (a.u.)
0.4
0.2
16
0.0
200 250 300 350 400
Wavelength (nm)
14 2
12
DOC and TOC (mg/L)
0.6
Absorbance (a.u.)
10 0.4
0.2
8 0.0
200 250 300 350 400
Wavelength (nm)
6 3
4 0.6
Absorbance (a.u.)
0.4
2 0.2
0 0.0
200 250 300
Wavelength (nm)
350 400
8:00 10:00 12:00 14:00 16:00 4
DOC TOC
Figure 18: Comparaison des paramètres COT, COD, spectres UV pour un événement de type
“douche” (08/11/10 – 16 échantillons‐ figure du haut) et « tempête » (27 et 28/02/10 – 24
échantillons – figure du bas) sur une parcelle avec épandage de lisier de porc
(adapté de Delpla et al., 2011b)
49
Etant donné les différences de comportement observé entre les exports de COD et de
COT, un fractionnement des échantillons a également été mené pour des périodes clés
du cycle hydrologique (première phase de la chasse d'eau, à débit fixe et lors de la
dernière phase de ruissellement). Trois fractions ont été étudiées: soluble (0,1 um),
colloïdale (0.1‐1.2μm) et de particulaire (1,2 pm). Les résultats obtenus lors de
l'événement de type « tempête » ont montré que le carbone organique est
principalement sous forme colloïdale et sous forme soluble au cours de la première
étape de l'événement, et ont suggéré un contrôle de l'intensité des précipitations sur le
transport de la fraction colloïdale (Figure 19).
8
Rainfall (mm/h)
6
4
2
0
100
Organic carbon (%)
80
60
40
20
0
140 182 365 500 518
Time (min)
Soluble (<0.1µm) Colloids (0.1‐1.2µm) Particulate (>1.2µm)
Figure 19 : Répartition du carbone organique entre les phases solubles, colloïdales et

particulaires lors d’un événement pluvieux (tempête Xynthia) (Delpla et al., 2011b)
2.2.2. A l’échelle de la rivière
Avant de passer à l’échelle du bassin versant, les recherches menées avec les
partenaires européens du projet ERA‐Net ont permis d’extrapoler les phénomènes de
transferts d’une parcelle (ou d’un sol) à la rivière à différents types d’environnements.
L’objectif était notamment d’appréhender le risque sanitaire lié à la formation de sous‐
produits de transformation liée au traitement de l’eau brute de rivière contenant une
certaine quantité de COD pour des événements particuliers (fortes pluies). Ces résultats
nous ont également permis d’aller plus loin dans l’interprétation des comportements
entre le COD et le COP au niveau des dynamiques dans le cours d’eau.
50
L’impact d’événements pluvieux sur l’export de la matière organique a été comparé pour
plusieurs types d’environnements très différents en France (Bretagne) et au Royaume‐
Uni (Wales), avec des compositions organo‐minérales très variables (tourbières,
environnements agricoles proche de la source de captage pour l’alimentation en eau
potable, pédologie plutôt minérale …). L’analyse des échantillons d’eau sur ces sites
(avant‐pendant et après l’épisode pluvieux) nous ont permis de montrer une
augmentation significative des teneurs en COD dans les cours d’eau (de 48 à 158%). En
utilisant le SUVA (rapport de l’absorbance à 254nm/COD), il a été démontré que les
composés aromatiques et de hauts poids moléculaire sont les plus abondants dans le
cours d’eau suite à l’événement pluvieux. Cependant, ces changements qualitatifs et
quantitatifs dans le cours d’eau ne semblent pas affecter de façon significative le
potentiel de formation des tri‐halométhanes (Delpla et al., 2015).
Les concentrations en COD pour les sites d'étude ont répondu différemment aux
événements de précipitations, bien que pour les 3 sites des boucles similaires
d'hystérésis concentrations/débits ont été générées (Figure 20).
Site A Site B Site D

20 20 20
Rising limb Rising limb
Descending limb Descending limb
15 15 15
DOC (mg/L)
10 10 10
5 5 5
0 0 0
0 5 10 15 20 0.0000 0.0025 0.0050 0.0075 0.0100 0 1 2 3 4
Flow rate (m 3 /s) Flow rate (m 3 /s) Flow rate (m 3 /s)
Figure 20 : Relations entre le COD et le débit de la rivière pour 3 sites d’étude pendant des
événements pluvieux (Delpla et al., 2015)
La rotation à gauche de l’hystérésis indiquerait que l'eau du sol, plutôt que celle issue du
ruissellement de surface et des eaux souterraines, est la plus importante source de COD
pour ces trois cours d'eau. Le flux de COD plus important sur la branche descendante de
l’hystérésis pourrait être expliqué par l’apport de matière organique dissoute terrestre
et un effet relativement faible de dilution en raison des événements de type « tempête ».
De plus, ces résultats ont montré un effet relativement prolongé des précipitations sur le
COD. En effet, pour le COD, un effet de longue durée des précipitations (plusieurs heures
après les pluies) a été observé, contrairement au COP. Ce travail a donc également
montré une différence notable entre le comportement de la matière organique dissoute
et particulaire et l’importance de déterminer les sources de particules pour le suivi des
teneurs dans les rivières.
51
2.2.3. A l’échelle du territoire (bassin versant)
Concernant la connaissance des dynamiques de transfert de la matière

organique dans l’eau, un pas de plus a été franchi lors de la thèse de Jean Causse qui a
permis d’aller poser ces questions au niveau de l’échelle du bassin versant.
Aujourd’hui, une contribution forte et partagée des différents partenaires de l’eau sur un
territoire fait particulièrement sens. Dans le cadre de la loi NOTRe, l’Assemblée
Nationale a adopté dès le 4 mars 2015 le transfert obligatoire des compétences eau et
assainissement aux collectivités territoriales, ce transfert étant efficient dès 2020.
L’enjeu est celui de la mise en œuvre, à l’échelle régionale, d’une politique intégrée et
plus cohérente, correspondant en Bretagne à la géographie de son bassin
hydrographique. L’expérience de la mise en œuvre du plan algues vertes en Bretagne a
montré toute la cohérence de cette approche.
Le travail de thèse de J. Causse (2012‐2015), co‐dirigé avec le LERES/EHESP (direction :

Olivier Thomas, co‐direction : Estelle Baurès) a été initié par CDFO (Coop de France
Ouest), regroupement d’entreprises agricoles de la région ouest, qui est l’une des parties
prenantes dans la mise en œuvre du plan national contre le développement des algues
vertes. Concernant cette problématique, la première interrogation était qu’en plus des
apports agricoles et des épandages d’élevages, il existe des rejets d’azote d’origine
humaine dont l’importance relative n’est pas réellement connue. Ces rejets sont ceux des
stations d’épuration et des dispositifs d’assainissement autonomes, et peuvent être
ponctuellement importants dans certaines zones côtières et surtout peu traités en tout
début de saison estivale le temps que les systèmes s’adaptent. La deuxième question
était liée aux effets des pluies qui sont moins nombreuses que le passé (en nombre de
jour), mais plus intenses localement. Le ruissellement qui en résulte lessive les sols
agricoles et urbains, et les réseaux d’assainissement peuvent être saturés
ponctuellement, ce qui peut provoque l’entraînement de fortes charges d’azote en
particulier.
Les questions posées à la recherche étaient donc les suivantes :
• Les efforts réalisés par les agriculteurs peuvent‐ils être quantifiés ?

• La dynamique des flux de nutriments centrée sur la période estivale est‐elle
suffisamment connue ? (périodes à risque)
• La qualité et la quantité des rejets azotés varient‐elles significativement dans le
temps et notamment lors de la saison estivale ?
• Quel est l’impact des changements climatiques (fréquence et intensité de la
pluviométrie, hausse des températures,…) ?
Suite à la publication d’un rapport interministériel6, CDFO s’est rapproché d’une équipe
de recherche pluridiscplinaire afin de s’impliquer dans la compréhension des
phénomènes de transferts de nutriments au sein des bassins versants.
6
CGAAER and CGEDD (2012) Bilan des connaissances scientifiques sur les causes de prolifération de
macrolagues vertes : applications à la situation de la Bretagne et propositions. Ce rapport incite au
développement de pistes de recherche pour une meilleure connaissance des flux de nutriments (N et P)
pendant la période de croissance des ulves et l’amélioration du savoir‐faire en génie écologique (intérêt et
52
Les objectifs du projet était de :
• Compléter les connaissances du phénomène d’export de nutriments à partir de

différentes sources sur le bassin versant et analyser ces variations à l’échelle
intra‐annuelle
• Contribuer à l’évaluation des flux de nutriments lors de situations à risque (fortes
pluies, forte affluence touristique, défaut d’infrastructures agro‐écologiques,…)
• Participer à l’évaluation des efforts des agriculteurs pour la réduction des exports
• Evaluer l’impact des changements climatiques et les effets de l’aménagement du
territoire
Le rapport interministériel confirmait de façon irréfutable que la réduction des nitrates

était le principal levier de lutte contre les algues vertes, tout en rappelant l’intérêt de
diminuer également l’export de phosphore. Une revue bibliographique internationale
(Causse et al., 2015) a montré l’importance d’une diminution conjointe d’autres
nutriments. L’influence du carbone organique dissous a notamment été assez peu
étudiée, malgré le fait que de nombreux paramètres lui soit lié. Cet état de l’art a
également permis de décrire les différents types de sources de contamination
(ponctuelles, diffuses, discontinues et continues) en distinguant les origines agricoles,
urbaines et naturelles (figure 21).
Figure 21 : Les différents types de sources d’azote dans l’environnement

(Causse et al., 2015)
limites de l’utilisation de diverses infrastructures écologiques, compréhension de leur mode

d’action/relations bassins versants écosystèmes côtiers dépendants)
53
C’est dans ce contexte que l’étude simultanée de différentes formes de nutriments a été
réalisée, à l’échelle de 2 bassins versants de taille moyenne (environ 100 km2), situés en
zone côtière et soumis au risque de marée verte dans leur partie littorale.
La méthodologie de travail choisie est innovante par le choix de l’échelle spatio‐

temporelle. En effet, très peu d’études en milieu agricoles considèrent les variations à
petites échelles de temps (événement voire épisodes inter‐crues). Ces variations posent
quant à elles pourtant beaucoup de questions scientifiques d’importance comme celle i)
du rôle exact des facteurs hydrologiques sur le transfert de nutriments, ii) de la
localisation spatiale des sources de nutriments dans les bassins versants, et iii) de la
nature des mécanismes par lequel le COD notamment est produit au niveau du sol.
Concernant le rôle de l'anthropisation et notamment des pratiques agricoles, il convient
de rappeler que les variations des teneurs et flux de COD aux échelles de temps courtes à
intermédiaires ont surtout été étudiées dans des contextes alpins et forestiers. De fait,
très peu d’études traitent de la dynamique des nutriments et en particulier du COD dans
des bassins versants agricoles, et ce alors que beaucoup des rivières concernées par des
augmentations long‐termes de leur teneur en COD drainent précisément des régions
d'agricultures intensives. Traiter la question des mécanismes impliqués dans le transfert
sol‐rivière du COD nécessite donc d'adapter judicieusement l'échelle temporelle
d'investigation à celle de la dynamique de ce transfert. Paradoxalement, peu d’auteurs se
sont intéressés à la dynamique du COD à l'échelle de la crue ‐ i.e. à l'échelle de l'heure
dans le cas de petits cours d'eau – alors que les crues constituent pourtant les principaux
épisodes au cours desquels du COD est transféré du sol vers les rivières.
L’étude a donc couvert les transferts de l’échelle de l’exutoire permettant de travailler à

l’échelle de l’événement (crue) jusqu’à l’échelle du bassin versant (à l’échelle
interannuelle) (Figure 22).
Figure 22 : Représentation des principales étapes nécessaires à l’évaluation de la variabilité

spatio‐temporelle de l’export de nutriments à l’échelle du bassin versant
(d’après Causse et al., en préparation)
Trois types d’expérimentation ont été conduits (campagnes de prélèvements

ponctuelles (pas de temps hebdomadaire sur 1 durée d’un an, prélèvements
automatiques (pas de temps horaire sur 1 saison), mesure haute fréquence (pas de
54
temps de 15 minutes sur 1 durée d’un an) (Tableau 2). Pour les mesures « haute
fréquence », l’originalité des analyseurs UV de terrain était la possibilité d’acquisition
d’un spectre UV avec 2 trajets optiques différents (3 et 30mm), permettant de
s’affranchir des problèmes de dilution pour la mesure des nitrates, tout en pouvant
« zoomer » sur la partie matière organique du spectre.
Tableau 2 : Plan d’expérimentations considéré et capacité à évaluer la variabilité spatiale (S) et

temporelle (T) de l’export de nutriments, ainsi que la représentativité des évènements à court
terme (R) (Causse et al., en préparation)
Une typologie des stations de prélèvements a pu être réalisée. En effet, 3 groupes de

stations ont été obtenus suivants les teneurs en nitrate et carbone organique dissous,
avec des profils différents, redémontrant la relation inverse entre ces 2 nutriments
(Tableau 3).
55
Tableau 3 : Exemples de groupes de station de mesures sur la base du rapport C/N (Thèse J.
Causse, 2015)
Le 3ème groupe correspond à des stations atypiques avec COD > 10 mg.L‐1 et ≈1
pour lequel la présence de STEP avec des mauvais fonctionnements induit globalement
une diminution des concentrations en nitrates et une augmentation en COD pour la
période étudiée.
Un schéma conceptuel associé à l’abattement des concentrations en NO3‐N pendant les

crues a été proposé pour expliquer les différences de comportement (Figure 23).
56
Figure 23 : Exemples de crues inférieure à débit modéré (< 2000 L.s‐1, 06/01/14) et fort débit (>
2000 L.s‐1, 06/02/14) et schéma conceptuel associé de l’abattement des concentrations en NO3‐
N pendant les crues associées. Les processus dominant pour l’abattement de la concentration en
NO3‐N sont indiqués en gras (thèse J. Causse, 2015)
Les nitrates et le COD sont en effet caractérisés par des dynamiques de transports
différents, les nitrates étant majoritairement transférés par les eaux souterraines après
lixiviation, alors que le COD est majoritairement exporté lorsque les eaux souterraines
atteignent les horizons de sols riches en matière organique, c'est‐à‐dire en période de
crue (hautes eaux). En période de crue, les processus de transformation dans ou près de
la rivière pourraient donc être prédominants. Pour le COD, l’impact de la connectivité
hydrologique entre la nappe souterraine et la rivière semble être important jusqu’à un
certain débit, puis une stabilisation des teneurs en COD peut être observé à cause du
carbone autochtone (représentant jusqu’à 80% du COD des rivières en fin d’été). Les
paramètres à prendre en compte pour aller plus loin dans l’interprétation seraient
l’augmentation des surfaces de contact entre la rivière et les zones de dénitrification
(zones humides), l’augmentation de la biodisponibilité du COD et l’augmentation de la
vitesse d’absorption des nitrates par les plantes en présence d’une forte teneur en COD.
L’exploitation statistique (matrices de corrélation) au niveau des données de qualité de
l’eau a également démontré une relation C/N très forte. Par exemple, le taux de
saturation en oxygène est corrélé positivement avec les nitrates et anti‐corrélé avec le
COD, ce qui confirme bien le lien entre la relation C/NO3 et les réactions biochimiques
dans les cours d’eau ou à proximité.
Si l’on considère la dynamique des nutriments de façon détaillée, des phénomènes

d’hystérésis, dont la forme est plus ou moins marquée suivant les crues, ont été
observés. Ces phénomènes sont interprétés comme la resuspension des sédiments de
rivière (sens horaire) ou au contraire comme l’apport par la contribution d’eau nouvelle
de sources de particules issues de l’érosion des sols (sens anti‐horaire) (Figure 24).
Comme pour les travaux de thèse de I. Delpla, on retrouve des comportements
57
sensiblement différents pour le COD et le COP suivants les mois (et les épisodes hydro‐
météorologiques considérés).
Figure 24 : schéma conceptuel de la dynamique des nutriments étudiés (concentrations en

fonction du débit. H : sens horaire. A : sens anti‐horaire. F : février M : mars A : Avril M : mai
J : juin) (oral thèse J. Causse, 2015)
Le dispositif de suivi actuel de la qualité des eaux des cours d’eau envisagé pour
contrôler l’efficacité du « Plan Algues Vertes » ne prévoit pas de mesures en continu de
la qualité des eaux basées sur la mise en œuvre de capteurs spécifiques de qualité de
l’eau. Certes on ne peut pas multiplier les capteurs pour des questions de coût.
Cependant leurs informations seraient d’autant plus précieuses que les paramètres de
qualité qui nous intéressent varient dans le temps et que de nombreuses inconnues dans
la compréhension des phénomènes subsistent. Bien entendu, il ne suffit pas d’avoir des
mesures en continu ; celles‐ci ne seront utiles que si elles sont exploitées à la hauteur de
ce qu’elles peuvent révéler sur la réponse du milieu naturel. Un travail concernant le
transfert des connaissances et le développement de recommandations pratiques au
niveau des utilisateurs (agriculteurs, syndicats de bassins versant, Agence de l’eau) est
en cours. Il permettra d’élaborer des fiches de « bonnes pratiques » tant pour les
recommandations au niveau de certains pratiques agricoles que pour assurer un suivi
pertinent des flux de nutriments. Comme il a déjà été annoncé, les questions des limites
des indicateurs actuels et du développement d’indicateurs pertinents de suivi se pose
également et fait l’objet du paragraphe 2.3.
2.3. Limites des indicateurs classiques et développement de nouveaux

indicateurs de risques sanitaires
2.3.1. Risque « Eutrophisation »
Parmi les techniques utilisées pour évaluer la variabilité de la qualité de l’eau, la

spectrophotométrie UV est une technique simple, rapide et peu coûteuse largement
utilisée dans les dernières décennies. L'exploitation des spectres UV pour les eaux
naturelles est basée sur deux domaines de longueur d’onde : entre 200 et 240 nm et
entre 250 et 400 nm. Le premier est lié à la présence de nitrates (pic centré sur 205 nm)
contenus dans la majorité des eaux naturelles. Le second domaine est lié à la présence
de matière organique (MO) et de matières en suspension. La présence de MO est
caractérisée par un léger épaulement vers 260‐270 nm. Le niveau d'absorption et la
pente générale du spectre dans la région 250‐350 nm peut varier selon la nature des
matières en suspension et la présence de colloïdes en forte concentration.
58
Pendant les travaux de thèse de I. Depla et J. Causse, l'utilisation de spectres UV a permis
de définir une typologie de la qualité des eaux de surface (Figure 25). Selon le rapport
existant entre le nitrate et le COD, les informations UV permettent une classification
simple entre les environnements « naturels » (1 & 2) et les environnements
« perturbés » par l’activité anthropique (3, 4 et 5). Cette représentation est assez
cohérente avec la publication de référence de Taylor et Townsend, Nature 20107.
Nitrate
5
Absorbance (a.u.)
6
3 2
0
3
200 250 300 350
Wavelength (nm)
Absorbance (a.u.)
0
200 250 300
Wavelength (nm)
350
Natural
3
1 2 4 5
Absorbance (a.u.)
3 3
DOC
1
Absorbance (a.u.)
Absorbance (a.u.)
2 2
0
200 250 300 350
Wavelength (nm)
1 1
0 0
200 250 300 350 200 250 300 350
Wavelength (nm) Wavelength (nm)
Figure 25 : Typologie des eaux de surface par spectroscopie UV

(présentation en réunion d’avancement projet européen ERA‐Net, 2011 non publié)
La méthode de normation consiste à imposer la même aire à un ensemble de spectres.

La somme des valeurs d’absorbance sur un intervalle de longueur d’onde précis (200 –
350 nm) est ainsi généralement ramenée à une valeur arbitraire choisie.
L’aire sous le spectre avant normation (AIRE) s’écrit :
[2]
où :
A( : valeur d’absorbance à une longueur d’onde avant normation.
: pas d’acquisition (1 nm dans le cas du spectrophotomètre utilisé).
L’absorbance normée s’écrit alors pour chaque longueur d’onde :
7 Taylor and Towsend, Stoichiometric control of organic carbon–nitrate relationships from soils to the sea,
Nature 464 (2010) 1178‐1181, doi:10.1038/nature08985.
59
∗ [3]
où :
A*( : valeur d’absorbance à une longueur d’onde après normation.
AN : aire relative de normation (valeur fixée généralement à 100).
Le facteur de normation (Fn) est le rapport de la somme des absorbances normalisées

(AN) sur la somme des absorbances du spectre. Entre 200 et 350 nm, il s’écrit donc :
∑ ∗
[4]
∑
Pour un échantillon donné (j), on peut alors calculer la concentration normée d’un soluté
via l’équation :
∗ [5]
où :
Cnj : concentration normalisée
Cj : concentration mesurée
Fn : facteur de normalisation du spectre j.
Cette méthode de normation permet, d’une part, une comparaison qualitative des
spectres de différentes origines et/ou conditions d’échantillonnage (prélèvements lors
de saisons différentes, d’un évènement exceptionnel, en différents points….) et, d’autre
part, de s’affranchir des effets de dilution ( Thomas et al., 2014). De plus, cette méthode
est utilisée ici afin de révéler la présence d’un point isobestique latent (figure 26) qui
met en évidence une conservation qualitative de la composition de l’eau, au cours du
temps et dans l’espace. Dans les eaux douces de surface, l’existence de ce point est
expliquée par la présence de 2 composés majoritaires, les nitrates et la matière
organique dissoute (évaluée par le COD) dont les proportions varient en fonction des
conditions météorologiques avec la relation suivante entre leur concentration :
1 [6]
60
Point isobestique
latent
Figure 26: Spectres bruts (a) et normés (b) d’échantillons d’eau de la rivière (Vilaine) et
relations entre les nitrates et le COD avant (c) et après (d) normalisation (relation linéaire après
normalisation R2 = 0,936, n = 55)
Le calcul de la dérivée seconde d’un spectre permet est un outil complémentaire

permettant quant à lui d’identifier la position exacte d’épaulements sur un spectre brut
parfois très difficiles à repérer dans le cas de légers d’épaulements sur un spectre brut
parfois très difficiles à repérer dans le cas de légers épaulements comme pour l’examen
de la matière organique entre 250 et 350 nm. Un épaulement correspond à un
changement de pente sur le spectre brut et se traduit par un minima ou un maxima sur
la dérivée seconde.
Le calcul de la dérivée seconde en une longueur d’onde se fait via l’équation suivante :
² 2 ∗ [7]
² ²
où :
Aλ : Absorbance à une longueur d’onde λ
h : pas de la dérivée
Le pas de la dérivée influe fortement sur les informations tirées du spectre. Dans la
plupart des cas, un pas intermédiaire de 10 nm a été choisi après plusieurs tests, ce qui
permet à la fois un premier lissage de la courbe et limite la perte d’informations. Dans le
cas d’un pas plus faible (ex : 5 nm), les variations sont difficilement interprétables.
61
Lors de la thèse de J. Causse, un traitement statistique des spectres UV a permis de
dégager plusieurs groupes de spectres. L’indice DSA270 (absorbance de la dérivée
seconde du spectre à 270nm) issu de la mesure de l’absorbance de la dérivée seconde à
270nm indiquerait la présence de carbone organique assimilable par les micro‐
organismes dénitrifiants d’origine naturel (Figure 27). Couplé avec l’indice DSA300
estimant la concentration en COD (résultat de la thèse de I. Delpla), il permettrait donc
l’identification de « zones clés » à restaurer en priorité pour permettre une efficacité
optimum des actions de remédiation dans la lutte contre le développement des algues
vertes. En effet, la restauration de paysages naturels ayant une capacité tampon
importante serait directement cartographiée par cet indice. De même, l’indice DSA280
pourrait être indicateur du carbone organique assimilable d’origine anthropique. Ces
résultats sont encore à confirmer pour voir la possibilité d’extrapolation au niveau
d’autres types d’eaux et d’effluents.
Figure 27 : Exemples de dérivées secondes de l’Absorbance (DSA) des spectres analysés lors de
la campagne de prélèvements ponctuels du 13/06/13 et indices associés : DSA270, DSA280,
DSA300 et DSA330 (thèse J. Causse, 2015)
2.3.2. Risque « cyanobactéries »
Le projet ORACLE illustre particulièrement le lien mis en évidence dans le cadre de

notre recherche entre les déterminants environnementaux et les risques sanitaires.
Cette étude a en effet permis de remettre en question certains indicateurs du contrôle
sanitaire « classique » (utilisé en routine notamment par les Agences régionales de la
santé, ARS) pour les eaux de baignade. Une corrélation entre des facteurs abiotiques
environnementaux (données morphologiques de lacs, situation géographique,
paramètres météorologiques et hydro‐climatiques …) avec les paramètres de risque sur
la santé liées aux crises de cyanobactéries (densité cellulaire et durée de dépassement
du seuil 3 de l’OMS, récurrence interannuelle et concentration en microcystine MC)
(tableau 4) a permis de démontrer qu’il était possible de développer des modèles
simples reliant qualité sanitaire des eaux, biomasse et paramètres abiotiques. La densité
cellulaire, choisie actuellement comme indicateur du point de vue du contrôle sanitaire a
62
été démontrée comme n’étant ni une mesure de biomasse, ni un descripteur de la place
occupée par les cyanobactéries dans l’ensemble de la flore planctonique. Ainsi, le
dépassement du seuil de niveau 3 de l’OMS ne constitue pas un indice de prédominance
des cyanobactéries.
Tableau 4 : corrélation des facteurs abiotiques avec les paramètres de risques sanitaires pour les
épisodes de crise de cyanobactéries (+/‐ test de pearson avec p < 0.05) (Baurès et al., 2013b)
L’influence significative des paramètres morphologiques montre qu’il est possible de

développer des indices de risques utilisables pour catégoriser les milieux en fonction de
leur aptitude à héberger des biomasses de cyanobactéries limitant certains usages
(rapport intermédiaire, projet Oracle). Des modèles décrivant les variations de densité
cellulaire (65% de la variance expliquée) en fonction de critères morphologiques, et de
certains paramètres hydrométéorologiques des sites d’étude (26 lacs répartis sur toute
la Bretagne) ont été développés dans le cadre de stages d’élèves ingénieurs de l’EME au
LERES (travaux de Caroline Farenc et Aurélie Fuchez, 3ème année du cycle ingénieurs).
De nouveaux dispositifs de surveillance et d’alerte sanitaires peuvent donc être basés

sur l’étude de déterminants environnementaux.
63
3. Evaluations des risques environnementaux et sanitaires
3.1. Evaluation des risques dans le contexte « santé au travail »
Suite à mon doctorat, j’ai intégré le service de santé au travail de Nancy

(ALSMT), en plein contexte de la réforme de la santé au travail, instaurant la
pluridisciplinarité technique, médicale et organisationnelle dans les instances de la
médecine du travail. Ma mission d’ingénieur prévention était double. D’une part, je
réalisais des prestations pour les entreprises adhérentes de l’association (définitions des
objectifs en hygiène, sécurité et toxicologie issus d’une demande ou déterminés en
fonction des audits et de l’analyse d’activité). Cette mission m’a permis d’appréhender la
mise en place de méthodologies de travail spécifiques et adaptées aux risques
professionnels et l’accompagnement des entreprises dans l’élaboration de leur
document unique. Ces méthodologies ont été particulièrement développées quant au
risque chimique, j’ai d’ailleurs pu me former auprès de l’INRS sur la métrologie de
mesure des polluants dans les atmosphères de travail. D’autre part, j’assurais des
conseils auprès des médecins du travail pour la mise en place d’une démarche vis‐à‐vis
des agents chimiques dangereux (dont agents cancérigènes, mutagènes et reprotoxiques
(CMR) en particulier). J’ai ainsi pu mettre en place une base de données de produits
chimiques utilisés dans les entreprises adhérentes au service, permettre la rédaction de
fiches de données de sécurité simplifiées et de fiches aide mémoire sur les produits
toxiques. J’ai ainsi pu développer mes compétences en toxicologie industrielle,
métrologie atmosphérique (réalisation de campagnes de prélèvements atmosphériques
en fonction des risques professionnelles), organisationnelles (répartition des taches au
travers des différentes disciplines du Pôle (hygiène, ergonomie… et encadrement de 2
techniciennes m’ayant assisté dans les mesures métrologiques aux postes de travail) et
de terrain en entreprise. Avec le responsable du Pôle technique, j’ai élaboré la démarche
d’audit mise en place a été élaborée lors de cette mission :
- Réalisation d’un prédiagnostic « risque chimique » : il s’agissait d’apprécier l’état

des connaissances des produits chimiques utilisés, sachant que la demande des
employeurs et sa volonté de s’engager étaient fondamentales dans notre action
de prévention. Ce prédiagnostic consistait en la mise en place d’un
entretien/questionnaire semi‐directif sur l’inventaire des produits chimiques,
l’état de mise à jour des fiches de données de sécurité, la création des notices au
poste et les consignes existantes en terme de stockage et de signalétique
- Réalisation d’une analyse d’activité fine : pour connaitre l’exposition
professionnelle des salariés au risque chimique (durée d’exposition, fréquence,
nombre d’opérateurs concernés…). Parallèlement à ces actions, je réalisais une
veille législative et technologique en lien direct avec la demande (veille
toxicologique, classification des dangers, réglementation en vigueur…)
- Evaluation des risques professionnels : d’après la méthodologie de l’INRS, je
réalisais des évaluations quantitatives des risques permettant d’attribuer un
score en fonction des dangers et des expositions identifiées
- Rédaction d’un rapport d’intervention : la restitution aux entreprises mentionnait
les conclusions issus de l’évaluation des risques et proposait des actions de
prévention hiérarchisées avec notamment une proposition de métrologie
atmosphérique, des conseils d’amélioration des procédés, pour les équipements
64
de protection individuelle et concernant l’affichage en entreprise et une
sensibilisation auprès des salariés
Lors de mon intégration à l’EME en 2007, j’ai pu poursuivre cette thématique de

recherche « santé au travail » sur 2 projets réalisés en parallèle : le projet RECORD et
l’étude réalisée pour la DCNS.
Projet RECORD8 (Recherches coopérative sur les déchets et l’environnement) (octobre

2007‐ avril 2009)
RECORD est un réseau ouvert à toute organisation publique ou privée. Il permet

la réalisation de recherches dans le cadre d’une coopération tripartite tout à fait
originale entre industries, organismes publiques et chercheurs. Outre les programmes
réalisés, cette coopération fait de RECORD un lieu privilégié d’échanges ainsi qu’un outil
de veille technologique et scientifique.
Le projet s’est inscrit dans une démarche de surveillance du personnel exposé

aux agents cancérigènes, mutagènes et reprotoxiques (CMR) qui constitue un enjeu
majeur en médecine du travail, concernant pleinement les personnels de l’industrie du
déchet. Dans ce cadre, l’utilisation des tests de génotoxicité peut représenter une
démarche intéressante à considérer et notre étude a visé à établir un état de l’art
concernant l’usage de ces tests pour la surveillance du personnel de l’industrie du
déchet.
Après avoir présenté la nature du danger et de l’exposition dans l’industrie du

déchet, qui inclut l’exposition à de nombreux composés CMR, j’ai pu exposer le cadre
réglementaire actuel s’appliquant à la surveillance du personnel. J’ai ensuite procédé à
des interviews d’experts industriels travaillant dans le domaine du déchet pour avoir
leur retour d’expérience concernant le suivi du personnel ; les réponses obtenues ont
indiqué un intérêt réel des industriels pour la mise en place de tests de génotoxicité, en
particulier pour caractériser les risques à l’échelle du poste de travail et par famille de
déchets manipulés. L’équipe INSERM partenaire9 a rappelé le principe et les atouts et
limites des principaux tests de génotoxicité. Leur application dans le cadre global de la
médecine du travail, et dans celui plus spécifique de l’industrie du déchet, a ensuite été
abordée par le biais d’une analyse bibliographique de la littérature et aussi d’interviews
réalisées auprès de spécialistes scientifiques et médicaux des tests de génotoxicité.
Les données recueillies ont montré que ces tests de génotoxicité ne sont pas, ou
de façon marginale, utilisés aujourd’hui dans l’industrie du déchet. La faisabilité
8
Comité de suivi : Christiane Baleux, Solvay; Isabelle Conche, Teris; Bénédicte Couffgnal,
présidente RECORD; Olivier Delanau, Socotec; Isabelle Deportes, Ademe; Jean‐François Gonnet,
Total; François Hébard, Séché; Gérard Keck, Ecole vétérinaire Lyon; Lucie Lambolez, Véolia; Yves
le Corfec, EDF
9
Partenariat recherche : INSERM U 620/SeRAIC (Signalisation et Réponses aux Agents
Infectieux et Chimiques, Université Rennes 1) : Prs. Olivier Fardel et Laurent Vernhet
65
technique d’une telle démarche semble pourtant avérée, même si de nombreux
paramètres concernant notamment l’interprétation des résultats, les facteurs
confondants, le coût économique, l’absence de cadre réglementaire et l’aspect éthique
peuvent limiter sa portée et son intérêt. En conclusion, les tests de génotoxicité ont déjà
fait l’objet de nombreuses applications en médecine du travail et il pourrait donc être
judicieux de les utiliser spécifiquement dans l’industrie du déchet, mais sans doute
aujourd’hui uniquement dans un cadre exploratoire et de recherche en santé au travail.
Ce travail a donné lieu à :
 une journée technique (avril 2008) organisée à l’Ecole des Métiers de

l’Environnement
 un rapport intermédiaire (juin 2008) soumis au comité de suivi et présenté
oralement à Paris (ADEME) en présence de l’ensemble des membres.
 un rapport final (avril 2009), restitué sous forme écrite et orale (ADEME, Paris)
 une restitution orale lors d’une journée RECORD à Paris pour l’ensemble des
études réalisées sur les 2 dernières années le 1er juillet 2010 à Paris.
Projet DCNS (mars 2008‐juin 2009)
L’objectif du projet commandé par la DCNS de Brest était de substituer leurs

produits CMR toutes catégories confondues (avec une priorité aux 1ère et 2ème
catégories) en tenant compte des exigences techniques des produits. Le cadre général de
l’étude est celui de la prévention concernant le risque chimique (prévue dans code du
travail) pour toutes les entreprises utilisatrices de CMR, renforcée par le règlement
européen REACH n°1907/2006.
En partant d’une base de travail de 44 substances précédemment identifiées par

la DCNS, le plan d’étude s’est organisé en 3 phases :
 étude de la toxicité des substances : mise à jour des fiches de données de sécurité
fournisseurs, veille sur les catégories de CMR, état de commercialisation des
produits (Phase 1)
 visite sur site afin de connaître les principales conditions habituelles d’usage des
produits en parallèle de l’évaluation des risques effectuée par la DCNS (Phase 2)
 recherche via des bases de données bibliographiques sur l’état de l’art
concernant les possibilités de substitution des composés dangereux et contact
avec des fournisseurs commercialisant ou pouvant potentiellement
commercialiser des produits non CMR (Phase 3)
Lors de ce travail, la base effective de produits à substituer à la DCNS de Brest a

été revue suite aux évaluations des risques réalisées par l’équipe prévention santé
sécurité au travail et aux contacts fournisseurs. En octobre 2008, la base finale de travail
correspondant à notre étude de substitution comportait (hors réactifs de laboratoire) :
 9 produits CMR catégorie 1 et 2 correspondants à 14 applications

 26 produits CMR catégorie 3 correspondants à 33 applications
66
L’étude a finalement montré que :
‐ pour les CMR de 1ère et 2ème catégories, 4 produits n’étaient pas potentiellement
substituables actuellement, 5 produits ont donné lieu à des propositions de substitution.
‐ pour les « 3ème catégorie », 14 produits n’étaient pas substituables, 2 produits

ont donné lieu à des échanges avec des fournisseurs pour envisager des substitutions
éventuelles. Pour 9 produits, des substituants potentiels ont pu être trouvés.
Dans le rapport d’étude final, il a été donné mention à la DCNS de partager son
expérience en terme de substitution, une fois les différents produits proposés testés et
validés sur le site de l’ANSES (ex AFSSET : http://www.enjeux‐cmr.fr).
3.2. Elaboration d’outils d’aide à la décision en environnement‐santé

3.2.1. Dans le champ de la pollution maritime : empreinte environnementale
En 2007 s’est également initié un rapprochement entre l’EME et le Pôle de

compétitivité Mer Bretagne. J’ai été sollicité en tant que responsable scientifique pour
l’EME du projet labellisé « NACRE »10 (NAvire Conduit dans le Respect de
l’Environnement), réalisé de septembre 2007 à janvier 2013. Ce projet m’a permis
d’acquérir des compétences en pollution maritime, et plus particulièrement au niveau de
méthodes d’évaluation des impacts environnementaux de navires. En effet, les navires
en activité produisent un certain nombre de déchets, solides, liquides et gazeux qui,
jadis, étaient rejetés dans l’environnement : ordures, eaux « grises » des éviers, lavabos,
lave‐vaisselle et lave‐linge, eaux « noires » des toilettes, eaux de cales hydrocarburées ou
non, eaux de nettoyage, gaz d’échappement des moteurs, émissions de ventilation des
citernes et soutes. La préoccupation environnementale se traduit par des normes et
règlements de plus en plus contraignants portés notamment par l’Organisation Maritime
Internationale. « Naviguer propre » devient aujourd’hui, pour les professionnels de la
construction et de la réparation navale, civiles et militaires, une nécessité d’ordre
règlementaire. Pour les armateurs, disposer d’une flotte respectueuse de
l’environnement est une question d’image et constitue un atout commercial qui leur
permettra de faire la différence sur leurs marchés et de naviguer en toutes zones.
10
Partenaires :
- DCNS, porteur du projet, maintenance et construction navale, établissements de

Brest (29) et de Lorient (56),
- Bertin Technologies, TARNOS (40), département « énergie, process, environnement»,
- Louis Dreyfus Armateurs, Suresnes (92),
- GENAVIR, armateur de flotte scientifique, Plouzané (29),
- Veolia Eau – Recherche et Innovation, Maisons Lafitte
- IPL santé environnement durables Bretagne (Laboratoire CGI), Ploemeur (56),
laboratoire d’analyse des eaux et de l’environnement,
- La Marine nationale : Direction de Service de Soutien de la Flotte, Brest (29).
- ENSTA Bretagne Brest, département « conception mécanique appliquée », Brest
(29)
- ENMM, Ecole Nationale de la Marine Marchande, Nantes (44),
- IFREMER, service Développements Mécaniques, Brest (29).
67
Actuellement, les navires existants s’équipent de matériels de stockage et de traitement
des différents types de déchets. Mais l’addition de ces matériels, variable selon les types
de transport, ne permet pas de prendre en compte globalement, pour chaque navire, la
question de ses rejets, et se traduit souvent par la mise en place, dans des volumes
nécessairement contraints, d’équipements encombrants voire peu adaptés.
NACRE proposait une approche environnementale globale, tant en termes de

diagnostic que de réponses technologiques. Il s’agissait, dans un premier temps, de
mesurer l’empreinte environnementale globale de différents types de navires en activité
en prenant en compte l’ensemble de leurs déchets, mais aussi leurs conditions
spécifiques d’exploitation. Dans un deuxième temps, NACRE proposait des solutions
techniques économiquement viables, adaptées à la configuration de l’espace et aux
modes de fonctionnement des différents types de navires : encombrements réduit,
consommation énergétique faible, compatibilité aux mouvements de plate‐forme. Ces
équipements devraient permettre de rejeter des effluents dont la qualité sera meilleure
que celle imposée par les normes de rejets existantes, afin d’anticiper l’évolution
règlementaire.
Les objectifs à l’entrée de projet pour l’EME étaient principalement :
‐ de développer une expertise technique en matière d’étude des gains

environnementaux pour les procédés de traitement proposés par les
développeurs associés au projet,
‐ de proposer des procédés de traitement complémentaires (méthanisation)
permettant la gestion des effluents d’un bâtiment en milieu contraint par des
essais réalisés au sein du laboratoire de l’EME.
‐ de proposer une méthodologie permettant d’établir un bilan des gains

environnementaux des procédés retenus en prenant en compte les données
pilotes des partenaires développeurs
Lors d’un premier travail bibliographique encadré en 2007/200811, une

description détaillée des rejets actuels sur différents types de navires et de leurs impacts
a permis d’identifier les moyens à mettre en œuvre pour répondre aux évolutions
possibles mais aussi aux contraintes des navires.
11
Brizais J., Saint Martin C. (2008) Les polluants émis par les navires, modes de dispersion,
effets sur l’environnement et la santé, Projet bibliographique Ingénieur 1ère année EME, 46 pp.
68
Les travaux d’ingénierie encadrés au niveau de la dernière année du cycle ingénieur12
ont permis de compléter cette étude permettant:
 Une analyse de la réglementation en matière de rejet de polluants

 Une analyse des polluants susceptibles d’être émis par un navire (bibliographie)
et inventaire des émissions/extractions (Pétavy et al., 2009)
 Une étude comparative de méthodes d’évaluation d’impact environnemental
utilisées dans d’autres domaines que le maritime
 L’élaboration d’une méthode d’identification des impacts environnementaux des
différents polluants conduisant à une caractérisation et une hiérarchisation des
impacts environnementaux adaptée aux navires (figure 28).
Figure 28 : méthode générale d’évaluation des impacts environnementaux mise en place

dans le projet Nacre
Pour l’étude du bilan des gains environnementaux réels, cette partie du projet avait pour
objectif de déterminer les gains environnementaux liés à l’installation de ces procédés à
bord des navires. La problématique était la suivante : « Quelle méthodologie est la plus
adaptée pour définir le gain environnemental des procédés mis en place sur les
navires ? ».
12 Burini J., Couturier R., Jaouen B., Jaubert P., Le Goff G. (2009) Élaboration d’une méthode
d’étude d’impact appliquée aux rejets des navires, Pôle Mer Bretagne, projet NACRE – partie 1,
DCNS Brest, 138 pp.
Carré A., Dubois F., Mary F., Saint Martin C., Segura C. (2010) Evaluation des impacts
environnementaux des rejets liés aux navires, Pôle Mer Bretagne, projet NACRE – partie 1.2, DCNS
Brest, 82 pp.
Blin E., Guirriec L., Kerbrat M. (2011) Études des gains environnementaux liés à de
nouveaux procédés de traitement installés à bord des navires, Travail d’ingénierie 3ème année
EME, projet NACRE – partie 1.3, DCNS Brest, 114 pp.
Beauchet S., Blanchet S., Savy M. (2012) Méthodologie d’évaluation environnementale de
nouveaux procédés de traitement des rejets des navires, Travail d’ingénierie 3ème année EME,
Projet nacre, DCNS Brest, 70 pp.
69
Cette question a obtenu des réponses dans le cadre du dernier travail d’ingénierie. Le
travail encadré s’est ensuite tourné vers le renforcement de la démarche sélectionnée
par les partenaires et l’élaboration d’une méthodologie finale, rigoureuse et pertinente,
permettant d’évaluer le bilan environnemental d’un navire. Le principal enjeu était de
faire en sorte que celle‐ci soit simple d’utilisation et livrée avec un guide, puisqu’elle
devait être appliquée par l’ensemble des acteurs du projet NACRE. Ce dernier point a été
réalisé dans le dernier livrable technique qui est un tableur excel regroupant l’ensemble
des objectifs globaux, dépendant de divers critères (ou sous‐objectifs), dont l’évolution
est suivie par des indicateurs et enregistrée par des paramètres (voir figure 29 pour un
exemple de résultat de bilan environnemental du procédé mis en place pour le
traitement des effluents aqueux‐partie eaux noires). Un rapport d’exécution final a
également été rédigé pour le Pôle de compétitivité Valorial.
Figure 29 : Exemple de résultat pouvant être obtenu grâce à l’outil d’aide à la décision mis en
place selon la méthodologie LBAT (Local Best Available Technique). Cette dernière a été conçue
dans le but d’évaluer les performances de techniques et procédés industriels en vue de les
valider comme MTD (Meilleures Techniques Disponibles), dans le cadre de la réglementation
2008/1/CE, dite IPPC (Beauchet et al., 2012).
3.2.2 Dans le champ de l’eau potable : indices de risques sanitaires
Un 2ème outil d’aide à la décision portant sur les risques associés à la

potabilisation de l’eau pour les petites et moyennes unités de traitement et de
distribution d’eau (UDI, desservant une population inférieure à 10000 habitants) a été
élaboré dans le cadre du dernier volet du projet ERA‐Net.
Les UDI, majoritaires en Europe, sont particulièrement sensibles aux variations

brutales de la qualité de l’eau brute. En effet, l’Organisation Mondiale de la Santé (OMS)
et la Commission Européenne ont récemment fait le constat d’une incapacité à satisfaire
les exigences sanitaires en distribution en raison de leur moyens financiers et
techniques plus limités. De plus, le processus de révision de la directive eau potable,
initié en 2003 et qui a conclu sur un abandon de la révision des textes, a également
70
proposé des recommandations encourageant les petites et moyennes unités de
distribution à fournir un effort supplémentaire et préconisé une approche non
législative orientée sur l’évaluation des risques. Il existe ainsi un réel besoin d’outils
complémentaires à destination des petites unités de traitement afin de leur apporter
une aide à l’anticipation des aléas climatiques, et leur permettre ainsi au maximum
d’éviter les dépassements des limites de qualité en prenant en compte à la fois le risque
microbiologique et le risque chimique. C’est avec ces objectifs qu’en partenariat avec
deux organismes anglais13 et un organisme hollandais14 que la réalisation d’un système
d’aide à la décision destiné aux gestionnaires des petites et moyennes UDI a pu être crée
(Delpla et al., 2014).
L'architecture a été élaborée suivant une approche séquentielle basée sur :
 L'analyse des caractéristiques des bassins hydrographiques et des conditions

climatiques
 L'impact du traitement de l'eau (technique, opérationnelle, la surveillance, ..)
 L'évaluation des risques de la santé (l'outil est conçu pour évaluer les risques
sanitaires liés aux substances potentiellement présentes après le traitement de
l'eau. Les substances sont choisies parmi la liste des substances sur la base de
questionnaires réponses)
 Les recommandations pour les gestionnaires permettant de diminuer leurs
scores de vulnérabilité (proposition d’actions correctives portant sur : la gestion
du bassin versant, la protection du captage, la conception, la maintenance, le suivi
de la performance des installations, le suivi de la qualité de l’eau).
Le système permet ainsi de prédire les concentrations en COD pour différents types de
bassins versants. En tenant compte des environnements spécifiques des domaines
étudiés par les différentes équipes du consortium, deux types d'environnements
naturels (pour Royaume‐Uni) et anthropiques (pour la Bretagne) ont été distingués dans
le système d’aide à la décision. Le COD permet de valider les questionnaires d’analyse
des risques auxquels les gestionnaires des petites et moyennes unités de traitement
doivent répondre afin d’évaluer la vulnérabilité de leur système. Les scores calculés
correspondent à la vulnérabilité du système de production d’eau potable face aux
risques chimiques, microbiologiques et climatiques (figure 30).
13
Université de Bangor et Centre d’Ecologie et d’Hydrologie de Lancaster
14
Université d’Ultrecht
71
Figure 30 : Arborescence du système d’aide à la décision (Delpla et al., 2014)
72
Conclusion et Perspectives
Dans le cadre de ma recherche actuelle et dans les perspectives de travail, j’aborde la

qualité de l’eau comme un des déterminants environnementaux représentant un des
leviers d’action pour promouvoir la santé via des politiques publiques adaptées.
A partir de questions d’ordre sanitaire liées à la qualité de l’eau, les différentes étapes de
travail passent par la caractérisation des milieux et la recherche d’indicateurs pertinents
permettent l’identification des situations à risques et l’élaboration de systèmes d’aide à
la décision à la fois pour les parties prenantes environnementales (gestionnaires de
bassins versants, traiteurs d’eau…) et sanitaires (Agences régionales de santé, ANSES,
InVS …) (figure 31).
Figure 31 : synoptique du fil conducteur de la recherche menée en terme d’évaluation

environnementale et sanitaire
Comme le mentionne15 Philippe Quénel, directeur du LERES, la surveillance sanitaire est

désormais couplée à des activités de recueil de données, de vérification, d'analyse,
d'évaluation et d'interprétation d'événements sanitaires pouvant constituer une menace
et non plus uniquement de données standardisées et structurées sur lesquelles repose la
surveillance qui peuvent être faussés en terme de pertinence scientifique. Ces activités
de recherche sont donc également directement liées à celles de la veille prospective.
15
MOOC Santé Publique, EHESP 2015
73
L’objectif de cette veille est de discerner, au sein du champ existant des connaissances
scientifiques et technologiques nouvelles produites quotidiennement, des signaux
faibles d’événements significatifs et qui pourrait constituer un risque émergent. Cette
veille n'est pas définie par une réactivité immédiate de la surveillance ou de la vigilance,
ou dans la temporalité à moyen ou long terme de l'observation. Elle s'inscrit dans une
démarche analytique prospective des signaux, dans un but d'anticipation des futurs
scénarios de risques sanitaires.
En ce sens, les intérêts de ma recherche se trouvent principalement dans l'étude des

microstructures environnementales (de l’échelle du nanomètre à celle des particules de
plusieurs micromètres), afin de mieux comprendre le comportement des systèmes
aquatiques (eaux de surface) de la ressource jusqu’à l’usage, afin de prévoir les risques
sanitaires liées au processus de potabilisation (figure 32). A cet effet, la recherche est
basée sur l’étude des processus dynamiques liés à la chimie physique et colloïdale,
effectuée à la fois en laboratoire et in situ. Parce que la plupart des composés à l’état de
trace sont associés à des colloïdes naturels, y compris des matières inorganiques solides
(particules par exemple d'argile) et de biopolymères (par exemple, les polysaccharides,
les composés humiques), des études spécifiques liées à la caractérisation de ces
colloïdes, de leurs sources (naturelles et anthropiques), de leurs interactions, et de la
nature des structures environnementales qu'ils forment seraient à poursuivre.
Figure 32: Sources et devenirs des colloïdes bio‐minéro‐organiques dans les eaux de
surface, de la ressource à l’usage (adapté de Jung et al., 2014)
Au‐delà de la compréhension des mécanismes liés à la chimie des colloïdes, mes travaux
s’orientent vers la recherche et le développement de nouveaux indicateurs pour
progresser dans l’action (perspectives 1 et 2 suivantes). D’après nos précédents
74
travaux , idéalement, un bon indicateur doit être biensûr validé scientifiquement, mais
au‐delà de cela, adapté à l’échelle spatiale, sensible aux changements attendus, fondé sur
des données fiables et facilement accessibles et enfin compris et partagé par l’ensemble
des acteurs de l’environnement et de la santé, les prescripteurs et les décideurs.
Enfin, il serait passionnant d’investiguer des champs de compétences transverses

(santé‐environnement‐travail) afin de croiser les approches et les méthodes
d’évaluation des risques encore trop considérées comme éloignées les unes des autres
(perspective 3).
Perspective 1 : poursuite des recherches sur les indicateurs de risque pour les
zones sensibles (relation C/N, discrimination de la matière organique naturelle de
la matière organique anthropique)
La surveillance des milieux repose actuellement sur des méthodes et des outils
permettant le diagnostic des effets des pressions anthropiques. En effet, que les impacts
soit directs (comme les rejets urbains et industriels ou les flux de nutriments en milieu
agricole) ou indirects (comme lors des événements pluvieux intenses), la cartographie
de ces pressions permet de mieux appréhender la variabilité spatio‐temporelle des flux
et leurs effets environnementaux et sanitaires. La détermination d’indices de risque
pourra contribuer à l’amélioration des pratiques environnementales, à confirmer par
exemple l’intérêt des zones humides naturelles et enfin à améliorer les types de
traitements utilisés par les stations. Un effort particulier de transfert des connaissances
au niveau des acteurs environnementaux (par exemple les coopératives agricoles mais
également les syndicats de bassins versants, les agences de l’eau dans le cas des
poursuites de la valorisation du travail de thèse de J. Causse) sera donné afin de traduire
les résultats du travail en terme de recommandations pratiques pour les principaux
intéressés.
Un des résultats majeurs de l’étude sur les Algues vertes (thèse J. Causse) montre
l’importance du carbone assimilable dans les milieux, expliquant le potentiel dénitrifiant
de certaines zones et/ou lors de certains événements climatiques. Ce nutriment étant
par ailleurs directement corrélé aux formes azotées comme nous l’avons vu
précédemment, on comprend l’importance de pouvoir caractériser les différentes
formes de la matière organique. L’indicateur « COD » serait donc un indicateur d’état
décrivant l’évolution des caractéristiques des milieux récepteurs en relation avec les
transferts de substances (thèse de I. Delpla), mais qui nécessite d’être explicité en terme
d’origine et de biodisponibilité (caractérisation de la fraction assimilable) afin de
traduire d’avantage l’indication de la pression exercée sur le milieu. L’étude de la
relation C/N permet d’aborder une question environnementale majeure : les relations
entre matières organiques et activité microbienne dénitrifiante dans les cours d’eau.
Lors des thèses de I. Delpla et J. Causse, nous avons clairement perçu que cette relation
n’est pas simplement due à un dénominateur commun d’ordre hydrologique, mais
correspond bien à un lien de cause à effet constitué par l’accentuation de la
dénitrification par l’apport de carbone assimilable par les communautés bactériennes.
Pour les résultats obtenus, la relation C/N a été trouvée très largement linéaire sauf
pour les épisodes de crues. On peut donc se questionner sur l’origine de cette non
linéarité : différence de comportement hydro‐morphologique de la réponse des bassins
versants lors des crues ? différence dans les pratiques environnementales au niveau des
75
parcelles ? nature du carbone différente (partition variable entre le carbone naturel et
anthropique) ? La nature exacte des mécanismes par lesquels le COD exporté du sol vers
les rivières se régénère au niveau du sol et la vitesse à laquelle cette régénération
s'opère font également débat. De fait, le réservoir de COD mobilisable par les
écoulements à un instant t semble être fini, et apparait considérablement plus petit que
le réservoir organique total du sol. Ceci est déduit du fait que les concentrations dans les
rivières diminuent généralement au cours d'une année hydrologique, pour réaugmenter
au début de l'année hydrologique suivante, les concentrations les plus fortes étant
généralement observées lors des premières crues d'automne. La répétition de ces deux
caractéristiques (épuisement apparent du réservoir COD en hiver et au printemps;
reconstitution apparente du réservoir en été) pose la question des mécanismes par
lesquels le COD se régénère, et la question de la vitesse de cette régénération et des
facteurs la favorisant avec, concernant ce dernier point la question clé du rôle des
facteurs hydro‐climatiques. Là encore, les hypothèses sont multiples et non tranchées.
Les principales conclusions scientifiques connues incluent: i) un effet de dessiccation du
sol en été rendant accessible à la dissolution des molécules organiques qui ne l'était pas
antérieurement; ii) la dégradation d'une partie de la litière de l'année; iii) la
solubilisation de débris de bactéries aérobies stimulées dans leur développement par la
sécheresse du sol et les températures élevées de l'été. Une autre question scientifique en
suspens porte sur le rôle de la production autochtone dans les petits cours d'eau.
Certains auteurs n’ont pu expliquer l’augmentation des concentrations en COD observée
en été dans certains petits cours d'eau autrement qu'en faisant appel à une source
autochtone de COD. Comprendre l'origine et la signification des variations saisonnières
des concentrations en COD dans les petits cours d'eau sont donc des questions
scientifiques d’importance que nous pourrions développer. La pertinence de l’indice UV
Ad270 serait à confirmer, ainsi que le développement d’autres indices permettant de
discriminer la part de matière organique naturelle de la part autochtone. Les études
réalisées pendant le doctorat ont également permis de montrer qu’avec des rapports
C/N différents, la matière organique de type « aquagénique autochtone » réagissait
différemment de la matière organique de type « pédogénique » en terme de signal de
fluorescence et de comportement avec des polluants du milieu. Cette technique pourrait
également être utilisée pour réaliser des cartographies de typologie de matières
organiques dans les phénomènes de transfert berges/rivières.
De plus, pour certains événements de type crues, la resuspension des sédiments de

rivière contrôle majoritairement les concentrations en nutriments. Une compréhension
fine de la cohésion dynamique des agrégats est la clé de l’explication ces transferts et
également de la turbidité, paramètre clé de la qualité de l’eau, notamment en terme de
risque sanitaire. Dans le transport particulaire, la matière organique a là aussi un rôle
fondamental. En effet, elle conserve la capacité de s’adsorber sur les surfaces minérales
des sédiments et des colloïdes minéraux. Dans le premier cas, elle augmente la capacité
d’adsorption de la phase immobile (particules, sédiments) en augmentant la teneur en
carbone organique. Dans le second, elle stabilise les agrégats colloïdaux. Des
perspectives de travail intéressantes pourraient porter sur un approfondissement de la
chimie des colloïdes, dans la poursuite de nos travaux déjà réalisés sur la formation, la
stabilité, et les modes d’association des particules. L’utilisation de particules
fonctionnalisées pour explorer les modes d’association de la MON ou l’évolution de sa
nature au cours du temps dans les systèmes aquatiques, pourrait être une idée à creuser.
En particulier, l’ajout de différents types de nanocolloïdes fluorescents de type « sondes
76
de fluorescence » à une eau de surface pourrait, après fixation, permettre de suivre les
abondances relatives des fonctionnalités chimiques par cytométrie de flux par exemple.
Lors des phénomènes de transfert, l’origine des particules serait donc en partie
explicitée. La spectroscopie de corrélation de fluorescence (FCS) et l'anisotropie de
fluorescence résolue en temps (TFA) sont également des techniques très utilisées
actuellement dans le domaine de la bio‐physique pour comprendre les dynamiques
d’assemblages moléculaires (étude des diffusions d’espèces et estimation quantitative
des interactions). Ces techniques permettraient de mieux caractériser le complexe bio‐
organique‐minéral. Quantifier les échanges de colloïdes individuels entre agrégats au
moyen de particules marquées, identifier le lieu de ces échanges entre agrégats serait
envisageable par microcoscopie confocale avec des particules fluorescentes. La
caractérisation des agrégats par microscopie électronique à transmission couplée à la
perte d’énergie permet quant à elle d’avoir des informations à l’échelle nanométrique
sur la composition élémentaire des agrégats, et notamment l’accès au rapport C/N au
niveau d’une particule. Ces 2 approches, locales et macroscopiques pourraient donc se
compléter de façon judicieuse. Une fois encore, cette thématique est dans le champ de la
santé environnementale, puisque l’identification des sources des nutriments
particulaires et l’explication de la turbidité touchent donc à la fois les problématiques
environnementales (lien avec l’aménagement du territoire et génie écologique) et
sanitaires (les particules sont des vecteurs de transfert des micropolluants et de
contamination microbienne ‐ voir perspective 2 ‐ ou dans le cas de certains nutriments,
comme par exemple le phosphore, peuvent conduire à des événements à risque sanitaire
majeur type bloom de cyanobactéries par exemple).
Perspective 2 : développement d’indicateurs de contamination bactérienne et

virale dans les eaux superficielles
La pollution microbienne dans les milieux aquatiques est l'une des questions cruciales à
l'égard de l'état sanitaire des masses d'eau utilisées pour l'approvisionnement en eau
potable, les activités récréatives et la récolte des fruits de mer en raison d'une
contamination potentielle par des bactéries pathogènes, protozoaires ou des virus. Pour
contrer ce risque, la surveillance de la contamination microbienne est généralement
évaluée par des mesures de turbidité effectuées à usines d'eau potable. Cités dans une
revue bibliographique récente (Jung et al., 2014), des recherches ont montré des
corrélations significatives de la contamination microbienne avec le risque de
gastroentérite endémique au travers de cet indicateur « turbidité ». Cependant, la
pertinence de la turbidimétrie peut être limitée car la présence de colloïdes dans l'eau
crée des interférences avec la réponse néphélométrique. D’autre part en France,
concernant le risque microbiologique en lien avec l’occurence de maladies hydriques, la
surveillance se fait essentiellement par la détection d’E.Coli, indicateur de contamination
fécale ne présageant en rien le risque de contamination virale. Ainsi, il y a un besoin de
développer un indicateur plus pertinent, simple et rapide pour la détection de la
contamination bactérienne et virale dans l'eau, en particulier dans la perspective du
changement climatique avec l'augmentation des épisodes de fortes précipitations. Tous
ces besoins de recherche doivent être considérés à leur juste valeur en raison de
l'importance de la gestion des coûts des maladies d'origine hydrique (telles que la
gastro‐entérite aiguë par exemple en France), en plus de l'impact sur la santé de la
77
population. La réduction des coûts, par une meilleure anticipation de la dégradation de
la qualité sanitaire des ressources en eau naturelle, est donc une question économique
importante. Pour le point de vue scientifique, la question principale serait de compléter
l'indicateur d'exposition principale (la mesure de la turbidité) avec des paramètres plus
spécifiques complémentaires, tenant compte des conditions hydrologiques. Ce projet
traduit aussi un réel besoin d'une approche multidisciplinaire entre les hydrologues,
chimistes, microbiologistes, gestionnaires de traitement de l'eau et des épidémiologistes.
La spectroscopie UV/Visible a beaucoup d'avantages pour étudier les interactions entre

les matrices naturelles et de polluants dans les compartiments dissoutes et colloïdales.
Le premier avantage de cette technique est d'être rapide, directe et de peu coûteuse. La
technique pourrait également donner des informations sur la taille de composés en
couplant spectroscopie UV/Visible à l'analyse de granulométrie laser. Dans la
littérature, certains travaux ont proposé une typologie des réponses UV de particules en
fonction de leur taille et un spectre de modèle de colloïdes minéraux, afin d'obtenir une
meilleure compréhension des phénomènes physiques d'absorption (voir revue de la
littérature Jung et al., 2014). Ils ont montré que les réponses UV de particules non
absorbantes peuvent être modélisées. Grâce à sa bonne sensibilité, cette technique
pourrait être susceptible de fournir des informations semi‐quantitatives sur les
interactions entre les phases par une surveillance directe, une préparation pré‐
analytique non‐spécifique et aux concentrations de milieux naturels.
Les résultats de ces études pourraient donner des arguments aux gestionnaires de
traitement d'eau pour modifier ou adapter les procédés de traitement de l'eau, ou d'une
manière différente de gérer leurs usines en prenant en compte une procédure d'alerte
proposée pour les situations dangereuses, sur la base de nouveaux indicateurs de la
contamination microbiologique des eaux de surface en entrée de station.
Cette proposition avait l’objet d’une demande à l’ANR (projet ISIS, 2013) qui avait été
retenue scientifiquement mais non financée. Ce même type de problématique pourrait
également être développée par rapport aux micropolluants, puisque pour ces
contaminants, la fraction colloïdale est également considérée comme un des principaux
vecteurs (ANR jeunes chercheurs, projet TURB’EAU, 2011 non financé).
Perspective 3 : recherche transversale « santé‐environnement‐travail »
Un 3ème champ de recherche serait celui de la santé environnementale, dans le cadre de

l’étude d’un environnement particulier : celui du poste de travail. Au niveau de cette
thématique, en cours de développement au sein de l’équipe « santé au travail » de
l’IRSET (Pr Roquelaure), des compétences au niveau de méthodes d’évaluation des
risques chimiques pourraient être apportées. L’approche développée par cette équipe
est celle de « l’ergo‐exposologie ». Cette approche est complémentaire à mes yeux de
celle actuellement utilisée au LERES pour la caractérisation de la « pollution » au sens
large. En effet, l’approche du LERES a une entrée « polluant moléculaire » alors que
l’ergo‐exposologie voit une entrée « activité » pour évaluer le risque d’exposition (Figure
33).
78
Figure 33 : Complémentarité des approches toxicologiques et ergonomiques pour l’évaluation
des risques chimiques au poste de travail (adapté d’Alain Garrigou16)
La notion d’exposition est une notion centrale en santé publique, mais polysémique.
L’ergo‐toxicologie est une méthode s’intéressant en parallèle aux composantes
techniques, humaines et organisationnelles. Il s’agit de développer des modèles opérants
pour décrire les situations d’exposition à des produits chimiques et d’identifier les
déterminants pour prévenir les expositions.
Un des objectifs de l’équipe « santé‐travail » est de faire entrer la prévention dans les
prises de décisions des directions d’entreprise ou d’organismes sur le plan
organisationnel, et dans le rôle du management. Pour ma part, le développement des
connaissances des situations de travail réelles, afin d'agir le plus en amont possible, peut
permettre d'outiller les entreprises au plan méthodologique, en s’inspirant des
démarches innovantes en cours, en produisant par exemple des « kits », des bases de
cas, des repères méthodologiques par secteur d'activité.
Je pourrai potentiellement accompagner les entreprises dans l'utilisation d'outils de

prévention au service de la caractérisation des expositions et de la réduction des
risques, notamment au niveau des méthodes d'évaluation des risques chimiques:
évaluation des émissions / des expositions, identification des sources : diagnostic in situ
(analyse en temps réel, analyseurs portables, méthodes de visualisation innovantes).
Comme le prévoit la réglementation REACH/CLP, les entreprises doivent faire remonter
aux fabricants de substances toutes les informations nécessaires permettant
l'élaboration des scénarios d'exposition, indexés notamment au niveau des fiches de
données de sécurité des produits, cette partie pourrait donc être également utile en ce
sens. Un autre objectif dans ce travail de la santé au travail est celui de la caractérisation
des écarts entre la théorie et la pratique managériale. Il pourrait être particulièrement
intéressant pour moi de qualifier l'écart réglementaire, et les pratiques du terrain au
niveau de la substitution des produits les plus dangereux par ailleurs.
La prévention de l’exposition aux produits nano‐technologiques, l'étude de la
16
Alain Garrigou (ISPED, INSERM U887, Laboratoire Santé Travail Environnement),
Environnement : expositions et santé des populations, séminaire EHESP‐IRSET, 10 mars
2015
79
contamination des surfaces sont également d'autres pistes de travail. La microscopie
électronique utilisée lors de ma thèse, pourrait par exemple servir en tant qu’outil à ce
niveau aussi.
Les techniques de métrologie utilisées en mesures environnementales sont

sensiblement les mêmes que celles utilisées en santé au travail, la différence venant
essentiellement au niveau de la cible d’exposition (objectifs différents en terme de seuils
réglementaires d’exposition). Mon expérience passée en tant qu’ingénieur prévention
pourrait également être utile pour faire le lien transversal entre la métrologie
environnementale actuellement déployée au LERES en mesures d’air intérieur et la
métrologie sanitaire utilisée au poste de travail.
La prévention est d’autant plus efficace qu’elle se situe en amont, comprendre et agir sur
le comportement humain et organisationnel permet à la fois de protéger la santé de
l’opérateur mais aussi celle de l’environnement et donc des populations. Il s’agit d’une
mission majeure de santé publique. Un travail a été initié en ce sens par l’intermédiaire
d’une étude d’ingénierie réalisée par des étudiants de l’EME avec le CHU de Rennes, sur
la problématique de la recherche de substances dangereuses dans les effluents
(réglementation RSDE2). Malgré les études déjà réalisés au CHU concernant la typologie
et l’utilisation des produits chimiques, un certain nombre de questions restent en
suspens quant à l’identification des sources précises des polluants et par conséquent de
la stratégie associée en terme de métrologie. En prenant en compte l’ensemble des
résultats de campagnes de prélèvements et des travaux antérieurs de recensements de
produits déjà réalisés (notamment recueil de fiches de données de sécurité), le groupe
de travail fait une synthèse et un recentrage des substances RSDE mises en avant. Cette
étude est faite sur la base d’un travail bibliographique concernant les sources de
polluants potentiels et les sous‐produits de dégradation/transformation notamment dus
aux effets « cocktails » (issus des mélanges de produits dont en particulier les
désinfectants). L’objectif est de produire un plan d’actions hiérarchisées afin de préparer
les campagnes de prélèvements à venir dans des zones stratégiques à définir, pour des
substances ciblées. Cette étude souhaite aller au‐delà d’un simple recueil de données
techniques et d’exploitation de connaissances scientifiques. En effet, les perspectives
seraient de confronter cette recherche bibliographique avec l’analyse des activités
réelles sur le terrain pour des services ciblés (repérage de produits n’apparaissant pas
forcément dans l’inventaire établi, analyse des pratiques, identification d’autres sources
de pollution (ex: lingettes, rejets atmosphériques…)) et définir le cas échéant des études
techniques complémentaires. Afin d’établir les objectifs fixés par la DREAL, l’étude
pourrait aussi se poursuivre afin de définir des pistes d’amélioration pour certaines
pratiques (selon les meilleures techniques disponibles) ou des propositions de
substitution de certains produits si existantes (étude technico‐économique à réaliser).
Ce projet pourrait donc faire le lien entre une problématique environnementale, mais
également sanitaire au travail (évaluation des risques au niveau des pratiques) et en
santé publique (risque lié aux techniques de traitement des effluents).
Ce sous axe de recherche s’intéresserait donc à l’étude de la toxicité des contaminants,

de leurs métabolites et produits de transformation, sur les écosystèmes et la santé des
populations humaines et sur la relation travail santé. Une approche multidisciplinaire
serait attendue pour : 1) étudier la notion d’exposome, les interactions entre différents
contaminants et leurs effets cumulatifs éventuels ; 2) caractériser les risques émergents
80
et établir des systèmes adéquats de surveillance, et analyser les conditions sociales
d’évaluation de ces risques, de débats et de prise de décision (les préoccupations de
« santé environnementale » s’expriment à travers le filtre de facteurs socio‐
économiques, socio‐techniques et socio‐éducatifs encore insuffisamment analysés et
pris en compte) 3) améliorer les capacités prédictives grâce à des approches
systémiques ; 4) comprendre les facteurs environnementaux, économiques et sociaux
qui déterminent ou modulent les expositions et les vulnérabilités des populations
humaines, la mobilisation des acteurs sociaux ; 5) développer des approches de
remédiation.
81
82
ANNEXES
Sélection de publications
83
ARTICLE IN PRESS
Water Research 39 (2005) 3849–3862

www.elsevier.com/locate/watres
Coagulation of humic substances and dissolved organic matter

with a ferric salt: An electron energy loss spectroscopy
investigation
A.-V. Junga,, V. Chanudeta, J. Ghanbajab, B.S. Lartigesa, J.-L. Bersillona
a
Laboratoire Environnement et Minéralurgie (LEM), UMR 7569 CNRS-INPL-ENSG, 15 avenue du Charmois, BP 40,
54501 Vandæuvre, France
b
Service commnun d’analyses (MET), Université Nancy I, BP 239, 54 506 Vandæuvre, France
Received 5 February 2005; received in revised form 7 July 2005; accepted 11 July 2005
Available online 19 August 2005
Abstract
Transmission electron microscopy (TEM) coupled with electron energy loss spectroscopy (EELS) and energy
dispersive X-ray spectroscopy (EDXS) was used to investigate the coagulation of natural organic matter with a ferric
salt. Jar-test experiments were first conducted with a reconstituted water containing either synthetic or natural extracts
of humic substances, and then with a raw water from Moselle River (France). The characterization of the freeze-dried
coagulated sediment by EELS in the 250–450 eV range, showed that Fe-coagulant species predominantly associate with
the carboxylic groups of organic matter, and that this interaction is accompanied by a release of previously complexed
calcium ions. The variation of Fe/C elemental ratio with iron concentration provides insightful information into the
coagulation mechanism of humic substances. At acid pH, Fe/C remains close to 3 over the whole range of iron
concentrations investigated, while a much lower atomic ratio is expected from the value of optimal coagulant dosage.
This suggests that a charge neutralization/complexation mechanism is responsible for the removal of humic colloids, the
aggregates being formed with both iron-coagulated and proton-neutralized organic compounds. At pH 8, the decrease
in Fe/C around optimal coagulant concentration is interpreted as a bridging of stretched humic macromolecules by Fe-
hydrolyzed species. Aggregation would then result from a competition between reconformation of humic chains around
coagulant species and collision of destabilized humic material. EELS also enabled a fingerpriting of natural organic
substances contained in the iron-coagulated surface water, N/C elemental analyses revealing that humic colloids are
removed prior to proteinic compounds.
r 2005 Elsevier Ltd. All rights reserved.
Keywords: Coagulation; Humic substances; Natural organic matter; Electron energy loss spectroscopy (EELS)
1. Introduction
Dissolved organic matter (DOM) represents the

Corresponding author. Private domicile, 84 chemin du heterogeneous mixture of organic colloids (excluding
Klettenberg, 68350 Brunstatt, Brunstatt, France. living organisms) and truly dissolved organic molecules
Tel.: +0033 6 32464105; fax: +0033 3 89647787. present in aquatic systems. It comprises biochemically
E-mail address: av1jung@hotmail.com (A.-V. Jung). defined compounds such as proteins and carbohydrates,
0043-1354/$ - see front matter r 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.watres.2005.07.008
ARTICLE IN PRESS
3850 A.-V. Jung et al. / Water Research 39 (2005) 3849–3862
operationally defined substances such as fulvic and chemical bonds within polymeric systems (Varlot et al.,
humic acids, and any supramolecular assembly of those 1998), and can then be used to fingerprint the various
molecules and macromolecules (Leenheer and Croué, organic macromolecules contained in coagulated organ-
2003). DOM has largely been studied in recent years, as ic matter. In environmental science, EELS was first used
it may play a major role in the mobility and bio- by biochemists to describe in situ natural materials
availability of anthropic contaminants in the environ- degradation with C/N ratio measurements (Villemin et
ment (Thurman and Malcolm, 1981; Tipping, 1993). In al., 1995; Watteau et al., 1996). It has also been used to
drinking water treatment, the main motivation for investigate the association of organic matter with clay
studying DOM stems from its role as precursor of particles (Furukawa, 2000).
carcinogenic chlorination by-products during the disin- In this paper, we present the results obtained by C-
fection stage (Oliver and Visser, 1980). However, most EELS. Fe-EELS results will be reported in a companion
problems encountered in the treatment process can be paper (Jung et al., 2005a). To avoid problems associated
related to the presence of DOM, including reduced with changes in the quality of the river water, experi-
adsorption capacity in activated carbon beds, taste and ments were first carried out with a reconstituted river
odour of the finished water, and biological regrowth in water containing either humic-like synthetic compounds
the distribution networks (Owen et al., 1995). or natural aquagenic humic extracts. In this way, the
The removal of DOM from water supplies is carried influence of pH on humics removal could be investi-
out in the first place by coagulation-flocculation with a gated. The method was then assessed on raw Moselle
hydrolyzable metal salt such as aluminum or iron river water.
(Dempsey et al., 1984; Edwards and Amirtharajah,
1985; Gregor et al., 1997). When added to water,
aluminum and iron coagulants yield hydrolysis products 2. Experimental section
that destabilize the organic material present in the raw
water. The destabilization mechanisms most often 2.1. Origin of natural organic matter (NOM) and
quoted in the literature include charge neutralization, extraction of humic substances
enmeshment, and adsorption (Dempsey et al., 1984).
Charge neutralization preferentially applies at acidic Three sources of organic matter were selected for this
pH, while sweep-flocculation and adsorption are com- study. A synthetic humic-like substance (SHS) was
monly referred to in pH-coagulant concentration prepared by polymerization of catechol with glycine
domains where abundant precipitation of metal hydro- (Andreux et al., 1980; Jung et al., 2005b). Catechol
xide is expected (Edwards and Amirtharajah, 1985). (6.6 g) was dissolved in 500 mL of a solution of sodium
However, DOM-specific characteristics, such as its phosphate buffer NaH2PO4/Na2HPO4 at pH 8 to obtain
molecular weight distribution, carboxylic acidity, and a 0.03 M catechol solution. This was then mixed with an
humic substance content, have been shown to signifi- equal volume of buffered glycine 0.03 M solution. The
cantly influence the efficiency of DOM removal (Collins resulting mixture was then allowed to react in a 2 L flask
et al., 1985; Bose and Reckhow, 1998; Matilainen et al., in the dark at 25 1C and under a pure oxygen constant
2002). Recent extended X-ray absorption fine spectro- pressure (0.9 bar) for 5 days. After dialysis, (Spectra/
scopy (EXAFS) at the Fe K-edge experiments also Pors 7 membrane- 8 kDa molecular weight cut-off—
questioned the coagulation mechanisms usually invoked Roth-Sochiel Lauterbourg, France), the synthetic mate-
in the literature, as poorly polymerized Fe species were rial in a sodium form was freeze-dried and stored at 4 1C
evidenced in a wide range of operational conditions in the dark. Extensive characterization of SHS and
(Vilgé-Ritter et al., 1999a, b; El Samrani et al., 2004). relevance to natural humic substances have been
As an optimized DOM coagulation is a key to better discussed in a previous work (Jung et al., 2005b).
drinking water treatment, the interaction behavior Elemental analysis gave a carbon to nitrogen ratio (C/N)
between organic compounds and coagulant species of 18.3 (%C w/w ¼ 40.7). Potentiometric titration
should be better understood. The approach taken in showed three acidities at pKA ¼ 4.6 (2.3 meq/g), 5.1
this work was to investigate, within flocs, both the (0.8 meq/g) and 7.2 (0.6 meq/g).
nature of DOM and the speciation of an iron-based A natural aquagenic humic substance, hereafter
coagulant by using electron energy loss spectroscopy referred to as NHS, was extracted from the Moselle
(EELS) combined with energy dispersive X-ray spectro- river suspended matter. The sampling site was located at
scopy (EDXS). EELS provides a quantitative analysis at Richardménil (France) in the vicinity of the pumping
the nanometer scale of light elements (carbon, nitrogen, facility supplying Grand Nancy Urban Community
oxygen) through the K-shell excitation, and of other drinking water treatment plant in surface water.
important elements in water treatment such as iron, Approximately, 12 m3 of river water were pumped
calcium, phosphorus, and sulfur, through the L-shell (1 m3/h) into a flow-through centrifuge operating at
excitation. It also enables the identification of organic 17,000 rpm. The humic substances contained in the
ARTICLE IN PRESS
A.-V. Jung et al. / Water Research 39 (2005) 3849–3862 3851
suspended material were then extracted following the The spectrometer entrance aperture, the half-angle of
procedure of Thurman and Malcolm (1981), adapted to collection and the dispersion were 2 mm, 10 mrad, and
yield the natural extract in a sodium form. NHS was 0.2 eV/channel, respectively. The energy resolution,
also purified by dialysis (Spectra/Pors 7 membrane— determined at the full-width at half-maximum of the
8 kDa molecular weight cut-off) and freeze-dried. zero-loss peak is 1.4 eV. The time acquisition was 2.5 s
Total exchangeable protons and C/N ratios determined for the low energy region and 8 s for the C K-edge
for NHS were 3.0 meq/g and 9.2 (%C w/w ¼ 44.2), region. The relative thickness (ratio of sample thickness
respectively. to the free electronic inelastic diffusion pathway), was
Grab samples of the Moselle river water (9.3 mg/L kept about 0.570.2 for all measurements. The electron
total organic carbon content) were also taken at the dose, calculated from the Egerton relationship (Egerton,
sampling site, thus providing a raw water with 1996), was about 101.9 C/m2 which is less than the
unfractionated NOM. critical value-inducing beam damage within organic
polymers (Varlot et al., 1998).
2.2. Coagulation procedure For each sample, 15 EEL spectra were recorded. All
data processing was carried out with GATAN EL/P 2.1
Synthetic waters were prepared by dissolving 30 mg of software. The spectra were first corrected for dark
humic material (either SHS or NHS) in 1 L of Milli-Qs current, gain variations, and double excitation by
ultra-pure water. 336 mg of NaHCO3 (4 103 mol/L) applying a Fourier-ratio deconvolution with weak loss
(Sigma-Aldrich—analytical grade) was also added to the energy spectra. Background fitting was performed with a
suspension to provide a carbonate alkalinity similar to power law in order to extract the core-excitation peaks
that of natural river waters. Prior to coagulant injection, for each element. The amounts of C, Ca, and N were
the pH of synthetic waters was adjusted to pH 6 or 8 then obtained by integrating the surface area of a 50 eV
with dropwise addition of 0.1 M HCl. The surface water energy window after having defined the lower-limit in
was used as collected (natural pH ¼ 7.4). Jar-tests were energy for each element, that is 283.8, 346.4 and
performed in 1 L baffled reactors of known power 401.6 eV for carbon (C K-), calcium (Ca L2,3-) and
dissipation characteristics (Lartiges et al., 1997). Suspen- nitrogen (N K-) edges, respectively. Integration results
sion was stirred at 100 rpm which corresponds to a mean are given by the software in atomic concentration/nm2.
velocity gradient of 135 s1. The coagulant, a 0.1 M Hence, Ca/C and N/C atomic ratios were directly
ferric nitrate solution prepared from crystallized calculated from EELS measurements, whereas Na/C
Fe(NO3)3, 9 H2O (95% pure—Sigma-Aldrich), was and Fe/C ratios were inferred from the amounts in Na,
injected under agitation using a 5000 mL micro-pipette Ca, and Fe, provided by EDXS microanalyses. When
(Eppendorf, Hamburg, Germany), and mixing was Ca was barely detectable on EEL spectra, which is the
continued for 20 min. The suspension pH was not re- case for most coagulation experiments carried out at pH
adjusted after coagulant addition. After 24 h of quies- 6, Fe/C and Na/C ratios obtained from EDXS spectra
cent settling in graduated Imhoff cones, the residual were corrected by a proportional constant determined
turbidity of the supernatant was measured on a HACH from EELS and EDXS results at pH 8.
Ratio XR turbidimeter. The settled volume was
recorded, and the sediment was then collected and
freeze-dried (48 h—EL105 Sentry) to perform TEM 3. Results
observations, EDXS and EELS analysis.
3.1. Jar-test results
2.3. Transmission electron microscopy (TEM)
Fig. 1 presents the results obtained from jar-tests
Electron microscopy observations were performed carried out at pH 6 and 8 with SHS and NHS. All
with a CM20 Philips transmission electronic microscope turbidity and sediment volume curves display a classical
(TEM) using a 200 kV accelerating voltage. The electron behavior: the residual turbidity steadily increases at low
microscope is coupled with an EDAX EDXS, and with a coagulant concentration to reach a maximum, and then
GATAN 666 EELS. The freeze-dried sediment was re- steeply decreases upon further addition of ferric
suspended in ethanol (99.5% spectroscopy grade) under coagulant. The point where the extrapolated steep
ultrasonication, and a drop of suspension was evapo- portion of the turbidity curve intersects the x-axis is
rated on a carbon network-like holey support film selected as the optimum coagulant concentration
placed on a 200 mesh copper grid (Euromedex— (OCC), and is indicated with an arrow on the graphs.
Mundolsheim, France). EDX spectra were recorded At higher iron concentrations, the residual turbidity
with a 10 nm spot diameter and a 30 s counting time. remains low and then strongly increases as the
EELS spectra were collected in the 250–450 eV region restabilization of the suspension is induced. The inverse
with the same spot size (analyzed area of about 80 nm2). pattern is observed for sediment volume: at low iron
ARTICLE IN PRESS
60 18
(e)
SHS pH 8 (d) 16 SHS pH 6
50
(c) 14
(f) (h)
40 (g)
(b) 12
10
30
8 (i)
20 6
(j)
4
10 (a)
2
0 0
0 2 10-4 4 10 -4
6 10 -4
8 10 -4
10-3
0 2 10-4 4 10-4 6 10-4 8 10-4
Iron concentration (M) Iron concentration (M)
25 12
NHS pH 8 NHS pH 6 (p)

(m) ( n)
10
20 (o)
8
15 (l)
6 (q)
(k)
10
4
5
2
0 0
0 2 10-4 4 10 -4
6 10 -4
8 10 -4
10-3 0 4 10-4 8 10-4 1.2 10-3
Iron concentration (M) Iron concentration (M)
Fig. 1. Evolution of residual turbidity (K) and sediment volume (’) with coagulant concentration for NHS and SHS at pHs 6 and 8.
Arrows indicate optimal coagulant concentrations. The letters refer to freeze-dried sediment examined by TEM.
concentration, no settleable aggregate is observed up to solutions of iron coagulant were prepared for each
the maximum in supernatant turbidity. The sediment experiment, this result could be due to the proteina-
volume then rapidly builds up just before the OCC, and ceaous nature of NHS (Jung et al., 2005b).
it levels off with further addition in coagulant. At pH 6,
the sediment volume gradually diminishes as the humic 3.2. Morphology of coagulated humic substances
suspension is restabilized.
OCC values, expressed in molar concentrations, and Typical micrographs of coagulated humic substances
pHs at OCC are given in Table 1. In all cases, the pH at are shown in Fig. 2. The sample preparation used in this
OCC falls within the 4.5–6 pH range for best removal of study precludes artifact free observations. Indeed,
organic substances (Dempsey et al., 1984). Surprisingly, freeze-drying is known to drastically modify the
OCCs obtained for NHS coagulated under acid and structure of aggregated colloids (Hung et al., 1996),
alkaline conditions are equivalent. As fresh stock whereas redispersion in ethanol may induce secondary
ARTICLE IN PRESS
Table 1
Optimum coagulant (iron) concentrations (OCC) determined by jar-test procedures
OCC pH 8.0 pH 6.0 pH 7.4
SHS NHS SHS NHS NOM
Molar concentration (mol/L) 8.3 104 6.4 104 2.2 104 8.1 105 1.7 104
‘‘Reduced’’ concentration (mg Fe/mg Corg ) 3.83 2.69 1.01 0.34 1.02
pH at the respective OCC 6.5 6.3 5.1 5.2 5.9
agglomeration effects upon solvent removal (Stevenson was not detected in our samples, the nitrogen K-edge
and Schnitzer, 1982). Actually, techniques such as was used instead.
atomic force microscopy in aqueous solution (Maurice An example of energy loss spectrum recorded in the
and Namjesnik–Dejanovic, 1999; Plaschke et al., 250–450 eV region (SHS coagulated at pH 8 with
1999) or TEM observation of ultracentrifuged [Fe] ¼ 7 104 mol/L), is illustrated in Fig. 3. Three
and resin-coated colloids (Wilkinson et al., 1999), main features can be observed at the carbon edge: the
seem more appropriate to provide a description of well-defined shoulder at 285 eV can be assigned to the
hydrated humic material. Nevertheless, as the prepara- 1s-p* transition of aromatic carbons (CQC); the
tion conditions remained identical for all samples, the small peak around 288 eV is attributed to the 1s-p*
morphology of coagulated humic substances could be transition of the CQO bond in carboxylic groups, while
compared. the broad peak centered at 300 eV is attributed to
Coagulated SHS appears as compact aggregates of overlapping s* transitions for CQC, C–C, C–H,
overlapping spheroids about 30 nm in size, regardless of CQO, and C–O bonds (Myneni, 2002). At higher
pH and iron concentration (Figs. 2a and b). The energy, the two peaks observed at 346 and 358 eV
appearance of NHS aggregated at pH 8 is similar with correspond to the scattered electrons of the L2–3 shell of
smaller spheroids (10–15 nm) (Fig. 2c). In contrast, calcium (Watteau et al., 1996). The presence of calcium
electron micrographs of NHS coagulated at pH 6 reveal corresponds to a contamination brought about by the
a network of filament or sheet-like structures at the addition of NaHCO3 ([Ca2+]2.7 104 mol/L). It is
micron scale (Fig. 2d). No significant change of the likely associated with the functional groups of SHS as
latter morphology could be detected with coagulant calcium has been shown to compete with trace elements
concentration. Despite the structural changes certainly for the binding sites of humic substances (e.g., Tipping,
induced by the preparatory technique, our TEM 1993; van den Hoop et al., 1995). Finally, in accordance
observations evidence individual spheroids which aver- with the elemental composition of SHS, a small amount
age dimension is consistent with the range of sizes of nitrogen can be detected at the N K-edge around
previously reported in the literature for humic colloids 401 eV.
(Wilkinson et al., 1999; Plaschke et al., 1999). Increasing the coagulant concentration at pH 8
significantly modifies the electron loss spectrum of
coagulated SHS (Fig. 4a). At the carbon edge, the sharp
3.3. EELS spectra of coagulated SHS and NHS peak at 288 eV assigned to p* (CQO) drastically
decreases, while the large peak at 300 eV broadens
To the authors knowledge, EELS investigations of further. In addition, the peak at the calcium edge rapidly
NOM remain relatively scarce in the literature (Watteau drops. A recent NEXAFS study of pedogenic humic
et al., 1996; Furukawa, 2000). The spectra of well- acid also showed a similar decrease of the p* resonance
characterized compounds such as Bovine serum albumin of the carboxyl group double bond upon aggregation
and Dextran (polysaccharide—M W ¼ 100; 000) were with Fe3+ at pH 2 (Thieme et al., 2002). Such
then recorded as standards, and the identification of observations can be readily explained by an exchange
EELS spectral features was conducted in agreement with of Fe-coagulant species with previously bonded Ca2+
recent near-edge X-ray absorption spectroscopy (NEX- ions onto carboxylic groups of SHS. The same interac-
AFS) studies (Myneni, 2002 and references herein). tion mechanism has also been inferred from infrared
EELS spectra were calibrated in energy using the spectroscopy results in the case of adsorption/desorption
346.4 eV Ca L2,3-edge as an internal reference. This of NOM on iron oxide particles (Gu et al., 1994),
yielded a 401.6 eV for the N K-edge energy which is coagulation of sewage with ferric chloride (El Samrani et
consistent with the attribution usually proposed for al., 2004), and more generally, such a reaction scheme is
nitrogen in NOM (Watteau et al., 1996). When calcium consistent with the strong influence of carboxylic group
ARTICLE IN PRESS
Fig. 2. Electron micrographs of freeze-dried coagulated sediment: (a) SHS pH 8—[Fe] ¼ 8.0 104 mol/L, (b) SHS pH 6—
[Fe] ¼ 2.0 104 mol/L, (c) NHS pH 8—[Fe] ¼ 7.0 104 mol/L, (d) NHS pH 6—[Fe] ¼ 8.0 104 mol/L.
content on coagulant demand (Lefebvre and Legube, supernatant after settling. For each iron concentration,
1990; Randtke, 1993; Collins et al., 1986; Specht et al., the spectrum closely resembles that of a sediment
2000). aggregated with a lower coagulant dosage, and can be
The inset of Fig. 4a presents the evolution of EELS characterized by a higher calcium content and by a
spectra of freeze-dried humic material remaining in the slightly sharper peak at 288 eV (p*CQO). This suggests
ARTICLE IN PRESS
Derivative of normalized intensities (A.U.)

293-294
299-300 Standard 1
Polysaccharide
286 Standard 2
Bovine serum
C -K albumin
346.4 401.6
296 - 297 eV
Normalized intensity (A.U.)
285 288 297 SHS

1s → σ∗
C-N, C=O, C-O
C=C, C-C, C-H
288 eV
275 310 345 380 415
1s → π∗
C=C C EELS Ca-L2,3 N-K
binding energies Energy Loss (eV)
285 eV
1s → π∗ Ca L2,3 -
C=C N-K
270 320 370 420

Energy Loss (eV)
Fig. 3. Typical EELS spectrum (SHS pH 8—[Fe] ¼ 8.0 104 mol/L). The inset shows the double derivative for (i) Bovine serum
albumin, (ii) Polysaccharide, (iii) SHS at pH 8 and at [Fe] ¼ 8.0 104 mol/L.
that the unsettled humic material contains a lesser Fig. 4b). In particular, the p* resonance of the CQO
amount of hydrolyzed Fe species. The EELS spectra of double bond appears as a peak at low iron concentra-
NHS coagulated at pH 8 as a function of coagulant tion, and this peak decreases at higher coagulant dosage.
concentration exhibit similar features (Fig. 4c): however, Likewise, the Ca edge which was virtually absent in the
the peak characterizing the 1s-p* transition of the coagulated sediment, can be clearly observed except at
CQO bond in carboxylic groups is much less resolved the highest iron concentration. Such observations
at low iron concentration, and it is gradually replaced by suggest that the residual turbidity is formed from a less
a weak shoulder at higher coagulant dosage. Similarly, coagulated humic material also at pH 6.
the two peaks at the Ca L2,3 edge, clearly defined at low
coagulant concentration, rapidly weaken with an 3.4. Elemental analysis of coagulated SHS and NHS
increase in iron concentration.
The EELS spectra of SHS and NHS coagulated at pH Fig. 5 shows the evolution of N/C, Ca/C, Na/C, and
6 as a function of iron concentration are shown in Figs. Fe/C atomic ratios of coagulated SHS and NHS with
4b and d. In that case, the relative intensities of the iron concentration at pHs 6 and 8. N/C remains
various contributions do not vary noticeably with constant as a function of coagulant concentration and
coagulant concentration. At the carbon edge, the 1s-p* pH, thus indicating that both the extraction procedure
transition of aromatic carbons (CQC) at 285 eV and the synthesis yielded a homogeneous humic material
remains as a distinct shoulder, whereas the contribution for coagulation experiments. N/C atomic ratios calcu-
at 288 eV (p* CQO) is now present as a weakly lated from EELS spectra are equal to 0.0770.02 and
resolved shoulder. The large dominant feature corre- 0.1770.02 for SHS and NHS, respectively. Such values
sponding to the s* transitions for CQC, C–C, C–H, are consistent with previous N/C elemental ratios—
CQO, and C–O bonds has shifted around 297 eV. 0.054 and 0.167—obtained from chemical analysis (Jung
Finally, a large contribution in the 325 eV region can be et al., 2005b). Similar trends are also observed for the
evidenced, but could not be assigned. However, the evolution of Ca/C and Na/C. At pH 8, Ca/C markedly
freeze-dried humic material from the supernatant yield decreases just before OCC to reach an almost constant
EELS spectra similar to those obtained at pH 8 (inset of elemental ratio at higher coagulant concentration,
ARTICLE IN PRESS
Fig. 4. EEL spectra of freeze-dried coagulated humic substances as a function of iron concentration: (a) SHS pH 8, (b) SHS pH 6,
(c) NHS pH 8, (d) NHS pH 6.
whereas Na/C remains close to 0.1 for both SHS and The evolution of Fe/C atomic ratios as a function
NHS. At pH 6, Ca is not detected within the coagulated of coagulant concentration follows two different
sediment, whereas Na/C continuously decreases with patterns according to pH. At pH 8, for both SHS
iron concentration in accordance with an exchange of and NHS, Fe/C is about 3 before OCC, it surprisingly
sodium ion for hydrolyzed Fe species onto anionic drops to a value close to 1 at OCC (0.8 (c) and 1.25 (m)),
groups of humic substances. The lower Na/C ratios and then re-increases to an intermediate value at
obtained at pH 8 are also consistent with an increased higher coagulant dosage. At pH 6, Fe/C slightly
deprotonation of SHS and NHS functional groups, increases with iron concentration from a value of
even though Na/C values may be modified during 1 at underdosage for NHS, to reach a Fe/C ratio
freeze-drying. of about 3 at OCC for both humic compounds, and
ARTICLE IN PRESS
Fig. 5. Evolution of elemental ratios as a function of coagulant concentration: N/C (’), Fe/C *1/10 (&), Na/C *1/10 for NHS pH 8
(K) and Ca/C (J).
values greater than 4 when the suspension is restabilized. pH 6, the same calculations yield Fe/C atomic ratios of
Interestingly, recent Fe K-edge EXAFS measure- 0.22 and 0.7 which is much less than the measured
ments carried out on Seine river NOM coagulated values. Such a discrepancy may be explained by the
with iron chloride (Vilgé–Ritter et al., 1999a, b), revealed aggregation behavior of humic substances with pH. The
that (i) iron hydrolysis is hindered to at most the hydrolysis of iron coagulant implies a decrease in pH;
trimer stage, (ii) each iron atom is surrounded at hence, at OCC, the final pH of the suspension initially
optimum dosage by three C and 1 C at pH 5.5 and coagulated at pH 6, was about 4.5. At the latter pH, a
pH 7.5, respectively. Such results would yield Fe/C small degree of aggregation of humic substances has
elemental ratios in the 1–3 range which is consistent with been clearly demonstrated from the measurement of
our data. their diffusion coefficients by fluorescence correlation
Assuming a complete removal of humic material by spectroscopy (Lead et al., 2000). In situ AFM observa-
coagulation, the determination of elemental ratios from tions of adsorbed humic colloids as a function of pH
chemical analysis predicts that, at pH 8, Fe/C should be (Plaschke et al., 1999; Balnois et al., 1999), also indicate
equal to 0.8 and 0.66 for SHS and NHS, respectively, an increased surface aggregation with a decrease in pH.
which is in relatively good agreement with the values The iron-coagulated sediment should then include some
obtained from EELS and EDX spectra. In contrast, at loosely attached humic colloids that may become rapidly
ARTICLE IN PRESS
volatilized under the electron beam; the microanalysis 299 eV peak, were also found (Fig. 6c). Such organic
being performed mainly with the iron coagulated humic matter is reminiscent of that observed by Furukawa in
substances. The almost constant Fe/C atomic ratio Jourdan river sediments (Furukawa, 2000). It should be
obtained at pH 6 is then in accordance with previous noted that, for NOM, the interpretation of EELS
studies suggesting a complexation mechanism between spectra at the carbon edge is complicated by the
oligomeric coagulant species and functional groups of presence of L2–3 peaks at 297 and 300 eV of potassium
fulvic and humic colloids around pH 5 (van Breemen et (Henke et al., 1993).
al., 1979; Jeckel, 1986; Lefebvre and Legube, 1990). Fig. 7 shows the jar-test results obtained with Moselle
The effect of pH on Fe/C elemental ratio can be river water coagulated with iron at natural pH (7.4). The
explained by a dynamic reorganization of humic supernatant turbidity continuously increases with iron
macromolecules upon interaction with Fe coagulant concentration up to [Fe]1.2 104 mol/L, then steeply
species. At pH 8, the behavior of deprotonated humic decreases at OCC ([Fe] ¼ 1.7 104 mol/L), slightly re-
material is generally compared with that of a flexible increases above OCC, and then decreases again at higher
anionic polymer in a stretched configuration (Gosh and coagulant dosage. The evolution with coagulant con-
Schnitzer, 1980; Vermeer et al., 1998). In that case, at centration of C-EELS spectra corresponding to humic-
low coagulant concentration, the partial neutralization like colloids is illustrated in Fig. 8a. This kind of
of humic macromocules by Fe-hydrolyzed species can material shows a similar reactivity to that of SHS and
induce a reconformation of polymer chains, thus leading NHS at pH 8, with a strong decrease of the p*CQO
to a compact structure and values of Fe/C close to 3. peak at 288 eV with iron concentration. Fig. 8b presents
Recent fluorescence spectroscopy using pyrene as a the variation of N/C, Ca/C, and Fe/C elemental ratios of
polarity-sensitive fluoroprobe also suggests such a coagulated NOM with iron concentration. Fe/C in-
rearrangement of humic macromolecules upon associa- creases continuously with coagulant dosage to reach a
tion with multivalent cations (Engebretson and von value close to 1 above OCC, which is consistent with the
Wandruszka, 1994; Kazpard et al., 2005). At higher results obtained with SHS and NHS. Likewise, Ca/C
coagulant dosage, the amount of hydrolyzed Fe species elemental ratio decreases in accordance with a substitu-
increases relative to the number of humic colloids, and tion of Ca for Fe-hydrolyzed species onto functional
aggregation between stretched polymer chains may groups of organic matter. N/C values reveal that NOM
proceed before folding and reconformation of humic removal occurs in two main stages: up to OCC, the N/C
material take place. In that case, the coagulant species atomic ratios are less than 0.03, thus suggesting a
may form bridges between stretched humic chains. In preferential coagulation of humic substances; above
accordance with such an aggregation mechanism, the OCC, N/C increases significantly to reach values in the
increase in sediment volume observed with pH above 0.1 range which indicates the presence of proteinic
OCC, is proportional to C/Fe ratios for both SHS and matter in the coagulated sediment. Such results are
NHS. Therefore, the increase in sludge volume at pH 8 consistent with previous pyrolysis-gas chromatography-
is here attributed to a change in shape factor of humic mass spectrometry (Py-GC-MS) experiments showing
colloids, and not to the formation of abundant iron that polyhydroxyaromatics are preferentially removed
hydroxide precipitate as predicted by a sweep-floccula- by ferric chloride (Vilgé–Ritter et al., 1999a, b; Jung,
tion mechanism (Gregory and Dupont, 2001). 2004). More generally, the hydrophobic fraction of
NOM is known to be more easily destabilized than
3.5. Coagulation of Moselle river raw water hydrophilic compounds (e.g., proteinic matter) (Croué
et al., 1993; Bose and Reckhow, 1998). In particular, the
Experiments with Moselle river water were run in slight increase of residual turbidity observed above OCC
duplicate, similar trends being observed for the two grab could correspond to the restabilization of some humic
samples. Four kinds of NOM compounds were identi- colloids (Fig. 7).
fied from TEM examination (Fig. 6): humic-like colloids
(Fig. 6a (A)), by far the most frequent compounds in the
freeze-dried coagulated sediment, were about 20 nm in
diameter with a N/C ratio in the 0.02–0.05 range; 4. Conclusion
proteinic substances (Fig. 6a (B)), appeared as large
grayish areas 300 nm in size with N/C values greater This study shows that TEM observations combined
than 0.1; polysaccharidic material was characterized by with EDXS and EELS analyses, represent a powerful
micron size spheroids with neither nitrogen nor iron tool to investigate the coagulation of NOM. The
detected, a CEEL spectrum equivalent to that of a sugar nanoscale resolution of EELS and EDXS is well-
backbone, and it was easily degraded under the electron adapted to the size range of organic colloids encoun-
beam (Fig. 6b); small organic colloids intimately tered in surface waters. In particular, a fingerprinting of
associated with smectite clay material and with a sharp coagulated NOM can be achieved, even though further
ARTICLE IN PRESS
C.K EELS EDXS

Elmt At%
A 1 O 55.6
289 300 A : humio-like A 2 Na 3.5
substances 0 : C Ka 6 P 1.0
1' : Fe La 7 S 1.1
8 Cl 2.5
285 TEM photo 0
401.6 1 9 K 2.6
2 89 10 Ca 2.1
10 11 Fe 31.6
67
250 300 350 400 450 0.7 2.1 3.5 4.9 6.3
B
Energy Loss (eV) A Energy (keV)
300 1 Elmt At%
B 1 O 81.6
B
346.4 2 Na 3.4
0.5 µm 0 : C Ka 6 P 1.1
285 401.6 1' : Fe La 7 S 1.1
8 Cl 2.6
0 1' 8 9
B : protein-like 9 K 2.5
2
67 10 10 Ca 6.1
substances 11
11 Fe 1.6
250 300 350 400 450

0.7 2.1 3.5 4.9 6.3
(a) Energy Loss (eV) Energy (keV)
C-K EELS TEM photo EDXS

300
0
Elmt (K) At%

1 (indicative)
285 0 C 55
1 0 45
0.5 µm
250 300 350 400 450 0.7 2.1 3.5 4.9 6.3
(b) Energy Loss (eV) Energy (keV)
C-K EELS TEM photo EDXS Elmt At%

299 1
298 1 O 58.9
0 : C Ka 2 Na 3.4
1' : O La 3 Mg 3.4
346.4 11
5 4 Al 6.6
285 5 Si 9.5
0 6 P 0.5
1' 4 0.5
3 7 S
2 8 9 8 Cl 2.8
67 10
9 K 1.8
10 Ca 2.2
0.2 µm 11 Fe 10.5
250 300 350 400 450 0.7 2.1 3.5 4.9 6.3
(c) Energy Loss (eV) Energy (keV)
Fig. 6. Typical NOM compounds identified by combining TEM observation with EDXS and EELS spectra: (a) Humic-like substances
(A) and protein-like material (B), (b) Polysaccharide-like substances, (c) Organo-mineral species.
ARTICLE IN PRESS
10
NOM natural pH 7.4
9 NOM natural pH 7.4
(u)
(v)
Normalizedintensity (A.U.)
8
7 (v)
(s)
6 (t) (u)
5
4 (t)
3
(s)
2
(r)
1 (r)
0 250 300 350 400 450

0 5 10-5 10-4 1.5 10-4 2 10-4 2.5 10-4 (a) Energy Loss (eV)
Iron concentration (M)
0.12
Fig. 7. Jar-test results for NOM at natural pH (7.4). (K NOM natural pH 7.4
residual turbidity (NTU) ’ sediment volume (mL)). (u)
Elemental atomic tal 0.1 (v)
work is still required to improve the distinction of (t)

natural organic substances. 0.08
EELS and EDXS examination of coagulated sediment (s)
also provides a useful insight into the destabilization 0.06
(r)
mechanism of humic substances and NOM: Fe-hydro-
lyzed species associate with carboxylic groups of humic 0.04
colloids at both pH 6 and 8, thus inducing a release of
previously complexed calcium ions. An overall charge 0.02
neutralization/complexation mechanism seems to be
responsible for the removal of humic substances in our 0
experimental conditions as no Fe-hydroxyde precipitate 1.1 10-4 1.4 10-4 1.7 10-4 2.0 10-4 2.3 10-4
could be detected in the coagulated sediment. The
(b) Iron concentration (M)
association of non-coagulated humic colloids within
flocs may also occur at pH 6. On the other hand, at Fig. 8. (a) EEL spectra for NOM as a function of coagulant
alkaline and neutral pH, our results suggest that a dosage. (b) Elemental atomic ratio: N/C (’), Fe/C *1/10 (&),
reorganization of humic macromolecules from a Na/C (K) and Ca/C (J).
stretched to a coiled conformation takes place upon
interaction with coagulant species. Such phenomenon is
worth exploring, as the optimal coagulant dosage should
then depend on the complex interplay between the rate References
of collisions of organic colloids/coagulant species and
the rate of macromolecules reconformation. Andreux, F., Golebiowska, D., Metche, M., 1980. Oxydative
polymerization of O-diphenols in presence or absence of
amino-acids. Topics on (catechol-glycocolle) and (catechol-
Acknowledgments diglycylglycine). General Assembly of Polyphenols Group
Report, Logrono, Spain, Bulletin, vol. 9, pp.178–188.
Balnois, E., Wilkinson, K.J., Lead, J.R., Buffle, J., 1999.
This work was achieved within the framework of
Atomic force microscopy of humic substances: effects of pH
‘‘Zone Atelier du bassin de la Moselle’’, with financial
and ionic strength. Environ. Sci. Technol. 33 (21),
support from Region Lorraine and CNRS (Contrat de 3911–3917.
Plan Etat-Région and Programme Environnement Vie et Bose, P., Reckhow, D.A., 1998. Adsorption of natural organic
Société). Agence de l’Eau Rhin-Meuse is gratefully matter on preformed aluminium hydroxide flocs. J. Envir-
acknowledged for providing the Field-centrifuge. The on. Eng. 124, 803–810.
authors would also like to thank T. Jacquin (Agence de Collins, M.R., Amy, G.L., King, P.H., 1985. Removal of
l’Eau Rhin Meuse, Rozerieules, France) for providing organic matter in water treatment. J. Environ. Eng. 11,
technical assistance during field measurements. 850–854.
ARTICLE IN PRESS
Collins, M.R., Amy, G.L., Steelink, C., 1986. Molecular weight investigations. C.R. Acad. Sci. Chim. submitted for
distribution, carboxylic acidity and humic substances publication.
content of aquatic organic matter: implications for removal Jung, A.-V., Frochot, C., Parant, S., Lartiges, B., Selve, C.,
during water treatment. Environ. Sci. Technol. 20 (10), Viriot, M.-L., Bersillon, J.-L., 2005b. Synthesis of amino-
1028–1032. phenolic humic-like substances and comparison with
Croué, J.-P., Lefebvre, E., Martin, B., Legube, B., 1993. natural aquatic humic acids: a multi-analytical techniques
Removal of dissolved hydrophobic and hydrophilic organic approach. Org. Geochem. in press.
substances during coagulation/flocculation of surface Kazpard, V., Lartiges, B.S., Jung, A-V., Frochot, C., Espinose
waters. Water Sci. Technol. 27, 143–152. de la Caillerie, J.B., Portal, J.M., Görner, T., Viriot, M.L.,
Dempsey, B.A., Ganho, R.N., O’Melia, C.R., 1984. The Bersillon, J.L., 2005. Aggregation of a humic macromole-
coagulation of humic substances by means of aluminium cules with polycations (Al13 and PEI): a combined pyrene
salts. J. Am. Water Works Assoc. 74, 141–150. fluorescence, electrophoretic mobility and SEC investiga-
Edwards, G.A., Amirtharajah, A., 1985. Removing color tions. Water Res. submitted for publication.
caused by humic acids. J. Am. Water Works Assoc. 77, Lartiges, B.S., Bottero, J.Y., Derrenger, L.S., Humbert, B.,
50–57. Tekely, P., Suty, H., 1997. Flocculation of colloidal silica
Egerton, R.F., 1996. Electron Energy-Loss Spectroscopy in the with hydrolysed aluminium: an 27Al solid state NMR
Electron Microscope, second ed. Plenum Press, New York investigation. Langmuir 13 (2), 147–152.
423pp. Lead, J.R., Wilkinson, K.J., Starchev, K., Canonica, S., Buffle,
El Samrani, A.G., Lartiges, B.S., Montargès-Pelletier, E., J., 2000. Determination of diffusion coefficients of humic
Kazpard, V., Barrès, O., Ghanbaja, J., 2004. Clarification substances by fluorescence correlation spectroscopy: role of
of municipal sewage with ferric chloride: the nature of solution conditions. Environ. Sci. Technol. 34 (7),
coagulant species. Water Res. 38 (3), 756–768. 1365–1369.
Engebretson, R.R., von Wandruszka, R., 1994. Microorganiza- Leenheer, J.A., Croué, J.-P., 2003. Characterizing aquatic
tion in dissolved humic acids. Environ. Sci. Technol. 28 (11), dissolved organic matter. Environ. Sci. Technol. 1 (1),
1934–1941. 19A–26A.
Furukawa, Y., 2000. Energy-filtering transmission electron Lefebvre, E., Legube, B., 1990. Iron (III) coagulation of humic
microscopy (EFTEM) and electron energy-loss spectro- substances extracted from surface waters: effect of pH and
scopy (EELS) investigation of clay-organic matter aggre- humic substances concentration. Water Res. 24 (5),
gates in aquatic sediments. Org. Geochem. 31 (7/8), 591–606.
735–744. Matilainen, A., Lindqvist, N., Korhonen, S., Tuhkanen, T.,
Gosh, K., Schnitzer, M., 1980. Macromolecular structures of 2002. Removal of NOM in the different stages of the water
humic substances. Soil Sci. 129, 266–276. treatment process. Environ. Int. 28 (11), 457–465.
Gregory, J., Dupont, V., 2001. Properties of flocs produced Maurice, P.A., Namjesnik-Dejanovic, K., 1999. Aggregates
by water treatment coagulants. Water Sci. Technol. 37, structures of sorbed humic substances observed in aqueous
35–43. solution. Environ. Sci. Technol. 33 (9), 1538–1541.
Gregor, J.E., Nokes, C.J., Fenton, E., 1997. Optimizing natural Myneni, S.C.B., 2002. Soft X-ray spectroscopy and spectro-
organic matter removal from low turbidity waters by microscopy studies of organic molecules in the environ-
controlled pH adjustment of aluminum coagulation. Water ment. Reviews in Mineralogy and Geochemistry
Res. 31 (12), 2949–2958. (Applications of Synchrotron Radiation in Low-Tempera-
Gu, B., Schmitt, J., Chen, Z., Liang, L., McCarthy, J.F., 1994. ture Geochemistry and Environmental Sciences) 49,
Adsorption and desorption of natural organic matter on 485–579.
iron oxide: mechanisms and models. Environ. Sci. Technol. Oliver, B.G., Visser, S.A., 1980. Chloroform production from
28 (1), 38–46. the chlorination of aquatic humic material: the effect of
Henke, B.L., Gullikson, E.M., Davis, J.C., 1993. X-ray molecular weight, environment and season. Water Res. 14
interactions: photoabsorption, scattering, transmission, (9), 1137–1141.
and reflection at E ¼ 50–30000 eV, Z ¼ 1–92. At. Data Owen, D.M., Amy, G.L., Chowdhury, Z.K., Paode, R.,
Nucl. Data Tables 54, 181–342. McCoy, G., Viscosil, K., 1995. NOM characterization and
Hung, W.T., Chang, I.L., Lin, W.W., Lee, D.J., 1996. treatability. J. Am. Water Works Assoc. 87, 46–63.
Unidirectional freezing of waste-activated sludges: effects Plaschke, M., Römer, J., Klenze, R., Kim, J.I., 1999. In situ
of freezing speed. Environ. Sci. Technol. 30 (7), 2391–2396. AFM study of sorbed humic acid colloids at different pH.
Jeckel, M.R., 1986. Interactions of humic acids and aluminium Colloids Surfaces A: Physicochem. Eng. Aspects 160,
salts in the flocculation process. Water Res. 20 (12), 269–279.
1535–1542. Randtke, S.J., 1993. Coagulation of NOM: an overview
Jung, A.-V., 2004. Interactions des micropolluants anthropi- of the science and U.S. practice. NOM in drinking
ques avec la matière organique naturelle et devenir lors des water, origin, characterization and removal. Workshop
opérations de traitement de l’eau. Ph.D. Thesis INPL, Proceedings of AWWA about Natural Organic Matter in
Nancy, 427pp. Drinking Water. AWWA Research Fundation Ed.,
Jung, A.-V., Chanudet, V., Ghanbaja, J., Lartiges, B.S., pp.155–163.
Abdelmoula, M., Bersillon, J.-L., 2005a. Iron populations Specht, C., Kumke, M., Frimmel, F., 2000. Characterization of
evolution during organic matter coagulation by a nitrate NOM adsorption to clay minerals by size exclusion
salt: energy-loss spectroscopy (EELS) and Mössbauer chromatography. Water Res. 34 (16), 4063–4069.
ARTICLE IN PRESS
Stevenson, I.L., Schnitzer, M., 1982. Transmission electron energy loss spectroscopy (EELS) of the brown cellular
microscopy of extracted fulvic and humic acids. Soil Sci. 133 substances of beech roots. Soil Biol. Biochem. 28 (10/11),
(3), 179–185. 1327–1332.
Thieme, J., Schmidt, C., Abbt-Braun, G., Specht, C., Frimmel, Wilkinson, K.J., Balnois, E., Leppard, G.G., Buffle, J., 1999.
F.H., 2002. X-ray microscopy studies of refractory organic Characteristic features of the major components of fresh-
substances. In: Frimmel, Abbt-Braun, Heumann, Hock, water colloidal organic matter revealed by transmission
Lüdemann, Spiteller (Eds.), Refractory Organic Substances electron and atomic force microscopy. Colloids Surfaces A:
in the Environment. Wiley-VCH, Weinheim, pp. 239–248. Physicochem. Eng. Aspects 155, 287–310.
Thurman, E.M., Malcolm, R.L., 1981. Preparative isolation of
aquatic humic substances. Environ. Sci. Technol. 15 (4),
463–466.
Tipping, E., 1993. Modeling the competition between alkaline
earth cations and trace metal species for binding by humic Further reading
substances. Environ. Sci. Technol. 27, 520–529.
Van Breemen, A.N., Nieuwstad, T.J., Van der MeentOlieman, Bruchet, A., Rousseau, C., Mallevialle, J., 1990. Pyrolysis-GC-
1979. The fate of fulvic acids during water treatment. Water MS for investigating high-molecular-weight THM precur-
Res. 13 (8), 771–780. sors and other refractory organics. J. Am. Water Works
Van den Hoop, M.A.G.T., van Leeuwen, H.P., Pinheiro, J.P., Assoc. 71, 66–74.
Mota, A.M., Simões Gonc- alves, M.L., 1995. Voltametric Hitchcock, A.P., Urquhardt, S.G., Rightor, E.G., 1992. Inner-
analysis of the competition between calcium and heavy shell spectroscopy of benzaldehyde, terephtalaldehyde,
metals for complexation by humic material. Colloids ethylbenzoate, terephtaloyd chlorides and phosphogen:
Surfaces A: Physicochem. Eng. Aspects 95, 305–313. models for core excitation of poly(ethylene) terephtalate.
Varlot, K., Martin, J.M., Quett, C., 1998. Physical and J. Phys. Chem. 96 (22), 8736–8750.
chemical changes in polystyrene during electron irradiation Johnson, P.N., Amirtharajah, A., 1983. Ferric chloride and
using EELS in the TEM: contribution of the dielectric alum as single and dual coagulants. J. Am. Water Works
function. J. Microsc. 191 (2), 187–194. Assoc. 54, 232–239.
Vermeer, A.W.P., van Riemsdijk, W.H., Koopal, L.K., 1998. Kikuna, J., Tonner, B.P., 1996. XANES spectra of a variety of
Adsorption of humic acid to mineral particles. 1-Specific widely used organic polymers at the C K-edge. J. Electron
and electrostatic interactions. Langmuir 14 (10), 2810–2819. Spectrosc. Relat. Phenom. 82 (1/2), 53–60.
Vilgé-Ritter, A., Masion, A., Boulangé, T., Rybacki, D., Lefebvre, E., Legube, B., 1993. Coagulation-flocculation by
Bottero, J.Y., 1999a. Removal of natural organic matter ferric chloride of some organic compounds in aqueous
by coagulation-flocculation: a pyrolysis-GC-MS study. solutions. Water Res. 27 (3), 433–447.
Environ. Sci. Technol. 33 (17), 3027–3032. O’Melia, C.R., 1972. Coagulation and flocculation. In: Weber,
Vilgé-Ritter, A., Rose, J., Masion, A., Bottero, J.-Y., Lainé, J.- Jr., W. (Ed.), Physicochemical Processes for Water Quality
M., 1999b. Chemistry and structure of aggregates formed Control. Wiley Interscience, New York 14pp.
with Fe-salts and natural organic matters. Colloids Surfaces Packam, R.F., 1965. Some studies of the coagulation of
A: Physicochem. Eng. Aspects 147 (3), 297–308. dispersed clays with hydrolysing salts. J. Colloid Sci. 20,
Villemin, G., Mansot, J.L., Watteau, F., Ghanbaja, J., Toutain, 81–92.
F., 1995. Etude de la biodégradation de la matière Rose, J., Vilgé, A., Lauquet-Olivie, G., Masion, A., Frechou,
organique végétale du sol par la spectroscopie des pertes C., Bottero, J.Y., 1998. Iron speciation in natural organic
d’énergie d’électrons transmis (EELS: Electron Energy Loss matter colloids. Colloids Surfaces A: Physicochem. Eng.
Spectroscopy): répartition du carbone, de l’azote et évalua- Aspect 136 (1/2), 11–19.
tion du rapport C-to-N au niveau ultrastructural in situ. Semmens, M.L., Field, T.K., 1980. Coagulation: experiences in
C.R. Acad. Sci. Paris 321 (IIa), 861–868. organics removal. J. Am. Water Works Assoc. 72, 476–483.
Watteau, F., Villemin, G., Mansot, J.L., Ghanbaja, J., Toutain, Stumm, W., O’Melia, C.R., 1968. Stoechiometry of coagula-
F., 1996. Localization and characterization by electron tion. J. Am. Water Works Assoc. 60, 514–539.
Organic
Geochemistry
Organic Geochemistry 36 (2005) 1252–1271
www.elsevier.com/locate/orggeochem
Synthesis of amino-phenolic humic-like substances

and comparison with natural aquatic humic acids: A
multi-analytical techniques approach
A.-V. Jung a,*, C. Frochot b, S. Parant c, B.S. Lartiges a,
C. Selve c, M.-L. Viriot b, J.-L. Bersillon a
a
Laboratoire Environnement et Minéralurgie (LEM), UMR 7569 CNRS-INPL-ENSG, 15 avenue du Charmois,
BP 40, 54501 Vandœuvre, France
b
Département de Chimie Physique des Réactions (DCPR-GRAPP), UMR 7630 CNRS-INPL-ENSIC,
1 rue Grandville, BP 451, 54500 Nancy, France
c
Laboratoire de Structure et Réactivité des Systèmes Moléculaires Complexes (SRSMC-LCPOC), UMR 7565 CNRS-UHP,
Université Henri Poincarré (Nancy I), 54500 Vandœuvre, France
Received 16 September 2004; accepted 27 April 2005

(returned to author for revision 1 March 2005)
Available online 20 June 2005
Abstract
Monitoring of the synthesis of low and high molecular weight humic-like molecules by oxidative reaction between
catechol and one amino-acid (glycine) or one dipeptide (triglycine) was conducted by Fourier transform infrared spec-
troscopy (FTIR) and fluorescence measurements. The results suggest that a well-defined energy threshold should be
reached to allow the formation of humic substances. This is substantiated by the occurrence of the humic-like fluores-
cence peak (around 350/450 nm for excitation–emission wavelengths), referred to as ‘‘alpha’’ component. In addition,
nitrogen incorporation was found to be one of the fundamental processes governing the behaviour of fluorescence sub-
stances. The study of the synthesis of these model molecules appears to be a promising route to obtain a variety of mate-
rials approaching the properties and the variability of natural substances but in a somewhat more reproducible way. In
this report, an approach relying on several complementary techniques is proposed. It involves elemental analysis, poten-
tiometric titration, Fourier transform infrared spectroscopy, high-performance liquid chromatography, pyrolysis gas
chromatography–mass spectrometry and UV/visible absorption spectrophotometry. Model synthetic molecules were
characterised and compared to natural humic substances extracted from the materials suspended in waters of a river.
All combined analytical results led to the conclusion that protein-like structures could be used to describe humic-like
structures and could be related to recent supramolecular association concepts proposed for the structural description of
humic substances. Synthetic molecules display the main chemical features of naturally occurring humic acids and could
therefore be suitable for investigating the reactivity of humic substances.
2005 Elsevier Ltd. All rights reserved.
*
Corresponding author. Fax: +33 389 647787
E-mail address: av1jung@hotmail.com (A.-V. Jung).
0146-6380/$ - see front matter 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.orggeochem.2005.04.004
A.-V. Jung et al. / Organic Geochemistry 36 (2005) 1252–1271 1253
1. Introduction precursors by chemical or enzymatic oxidation in the

presence of a nitrogen source, generally ammonia, ami-
1.1. Humic aquatic organic matter no-acids, or proteins (Flaig et al., 1975; Mathur and
Schnitzer, 1978). Stevenson (1994) extensively describes
Humic substances (HSs) account for 30–50% of the pathways of formation of humic substances such
aquatic natural organic matter (NOM). They are well as the lignin and polyphenol theories as well as Mail-
known for their ability to interact with organic and inor- lardÕs route. Until now, few models of humic substances
ganic contaminants and to affect the environmental have been synthesised. Most have been obtained via the
behaviour of pollutants in natural waters (Chiou et al., Maillard reaction (Jokic et al., 2004) which consists of a
1986, 1987). Despite their prominent environmental reduction of amino-acids (melanoidin, glycine and mel-
role, their molecular properties and conformational anoin) by sugars (xylose). In the present work, the pol-
structures are far from being elucidated and are still a yphenol condensation route theory (Stevenson, 1994)
matter of debate (Wershaw et al., 1977; Wershaw, was chosen. Previous communications by Andreux
1986; Schulten and Schnitzer, 1993, 1995, 1997; Sein et al. (1980), Andreux (1981), Andreux and Munier-
et al., 1999; Piccolo, 2001). Lamy (1994) were used as guidelines for the synthesis
In particular, the hydrophilic/hydrophobic balance of the humic material. Briefly, the process comprises
could account for the complexation and transport of an oxidative reaction using catechol and one amino-acid
non-polar pollutants in natural media, such as polycyclic (glycine) or one dipeptide (triglycine) to produce model
aromatic hydrocarbon components (PAH) (Karickhoff molecules with various N/C elemental atomic ratios.
et al., 1979; Laor et al., 1998). Many attempts have been
made to quantitatively describe the hydrophobic proper- 1.2. Analytical and characterisation methods
ties of HSs in order to relate them to biological activity
or to their environmental fate. The octanol/water parti- The use of a suite of powerful analytical techniques
tion coefficient (Kow) is of particular importance in pre- enables valuable qualitative and quantitative informa-
dicting pollutant solubility and has been correlated with tion for the characterisation of HSs to be obtained.
bioconcentration factors (Davies and Dobbs, 1984) or Techniques used include
with adsorptive capacities on activated carbon (Blum
et al., 1993). high-performance size-exclusion chromatography
Although a general qualitative description of the (HPSEC) (Conte and Piccolo, 1999a,b; Piccolo
complexing functional groups exists, their quantification et al., 2002);
is still difficult despite powerful nuclear magnetic reso- elemental analysis (Thurman and Malcolm, 1981);
nance (NMR) studies that have recently been performed pyrolysis gas chromatography–mass spectrometry
to obtain information on the nature and number of the (Py/GC–MS) (Gadel and Bruchet, 1987);
different functional groups comprising HSs (Conte et al., potentiometric titration (Brunelot et al., 1989);
2002). For a long time, elemental analysis such as the Fourier transform infrared (FTIR) (Niemeyer et al.,
nitrogen to carbon atomic ratio (N/C) was used to dis- 1992);
criminate the aquagenic from the pedogenic origin of NMR (Clair et al., 1996; Chen et al., 2002; Conte
the humic substances (Aiken et al., 1985). According et al., 2004);
to these authors, this difference of origin may be related fluorescence correlation spectroscopy (FCS) (Lead
to differences in environmental behaviour and reactivity. et al., 1999; Egeberg et al., 2002) and
Moreover, very few analytical techniques that respect electrospray ionisation-mass spectrometry (ESI-MS)
the nature and structural organisation of such complex (Kujawinski et al., 2002).
molecules (non-destructive methods) can be used to
characterise the natural ‘‘dissolved organic matter’’ of Reverse-phase chromatography is commonly used
a river (Belin et al., 1993; Matthews et al., 1996; Lead for the determination of hydrophobic properties. Brau-
et al., 1999; Mounier et al., 1999). mann (1986) found a correlation between the coefficient
One possible approach consists of synthesising well- Kow and the binding capacity on different chromato-
defined molecules, under controlled conditions, that ex- graphic columns of several organic compounds. Differ-
hibit the main characteristics and properties of natural ent molecular weight acid or fulvic fractions isolated
humic substances or their precursors. The better known from the field and concentrated were characterised else-
theories concerning the formation of humic substances where by their retention times using high-performance
are based on oxidative processes occurring for the liquid chromatography (HPLC; Münster, 1999; Wu
numerous phenolic molecules released by plant residues et al., 2002). Several elution and column conditions were
and soil microorganisms (Stevenson, 1982). Several at- tested (Hayase and Tsubota, 1984; Norden and Dabek-
tempts have been made in the past to synthesise Zlotorzynska, 1996). The development of fluores-
humic-like substances from representative phenolic cence spectroscopy methods enabled highly sensitive
1254 A.-V. Jung et al. / Organic Geochemistry 36 (2005) 1252–1271
and non-destructive descriptions of the studied mole- 2. Materials and methods

cules in aqueous solutions to be made. This technique
has been used widely to characterise the proteinaceous 2.1. Synthetic and natural humic substances
and the humic parts of NOM (Chen and Bada, 1992;
Mopper and Schultz, 1993; Coble, 1996; Parlanti et al., Catechol (99% purity analytical grade), glycine
2000). In recent years, several correlations between or- (>97% pure) and triglycine (>97% pure), NaH2PO4
ganic matter characteristics like hydrophobicity (Fang (99% purity analytical grade) and Na2HPO4 (99% purity
et al., 1998), condensation indices (Zolsnay et al., analytical grade) were purchased from Sigma–Aldrich
1999), dissolved and total organic content (Buffle Chimie (Lyon, France) and stored at 5 C in the dark.
et al., 1982; Mounier et al., 1999) and, biological oxygen For all sample preparation and dilutions, Milli-Q ul-
demand (Reynolds and Ahmad, 1997) have been made tra-pure water from Millipore (Billerica, MA, USA)
for NOM. Synchronous scan fluorescence spectroscopy was used.
(SFS) has been used for the characterisation and identi- The general procedure described by Andreux et al.
fication of natural organic matter from different origins (1980) was applied for the synthetic substances prepared
(Miano et al., 1988; Senesi et al., 1991; Coble, 1996; De with catechol and either glycine or triglycine. Catechol
Souza-Sierra et al., 1997). It has also been used in (6.6 g) was dissolved in 500 ml of a solution of sodium
tracing sources of dissolved organic matter in natural phosphate buffer NaH2PO4/Na2HPO4 at pH 8 to obtain
water from different watersheds and in differentiating a 0.03 M catechol solution. This was then mixed with an
between organic substances in extracts obtained from var- equal volume of buffered glycine or triglycine 0.03 M
ious decayed plant materials (Cabaniss and Schuman, solution. The resulting mixture was then allowed to re-
1987). act in a 2 l flask in the dark at 25 C and under a pure
Humic substances are also viewed as complex and oxygen constant pressure (0.9 bar) for 5 days. One milli-
polydisperse mixtures, whose properties depend primar- liter aliquots were withdrawn at 0, 1, 6, 24, 48, 72 and
ily on their natural conditions of formation (Andreux, 120 h in order to assess reaction kinetics and comple-
1981; Aiken et al., 1985). In accounting for the proper- tion. Five hundred microliters of the mixture were
ties of HS, classical polymeric views (Hayes et al., freeze-dried to perform FTIR spectroscopy measure-
1989; Schulten and Schnitzer, 1993, 1995, 1997) are chal- ments. Five hundred microliters were diluted 200 fold
lenged by the ‘‘supramolecular concept’’ (Kingery et al., to perform the UV/visible absorbance and the excita-
2000; Piccolo, 2001; Simpson, 2001, 2002; Buurman tion–emission matrices (EEM) fluorescence measure-
et al., 2002; Simpson et al., 2004) supported by chro- ments. This dilution was necessary to avoid UV/visible
matographic and spectroscopic evidence of low molecu- spectrophotometer saturation and to limit the inner filter
lar weight associations in humic structure. In order to effects for fluorescence measurement. At the end of the
investigate important questions such as the binding po- reaction, the solutions were transferred to dialysis bags
tential of aquatic humic materials with respect to other Spectra/Por 7 (Roth Sochiel, Lauterbourg, France).
pollutants, a standard material exhibiting constant char- The fraction >8 kDa molecular cut-off was collected
acteristics is required. In addition, the possibility of for the reaction product prepared with glycine (CGHA).
varying some of the properties of the standard material Successive dialysis steps with three different molecular
is desirable. The extraction of large amounts of such weight cut-offs were used to isolate two different trigly-
materials from water does not offer a suitable solution cine molecular weight fractions (1–3.5 kDa, CTHA1;
since the stock will eventually be consumed. Another above 8 kDa, CTHA2). The fraction below 1 kDa, com-
extraction would most probably yield materials with a prising non-reactive compounds (chloride and phos-
completely different set of properties. Model synthetic phate salts) or remaining free monomeric and
molecules avoid this inconvenience, assuming that a reli- oligomeric products, was not kept. The CTHA1 and
able synthetic route is followed. the CTHA2 fractions were collected to obtain a fulvic-
The objectives of the present work are like and a humic-like range of molecular weight organic
compounds, respectively (Schnitzer and Khan, 1972).
1. to synthesise models of some organic humic-like The final ionic forms of all synthetic humic acids were
substances, sodium saturated. The different fractions of organic
2. to characterise the final synthetic substances and to solutions were freeze-dried and weighed. The powders
compare their properties with those of humic acid were carefully stored in glass bottles at 4 C in the
extracts from aquatic suspended matter, and dark.
3. to correlate the data from spectrofluorimetric tech- Natural suspended matter samples were collected
niques with that derived from more conventional with a 17,000 rpm flow-through (1 m3/h) centrifuge at
analytical tools (elemental analysis, potentiometric the water intake of the Urban Community of Nancy,
titration, FTIR spectroscopy, HPLC, Py/GC–MS France. This sampling point is located between the
and UV/visible absorption spectrophotometry). pumping station in the Moselle River and the retention
basins used as an alternative water resource. The water ature inside the quartz tube was controlled with a
was pumped from 20 cm below surface level. The sus- K-type thermocouple minithermometer and maintained
pended solids were collected on Teflon plates covering at 625 ± 5 C. Pyrolysis products were separated on a
the inner surface of the centrifuge. Each sample was 30 m DB Wax fused silica capillary column from Agilent
placed in a glass bottle and stored at 4 C during trans- (Palo Alto, CA, USA) programmed from 30 to 220 C at
portation and storage prior to treatment. The extraction a rate of 3 C/min. Identification was performed using a
of humic substances began within 24 h after arrival at mass spectrometer operated with and electron energy of
the laboratory. Two samples were collected, NHA1 un- 70 eV and scanning from 30 to 350 m/z at 1 scan/s. The
der dry and ‘‘hot’’ summer conditions, and NHA2 under quantitative method of characterisation described by
rainy and cold fall weather conditions. Extraction was Bruchet et al. (1990) was used to assist sample compar-
carried out according to the International Humic Sub- ison. Each pyrolysis fragment was assigned to one of the
stances Society (IHSS) standard procedure for the isola- following four types of organic compound families: car-
tion of humic fraction of the bulk solid earth materials bohydrates, proteinaceous materials, amino-sugars or
with slight modification (Thurman and Malcolm, polyhydroxy-aromatic substances. Fragments that could
1981). During the dissolution step of the humic acid not be assigned to one of the four previous categories
fraction, NaOH 0.1 M was used instead of KOH. This were lumped into a ‘‘miscellaneous’’ category.
base was chosen in order to obtain the sodium form of Acidic titration was performed using a potentiometer
the final humic acid, so as to better compare these ex- (precision of 0.1 mV or 0.001 units of pH) and a Tacussel
tracts with the synthetic products. For the same reason, combination glass electrode both from Radiometer
crystallised NaCl was added to flocculate the mineral (Villeurbanne, France). The outer chamber of the
fraction (instead of KCl). Due to the difference in the electrode was filled with saturated KCl solution. The elec-
flocculation potential of these two salts, the NaCl dose trode was calibrated from a potentiometric titration of a
was recalculated according to results from Bonneau known volume of a standard 0.1 M NaOH solution. Both
and Souchier (1977). Moreover, Engebretson and von solutions were made in 0.1 M NaCl. The temperature was
Wandruska (1999) mentioned that the HF treatment controlled at 25.0 ± 0.1 C. The titrations of 400 mg/l HA
could totally change the hydrophobic characteristics of solution were performed with an automatic burette
humic substances. This is the reason why no purification (precision 15 ll) under pure N2 atmosphere. All titrant
(HCl/HF treatment) was performed so as to preserve the solutions were CO2-free. The initial volume of HCl
final structure of the macromolecules. Dialysis was per- (1 M), added to bring the pH to 2.6, was included in the
formed against distilled water with an 8000 molecular determination of acidities. The total titration curves were
weight cut-off (Spectra/Por 7, Roth). It should be men- performed within the 3–13 pH range. The equivalent
tioned that no UV/visible detectable organic compounds point determinations and quantitative acidity dosages
under this molecular weight cut-off were obtained. were obtained using the derivative isotherm summation
(DIS) method (Villieras et al., 1997; Prélot, 2001).
2.2. Overall characterisation methods Hydrophilic/hydrophobic characterisation of the hu-
mic acid fractions was performed with a Shimadzu
Elemental analyses of humic acid fractions were per- HPLC (DGU-14A, Kyoto, Japan), equipped with a
formed on a 1108 Carlo Erba (Orono, ME, USA) auto- C18 reversed-phase column (5 lm, 150 · 4 mm) from
analyser. Dissolved organic carbon (DOC) was measured Lichrosorb (Agilent, Palo Alto, CA, USA), a fluorescent
with a 5000-A Carbon Analyser from Shimadzu (Kyoto, RF-10AXL detector, an LC-10AT UP pump and an
Japan) using high temperature (680 C) and catalytic oxi- FCV-10ALVP auto-sampler (all from Agilent). The
dation. For 0.45 lm filtered solutions (PC Whatman), reversed-phase HPLC conditions were chosen according
DOC was obtained by difference between the total carbon to Wu et al. (2002), who studied different elution condi-
and the inorganic carbon contents (measured after acidi- tions and HPLC columns to define hydrophobic/hydro-
fication to pH 2 with HNO3). FTIR spectroscopy of 1 mg philic indices for molecular weight fractionated aqueous
of the HA compound in 100 mg KBr pellets was deter- natural and commercial fulvic acids. The injection
mined on a IFS 55 Bruker spectrometer (Billerica, MA, volume was 50 ll and the detector was set at excita-
USA) with a deuterated triglycerine sulfate detector. tion–emission (Ex–Em) 300/380 nm wavelengths for
Samples were first carefully dried in a dessicator contain- the fluorescence measurements. Aqueous solutions of
ing silica gel and measured at the rate of one point every 30 mg/l HA were all adjusted to pH 8.0 with HCl
4 cm1. 0.1 M before measurement. For the eluent, acetonitrile
For the Py/GC–MS measurements, 0.4 mg aliquots was used as solvent A and Milli-Q ultra-pure water
of humic acid powder were used. Each quartz sample as solvent B. A gradient was set from 0% A to 10%,
holder was then inserted into a filament pyrolyser 30%, 70% and 100%, respectively, in 5, 9, 13 and
Pyroprobe 2000 from CDS (Oxford, PA, USA) and then 17 min time periods, and continued to 23 min with iso-
heated to 720 C at a rate of 20 C/ms. The final temper- cratic conditions (total 40 min). The components were
eluted with constant flow of 0.8 ml/min. UV/visible served after a few minutes and seen to persist for several
absorption data were collected on a 2501 PC com- hours until a brown colour developed and reached a
puter-driven model spectrophotometer from Shimadzu maximum intensity after about 100 h at pH 8. Kinetic
(Kyoto, Japan). Absorption of HA solutions was mea- studies of the absorption spectra in UV/visible light
sured from 200 to 700 nm for the inner filter effects cor- (Fig. 1) show a progressive decrease of the absorption
rections. Whatman quartz liquid sample holders with a band of catechol at 280 nm, followed by a sharp de-
light path length of 10 mm were used for both absorp- crease after 24 h. The 320–330 nm shoulder was attrib-
tion and fluorescence measurements. uted to glycine combination with catechol during the
reaction process. The final CGHA UV/visible spectrum
2.3. Fluorescence characterisation methods was found as featureless as in humic substances (Hayes
and Swift, 1978).
Fluorescence was measured either in 2D emission Fig. 2 illustrates the FTIR striking difference in gly-
acquisitions at one specific wavelength or in 3D excita- cine and triglycine incorporation, essentially occurring
tion–emission matrices (EEM) acquisitions or with syn- in the 3450–3050 cm1 region. According to diol and
chronous fluorescence scans (SFS). 1,2 polyol vibration band attribution (Wojtkowiak and
All EEM spectra were collected using the spectroflu- Chabanel, 1977), the fine 3450 cm1 band was assigned
orometer Fluorolog Spex 3 from Jobin Yvon (Stanmore, to intramolecular AOH valence vibration (mOH) and
UK). A 450 W xenon arc lamp was used as the excita- the larger 3326 cm1 band to the conjugated chelation
tion source and monochromators to select excitation of mOH. In the CTHA synthesis, according to the same
and emission wavelength. Series of emission spectra authors, the triglycine incorporation can be followed
were collected over a range of excitation wavelengths through the monitoring of the free mNH for amines
to generate EEM spectra. The wavelength step size and amides at 3382 cm1 and the associated mNH at
was 5 nm and the speed of acquisition scanning was 3050 cm1. The high resolution of these (normally
4 nm/s. The excitation wavelength range was 250– broad) OH stretching peaks suggests very well-struc-
450 nm. The emission wavelength range was set from tured N-quinone derivatives during the first step of
270 to 700 nm. The entrance and exit slits of the mono- CTHA formation. During the catechol/glycine forma-
chromators were set to 2 nm. The sample compartment tion, a substituted amino-quinone or amino-phenol
temperature was maintained at 25.0 ± 0.1 C by an moietie is formed and is found in the structure, glycine
LTD6G temperature control unit from Grant (Sarose, is also better solvated by the quinone derivatives than
UK). For the EEM spectra, Raman and Rayleigh scat- triglycine.
terings were corrected by an internal correction of the EEM monitoring of the synthetic CGHA and CTHA
DATAMAX software (Orlando, FL, USA). are respectively presented in Fig. 3(a) and (b). An imme-
Synchronous fluorescence scans (SFS) were run at the diate loss of the double fluorescence band for the two
constant 18 nm offset value (Dk = kem kex). The excita- main benzene-diol peaks (centred at 270/310 and 270/
tion and emission slits and the excitation wavelength 650 nm) was observed for CGHA as well as for CTHA
were, set at 5, 10 and 254 nm, respectively. Fluorescence during the first step of the monitoring (Figs. 3(a), (b)
spectra were obtained with the Fluorolog Spex 3 using a and 4 for catechol EEM spectrum). With regards to
1 cm pathlength quartz sample holder. In the hope of dis- CGHA formation, an intermediate fluorescence state
tinguishing the chemical behaviour of the various fluoro- was measured after 30 h reaction time. The correspond-
phores in humic fractions, the emission spectra were ing fluorescence peak is centred in the 250/450 nm (exci-
recorded over a 2–10 pH range. The pH was adjusted tation–emission wavelengths, respectively). No more
using only HCl or NaOH to avoid possible interference fluorescence signals were recorded in the 270/310 and
from buffers during fluorescence experiments. 270/650 nm regions at this stage of the process. The hu-
In order to compare the fluorescence intensities of all mic-like fluorescence peak (350/450 nm), usually referred
fluorescence spectra (emission, synchronous and EEM to as ‘‘alpha’’ component (Parlanti et al., 2000), only ap-
spectra) and to avoid equipment baseline time drift, the peared after 72 h. For the CTHA reaction, the main hu-
fluorescence was normalised with a 0.16 mg/l fluorescein mic acid fluorescence broad band appeared after a
solution buffered at pH 10. The maximum emission of this reaction time of 24 h, it then enlarged and grew with
reference solution (512 nm for an excitation wavelength time. Additional signals at 270/310 and 270/650 nm per-
at 472 nm) was determined between measurements. sisted right to the end of the reaction.
Elemental compositions, expressed as percent by dry
weight of freeze-dried final material, are listed in Table
3. Results 1. Elemental atomic H/C and N/C ratios are reported
in Table 2. These ratios range from 0.56 for CGHA to
In the reaction flask, an intense pink colour, attrib- 1.87 for NHA1 and from 0.89 for CGHA to 4.92 for
uted to the formation of N-substituted quinines, was ob- CTHA1. All data lie within ranges usually reported,
Fig. 1. UV/visible for CGHA and CTHA formation monitoring.
Fig. 2. FTIR for CGHA and CTHA formation monitoring.
for humic acid, in the literature (Thurman and Malcolm, CGHA pyrolysis is six times more abundant than for
1981) except for CTHA1 which has a lower carbon the CTHA fractions. In CTHA, the benzene-diol func-
content. tions are more abundant in the final polymerisation
The acidity dissociation constants and corresponding product than in CGHA. When considering the protein
acidity amounts obtained are given in Table 1. The and amino-sugar contents in the synthetic polymers,
lower acidity value (pKA1) was attributed to carboxylic acetamide and methyl-acetamide are found to be three
acidity, the second (pKA2) to amino-phenolic acidity times more abundant in the CTHA2 fraction than in
and the third one (pKA3) to phenolic acidity (Andreux the CTHA1 fraction. Benzene-diol is detected only for
et al., 1977). the CTHA fractions.
Pyrolysis GC–MS results are reported in Table 3. Fig. 5 shows the FTIR spectra of freeze-dried humic
The polyhydroxy-aromatic, amino-sugars and protein acids in their sodic form, phosphate-free. OH stretch-
contents of CTHA1 and CGHA fractions were found ing occurs in the FTIR spectra at 3380 cm1. The
to be similar by Py/GC–MS. The CO2 released by 2920 and 2860 cm1 absorption bands attributed to
Fig. 3. EEM for CGHA (a) and EEM for CTHA (b) formation monitoring.
the asymmetric and symmetric stretching vibrations of CTHA1 to CTHA2 and to CGHA, concomitant with
aliphatic CAH bonds in methyl or methylene units, the presence of longer aliphatic chains in CTHA com-
respectively, show a decrease in absorbance from pared with CGHA. The C@O stretching vibration
Fig. 4. EEM (30 mg/l, pH 8). (1±10 nm.)
Table 1 P
Humic substances: description, elemental analysis (all in wt%) (O = 100% C, H, N) and acid functional groups values
Type of HA %C %H %O %N pKA1 (meq/g) pKA2 (meq/g) pKA3 (meq/g)
NHA1 48.6 7.6 34.4 9.5 3.7 (2.1) 8.9 (0.4) 10.1 (10.6)
NHA2 44.2 5.9 44.3 5.6 3.2 (2.7) 8.5 (0.3) 9.5 (15)
CGHA 40.7 1.9 54.8 2.6 4.6 (2.3) 5.1 (0.8) 7.2 (0.6)
CTHA2 41.2 2.7 46.3 9.8 3.9 (4.1) 7.7 (0.2) 9.8 (1.3)
CTHA1 25.3 2.9 62.9 8.9 3.8 (3.8) 7.6 (0.1) 9.6 (1.5)
Table 2
Environmental significance of synthetic substances compared with natural extract: aromaticity and condensation degree
Type of HA H/C N/C If/DOC E2/E3 E4/E6 HIRP (%)
NHA1 1.87 0.17 68,615 2.59 18.54 51.1
NHA2 1.61 0.11 15,082 2.82 11.45 52.2
CGHA 0.56 0.05 11.46 1.90 19.76 84.5
CTHA2 0.78 0.20 202 2.37 26.10 82.8
CTHA1 1.38 0.30 1844 2.82 37.57 76.6
H/C and N/C: atomic ratios; If: fluorescence maxima intensities obtained from EEM for 30 mg/l aliquots; DOC: dissolved organic
carbon in mg/l; E2/E3 (UV measurements) calculated for 100 mg/l aliquots (see Fig. 6); E4/E6 (UV measurements) calculated for
100 mg/l aliquots (see Fig. 6); HIRP: hydrophobic indexes determined by reverse-phase chromatography as areas ratios of ‘‘B’’ peaks
to ‘‘A’’ ones (see Fig. 6).
Table 3
Humic substances Py/GC–MS results
Pyrolysed fragment Classification Areas
CTHA1 CTHA2 CGHA NHA1 NHA2
Acetonitrile Proteins 225 248 760 404 18
Toluene 13 2 30 765 116
Pyridine 3 1 8 20 9
Methylpyridine 1 1 6 10 4
Pyrrole 3 4 48 32 1
Methylpyrrole 2 0 21 26 0
Benzonitrile 0 2 12 0 13
Phenylacetonitrile 1 0 4 9 0
Indole 0 0 0 61 0
Methylindole 0 0 0 8 0
Total distribution (%) 39 25 44 25 16
Hydroxypropanone Polysaccharides 2 3 11 5 20
2-Methylfurane 1 1 5 0 0
2-Furaldehyde 1 2 6 0 0
Indene 2 0 2 8 3
Acetamide Amino-sugars 12 60 37 32 19
Methyl-acetamide 0 15 0 4 2
Phenol Hydroxy-polyaromatics (HPA) 11 10 46 62 127
p-Cresola 3 5 14 402 0
m-Cresola 0 0 0 0 149
Benzene Miscellaneous 7 0 12 191 9
Benzene-diolb 115 120 0 0 0
Styrene 2 2 12 16 10
Acetic acid 20 31 90 36 23
Propionic acid 0 1 9 0 67
CO2 575 488 2850 3765 638
a
Both attributable to proteins and HPA parts (see the text for discussion).
b
Catechol included.
band at 1730 cm1 is primarily due to carboxyl groups is swamped by a peak that exhibits a maximum at
and is only easily visible for CGHA and CTHA1. The 1591 cm1. In CGHA, the amide II band due to NH
most significant differences among all our FTIR spectra deformation and contribution of CAN stretching is
occur in the 1650–1540 cm1 region. The 1650 cm1 found at 1480 cm1. The 1380–1410 cm1 band may
band is assigned to the aromatic C@C double bonds be attributed to CAH bond and the 1250 cm1 region
conjugated with C@O or COOA (Schnitzer and Skin- to CAOAH deformation and CAO stretching of phe-
ner, 1968) and the 1540 cm1 one to the amide II band. nolic compounds added to the asymmetric CACAO
These two bands are present in both CTHA spectra. A stretching of ester groups. The exact wavenumber of
decrease of the amide band at 1540 cm1 in CTHA1 this band varies from one humic acid to another and
compared with CTHA2 can be noticed. In the CTHA2 could depend either on the attached phenolic substi-
spectra, the 1603 cm1 band could be attributed to the tuted groups or on another structural effect. The
aromatic C@C pure skeletal stretching of the aromatic 1230 cm1 peak relates to the amide III band and is
ring of phenolic compounds. However, in the CGHA due to the interaction between CAN stretching and
spectrum, this region is very different: the amide I band in-plane CANAH bending of proteins. The 1120–
(C@O stretching + contribution from CAN stretching) 1160 cm1 broad band is attributed to the ACAOH
Fig. 5. FTIR spectra of humic substances.
stretch of aliphatic OH, the 956, 875 and 806 cm1 to EEM with final synthetic EEM spectra shows that dial-
the aromatic out-of-plane CH bending (Niemeyer ysis purification efficiently removed excessive residual
et al., 1992). Overall, the IR absorption spectra exhibit catechol and unreacted quinone moieties. Moreover,
similar primary features to those encountered in humic lyophilisation did not alter the fluorescence properties
materials (Andreux et al., 1977; Aiken et al., 1985; Nie- of the humic substances since fluorophores apparently
meyer et al., 1992). remain preserved from the end of the synthesis reaction
Three elution domains are observed in the HPLC (Fig. 3(a) and (b)) to the resuspension of the freeze-dried
chromatograms (Fig. 6): components with a retention substances in aqueous solutions (Fig. 4).
time below 11 min were considered as ‘‘hydrophilic’’ The alpha fluorescence intensity to dissolved organic
moieties since they are essentially eluted with the polar content ratio (If/DOC) is shown in Table 2. This ratio
elutant (Milli-Q ultra-pure water). Small peaks ob- may be used to discriminate synthetic HA from natural
tained below 5 min retention time were attributed to HA. Indeed, the If/DOC ratio is more than 10 times low-
salts (NaCl) or other traces of small constituents. Peaks er in CGHA, CTHA1 and CTHA2 than in NHA1 and
with elution times over 20 min (which corresponds to NHA2. The values (height, intensity) for natural ex-
isocratic acetonitrile conditions) were included in the tracts are higher than those determined for surface river
hydrophobic contribution quantification. Hydrophobic waters by Buffle et al. (1982).
indices obtained by reverse liquid phase chromatogra- Fig. 8 depicts alpha, alpha 0 and tryptophan maxi-
phy (HIRP) are reported in Table 2. mum fluorescence intensities as a function of pH. For
HA UV/visible spectra are presented in Fig. 7. For all CGHA, fluorescence signals drastically increase from
HAs, a 270 nm chromophoric band occurs. For syn- pH 2 to 6, then remain stable for higher pH values.
thetic substances, glycine or triglycine/catechol combi- The two CTHA fractions have a totally different fluores-
nations are revealed by the 320 nm absorption band. cent behaviour: in the fluorescence increase, two steps
This chromophore cannot be found in NHA1 and can be distinguished. The first one occurs for pH < 4
NHA2 natural extracts. For these fractions, a 400– and the second for pH > 6. The NHA2 results are very
410 nm absorption band is attributed to conjugated similar to those for CTHA, whereas NHA1 evolution
multiple double bonds (Pretsch et al., 2000). All of the is quite different from all other HAs. For this last case,
HA UV/visible spectra describe similar shapes and are alpha prime (alpha 0 ) compound evolution follows the
consistent those previously reported for humic sub- tryptophan pattern. The non-linear increase of intensity
stances (Aiken et al., 1985). E2/E3 (250/365 nm) and of these characteristic HA peaks (alpha, alpha 0 and tryp-
E4/E6 (465/665 nm) absorbance ratios were calculated tophan) that occurs with increasing pH is similar to that
for each HA fraction and the values are reported in Ta- found in previous works on DOM (Laane, 1982; Mat-
ble 2. thews et al., 1996).
Excitation–emission matrices (EEM) of the final As shown in Fig. 9, the synchronous spectrum of
materials and the corresponding peak wavelength values each humic fraction exhibits a different spectral line
are presented in Fig. 4. The comparison of catechol shape at various pH (2–10). This technique allows a
Fig. 6. RP-HPLC chromatograms of natural and synthetic humic substances: (A) divided by 2 and (B) divided by 2/5.
Fig. 7. UV/visible spectra of natural and synthetic humic substances (100 mg/l, pH 8).
better discrimination between the different fluoro- is composed of more resolved peaks. A 520 nm peak
phores. CTHA1 shows a simple spectrum with a single appears above pH 4 and increases with increasing
peak at 350 nm for pH < 10 and exhibits a second peak pH. The pH influence is very different for CGHA
at 410 nm for pH 10. CTHA2 exhibits the same two and CTHA, since the overall signal intensity increases
peaks and similar behaviour with respect to pH. An for CGHA whereas it decreases for both CTHAs.
additional peak at 320 nm may be observed, which dis- Moreover, relative fluorescence intensity values are
appears with increasing pH. The CGHA SFS spectrum stronger for CTHA than for CGHA over the whole
Fig. 8. pH-dependence of alpha, alpha 0 and tryptophan humic substances components.
wavelength range. The two NHAs present more 4. Discussion

decreasing fluorescence spectra with increasing excita-
tion wavelength and the overall intensity decreases with 4.1. Synthetic reaction monitoring: incorporation of
increasing pH. The NHA1 peak (450–470 nm) is espe- glycine and triglycine derivatives in quinonic moieties
cially pH-dependent and decreases in intensity with derived from catechol
increasing pH. All SFS spectra are consistent with pre-
vious works on humic substances extracted from soils The observed phenomena are attributed to several
or water (Cabaniss and Schuman, 1987; Fang et al., resonance steps occurring for quinonic derivative forma-
1998). tion in the early steps of the process. Tyson and Martell
1000000
Fluorescence intensity (cps)
pH 2
CTHA1 pH 4
600000 pH 6
pH 8
pH 10
200000
300 350 400 450 500 550

Wavelength (nm)
1500000
350000

NHA1
1000000 CTHA2
200000
500000
500000
0
0
300 350 400 450 500 550 300 350 400 450 500 550
50000
1000000
CGHA NHA2
30000 600000
10000 200000
0
0 300 350 400 450 500 550
300 350 400 450 500 550
Wavelength (nm) Wavelength(nm)
Fig. 9. SFS of humic substances (30 mg/l, pH 8).
(1972) proposed an ortho-benzoquinone formation from has been fully incorporated in the structure. Thus, the
catechol oxidation by the transfer of two electrons, CGHA fluorescent aromatic center is suspected to be
accompanied by H2O2 release. The system resonance more homogeneous and compact than in the CTHA.
stability also evolves during the early steps of amino- In other words, to reach the same stabilisation reso-
acid incorporation. In the case of CGHA formation, nance energy level (i.e., comparable excitation, emission
the total disappearance of ‘‘free’’ remaining quinonic wavelengths and fluorescence intensity), CTHA moieties
fluorescent moieties after 30 h suggests that catechol need 1/3 of the time taken by CGHA formation (72 h) to
organise themselves into humic-like molecular assem- It was shown that fluorescence is a sensitive tool for
blies. For CGHA, the system apparently needs to go the study of HA internal microdomains. Also, amino-
through a high energy barrier (intermediate state at group incorporation in humic-like structures drastically
48 h) before organising itself into structured humic-like changes the fluorescence properties of these macromole-
assemblies. These results are consistent with Andreux cules favouring the production of long wavelength-emis-
et al. (1980) who used gaseous exchange kinetic mea- sive fluorophores.
surements during the oxidation of catechol/glycine, cat-
echol/triglycine and other amino-acids such as L and
4.2. Comparative characterisation of synthetic and natural
b-alanine and proline mixtures. These authors con-
humic acids compounds
cluded that, contrary to most amino-acids that poorly
stimulate the oxidation of catechol, glycine significantly
4.2.1. Elemental contents
modifies the reaction kinetics due to an easier oxidative
The H/C atomic ratio is a useful qualitative parame-
decarboxylation and deamination (i.e., the Trautner cy-
ter when comparing the aromaticity and the hydropho-
cle). The present work confirms that this process can
bicity of humic substances (Chen et al., 1977). H/C
occur less easily when oligomeric reactants (like trigly-
values are found to be around 1 for aqueous counter-
cine) are used. The triglycine is less subject to Trautner
parts, whereas higher ratios (1:2) are generally found
cycle reactions, resulting in a larger proportion of en-
in terrestrial humic acids (Ishiwatari, 1975). The syn-
tire molecules being incorporated into CTHA, much
thetic material aromaticity also increases from CTHA1,
before secondary aggregation of oligomeric substituted
CTHA2 and finally CGHA. The variability in the nitro-
quinines takes place. Due to fluorescence monitoring,
gen content is noticeable between the two natural humic
the CGHA formation is also viewed as a ‘‘directed
extracts: the high nitrogen content of the summer extract
assembly’’, as mentioned by Zhang et al. (2002), and
may be due to the incorporation of protein decomposi-
is more energy-demanding than the CTHA formation
tion products.
reaction. This last reaction consists of a ‘‘self-assem-
bly’’ of subunits, like those observed in some protein-
like structures formation (Zhang et al., 2002). These 4.2.2. FTIR
results also suggest that humic-like fluorescent com- The 3080 cm1 aromatic CH stretch, more marked in
pounds do not appear in the oxidation step but require CTHA than in CGHA, suggests that the aromatic parts
oligomeric quinonic moieties to be formed. A critical of these two models are more reactive than in the other
energy threshold of reaction, required to stabilise the molecules since they have been less overwhelmed by the
overall electronic energy and to allow fluorescence, other functional constituent bands. These aromatic moi-
was also observed: the stability of humic substances eties may be more accessible for hydrophobic interaction
conformation could be attributed to the entropy-driven processes. CGHA has a rather different spectrum, due to
exclusion of water from the self-associated hydropho- the absence of peptidic chains, and higher aromaticity
bic domains, thus decreasing the total molecular free than in any other compounds of the study. The natural
energy. Later in the reaction process, after 120 h of extract FTIR absorption spectra are similar to those ob-
reaction, the main fluorescence peak increases in inten- tained with the synthetic products, although the ali-
sity and widens to the usual widths reported in the lit- phatic CH3/CH2 groups are more prominent.
erature for excitation and emission range of humic Moreover, the relative absorption intensity at about
materials (Matthews et al., 1996). In addition, these 1620 cm1 (attributed to aromatic C@C and conjugated
findings confirm the previously mentioned view: HS C@O) is comparable in both types of substances. The
structure is also directly governed by ways of small main differences between the two NHA extracts is that
molecule and oligomeric compound association, lead- for the first one, the NH elongation band is much more
ing to ‘‘supramolecular’’ macromolecules. This present pronounced than in the NHA second extract (enlarge-
work shows that humic substances in solution are ment of the near 3400 cm1 peak) due to the high contri-
loosely bound self-associations of different fluorescent bution of the protein part in this extract. Furthermore,
and non-fluorescent chromophores and that intermo- the aromatic structure (3080 cm1 aromatic CH
lecular hydrophobic interactions can be assumed as stretches and near 800 cm1 to the aromatic out-of-
the predominant binding forces. Two parallel mecha- plane CH bending) is more pronounced, perhaps due
nisms are proposed: to a higher content in structured aromatic moieties.
With regards to secondary amide bands, it should be
(a) fluorophores ‘‘moving’’ into solvating regions that mentioned that they also exist before reaction but only
favour fluorescence and in the triglycine moieties; after the reaction, they subsist
(b) molecular structure stiffening and stabilisation of and prevail in CTHA substances. Those appearing in
the electronic energy due to the condensation CGHA result from the nucleophilic addition of ANH
process. on the rings. Their detection is not easy and here yields
to smaller and different bands. The differences observed ment, is the p-cresol to phenol ratio. The lower this
for amide band evolution among CTHA fractions are ratio, the more likely a polyhydroxy-aromatic origin
consistent with more peptidic links being present in the for the phenolic part. Since NHA2 shows a larger
higher molecular weight fractions. Séné et al. (1994) ob- amount of aromatic moieties and a lower protein con-
served such features in peptidic extracts from cell walls tent than NHA1, a change in protein nature from ‘‘hot
of five angiosperm plants. The corresponding wavenum- weather’’ to ‘‘cold weather’’ conditions can be suspected.
bers in the FTIR spectra of their extracted biopolymers Natural and synthetic humic substances are discrimi-
were attributed to a ‘‘secondary structure effect’’, occur- nated by the aromatic to protein ratio: the natural sub-
ring in proteins. Such an effect can be expected in the hu- stances contain a higher amount of aromatic structures
mic substances studied in this work. and a lower amount of proteins than the synthetic
materials.
4.2.3. Py/GC–MS
The important quantity of pyrrole in the CGHA 4.2.4. Acidities
fraction indicates that the cyclisation reactions neces- The two CTHA molecular weight fractions have sim-
sary to produce such heterocycles is favoured. This is ilar pKA values, which is not surprising since they result
due to a closer spatial proximity between the amino- from the same polymerisation process. Therefore, the
and phenolic-reactant groups or to a larger resonant same functional acidic groups are present for the final
electronic delocalisation effect (abundance of phenolic products. The most surprising result is that their acidic
reactive sites). Only the ‘‘polysaccharide’’ part, essen- content is similar. Such differences remain to be investi-
tially composed of carboxylic acids (acetic acid) and gated, in particular by studying the relaxation during the
in lesser amounts by furan derivatives and keto-cyclic titration process. CGHA has a very different acidity
pyrolysis fragmentation products, differentiates the titration feature as compared to CTHA. An explanation
two fractions. In CGHA fractions, due to the high can be found in the different mode of glycine incorpora-
amount of CO2 evolved, the aliphatic carboxylic parts tion during the polymerisation step. The two first acidi-
are also likely to be more numerous and labile than in ties are close to those obtained for benzoic derivative
CTHA. A similar molecular model could be proposed acids; the third one could correspond to charge delocal-
for the synthetic polymers, composed of a rather com- isation in phenolic groups (Aiken et al., 1985). Carbox-
pact polyhydroxy-aromatic nucleus and labile aliphatic ylic groups due to glycine incorporation are more
chains of variable lengths. In CGHA, phenolic func- abundant in the synthetic polymers whereas phenolic
tions could be involved in intermolecular bonds or groups are 10 times more numerous in natural sub-
could be totally oxidised into quinone derivatives dur- stances than in synthetic products. These results clearly
ing the first step of the process, so that no residual show that CTHA exhibits the carboxylic and phenolic
‘‘free’’ catechol moieties are isolated from the structure acidities that are also found in natural humic com-
by Py/GC–MS. Bruchet et al. (1990) studied pentagly- pounds from the Moselle River. The pKA1 and pKA3 val-
cine pyrolysis and concluded that unusual acetamide ues (Table 1) are consistent with literature data
derivative proportions as compared with general pep- (Banerjee, 1979; Aiken et al., 1985). The pKA2 deter-
tide composition were due to glycine properties. In mined to fit the DIS method is consistent with Andreux
CTHA2, the triglycine is also fully incorporated, et al. (1977) findings, attributing this second acidity to
mainly in non-branched or substituted peptidic chains. combinations of glycine with carboxylic-like structures.
Conversely, in CTHA1 fraction, the triglycine part is This pKA2 only accounts for less than 5% of total acidity
involved with other peptidic substituents and is not amounts.
fully incorporated in the final product. In natural hu-
mic extracts, the phenolic peak may originate from 4.2.5. Hydrophobicity
tyrosine. However, the unusually high amount of p- The synthetic fractions exhibit significantly higher
cresol in the first extraction (NHA1) is not fully ex- HIRP than the two natural extracts. Using this separa-
plained. Other bacterial or plant phenolic derivatives tive technique, the classification of HA fractions from
(especially lignin or tannin pigments) may occur, due more to less hydrophobic is the following: CGHA,
to the exceptionally dry and warm weather conditions CTHA2, CTHA1, NHA2 and NHA1. As mentioned be-
during the sampling. These compounds also contribute fore, this is a global parameter, since the nature of the
to p-cresol formation. Other protein derivatives like hydrophilic/hydrophobic moieties cannot be taken into
tryptophan, revealed by the indoles, are also present. account, as the different peaks found in the two domains
An explanation of this high tryptophan contribution correspond to several constituents. The relative reten-
is found in much diversified proteic nature of the nat- tion times are greater than those reported by Wu et al.
ural extract. One criterium determined by Bruchet et al. (2002). The very hydrophilic constituents (fulvic part)
(1990), to distinguish between proteic and polyhy- were not present in the compounds studied in the pres-
droxy-aromatic origins for the pyrolysed phenolic frag- ent work.
4.2.6. UV/visible More generally, it can be concluded that atomic ele-

For synthetic substances, the 270 nm absorption mental humic substance composition is not a determin-
band can be attributed to phenolic moieties and, in nat- ing factor as far as fluorescence behaviour is concerned.
ural extracts, to both phenolic and partly complexed On the other hand, the structural organisation of humic
protein bonds (Pretsch et al., 2000). substances, especially the peptidic bonds placed around
E2/E3 and E4/E6 ratio values are comparable to those the fluorophores (cf. kinetic results), influence the over-
obtained by Kononova (1966), Banerjee (1979) and all fluorescence pattern in such a way that fluorescent
Schnitzer (1977) for different molecular weight HAs. moieties are masked or, on the contrary, revealed. Fur-
Opposite variations for these two ratios are found for thermore, direct correlation between fluorescence inten-
all HAs. Among synthetic HAs, CGHA has the lowest sity and DOC should also be carefully established as
E4/E6 ratio. Therefore, the ‘‘condensation degree’’ of already noticed by Parlanti et al. (2000). Then, if the
the CGHA aromatic carbon network is higher than for majority of DOC is attributed to fluorescent structures,
the other synthetic polymers. Conversely, the high E4/ the HA reconfirmation capacity should be taken into
E6 ratio found in CTHA reflects the presence of rela- account.
tively large proportions of aliphatic structures in these The SFS technique makes it possible to confirm that
synthetic substances. As suggested by all previous ana- CGHA has spectral characteristics closer to the soil hu-
lytical data, the low value of the E4/E6 ratio obtained mic extract derivatives found in natural river waters
for NHA1 and NHA2 cannot be strictly indicative of and that CTHA is closer to aquatic humic substances
relatively high degrees of condensation. The E2/E3 ratio (Galapate et al., 1998; Fukushima et al., 2001) and
leads to the same conclusions and it is correlated appro- from natural humic acids studied in this work. It is
priately to the H/C ratio, as shown by Peuravuori and suggested that the observed fluorescence intensity
Pihlaja (1997). enhancement results from conformational changes in
humic materials.
4.2.7. Fluorescence
Different fluorophores may be identified using previ- 4.2.8. Combined analytical results
ously obtained data on organic matter fluorescence In integrating a number of analytical approaches,
characterisation (Matthews et al., 1996; Parlanti et al., several correlation results should be highlighted: the
2000). The alpha 0 (250–260/420–480 nm) component is aromaticity characteristics determined by elemental
present in CTHA1, NHA1 and NHA2. Alpha (330– and FTIR correlate with HIRP results (Table 2) and hy-
350/450–490 nm) peak is present in all HA synthetic droxy-polyaromatics amounts found by Py/GC–MS
and natural fractions. For NHA1, at least two other (Table 3; all fractions taken into account). For the nat-
fluorophores are detected, confirming the complex mix- ural fraction NHA1, according to MathewÕs findings
ture of fluorophoric compounds formed under sun (1996) and to our Py/GC–MS results, tyrosin has been
bleaching conditions or extracted with the humic acids suspected to contribute to the phenolic emission fluores-
of the ‘‘hot weather’’ conditions sample. cence peak. This last result is relatively new, since even
The large amount of phenolic groups (acidity deter- in very protein-rich humic compounds, the contribution
mination and Py/GC–MS) in natural extracts can of tyrosine is likely to be much lower than that of lignin
partially explain the If/DOC results. For CTHA1, structural units based on general results from the degra-
the If/DOC ratio is very close to the phenol value dation of natural organic products (Aiken et al., 1985).
determined by Buffle et al. (1982). In this fraction, Finally, the most interesting combined results are those
phenol moieties with a low level of substitution are obtained from classical techniques and those obtained
abundant. For CTHA2 and CGHA, the fluorophoric with fluorescence. First, the acidity titration shows the
signals may be inhibited by the high compactness of ionisable fluorophore contribution to the global fluores-
the aromatic nucleus. In the case of CGHA, the low cence signal. It can also be assumed that the CGHA al-
If/DOC ratio is attributable to more substituted pha component signal is mostly governed by carboxyl
groups on aromatic cycles than in other HAs. This proton removal occurring upon glycine molecule incor-
is consistent with the mechanistic synthesis model here poration into the polymers. Moreover, this result sug-
described: the lower steric cluttering of glycine leads to gests that carboxylic moieties are sufficiently close to
a higher compactness of the aromatic center, thereby aromatic rings to allow an electron charge delocalisa-
explaining the observed fluorescence ‘‘self-quenching’’. tion. This is sufficient to account for the fluorescence in-
Moreover, when separately comparing this If/DOC ra- crease since aliphatic carboxylic groups cannot
tio with H/C and N/C ratios (Table 2) for synthetic fluoresce. CTHA fluorescence is governed by carboxylic
substances and natural substances, a strong correlation proton exchange in the lower pH range. The second
between these ratios may be observed: H/C and N/C fluorescence increase (pH > 6) can be attributed to the
evolutions are similar to that of If/DOC among the removal of protons from phenolic moieties. The same
organic fractions. mechanism can be suggested for NHA2 fluorescence
evolution, in a quantitatively different way. NHA1 Potentiometric titration reveals that the synthetic
behaviour can be explained by other amino-aromatic substances exhibit the two primary carboxylic and phe-
group contributions. As suggested by Matthews et al. nolic acidities even though phenolic moieties contribute
(1996), protein fluorescence is directly attributable to to a higher proportion in the total acidity for natural
bacterial activity in natural waters. Moreover, fluores- materials. This suggests that the aromatic parts of the
cent humic substances may be bleached by sunlight in model molecules have a higher degree of polymerisation
surface waters. It is also not surprising to have this kind than the natural compounds isolated in this work. The
of contribution for NHA1 extract collected under ‘‘hot model molecules look close to final products of the oxi-
weather’’ conditions. In summary, as pH increases, car- dation evolution of natural substances. The oxygen pres-
boxylic and phenolic acidities are dissociated. Fluores- sure could also be optimised for future model molecule
cence related to p electrons of the nucleus constitutes a synthesis.
baseline. Since acid site dissociation locally increases The hydrophilic/hydrophobic balance ratio is found
electron delocalisation, fluorescence will increase with to be lower (more hydrophobic) for the synthetic mole-
increasing pH. As mentioned before, a high condensa- cules than for the natural ones. This is consistent with
tion degree, like in the case of CGHA, or an aggregation the previous conclusion about the higher aromatic con-
state like in the case of CGHA at pH 2, may lead to fluo- tribution in model molecules. The hydrophobic charac-
rescence intensity inhibitions due to self-quenching of ter of synthetic humic-like substances is also confirmed.
the fluorophores within the macromolecular structure. Chromophoric moieties absorbing UV/visible light
With increasing pH, fluorophores become ‘‘free’’ to ex- are similar in both synthetic and natural humic sub-
hibit their signals. Acid titration works on CGHA allow stances, despite differences in their optical density ratios.
the assignment of the 520 nm peak to phenolic moieties The carbon network of CGHA is found to have a higher
combined with glycine. In CTHA structures, the 410 nm aromaticity than the other organic species.
peak can be assigned to phenol moieties. By comparison, Fluorescence investigations show that the 3D humic
the 450–470 nm NHA1 peak evolution can be assigned structure layout is the determining parameter for hu-
to phenolic structures associated with tryptophan. The mic-like substance fluorescence properties. The fluores-
poor NHA peak resolution confirms that, even if natural cence emission–pH studies established the qualitative
extracts contain large amounts of aromatic moieties, ionisable fluorophore contribution. According to the
they are loosely structured. evolution of the overall fluorescence pattern with pH,
triglycine based model molecules are closer to natural
aquagenic materials than glycine based substances. The
5. Conclusions comparison between CGHA and CTHA shows that
smaller peptidic chains such as those in glycine based
Elemental analysis shows that glycine and triglycine molecules lead to fluorescence inhibition in the alkaline
based model molecules are very similar in elemental to precipitation pH range.
composition to natural organic materials isolated from Integrating all data, humic-like substances can be
river suspended solids, regardless of the season of sam- viewed as an association of individual subunits consti-
ple collection. All values are found to be within the tuting the final molecular architecture. Subunits are phe-
range reported in the literature, with the exception of tri- nolic, quinoid and benzene carboxylic acids. This
glycine based lower molecular weight (1–3.5 kDa), interpretation is closely related to protein-like self-
which has a somewhat lower carbon content than ex- assembly organisation already suggested by Wershaw
pected. According to elemental analysis, CTHA1 sub- et al. (1977), leading to the recent supramolecular
stances also represent the end-point of the evolution of description for humic substances (Piccolo, 2001; Simp-
humic substances in natural media, where microbial son et al., 2004). Investigating the self-assembly of natu-
derived material constitutes the major component. Syn- rally occurring organic materials is also likely to provide
thetic CGHA substances show an elemental composi- new opportunities to unravel complex naturally occur-
tion related to a soil humic material prior to any ring phenomena such as NOM formation or organic
microbial proteinous contribution. These ratios also and mineral pollutant entrapment by these organic
indicate that NHAs are more aliphatic than the syn- materials.
thetic materials.
FTIR and Py/GC–MS analyses reveal that synthetic
and natural organic materials characterised in this work Acknowledgements
have molecular structures close enough to previously
studied NOMs to give a similar partition of molecular This work was made within the framework of the
fragments. This reinforces the previous statements about programme ‘‘Zone Atelier du bassin de la Moselle’’,
the similarity in the molecular structures of these with grants from Région Lorraine and CNRS (Contrat
products. de Plan Etat – Région and Programme ‘‘Environnement
Vie et Société’’). Field-centrifuge and material sample (Eds.), Constituants Et Propriétés Du Sol, Pédologie 2.
collections provided by Agence de lÕEau Rhin-Meuse Masson, Paris, pp. 105–142.
are gratefully acknowledged. The authors would like Andreux, F., Golebiowska, D., Chone, T., Jacquin, F., Metche,
to thank T. Jacquin (Agence de lÕEau Rhin-Meuse, Roz- M., 1977. Characterisation and transformation in mud
middle of a humic model issued from autoxidation of
erieules, France) for providing technical assistance dur-
catechol/glycine system and selectively marked with 14C. In:
ing field measurements, A. Bruchet and G. Filippi for Anderson, M.A. (Ed.), Soil Organic Matter Studies, vol. 2.
Py/CG-MS analysis (CIRSEE – Ondeo Service, Le Pecq, IAEA, Vienna, pp. 43–57.
France), O. Barrès and P. Dedonato for FTIR spectros- Andreux, F., Golebiowska, D., Metche, M., 1980. Oxidative
copy measurements, J.-M. Portal and B. Gérard polymerization of O-diphenols in presence or absence of
(CRPG, Nancy) for humic acids synthesis elaboration amino-acids. Topics on (catechol–glycocolle) and (catechol–
and elemental analysis. diglycylglycine). General Assembly of Polyphenols Group
Dr. Pellegrino Conte is faithfully acknowledged for Report, Logrono (Spain), Bulletin 9, pp. 178–188
spending a great deal of time reviewing the present Banerjee, S.K., 1979. Acidity, quotient values, and metal
paper. retention power of humic acids of varying molecular weight.
Journal of Indian Society of Soil Science 27, 197–205.
Belin, C., Quelec, C., Lamotte, M., Ewald, M., Simon, P., 1993.
Characterization by fluorescence of the dissolved organic
Appendix matter in natural water. Application to fractions obtained
by tangential ultrafiltration and XAQ resin isolation.
Environmental Technology 14, 1131–1144.
CGHA catechol/glycine humic acid (molecular
Blum, D.J.W., Duguet, J.P., Suffet, I.H., 1993. Estimating the
weight > 8 kDa) activated carbon adsorption of organic chemicals in water.
CTHA1 catechol/triglycine humic acid (molecu- Critical Reviews in Environmental Science and Technology
lar weight 1–3.5 kDa) 23, 193–215.
CTHA2 catechol/triglycine humic acid (molecu- Bonneau, M., Souchier, B., 1977. Constituants Et Propriétés
Du Sol, second ed. Masson, Paris, pp. 1–230.
lar weight > 8 kDa)
Braumann, T., 1986. Determination of hydrophobic parameters
EEM excitation–emission matrices by reversed-phase liquid chromatography: theory, experi-
HA humic acid mental techniques and application in studies on quantitative
HIRP hydrophobicity index determined by re- structure–activity relationships. Journal of Chromatogra-
verse-phase liquid chromatography phy A 372, 191–225.
Bruchet, A., Rousseau, C., Mallevialle, J., 1990. Pyrolysis-GC–
HS humic substance
MS for investigating high-molecular-weight THM precur-
NHA1 natural humic acid (‘‘hot weather’’ ex- sors and other refractory organics. Journal American Water
tract) Works Association 71, 66–74.
NHA2 natural humic acid (‘‘cold weather’’ ex- Brunelot, G., Adrian, P., Rouiller, J., Guillet, B., Andreux, F.,
tract) 1989. Determination of dissociable acid groups or organic
compounds extracted from soils, using automated potenti-
NOM natural organic matter
ometric titration. Chemosphere 19, 1413–1419.
SFS synchronous scan fluorescence spectros- Buffle, J., Deladoey, P., Zumstein, J., Haerdi, W., 1982.
copy Analysis and characterisation of natural organic matters
in freshwaters I. Study of analytical techniques. Schweize-
Associate Editor—Ian Bull rische Zeitschrift für Hydrologie 44, 325–362.
Buurman, P., Van Lagen, B., Piccolo, A., 2002. Increase in
stability against thermal oxidation of soil humic substances
as a result of self association. Organic Geochemistry 33,
References 367–381.
Cabaniss, S.E., Schuman, M.S., 1987. Synchronous fluores-
cence spectra of natural waters: tracing sources of dissolved
Aiken, G.R., McKnight, D.M., Wershaw, R.L., 1985. Humic
organic matter. Marine Chemistry 21, 37–50.
substances in soil, sediment, and water. In: Aiken, G.R.,
Chen, R.F., Bada, J.L., 1992. The fluorescence of dissolved
McKnight, D.M., Wershaw, R.L., MacCarthy, P. (Eds.),
organic matter in the seawater. Marine Chemistry 37, 191–
Geochemistry, Isolation and Characterisation. Wiley-Inter-
221.
science, New York, pp. 1–692.
Chen, Y., Senesi, N., Schnitzer, M., 1977. Information provided
Andreux, F., 1981. Utilisation de molécules modèles de
on humic substances by E4/E6 ratios. Soil Science Society
synthèse dans lÕétude des processus dÕinsolubilisation et de
American Journal 41, 352–358.
biodégradation des polycondensats humiques. Bulletin
Chen, J., Gu, B., Leboeuf, E.J., Pan, H., Dai, S., 2002.
de lÕAssociation Française de lÕEnvironnement et du Sol
Spectroscopic characterisation of the structural and
(AFES), Science du Sol 4, 271–292.
functional properties of natural organic matter fractions.
Andreux, F., Munier-Lamy, C., 1994. Genèse et propriétés des
Chemosphere 48, 59–68.
molécules humiques. In: Souchier, B., Duchaufour, Ph.
Chiou, C.T., Malcolm, R.L., Brinton, T.I., Kile, D.E., 1986. plants in sediments and waters. Water Research 21, 1195–
Water solubility enhancement of some organic pollutants 1206.
and pesticides by dissolved humic and fulvic acids. Envi- Galapate, R.P., Baes, A.U., Ito, K., Mukai, T., Shoto, E.,
ronmental Science and Technology 20, 502–508. Okada, M., 1998. Detection of domestic wastes in Kurose
Chiou, C.T., Kile, D.E., Brinton, T.I., Malcolm, R.L., Leenh- River using synchronous fluorescence spectroscopy. Water
eer, J.A., MacCarthy, P., 1987. A comparison of water Research 32, 2232–2239.
solubility enhancements of organic solutes by aquatic humic Hayase, K., Tsubota, H., 1984. Reversed-phase liquid chroma-
materials and commercial humic acids. Environmental tography of molecular weight-fractionated sedimentary
Science and Technology 21, 1231–1234. fulvic acid. Journal of Chromatography A 295, 530–532.
Clair, T.A., Sayer, B.G., Kramer, J.R., Eaton, D.R., 1996. Hayes, M.H.B., Swift, R.S., 1978. The chemistry of soil organic
Seasonal variation in the composition of aquatic organic colloids. In: Greenland, D.J., Hayes, M.H.B. (Eds.), The
matter in some Nova Scotian brownwaters: a nuclear Chemistry of Soil Constituents. Wiley-Interscience, New
magnetic resonance approach. Hydrobiologia 317, 141– York, pp. 179–230.
150. Hayes, M.B.H., McCarthy, P., Malcolm, R.L., Swift, R.S.,
Coble, P.G., 1996. Characterisation of marine and terrestrial 1989. The search for structure: setting the scene. In: Hayes,
DOM in seawater using excitation–emission matrix spec- M.B.H., McCarthy, P., Swift, R.S. (Eds.), Humic Sub-
troscopy. Marine Chemistry 51, 325–346. stances II: in Search of Structures. Wiley-Interscience, New
Conte, P., Piccolo, A., 1999a. High pressure size exclusion York, pp. 630–664.
chromatography (HPSEC) of humic substances. Molecular Ishiwatari, R., 1975. Chemical nature of sedimentary humic
sizes, analytical parameters, and column performance. acid. In: Povoledo, D., Golterman, H.L. (Eds.), Humic
Chemosphere 38, 517–528. Substances, their Structure and Function in the Biosphere.
Conte, P., Piccolo, A., 1999b. Conformational arrangement of Center for Agricultural Publications and Documentation,
dissolved humic substances. Influence of solution composi- Wageningen, pp. 109–121.
tion on association of humic molecules. Environmental Jokic, A., Wang, M.C., Liu, C., Frenkel, A.I., Huang, P.M.,
Science and Technology 33, 1682–1690. 2004. Integration of the polyphenol and Maillard reactions
Conte, P., Piccolo, A., Van Lagen, B., Buurman, P., Hem- into a unified abiotic pathway for humification in nature:
minga, M.A., 2002. Elemental quantitation of natural the role of d-MnO2. Organic Geochemistry 35, 747–762.
organic matter by CPMAS 13C-NMR spectroscopy. Solid Karickhoff, W., Brown, D.S., Scott, T.A., 1979. Sorption of
State Nuclear Magnetic Resonance 21, 158–170. hydrophobic pollutants on natural sediments. Water
Conte, P., Spaccini, R., Piccolo, A., 2004. State of the art of Research 13, 241–248.
CPMAS 13C-NMR spectroscopy applied to natural organic Kingery, W.L., Simpson, A.J., Hayes, M.H.B., Locke, M.A.,
matter. Progress in Nuclear Magnetic Resonance Spectros- Hicks, R.P., 2000. The application of multidimensional
copy 44, 215–233. NMR to the study of soil humic substances. Soil Science
Davies, E., Dobbs, M., 1984. The prediction of bioconcentra- 165, 483–494.
tion in fish. Water Research 18, 1253–1263. Kononova, M.M., 1966. Soil Organic Matter, second ed.
De Souza-Sierra, M.M., Donard, F.X., Lamotte, M., 1997. Pergamon Press, Oxford, pp. 101–104.
Spectral identification and behaviour of dissolved organic Kujawinski, E.B., Freitas, M.A., Zang, X., Hatcher, P.G.,
fluorescent material during estuarine mixing processes. Green-Churuch, K.B., Jones, R.B., 2002. The application of
Marine Chemistry 58, 51–58. electrospray ionisation mass spectrometry (ESI-MS) to the
Egeberg, P.K., Christy, A.A., Eikenes, M., 2002. The molecular structural characterisation of natural organic matter.
size of natural organic matter (NOM) determined by Organic Geochemistry 33, 171–180.
diffusimetry and seven other methods. Water Research 36, Laane, R.W.P.M., 1982. Influence of pH on the fluorescence of
925–932. dissolved organic matter. Marine Chemistry 11, 295–401.
Engebretson, R.R., von Wandruska, R., 1999. Effects of humic Laor, Y., Strom, P.F., Farmer, W.J., 1998. Bioavailability of
acid purification on interactions with hydrophobic organic phenanthrene sorbed to mineral-associated humic acid.
matter: evidence from fluorescence behaviour. Environmen- Water Research 7, 1719–1729.
tal Science and Technology 33, 4299–4303. Lead, J.R., Balnois, E., Hosse, M., Menghetti, A., Wilkinson,
Fang, F., Kanan, S., Patterson, H.H., Cronan, C.S., 1998. A K.J., 1999. Characterisation of Norwegian natural organic
spectrofluorimetric study of the binding of carbofuran, matter: size, diffusion coefficients and electrophoretic mobil-
carbaryl and aldicarb with dissolved organic matter. Anal- ities. Environment International 25, 245–258.
ytica Chimica Acta 373, 139–151. Mathur, S.P., Schnitzer, M., 1978. A chemical and spectro-
Flaig, W., Beutelspacher, H., Rietz, E., 1975. Chemical scopic characterisation of some synthetic analogues of
composition and physical properties of humic substances. humic acids. Soil Science Society of America Journal 42,
In: Gieseking, J.E. (Ed.), Soil Components, Organic Com- 591–596.
ponents. Springer, Berlin, pp. 1–211. Matthews, B.J.H., Jones, A.C., Theodorou, N.K., Tudhope,
Fukushima, T., Ishibashi, T., Imai, A., 2001. Chemical char- A.W., 1996. Excitation–emission-matrix fluorescence spec-
acterisation of dissolved organic matter in Hiroshima Bay, troscopy applied to humic acid bands in coral reefs. Marine
Japan estuary. Coastal and Shelf Science 53, 51–62. Chemistry 55, 317–332.
Gadel, F., Bruchet, A., 1987. Application of pyrolysis-gas Miano, T.M., Sposito, G., Martin, J.P., 1988. Fluorescence
chromatography–mass spectrometry to the characterisa- spectroscopy of model humic acid-type polymers. Soil
tion of humic substances resulting from decay of aquatic Science Society of American Journal 52, 1016–1027.
Mopper, K., Schultz, C.A., 1993. Fluorescence as a possible approaches to the lowest energy conformer. Environmental
tool for studying the nature and water column distribution Science and Technology 33, 546–552.
of DOC components. Marine Chemistry 41, 229–238. Séné, C.F.B., MacCann, M.C., Wilson, R.H., Grinter, R., 1994.
Mounier, S., Patel, N., Quilich, L., Benaim, J.Y., Benamou, C., Fourier-transform Raman and Fourier-transform infrared
1999. Three-dimensional fluorescence of the dissolved spectroscopy. An investigation of five higher plant cell walls
organic carbon in the Amazon River. Water Research 33, and their components. Plant Physiology 106, 1623–1631.
1523–1533. Senesi, N., Miano, T.M., Provenzo, M.R., Brunetti, G., 1991.
Münster, U., 1999. Amino acid profiling in natural organic Characterisation, differentiation and classification of humic
matter isolated by reverse osmosis from eight different substances by fluorescence spectroscopy. Soil Science 152,
boreal freshwaters. Environment International 25, 209–224. 259–271.
Niemeyer, J., Chen, Y., Bollag, J.-M., 1992. Characterisation of Simpson, A.J., 2001. Multidimensional solution state NMR of
humic acids, composts, and peat by diffuse reflectance humic substances: a practical guide and review. Soil Science
Fourier-transform infrared spectroscopy. Soil Science Soci- 166, 795–809.
ety of America Journal 56, 135–140. Simpson, A.J., 2002. Determining the molecular weight, aggre-
Norden, M., Dabek-Zlotorzynska, B., 1996. Study of metal– gation, structures and interactions of natural organic matter
fulvic acid interactions by capillary electrophoresis. Journal using diffusion ordered spectroscopy. Magnetic Resonance
of Chromatography A 739, 421–429. in Chemistry 40, S72–S82.
Parlanti, E., Wörz, K., Geoffroy, L., Lamotte, M., 2000. Simpson, A.J., Lefebvre, B., Naser, A., Williams, A., Larim, N.,
Dissolved organic matter fluorescence spectroscopy as a Kvaska, M., Kingery, W.L., Kelleher, B., 2004. Identifying
tool to estimate biological activity in a coastal zone residues in natural organic matter through spectral predic-
submitted to anthropogenic inputs. Organic Geochemistry tion and pattern matching of 2D NMR datasets. Magnetic
31, 1765–1781. Resonance in Chemistry 42, 14–22.
Peuravuori, J., Pihlaja, K., 1997. Molecular size distribution Stevenson, F.J., 1982. Humus Chemistry, Genesis, Composi-
and spectroscopic properties of aquatic humic substances. tion, Reactions. Wiley-Interscience, New York, pp. 195–213.
Analytica Chimica Acta 337, 133–149. Stevenson, F.J., 1994. Humus Chemistry: Genesis, Composi-
Piccolo, A., 2001. The supramolecular structure of humic tion, Reactions, second ed. Wiley and Sons, New York, pp.
substances. Soil Science 166, 810–832. 1–496.
Piccolo, A., Conte, P., Trivellone, E., Van Lagen, B., Buurman, Thurman, E.M., Malcolm, R.L., 1981. Preparative isolation of
P., 2002. Reduced heterogeneity of a lignite humic acid by aquatic humic substances. Environmental Science and
preparative HPSEC following interaction with an organic Technology 15, 463–466.
acid. Characterization of size-separates by Pyr-GC–MS and Tyson, C.A., Martell, A.E., 1972. Kinetic studies on the
1H-NMR spectroscopy. Environmental Science and Tech- autoxidation of 3,5-dibutylpyrocatechol. Journal of Physi-
nology 36, 76–84. cal Chemistry 74, 2601–2609.
Prélot, B., 2001. Mesure et modélisation de lÕhétérogénéité Villieras, F., Michot, L.J., Bardot, F., Cases, J.M., François,
énergétique à lÕinterface oxyde/électrolyte/métaux. Ph.D. M., Rudzinski, W., 1997. An improved derivative method to
thesis, Institut National Polytechnique de Lorraine, pp. 1– study sulfate heterogeneity of clays minerals. Langmuir 13,
246 1104–1117.
Pretsch, E., Bühlmann, P., Affolter, C., 2000. Structure Deter- Wershaw, R.L., 1986. A new model for humic materials and
mination of Organic Compounds. Springer, Berlin, pp. 1– their interactions with hydrophobic organic chemicals in
240. soil–water or sediment–water systems. Journal of Contam-
Reynolds, D.M., Ahmad, S.R., 1997. Rapid and direct deter- inant Hydrology 1, 29–45.
mination of wastewater BOD using fluorescence technique. Wershaw, R.L., Pinckney, D.J., Booker, S.E., 1977. Chemical
Water Research 31, 2012–2018. structure of humic acids. Part 1 – a generalized structural
Schnitzer, M., 1977. Recent Findings on the Characterisation of model. Journal of Research of US Geological Survey 5,
Substances from Soils from Different Climatic Zones. 565–569.
International Atomic Energy Agency, Vienna, p. 117. Wojtkowiak, B., Chabanel, M., 1977. Spectrochimie Molécu-
Schnitzer, M., Khan, U.S., 1972. Humic Substances in the laire. Technique et Documentation. Lavoisier, Paris, pp. 1–
Environment. Marcel Dekker, New York, pp. 1–327. 310.
Schnitzer, M., Skinner, S.I.M., 1968. Alkali versus acid Wu, F.C., Evans, R.D., Dillon, P.J., 2002. High-performance
extraction of soil organic matter. Soil Science 105, 392–396. liquid chromatographic fractionation and characterisation
Schulten, H.R., Schnitzer, M., 1993. A state of the art of the of fulvic acid. Analytica Chimica Acta 464, 47–55.
structural concept for humic substances. Naturwissenschaf- Zhang, S., Marini, D.M., Hwang, W., Santoso, S., 2002. Design
ten 80, 29–30. of nanostructured biological materials through self-assem-
Schulten, H.R., Schnitzer, M., 1995. Three dimensional models bly of peptides and proteins. Current Opinion in Chemical
for humic acids. Naturwissenschaften 11, 487–498. Biology 6, 865–871.
Schulten, H.R., Schnitzer, M., 1997. Chemical model structures Zolsnay, A., Baigar, E., Jimenez, M., Steinweg, B., Sacco-
for soil organic matter and soils. Soil Science 162, 115–130. mandi, F., 1999. Differentiating with fluorescence spectros-
Sein, L.T., Varnum, J.M., Jansen, S.A., 1999. Conformational copy the sources of dissolved organic matter in soils
modelling of a new building block of humic acid: subjected to drying. Chemosphere 38, 45–50.
Environment International 35 (2009) 1225–1233
Contents lists available at ScienceDirect
Environment International
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / e n v i n t
Review article
Impacts of climate change on surface water quality in relation

to drinking water production
I. Delpla, A.-V. Jung, E. Baures, M. Clement, O. Thomas ⁎
School of Public Health (École des Hautes Études en Santé Publique), Laboratoire d'Étude et de Recherche en Environnement et Santé (LERES), Avenue Professeur Léon Bernard,
35043 Rennes Cedex, France
a r t i c l e i n f o a b s t r a c t
Article history: Besides climate change impacts on water availability and hydrological risks, the consequences on water
Received 13 March 2009 quality is just beginning to be studied. This review aims at proposing a synthesis of the most recent existing
Accepted 3 July 2009 interdisciplinary literature on the topic. After a short presentation about the role of the main factors
Available online 29 July 2009
(warming and consequences of extreme events) explaining climate change effects on water quality, the focus
will be on two main points. First, the impacts on water quality of resources (rivers and lakes) modifying
Keywords:
parameters values (physico-chemical parameters, micropollutants and biological parameters) are consid-
Climate change
Water quality
ered. Then, the expected impacts on drinking water production and quality of supplied water are discussed.
Warming The main conclusion which can be drawn is that a degradation trend of drinking water quality in the context
Extreme events of climate change leads to an increase of at risk situations related to potential health impact.
Floods © 2009 Elsevier Ltd. All rights reserved.
Droughts
Organic carbon
Nutrients
Micropollutants
Cyanotoxins
Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1226
1225
2. Impacts on water quality parametes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1226
2.1. Basic parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1226
2.2. Dissolved Organic Matter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1226
2.3. Nutrients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1228
2.4. Inorganics micropollutants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1228
2.5. Organics micropollutants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1229
2.6. Pathogens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1229
2.7. Cyanobacteria and cyanotoxins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1229
2.8. Water quality indicators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1229
2.9. Synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1229
3. Expected impacts on drinking water production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1230
3.1. DBPs determinants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1230
3.2. Potential impacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1230
3.3. Monitoring and modeling of impacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1231
3.4. Synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1231
4. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1232
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1232
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1232
⁎ Corresponding author. Tel.: +33 2 99 02 29 20; fax: +33 2 99 02 29 29.

E-mail address: Olivier.Thomas@ehesp.fr (O. Thomas).
0160-4120/$ – see front matter © 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.envint.2009.07.001
1226 I. Delpla et al. / Environment International 35 (2009) 1225–1233
1. Introduction (Soh et al., 2008). Phototransformation should be seriously taken into

account when evaluating the possibility of formation of UV transfor-
Floods and droughts are the main impacts of climate change on mation products from organic micropollutants such as pharmaceu-
water availability. Besides these quantitative impacts, surface water ticals (Canonica et al., 2008). Many papers consider pharmaceuticals
quality is also affected by climate change. For example, it seems to be photo reactive substances (Boreen et al., 2003; Buerge et al.,
obvious that a drought may imply at least a modification of surface or 2006; Petrovic and Barceló, 2007).
ground water quality (concentration) sometimes leading to water This paper aims at reviewing the main impacts on water quality
supply limitation. If surface water withdrawal can be directly affected parameters, generally described for surface water (rivers and lakes),
by water quality degradation, wells pumping can be cut off for sanitary and the expected impacts on drinking water production.
reasons (groundwater quality) as well as for security reasons (floods
threats). However, even if these facts are well known, few scientific 2. Impacts on water quality parametes
works have been published until recently on the impacts of climate
change on water quality modification. Water quality parameters can be classified according to i) physico-
Actually, climate change is not the only factor affecting water quality. chemical basic parameters (temperature, pH, dissolved oxygen,
Integrated into the global change concept, land use evolution, dissolved organic matter…) and nutrients, ii) micropollutants
deforestation, urban spreading and area waterproofing may also (inorganic and organic) including metals, pesticides and pharmaceu-
contribute to water quality degradation. But more often, water pollution ticals, and iii) biological parameters with pathogens microorganisms,
is directly linked to human activities of urban, industrial or agricultural cyanobacteria and water quality proxies (Table 1).
origin, and climate change could lead to degradation in surface water
quality as an indirect consequence of these activities. When point source 2.1. Basic parameters
pollution is reduced in many countries (even if wastewater treatment
plants begin to reach their capacity limits), climate (global) change A rise in surface water temperatures was observed since the 1960s in
impacts could tend to increase the diffuse pollution with for example Europe, North America and Asia (0.2–2 °C), mainly due to atmospheric
urban or agricultural runoff. The climate change determinants affecting warming in relation to solar radiation increase (Bates et al., 2008). In
water quality are mainly the ambient (air) temperature and the increase European rivers, Zwolsman and van Bokhoven (2007), and Van Vliet and
of extreme hydrological events. Soil drying-rewetting cycles and solar Zwolsman (2008) observed an average increase in water temperature of
radiation increase may also be considered. around 2 °C respectively in Rhine and Meuse rivers after the severe
First of all, temperature (in general) must be viewed as the main drought of 2003, with a pH increase (reflecting a decrease in CO2
factor affecting almost all physico-chemical equilibriums and biologi- concentration), and a decrease in dissolved oxygen (DO) solubility
cal reactions. It is well known that all physico-chemical “constants” reflecting a lower DO solubility under higher water temperatures. A DO
vary with temperature, and frequently increasing endothermic decrease can also be associated to an increase in DO assimilation of
reactions. According to Arrhenius relation, kinetic of a given chemical biodegradable organic matter by microorganisms (linked to an increase
reaction can be doubled for a temperature increase of 10 °C. in Dissolved Organic Carbon (DOC)) (Prathumratana et al., 2008). In the
Consequently, several transformations or effects related to water will same study dealing with the surface water quality in the lower Mekong
be favoured by water temperature increase such as dissolution, River, negative significant correlations were generally found between
solubilisation, complexation, degradation, evaporation, etc. This precipitation (or discharge flow) and DO, pH and conductivity (from 0.2
phenomenon globally leads to the concentration increase of dissolved to 0.9). In several lakes in Europe and Northern America, the stratified
substances in water but also to the concentration decrease of dissolved period has lengthened by 2–3 weeks and water temperatures have risen
gazes. This last point is very important with respect to dissolved of 0.2 to 1.5 °C, which have an influence on thermal stratification
oxygen in water. In fact, its saturation concentration decreases of (Komatsu et al., 2007) and lakes hydrodynamics (Bates et al., 2008).
almost 10% with a 3 °C increase (10 mg/L at 15 °C). Remind that, Computer models predict an increase of around 2 °C by 2070 in
whatever the IPCC scenario, the average global air temperature should European lakes, although this rise will also depend on lake character-
increase between 1.8 and 4.0 °C (Bates et al., 2008) during the 21st istics and season (George et al., 2007; Malmaeus et al., 2006). It has been
century. Moreover, a drying tendency in summer is expected, demonstrated that shallow lakes are likely to be the most vulnerable to
particularly in subtropics, low and mid-latitudes, in addition with an climate change. Water temperatures have an impact on internal lake
extreme events increase in general (Bates et al., 2008). processes like diffusion, mineralization and vertical mixing (Malmaeus
Floods and droughts will also modify water quality by direct effects et al., 2006). Residence time of lakes would probably increase in summer
of dilution or concentration of dissolved substances. For low river flow by 92% in 2050 for lakes with short residence times George et al. (2007).
rates, the main effect on water quality is as for a temperature increase, Also, it is predicted that especially shallow lakes will experience an
a concentration increase of dissolved substances in water but a increase of temperature in epilimnion and hypolimnion during summer
concentration decrease of dissolved oxygen (Prathumratana et al., (Jöhnk et al., 2008), although man-made lakes (in The Netherlands)
2008; Van Vliet and Zwolsman, 2008). A correlative positive effect is respond even more directly to weather variations (Mooij et al., 2005).
the concentration decrease of some pollutants due to a low water Nevertheless, deepest lakes are most sensitive to climate warming on a
velocity (aquatic plants assimilation of nutrients and adsorption/ long period of time due to their greater heat storage capacity and will
complexation of heavy metals on suspended matter and settling). consequently show the highest winter temperatures (George et al.,
These phenomenons will be detailed hereafter. For heavy rain falls and 2007). An increase in water temperature has also an impact on lakes
strong hydrologic conditions, runoff and solid material transportation chemical processes with increases in pH and greater in-lake alkalinity
are the main consequences. For countries in the temperate zone, generation (Psenner and Schmidt, 1992). Concerning the impacts of the
climate change will decrease the number of rainy days but increase predicted increase in winter precipitations on lake waters, it depends on
the average volume of each rainfall event (Brunetti et al., 2001; Bates the lake size. Small lakes with short residence times will be particularly
et al., 2008)). As a consequence, drought–rewetting cycles may impact sensitive to a change in rainfalls (George et al., 2007).
water quality as it enhances decomposition and flushing of organic
matter into streams (Evans et al., 2005). 2.2. Dissolved Organic Matter
Solar irradiation increase could also alter water quality and
especially characteristics of natural organic matter in freshwaters Dissolved Organic Matter (DOM) affects the functioning of aquatic
systems both by warming and UVB radiation (increasing photolysis) ecosystems through its influence on acidity, trace metal transport,
Table 1
Impacts of climate change on water quality parameters.
Water quality parameter CC factors Water body Reference Comments

affecting WQ
Physico- Basic pH Droughts River Van Vliet and Zwolsman, 2008 Higher maximum values in Meuse river.
chemicals parameters Temperature Lakes Bates et al., 2008 Increase in pH.
increase
Rainfalls River Prathumratana et al., 2008 Negatively correlated (Mekong).
DO Droughts and Rivers and lakes Zwolsman and van Bokhoven, 2007; Lower dissolved oxygen solubility and concentration.
temperature Komatsu et al., 2007;
increase Van Vliet and Zwolsman, 2008
Rainfalls River Prathumratana et al., 2008 Negatively correlated (Mekong).
Temperature Droughts and Rivers Zwolsman and van Bokhoven, 2007; Warming of the water column.
temperature Van Vliet and Zwolsman, 2008;
increase Ducharne et al., 2007; Bates et al., 2008
Lakes Jöhnk et al., 2008; Komatsu et al., 2007; Increase in the epilimnion and hypolimnion, especially in shallow lakes.
Malmaeus et al., 2006; Mooij et al., 2005; Increase of temperature, stability of the water column and summer water residence times.
George et al., 2007
DOC Temperature and Streams and lakes Monteith et al., 2007; Evans et al., 2005; DOC increase (UK, Scandinavia, Czech Republic, Northeastern USA and Canada).
rainfalls increase (peatlands) Worral et al., 2004; Hejzlar et al., 2003
Clark et al., 2008; Prathumratana et al., 2008 DOC flux increase during storm events (UK and Mekong basin).
Nutrients Droughts River Van Vliet and Zwolsman, 2008; Sediment release during droughts. Ammonium increase.
Zwolsman and van Bokhoven 2007
Temperature Surface and Wilhelm and Adrian, 2008 Increase mineralization and release of N, C and P from soil OM.
increase groundwater
Lakes Wilhelm and Adrian, 2008; Jackson et al., 2007; Sediment nutrients release and pulses into the euphotic zone.
Malmaeus et al., 2006; Komatsu et al., 2007; Strong changes in nutrient loading in shallow lakes.
Petterson et al., 2003 Elevated phosphate and ammonium concentrations in the hypolimnion during the warm period.
Temperature and River basins, lake, Kaste et al., 2006; Arheimer et al., 2005; Increase N river loading (Norway) and nitrate load (UK river and Seine groundwater).
rainfalls increase fjord and Ducharne et al., 2007; Zhu et al., 2005; Increase N and P (Sweden) and nutrients (Canada) loading.
groundwater Wilby et al., 2006; Weyhenmeyer, 2008 Climate change effects could be comparable between small and large lakes.
Heavy rainfalls Stream and lakes Drewry et al., 2009; Mooij et al., 2005; Australia, Mekong basin. Increased P loading of Dutch lakes.
Prathumratana et al., 2008
Forested watershed Bhat et al., 2007 Total Kjeldahl nitrogen (TKN) loading increase (USA).
Micropollutants Inorganics Metals Droughts River Van Vliet and Zwolsman, 2008; Selenium, Baryum, mercury, zinc, cadmium, lead and nickel increase.
Zwolsman and van Bokhoven, 2007
Temperature High alpine lakes Thies et al., 2007 Increasing amount of snowmelt lead to a micropollutants increase (European Alps).
increase
Heavy rainfalls Streams Rothwell et al., 2007 Correlations between DOC and metals mobilisation (UK).
Temperature and Stream Pédrot et al., 2008 Humic and fulvic acids mobilized various trace elements (Brittany).
rainfalls increase Olivie-Lauquet et al., 2001 Trace element release coincide with a decline in redox potential and increase
I. Delpla et al. / Environment International 35 (2009) 1225–1233
of organic carbon content (Brittany)

Organics Pesticides Temperature and Surface & Bloomfield et al., 2006 Changes in rainfalls intensity and seasonality and increased temperatures are main
rainfalls increase groundwater climate drivers for changing pesticides fate and behaviour.
Drying and Lennartz and Louchart, 2007 Soil drying increased the biding of herbicidal compounds.
rewetting cycles
Rainfalls increase Streams Probst et al., 2005 Increased precipitations lead to an increase of pesticides flux.
Pharmaceuticals Temperature Groundwater Massman et al., 2006 Artificial recharge pond. Change in redox conditions (anaerobic conditions)
increase with an increase in temperature (Deutchland).
Rainfalls Streams Lissemore et al., 2006 Correlations between DOC and pharmaceuticals concentrations in water.
Oppel et al., 2004 Clofibric acid and iopromide are very mobile and could contaminate groundwaters through rivers.
Biologicals Pathogens Temperature and Surface waters Charron et al., 2004; Curriero et al., 2001 Half the waterborne disease outbreaks in USA during the last half century followed a period of
rainfalls increase Hunter, 2003; Pednekar et al., 2005 extreme rainfall.
Nearly 70% of the variability in the coliform record is due to seasonal and
interannual variability in local rainfall.
Cyanobacteria Temperature and Lakes Arheimer et al., 2005; Jöhnk et al., 2008; Summer heatwaves boost the development of Cyanobacteria blooms (Nederlands and Sweden).
rainfalls increase Hunter, 2003
Cyanotoxins Temperature Lakes Brient et al., 2008; Wiedner et al., 2007 Cylindrospermopsis Raciborskii northern spread. Microcystin blooms during hot summers.
increase
Fishes, green algae, diatoms Temperature Freshwaters Daufresne et al., 2003; Daufresne and Boët, 2007 Pollutants uptake rate increase due to an increased metabolic rate and decrease in oxygen
increase solubility. Species predominance and abundance changes.
1227
Others Temperature Soils Sardans et al., 2008 Increase enzymatic activity.

increase
light absorbance and photochemistry and, energy and nutrient supply in stratified lakes is expected to increase, due to declining oxygen
(Evans et al., 2005). The principal source of DOM in surface waters is concentrations in the bottom waters (Wilhelm and Adrian, 2008).
soil leaching (Hejzlar et al., 2003). Furthermore, positive spatial Regional and global climate scenarios and models are useful tools
relationships between Dissolved Organic Carbon (DOC) export and to produce data inputs for hydrological models in order to understand
wetland areas like peatlands have been demonstrated (Evans et al., and predict the potential effects of climate change on water bodies. An
2005). Since the 1980s, various studies have shown significant DOC increase in dry summertime frequency may lead to gradually mobilize
increases in Northern Europe (Evans et al., 2005; Monteith et al., nitrogen in soils that would be flushed into streams at the beginning
2007; Worrall et al., 2004), Central Europe (Hejzlar et al., 2003) and of the wet season and cause higher rivers nitrate concentrations
Northern America (Monteith et al., 2007). Many potential factors (air (Wilby et al., 2006). Ducharne et al. (2007) predict an increase in
temperature, increase in rainfalls intensity, atmospheric CO2 increase nitrate concentration in the Seine basin aquifer layers for the years
and decline in acid deposition) have been proposed to explain these 2050 and 2100 due to an increase in precipitations and consequently
trends in DOC, although there is no scientific consensus. Evans et al. in soil leaching. Kaste et al. (2006) and Arheimer et al. (2005)
(2005) have shown that recovery from acidification and water respectively predict a 40–50% increase in nitrate flux by 2070–2100 in
temperature are potential drivers, since many compounds forming a Norwegian river basin, and an increase in phosphorus (50%) and
part of DOC are acidic. In fact, a decrease in acid deposition is observed nitrogen (20%) in a lake. Correlations between precipitations, air
resulting partly from a decrease in anthropogenic sulphur emissions temperature, discharge flow and phosphates, nitrates and Total
(industries, passengers/goods transportation…) (Monteith et al., Phosphorus (TP) in the Mekong River have also been observed
2007; Evans et al., 2008). This could lead to an increase in soil pH (Prathumratana et al., 2008). These results are in accordance with
and consequently to an organic acids increase permitted by new redox Bhat et al. (2007) who found that 73% of the total Kjeldhal nitrogen
conditions. Nevertheless, trends in DOC are probably resulting from a load at a forested watershed outlet was exported by surface runoff
combination of various factors, including acid deposition, since during storm events. Drewry et al. (2009) also found positive
increasing trends have begun in a few places before reduction in correlations between TP, total nitrogen, suspended solids and flow. It
acid deposition (Worrall and Burt, 2007). was also suggested that a major part of the phosphorus is adsorbed
According to Clark et al. (2008), a variation in stream flow can be a onto suspended solids.
good indicator of changes in DOC concentration in streams draining For lakes, higher phosphate and ammonium concentrations in the
organo-minerals soils, although the same is false for peat soils (in this hypolimnion are frequently observed during the warm period in
case, temperature is better). Finally, Prathumratana et al. (2008) temperate countries (Petterson et al., 2003). Climate change impact
shows that COD (Chemical Oxygen Demand), used as an indicator of these ecosystems with various manners: changes in temperature, ice-
Natural Organic Matter (NOM), have weak to fair significant cover, wind and precipitation (Mooij et al., 2005). P loading
correlations with precipitations and discharge flows for the Mekong exportation, which is governed by discharges following heavy rain
River (from 0.3 to 0.4). falls, will tend to increase with climate change and consequently have
Finally, Prathumratana et al. (2008) shows that COD (Chemical an impact on lakes (Mooij et al., 2005). Conversely, streams nitrogen
Oxygen Demand), used as an indicator of Natural Organic Matter concentrations are less dependent on stream discharge (Mooij et al.,
(NOM), have weak to fair correlations with precipitations (0.295– 2005). Increasing temperatures is supposed to decrease nitrate
0.426) and discharge flows (0.312–0.324). concentrations in lakes with an increase in denitrification rate and
higher N losses in upstream-situated soil and surface waters (Mooij et
2.3. Nutrients al., 2005). On the contrary, the internal P loading increases thanks to
microbial decomposition of lake sediments (Jackson et al., 2007).
An increase of N mineralization in soil due to an increase in mean Accumulation of soluble hypolimnetic phosphorus depends on
soil temperature is expected (Ducharne et al., 2007). Moreover, thermocline depth and hypolimnetic temperatures (Wilhelm and
droughts increase the soil extractible Total Organic Carbon (TOC) Adrian, 2008). Indeed, higher hypolimnetic temperatures increase
concentration in winter and warming increases extractible nitrate in both mineralization of organic hypolimnetic matter and phosphorus
summer and autumn and extractible ammonium in winter. A release from sediments. Dramatic nutrient pulses into the euphotic
moderate increase in soil temperature (spring, summer and winter) zone could be observed after heatwaves (Wilhelm and Adrian, 2008).
could lead to a large increase in enzymatic activity. Temperature is Hence, alternating mixing and long stratification events threat more
positively correlated with nitrification process (increasing phospha- especially polymictic lakes than dimictic lakes (Wilhelm and Adrian,
tases activity and P mobilisation in soils). Changes observed in 2008). P increases in the surface layer, fueling phytoplankton growth
enzymatic activity are linked with direct effect of soils warming which (Jackson et al., 2007), leading to algal blooms and a deterioration of
stimulates biological activity and increases N availability (Sardans et water quality (Komatsu et al., 2007). Lastly, concerning Total Phos-
al., 2008). Soil warming increases soil extractable nitrates concentra- porous (TP) concentrations, higher temperatures may impact mainly
tion in summer and autumn (N losses facilitated) and concentration of lakes with long residence times (Malmaeus et al., 2006), even though
extractable ammonium in winter. rates of change of phosphate and nitrates concentrations seem to be
Water bodies quality is subjected to weather seasonality which has independent of lake morphometry (Weyhenmeyer, 2008).
an important impact on their nutrient patterns (Zhu et al., 2005). A
warmer climate will create indirect impacts on water bodies like an 2.4. Inorganics micropollutants
increase nutrients load in surface and groundwater (Van Vliet and
Zwolsman, 2008) and counteract policies effects of external nutrient In Western Europe, metal concentrations in rivers have greatly
loading reduction (Wilhelm and Adrian, 2008). Indeed, higher decreased in the past decades with industrial and urban wastewater
temperatures will increase mineralization and releases of nitrogen, treatment efforts. Nevertheless, droughts may have impacts on river
phosphorus and carbon from soil organic matter. Moreover, an water quality (Zwolsman and van Bokhoven, 2007; Van Vliet and
increase in runoff and erosion due to greater precipitations intensity Zwolsman, 2008), depending on the compound properties that could
should result in an increase in pollutants transport, especially after a be as well either negative or positive. First of all, significantly higher
drought period. Higher ammonium concentrations could be observed concentrations for barium, selenium and nickel were observed in river
in rivers with a reducing dilution capacity caused by droughts Meuse during the drought of 2003 (Van Vliet and Zwolsman, 2008).
(Zwolsman and van Bokhoven, 2007; Van Vliet and Zwolsman, Conversely, significantly lower concentrations of total lead, chromium,
2008). Furthermore, release of phosphorus from bottom sediments mercury and cadmium were measured within the same period. These
I. Delpla et al. / Environment International 35 (2009) 1225–1233 1229
differences are mainly due to dissimilarities between adsorption increase in the environment, although there is still no clear evidence
capacities by suspended solids but discrepancies exist between (Hunter, 2003).
studies. Indeed, in the Rhine river, it was observed that droughts Floods often led to a contamination of groundwater and additional
have a negative impact on metal concentrations of cadmium, disease outbreaks like Acanthamoeba keratitis in Iowa (USA) in 1994
chromium, mercury, lead, copper, nickel and zinc which were higher (Hunter, 2003). According to Curriero et al. (2001), half of the
during the 2003 drought than during reference periods (Zwolsman waterborne disease outbreaks in the US during the last half century
and van Bokhoven, 2007). followed a period of extreme rainfall. Even though the risk of diseases
Thies et al. (2007) have studied high alpine lake waters (Alps) outbreaks linked to mains drinking waters is low in developed
response to climate warming and observed a solute release from an countries, private supplies would be at risk (Hunter, 2003). In
active rock glacier ice. Surface waters over metamorphic rocks were addition, an increase in temperature threats water quality with regard
affected by the rising export of ions and heavy metals from meltwater. to waterborne diseases especially cholera disease in Asia and South
They predicted that high mountain freshwater will thus become America (Hunter, 2003). Lastly, it was shown that with increased UV
increasingly affected by climate warming. radiation due to ozone layer depletion, NOM trap higher levels of UV
Furthermore, a strong complexation of some metals by DOC could energy and breaks down to more bioavailable organic compounds,
lead to a transport of dissolved lead, titane and vanadium in peatland minerals and micronutrients. All these processes could stimulate
systems after a stormflow (Rothwell et al., 2007). A seasonal change in bacterial activity in aquatic ecosystems (Soh et al., 2008).
dissolved metal concentrations was also observed for various trace
elements (Fe, Mn, Al, La, U, Th, Cd and As). An increase of organic 2.7. Cyanobacteria and cyanotoxins
carbon content and a decline in redox conditions seem to be related
with a trace elements release. A positive correlation is also found Competition between phytoplankton and cyanobacteria could be
between storm events and trace element concentrations in streams switch in favor of cyanobacteria in a warmer climate (Arheimer et al.,
(Olivie-Lauquet et al., 2001). In fact, organic and inorganic colloids 2005) and could also increase their dominance. A higher phosphorus
could play an important role in trace elements mobilisation in soils flux in the epilimnion can promote phytoplankton growth in the
and water (Pédrot et al., 2008). euphotic layer and lead to an evolution from a macrophyte-dominated
clear water state to a phytoplankton-dominated turbid state. Increase in
2.5. Organics micropollutants water temperatures and nutrient concentration causes massive cyano-
bacteria bloom in many waterbodies (Hunter, 2003). Summer heat-
Surface waters are the main receptors for pesticides contamination waves could also boost the cyanobacteria development in lakes through
from the agricultural use. Bloomfield et al. (2006) observed that reducing vertical turbulent mixing and increasing growth rates (Jöhnk
changes in rainfall seasonality and intensity and increased air et al., 2008). Moreover, new cyanobacterium species as Cylindrosper-
temperatures are the main climate drivers for changing pesticides mopsis raciborskii have colonized northern habitats due to effects of
fate and behaviour, although effects of climate change are likely to be rising temperatures. This tropical cyanobacterium, known to produce
variable and difficult to predict. Cylindrospermopsin, is now detected in South and Western Europe
Lennartz and Louchart (2007) has studied the physico-chemical freshwaters (Italy, Spain and France) (Brient et al., 2008) and has been
interactions between soil organic matter and herbicidal compounds detected in German lakes (Wiedner et al., 2007). Moreover, an earlier
(diuron and terbuthylazine) after drying and rewetting cycles to annual warming in temperate countries permits an earlier and more
examine impacts of climate induced soil water status variations. important growth of this alga (Wiedner et al., 2007). Lastly, other
Results show that variations in soil water contents modify the soil cyanobacteria, like Microcystis which can produce microcystin, could
organic matter structure, which hinder diffusion and trap pesticides. become invasive with climate warming (Jöhnk et al., 2008).
An increase in extreme events with climate change will probably
counteract pesticides reduction measures. Probst et al. (2005) 2.8. Water quality indicators
simulate pesticides entries into streams and found, with a heavy
rainfalls scenario (precipitation increase from 10 to 20 mm/day), that Fishes, green algae and diatoms are often used as water quality
isoproturon and bifenox could potentially present a greater risk due to indicators. Daufresne and Boët (2007) observed an increase related to
their ecotoxicity. global warming in total abundance and in proportions of warm-water
For pharmaceuticals, in a Southern Ontario watershed, Lissemore species and size-structures changes in fish communities in French
et al. (2006) found significant correlations between DOC and some rivers. Southern thermophilic fish species progressively replaced
active substances frequently detected in water (monensin and northern cold water species in the upper Rhône River (Daufresne
carbamazepine), with concentration variations of monensin, linco- et al., 2003). Furthermore, high temperature and low turbulent
mycin, sulfamethazine, trimethoprim and carbamazepine depending diffusivities in lakes could suppress the population abundances of
on flow rate and precipitation amount. Furthermore, clofibric acid and green algae and diatoms (Jöhnk et al., 2008). High temperatures seem
iopromide are found to have an important leaching potential which to favor the cyanotoxins dominance, as Microcystis, over diatoms and
could represent a long term risk for groundwater contamination from green algae (Jöhnk et al., 2008).
river water through sediment and subsoil (Oppel et al., 2004),
especially in case of heavy rainfalls events. 2.9. Synthesis
2.6. Pathogens The climate change impacts on surface water quality can be
summarized in Fig. 1, which consider the effects (droughts and floods)
Waterborne pathogens could be spread within the freshwater after of the two main factors (temperature and rainfalls). These impacts
a contamination by animal or human waste due to heavy rainfall depend on natural or man built environment, and the consequences
discharge in combined sewer systems (CSS). When the flow exceeds can be different according to water body type (rivers, lakes, dams,
the CSS capacity, the sewers overflow directly into surface water body ponds, wetlands...) and characteristics (water residence times, size,
(Charron et al., 2004). Pednekar et al. (2005) have studied coliform shape, depth…). For streams, the main parameters affected are DOM
load in a tidal embayment and shown that stormwater coming from and nutrients meanwhile pathogens and cyanobacteria/cyanotoxins
the surrounding watershed is a primary source of coliform. Moreover, are more related to lakes. In between, micropollutants, inorganic or
higher water temperatures will probably lead to a pathogen survival organic are also frequently affected.
this general trend should be moderated, since their decomposition

could increase with temperature.
In terms of water quality, it has been established that fulvic and
humic constituents of organic matter constitute important precursors
for THMs (Christman et al., 1990). Total organic carbon (TOC), as well
as UV absorbance, have been used as indicators of the presence of
organic matter in drinking water (Thomas, 2007). Some authors
pointed out that the minimum effective alum dose shows a strong
stoichiometric relationship with DOC concentrations in model waters
(Shin et al., 2008). Moreover, several studies have mentioned that
dissolved organic carbon (DOC) concentration in alum or iron treated
water was directly related to the THM formation potential (van
Leeuwen et al., 2005; Uyak and Toroz, 2007). Some research projects
based on laboratory bench-scale and field data have shown that the
higher values are for these parameters, the higher the concentrations
of THMs formed (Rodriguez et al., 2000; Golfinopoulos et al., 1998;
Garcia-Villanova et al., 1997; Montgomery, 1993). For DBPs formation,
the determinant factor could be the aromatic part or hydrophobic
Fig. 1. Impacts of climate change on water resources and drinking water quality. fraction of NOM and molecular weight distribution (Randtke and
Jepsen, 1981; Bose and Reckhow, 1998; Croue et al., 1999; Singer,
1999). Moreover, some studies have already showed that for waters
3. Expected impacts on drinking water production that are not governed by sweep flocculation, coagulant doses are
determined by the concentrations of both NOM and particles, silica
Research undertaken in the 1970s, indicated the presence of being the dominant factor for coagulant demand at high particle
disinfection by-products (DBPs) in drinking water (Rook, 1974; concentrations (N100 mg/L) (Shin et al., 2008). Consequently,
Symons et al., 1975). Special attention was given to the concentration mineral-bound humic substances increase the intrinsic complexation
of trihalomethanes (THMs) because of their potential carcinogenic properties of mineral substances for organic and inorganic pollutants
effects (Singer, 1993). The study on the occurrence of DBPs in drinking (Murphy and Zachara, 1995) and directly impact on DBPs formation
water distribution systems has increased in the recent years, focusing potential.
firstly on natural organic matter transformation. Concerning emerging
DBPs linked to pharmaceuticals and new pesticides, very few studies 3.2. Potential impacts
have been published in order to understand their formation and fate
during water treatment. For pharmaceutical by-products, most of Concerning climate change issues on DBPs formation, past
studies are limited to parent's pharmaceutical products (PPs) (Fent investigations have already observed that the occurrence of THMs in
et al., 2006; Mompelat et al., 2009). chlorinated water may vary significantly according to season and
This part aims at reviewing the main known determinants on DBPs geographical location in the distribution system (Williams et al., 1997;
formation under usual water treatment conditions. Then we consider Garcia-Villanova et al., 1997; Arora et al., 1997; Singer et al., 1995;
the expected impacts of climate change on these parameters and the Clark, 1994). These temporal and spatial variations are due to changes
degradation of drinking water quality. Lastly, we present some further in raw and treated water quality as well as in operational parameters
monitoring needs for a better knowledge. (pH, chlorine dose, contact time…) related to chlorination.
Rainstorm events lead to elevated levels of turbidity and organic
3.1. DBPs determinants matter found in river waters which cause deterioration in treatment
performance. However, it has been shown that this effect is not
Several factors such as temperature, dissolved organic carbon uniform. This could be due to a combination of lower water
(DOC), pH, bromide concentrations, and operational factors or temperatures and a change in the nature and increased concentrations
chlorine doses and contact time were reported to significantly affect of NOM in the natural water (Hurst et al., 2004). This could also
the formation of DBPs (Nikolaou et al., 2004; Teksoy et al., 2008). explain why these authors observed that seasonal differences have a
Temperature and organic matter matrix, influenced by climate significant impact on process robustness, independent of the raw
change, are considered hereafter. water turbidity. Rodriguez and Serodes (2001) showed that when
Concerning the influence of the water temperature on the DBPs water temperature is lower than 15 °C, THMs in treated water will not
formation, the general trend shows that for natural surface water be higher than initial THM concentration, even if these ones are high
temperatures (5–30 °C), increased temperatures promote the DBPs (60 µg/L). The latter situation may be typical in spring or fall, when
formation rate. Some studies (Rodriguez and Serodes, 2001) showed the organic content of raw water and treated water tends to increase
that THM concentrations vary significantly (from 1.5 to 2 times, following rain or field runoff. For typical summer water temperatures
depending on the utility) between drinking water plant and tap (the (N18 °C), the THM concentration within the system treatment may
most distant). When water temperature exceeds 15 °C, spatial THM rise from 2 to 4 times, depending on the utility (Rodriguez and
variations are particularly high (from 2 to 4 times, depending on the Serodes, 2001).
utility). In the same way, others authors reported that increasing DOC nature and concentration are not the only parameters which
temperature (10–33 °C) generally increased the formation of bro- drastically change during rainstorm events and the biological
moorganic DBPs (Zhang et al., 2005). However, this general trend compartment contribution should be taken into account too. Chen
should be moderated for certain unstable DBPs. Indeed, Yang et al. and Zhang (2008) globally showed that algae contributed much more
(2007) studied the formation of DBPs after 3 days of chloramination to the HAA (haloacetic acids) formation than the THM during the
with monochloramine (NH2Cl) at three temperatures (10 °C, 20 °C summer and autumn blooms. During these special events, when the
and 30 °C). They showed that chloroform formation increases with algal concentration is 20–80 million per liter, the DBP precursors
temperature from 10 to 30 °C. However, for more unstable DBPs as originating from algae would account for about 20% to 50% of the total
dichloroacetonitrile (DCAN) and 1,1-dichloro-2-propane (1,1-DCP), formation potential.
Variations of temperature, pH and aqueous composition occurring instance. Other models have already been developed that relate
during climate change could also have an influence for contaminants coagulant dose to the concentration and character of organics present
on their sorption on mineral phases. When assessing the leaching in natural waters. These models enable prediction of inorganic
behaviour of anthropogenic compounds, the influence of the proper- coagulant doses that maximize removal of organics at a particular
ties of soils has to be taken into account (Oppel et al., 2004; Yu et al., coagulation pH (van Leeuwen et al., 2005). As we have seen before,
2009). Moreover, especially during rainfall events, mineral particles model development studies taking into account temperature should
leaching could lead to high concentrations in natural waters, having a be continued. The complexity of DBP formation reactions makes it
direct impact on coagulant demand during water treatment as seen difficult to develop universally applicable models. This research field
before (Shin et al., 2008) and on DBPs formation. should however be considered with special attention for the future.
Concerning the occurrence and fate of micropollutants with A last point to consider is the analytical development for emerging
respect to drinking water treatment, the main (recent) studies are substances and by-products. Concerning pharmaceuticals and pesti-
related to pharmaceuticals. Actually, studies on pharmaceuticals are cides, there is a real need for identification and toxicity assessment on
principally linked to wastewater treatment, the efficiency of which the degradation byproducts formed during water treatment. The
may affect the quality of water resources downstream a treated removal of pharmaceuticals and other polar micro-pollutants can only
effluent discharge. Even if they are partially removed, residual be assured using advanced techniques such as ozonation, activated
quantities may remain in treated water, and have been found in carbon or membrane filtration (Ternes et al., 2002) or eventually UV
drinking (tap) water (Al-Ahmad et al., 1999; Hernando et al., 2006). treatment (Canonica et al., 2008). However, developing only the best
The efficiency of pharmaceuticals removal varies with treatment available treatment techniques to remove these substances without
processes and also with temperature and weather (Choi et al., 2008). taking into account DBPs formation is not the actual challenge.
For instance, diclofenac showed largely different elimination rates Comparison on emerging substances consumption (such as pharma-
between 17% (Heberer et al., 2002), 69% (Ternes et al., 1998), and 100% ceuticals) and occurrence in water based on a reference methodology,
(Thomas and Foster, 2004) depending on these two last parameters. health and ecotoxicological risk assessments should be developed in
Finally, for pesticides elimination in conventional physi-cochemical parallel with analytical methods permitting identification and quan-
drinking water treatment processes, such as flocculation, sedimenta- tification of by-products.
tion, filtration, or lime softening, only certain lipophilic substances are
removed adequately (Baldauf, 2006).
Natural micropollutants, mainly represented by cyanotoxins may 3.4. Synthesis
also have a huge impact on drinking water treatment. Chlorination,
micro-/ultrafiltration and especially ozonation are the most effective The climate change impacts on drinking water treatment issues
water treatment procedures in destroying cyanobacteria and in can be summarized in Fig. 2. Remind that climate change may cause at
removing microcystins (Hitzfeld et al., 2000). During cyanobacterial the resource level (surface water), huge hydrologic variations, water
bloom events, ozonation may be an appropriate process to eliminate temperature rise and increases of pollution load (chemical and
peptide toxins like microcystin-LA and-LR (Rositano et al., 1998, 2001; microbiological). For treatment plants, considering that all remedia-
Brooke et al., 2006). Lots of studies concerning removal of cyano- tion actions have been made (pollution source reduction, run off
bacterial toxins from water showed that the effectiveness of the limitation, fertilizers and pesticides reduction management, etc.),
oxidation process is not only dependant on the reactant concentration, adaptation measures must be envisaged for a better efficiency,
but also on temperature, pH, ionic composition (Rositano et al., 1998; particularly with regards to extreme events (heavy rainfalls and
Shawwa and Smith, 2001) and NOM concentration (Al Momani et al., droughts). These measures integrate complementary treatment steps
2008). Although very few studies reported the relationship between and process control even for small water supply systems. Moreover,
algae and DBPs precursors, some authors pointed out the algal water quality monitoring with analysis of micropollutants among
contribution to some DPBs formation (Chen and Zhang, 2008). Some which emerging substances and treatment by products must be carry
studies tried to improve new by-products identification for cyanotox- out, as well as health risk assessment (following the water safety plan
ins (Rodriguez et al., 2007; Merel et al., 2009). However, DBPs procedure). Obviously, in case of severe floods, transportation of
formation has largely not been investigated. bottles or tanks may be the only solution for safe drinking water
supply.
3.3. Monitoring and modeling of impacts
Face to the previous expected impacts on the degradation of

(drinking) water quality, several monitoring tools are proposed.
The first way is to incorporate on-line TOC (or DOC) measurements
into coagulation control algorithm for pH control and coagulant dose
to prevent uncontrolled variation especially during storm/raining
events (Hurst et al., 2004). In the same area, the development of field
monitoring procedures or system may be useful for improving DOM
knowledge in order to evaluate bulk properties mainly influencing
disinfection by products formation such as biological activities (algal
contribution, photosynthetic evolution…) or molecular polarity.
Fluorescence analysis could help in the assessment of DOM sources
by spectral signatures related to waters affected by microbial activity,
either by wastewater influence or by autochthonous processes and
could correlate some of these data with DOC for instance (Parlanti
et al., 2000; Jung et al., 2005; Rosario-Ortiz et al., 2007).
Another solution is the prediction of the occurrence or fate of some
physico-chemical parameters through modeling. Uyak and Toroz
(2007) have proposed a model for the concentration estimation of
THM and HAA in chlorinated raw water of surface water supplies for Fig. 2. Climate change impacts and drinking water treatment issues.
4. Conclusion Choi K, Kima Y, Park J, Park CK, Kim M, Kim HS, et al. Seasonal variations of several
pharmaceutical residues in surface water and sewage treatment plants of Han
River. Sci Total Environ 2008;405:102–28.
The main outcome of this literature review on climate change impact Christman RF, Kronkerg L, Sing R, Ball LM, Johnson JD. Identification of mutagenic
on surface water quality (from resources to tap) is that there is a byproducts from aquatic humic chlorination. Denver: AWWA Research Foundation
and AWWA; 1990.
degradation trend of drinking water quality leading to an increase of at Clark RM. Modelling water quality changes and contaminant propagation in drinking water
risk situations with regard to potential health impact, mainly during distribution systems: a US perspective. J Water Supply Res Technol 1994;43:133–43.
extreme meteorological events. Among water quality parameters, Clark JM, Lane SN, Chapman PJ, Adamson JK. Link between DOC in near surface peat and
stream water in an upland catchment. Sci Total Environ 2008;404:308–15.
dissolved organic matter, micropollutants and pathogens are susceptible Croue JP, Debroux JF, Aiken G, Leenheer JA, Amy GL. Natural organic matter: structural
to rise in concentration or number as a consequence of temperature and reactive properties. In: Singer PC, editor. Formation and control of disinfection
increase (water, air and soil) and heavy rain falls in temperate countries. by-products in drinking water. Denver, CO: American Water Works Assn; 1999.
Curriero FC, Patz JA, Rose JB, Lele S. The association between extreme precipitation and
Another conclusion is the lack of information on micropollutants
waterborne disease outbreaks in the United States, 1948–1994. Am J Public Health
occurrence and fate with regard to climate change impacts and 2001;91:1194–9.
treatment efficiency, including potentially association and transporta- Daufresne M, Boët P. Climate change impacts on structure and diversity of fish
tion with natural organic matter. Disinfection by products of communities in rivers. Glob Chang Biol 2007;13:2467–78.
Daufresne M, Roger MC, Capra H, Lamouroux N. Long-term changes within the inver-
micropollutants not removed during treatment and of residues must tebrate and fish communities of the Upper Rhône River: effects of climatic factors.
be identified and their toxicity assessed. Last conclusion concerns Glob Chang Biol 2003;10:124–40.
water borne diseases potentially highly linked to climate change Drewry JJ, Newham LTH, Croke BFW. Suspended sediment, nitrogen and phosphorus
concentrations and exports during storm-events to the Tuross estuary, Australia.
impacts but still rarely studied at least for temperate countries. Finally, J Environ Manag 2009;90:879–87.
there is a huge need for water quality monitoring and predictive tools Ducharne A, Baubion C, Beaudoin N, Benoit M, Billen G, Brisson N, et al. Long term
as models and decision support systems mainly with the aim of health prospective of the Seine river system: confronting climatic and direct anthro-
pogenic changes. Sci Total Environ 2007;375:292–311.
risks assessment and remediation and adaptation actions. Evans CD, Monteith DT, Cooper DM. Long-term increases in surface water dissolved
organic carbon: observations, possible causes and environmental impacts. Env Poll
Acknowledgments 2005;137:55–71.
Evans CD, Monteith DT, Reynolds B, Clark JM. Buffering of recovery from acidification by
organic acids. Sci Total Environ 2008;404:316–25.
The authors would like to thank Sophie Mompelat and Sylvain Fent K, Weston AA, Caminada D. Review: ecotoxicology of human pharmaceuticals.
Merel, PhD students of the Laboratoire d'Etude et de Recherche en Aquat Toxicol 2006;76:122–59.
Garcia-Villanova J, Garcia C, Gomez JA, Garcia MP, Ardanuy R. Formation, evolution and
Environnement et Santé (LERES), for their fruitful discussions on
modelling of trihalomethanes in the drinking water of a town: II. In the distribution
cyanotoxins and pharmaceuticals occurrence and persistence in the system. Water Res 1997;31:1405–13.
environment. They also thank André Lavoie from Sherbrooke George G, Hurley M, Hewitt D. The impact of climate change on the physical characteristics
University for his help as well as the reviewers of the manuscript, of the larger lakes in the English Lake District. Freshw Biol 2007;52:1647–66.
Golfinopoulos SK, Xilourgidis NK, Kostopoulou MN, Lekkas TD. Use of a multiple
suggesting us very relevant improvements. regression model for predicting trihalomethane formation. Water Res 1998;32:
2821–9.
Heberer T, Reddersen K, Mechlinski A. From municipal sewage to drinking water: fate
References and removal of pharmaceutical residues in the aquatic environment in urban areas.
Water Sci Technol 2002;46:81–8.
Al Momani F, Smith DW, El-Din MG. Degradation of cyanobacteria toxin by advanced Hejzlar J, Dubrovsky M, Buchtele J, Ružička M. The apparent and potential effects of
oxidation process. J Hazard Mater 2008;150:238–49. climate change on the inferred concentration of dissolved organic matter in a
Al-Ahmad A, Daschner FD, Kummerer K. Biodegradability of cefotiam, ciprofloxacin, temperate stream (the Malše River, South Bohemia). Sci Total Environ 2003;310:
meropenem, penicillin G, and sulfamethoxazole and inhibition of waste water 143–52.
bacteria. Arch Environ Contam Toxicol 1999;37:158–63. Hernando MD, Mezcua M, Fernandez-Alba AR, Barceló D. Environmental risk
Arheimer B, Andreásson J, Fogelberg S, Johnsson H, Pers CB, Persson K. Climate change impact assessment of pharmaceutical residues in wastewater effluents, surface waters
on water quality: model results from southern Sweden. Ambio 2005;34(7):559–66. and sediments. Talanta 2006;69:334–42.
Arora A, LeChevallier MW, Dixon KL. DBP occurrence survey. J Am Water Works Assn Hitzfeld BC, Höger SJ, Dietrich DR. Cyanobacterial toxins removal during drinking water
1997;89:60–8. treatment and human risk assessment. Environ Health Perspect 2000;108:113–22.
Baldauf G. Removal of pesticides in drinking water treatment. Acta Hydrochim Hunter PR. Climate change and waterborne and vector-borne disease. J Appl Microbiol
Hydrobiol 2006;21:203–8. 2003;94:37S–46S.
Bates BC, Kundzewicz ZW, Wu S, Palutikof JP. Climate change and water. Geneva: Hurst AM, Edwards MJ, Chipps M, Jefferson B, Parsons SA. The impact of rainstorm
Technical paper of the Intergovernmental Panel on Climate change. IPCC events on coagulation and clarifier performance in potable water treatment. Sci
Secretariat; 2008. Total Environ 2004;321:219–30.
Bhat S, Hatfield K, Jacobs JM, Lowrance R, Williams R. Surface runoff contribution of Jackson LJ, Lauridsen TL, Søndergaard M, Jeppesen E. A comparison of shallow Danish
nitrogen during storm events in a forested watershed. Biogeochemistry 2007;85: and Canadian lakes and implications of climate change. Freshw Biol 2007;52:
253–62. 1782–92.
Bloomfield JP, Williams RJ, Gooddy DC, Cape JN, Guha P. Impacts of climate change on Jöhnk KD, Huisman J, Sharples J, Sommeijer B, Visser PM, Stroom JM. Summer heatwaves
the fate and behaviour of pesticides in surface and groundwater—a UK perspective. promote blooms of harmful Cyanobacteria. Glob Chang Biol 2008;14: 495–512.
Sci Total Env 2006;369:163–77. Jung A-V, Frochot C, Parant S, Lartiges BS, Selve C, Viriot M-L, et al. Synthesis of amino-
Boreen AL, Arnold WA, McNeill K. Photodegradation of pharmaceuticals in the aquatic phenolic humic-like substances and comparison with natural aquatic humic acids:
environment: a review. Aquat Sci 2003;65(4):320–41. a multi-analytical techniques approach. Org Geochem 2005;36:1252–71.
Bose P, Reckhow DA. Adsorption of natural organic matter on preformed aluminium Kaste Ø, Wright RF, Barkved LJ, Bjerkeng B, Engen-Skaugen T, Magnusson J, et al. Linked
hydroxide flocs. J Environ Eng 1998;124:803–11. models to assess the impacts of climate change on nitrogen in a Norwegian river
Brient L, Lengronne M, Bormans M, Fastner J. Short communication: first occurrence of basin and fjord system. Sci Total Environ 2006;365:200–22.
Cylindrospermopsin in freshwater in France. Environ Toxicol 2008;24(4):415–20. Komatsu E, Fukushima T, Harasawa H. A modeling approach to forecast the effect of
Brooke S, Newcombe C, Nicholson B, Klass G. Decrease in toxicity of microcystins LA and long-term climate change on lake water quality. Ecol Model 2007;209:351–66.
LR in drinking water by ozonation. Toxicon 2006;48:1054–9. Lennartz B, Louchart X. Effect of drying on the desorption of diuron and terbuthylazine
Brunetti M, Maugeri M, Nanni T. Changes in total precipitation rainy days and extreme from natural soils. Environ Pollut 2007;146:180–7.
events in northeastern Italy. Int J Climatol 2001;21:861–71. Lissemore L, Hao C, Yang P, Sibley PK, Mabury S, Solomon KR. An exposure assessment
Buerge IJ, Buser HR, Poiger T, Müller MD. Occurrence and fate of the cytostatic drugs for selected pharmaceuticals within a watershed in Southern Ontario. Chemo-
Cyclophosphamide and Ifosfamide in wastewaters and surface waters. Environ Sci sphere 2006;64:717–29.
Technol 2006;40:7242–50. Malmaeus JM, Blenckner T, Markensten H, Persson I. Lake phosphorus dynamics and
Canonica S, Meunier L, von Gunten V. Phototransformation of selected pharmaceuticals climate warming: a mechanistic model approach. Ecol Model 2006;190:1–14.
during UV treatment of drinking water. Water Res 2008;42:121–8. Merel S, Lebot B, Clement M, Seux R, Thomas O. MS identification of microcystin-LR
Charron DF, Thomas MK, Waltner-Toews D, Aramini JJ, Edge T, Kent RA, et al. Vul- chlorination by-products. Chemosphere 2009;74:832–9.
nerability of waterborne diseases to climate change in Canada: a review. J Toxicol Mompelat S, Lebot B, Thomas O. Occurrence and fate of pharmaceutical products and
Environ Health Part A 2004;67:1667–77. by-products, from resource to drinking water. Environ Int 2009;35:805–14.
Chen C, Zhang X-J, Zhu L-x, Liu J, He W-j, Han H-d. Disinfection by-products and their Monteith DT, Stoddard JL, Evans CD, de Wit HA, Forsius M, Høgåsen T, et al. Dissolved
precursors in a water treatment plant in North China: seasonal changes and fraction organic carbon trends resulting from changes in atmospheric deposition chemistry.
analysis. Sci Total Environ 2008;397:140–7. Nature 2007;450:537–41.
Montgomery W. Mathematical modeling of the formation of THMs and HAAs in Singer JJ. Formation and characterization of disinfection by-products. In: Craun GF, editor.
chlorinated natural waters. Denver, CO: Am. Water Works Ass; 1993. Safety of water disinfection: balancing chemical and microbial risks. Washington: ILSI
Mooij WM, Hülsmann S, De Senerpont Domis LN, Nolet BA, Bodelier PLE, Boers PCM, Press; 1993.
et al. The impact of climate change on lakes in the Netherlands: a review. Aquat Singer PC. Humic substances as precursors for potentially harmful disinfection by-
Ecol 2005;39:381–400. products. Water Sci Technol 1999;40:25–30.
Murphy EM, Zachara JM. The role of sorbed humic substances on the distribution of Singer PC, Obolensky A, Greiner A. DBPs in chlorinated North Carolina drinking waters.
organic and inorganic contaminants in groundwater. Geoderma 1995;67:103–24. J Am Water Works Assn 1995;87:83–92.
Nikolaou AD, Golfinopoulos SK, Lekkas TD, Arhonditsis GB, Kolovoyiannis V, Lekkasa TD. Soh YC, Roddick F, van Leeuwen J. The future of water in Australia: the potential effects
Modelling the formation of chlorination by-products in river waters with different of climate change and ozone depletion on Australian water quality, quantity and
quality. Chemosphere 2004;55:409–20. treatability. Environmentalist 2008;28:158–65.
Olivie-Lauquet G, Gruau G, Dia A, Riou C, Jaffrezic A, Henin O. Release of trace elements Symons JM, Bellar TA, Carswell JK, DeMarco J, Kroop KL, Robeck GG, et al. National
in wetlands: role of seasonal variability. Water Res 2001;35(4):943–52. organics reconnaissance survey for halogenated organics. J Am Water Works Assn
Oppel J, Broll G, Löffler D, Meller M, Römbke J, Ternes Th. Leaching behaviour of 1975;67:634–47.
pharmaceuticals in soil testing systems: a part of an environmental risk assessment Teksoy A, Alkan U, Baskaya HS. Influence of treatment process combinations on the
for groundwater contamination. Sci Total Environ 2004;328:265–73. formation of THM species in water. Sep Purif Technol 2008;61:447–54.
Parlanti E, Wörz K, Geoffroy L, Lamotte M. Dissolved organic matter fluorescence Ternes T, Hirsch R, Mueller J, Haberer K. Methods for the determination of neutral drugs
spectroscopy as a tool to estimate biological activity in a coastal zone submitted to as well as betablockers and alpha2-sympathomimetics in aqueous matrices using
anthropogenic inputs. Org Geochem 2000;31:1765–81. GC/MS and LC/MS/MS. Fresen J Anal Chem 1998;362:329–40.
Pednekar AM, Grant SB, Jeong Y, Poon Y, Oancea C. Influence of climate change, tidal Ternes T, Meisenheimer M, McDowell D, Sacher F, Brauch H-J, Haist-Glude B, et al.
mixing, and watershed urbanization on historical water quality in Newport Bay, a Removal of pharmaceuticals during drinking water treatment. Environ Sci Technol
saltwater wetland and tidal embayment in southern California. Environ Sci Technol 2002;36:3855–63.
2005;39:9071–82. Thies H, Nickus U, Mair V, Tessadri R, Tait D, Thaler B, et al. Unexpected response of high
Pédrot M, Dia A, Davranche M, Bouhnik-Le Coz M, Henin O, Gruau G. Insights into colloid- Alpine Lake waters to climate warming. Environ Sci Technol 2007;41:7424–9.
mediated trace element release at the soil/water interface. J Colloid Interface Sci Thomas O, Burgess C. Elsevier ed. UV–Visible spectrophotometry for water and
2008;325:187–97. wastewater; 2007.
Petrovic M, Barceló D. LC-MS for identifying photodegradation products of pharma- Thomas PM, Foster GD. Determination of nonsteroidal anti-inflammatory drugs,
ceuticals in the environment. Trends Anal Chem 2007;26(6):486–93. caffeine, and triclosan in wastewater by gas chromatography-mass spectrometry.
Petterson K, Grust K, Weyhenmeyer G, Blenckner T. Seasonality of chlorophyll and J Environ Sci Health A 2004;39:1969–78.
nutrients in Lake Erken—effects of weather conditions. Hydrobiologia 2003;506– Uyak V, Toroz I. Disinfection by-product precursors reduction by various coagulation
509:75–81. techniques in Istanbul water supplies. J Hazard Mater 2007;141:320–8.
Prathumratana L, Sthiannopkao S, Kim KW. The relationship of climatic and hydro- Van Leeuwen J, Daly R, Holmes M. Modelling the treatment of drinking water to
logical parameters to surface water quality in the lower Mekong River. Environ Int maximize dissolved organic matter removal and minimize disinfection by-product
2008;34:860–6. formation. Desalination 2005;176:81–9.
Probst M, Berenzen N, Lentzen-Godding A, Schulz R. Scenario-based simulation of Van Vliet MTH, Zwolsman JJG. Impact of summer droughts on the water quality of the
runoff-related pesticide entries into small streams on a landscape level. Ecotox Meuse river. J Hydrol 2008;353:1–17.
Environ Safe 2005;62:145–59. Weyhenmeyer GA. Rates of change in physical and chemical lake variables—are they
Psenner R, Schmidt R. Climate-driven pH control of remote Alpine lakes and effects of comparable between large and small lakes? Hydrobiologia 2008;599:105–10.
acid deposition. Nature 1992;356:781–3. Wiedner C, Rücker J, Brüggemann R, Nixdorf B. Climate change affects timing and size of
Randtke SJ, Jepsen CP. Chemical pre-treatment for activated carbon adsorption. J Am populations of an invasive cyanobacterium in temperate regions. Oecologia
Water Works Assn 1981;73:411–20. 2007;152:473–84.
Rodriguez MJ, Serodes JB. Spatial and temporal evolution of trihalomethanes in three Wilby RL, Whitehead PG, Wade AJ, Butterfield D, Davis RJ, Watts G. Integrated modelling
water distribution systems. Water Res 2001;35:1572–86. of climate change impacts on water resources and quality in a lowland catchment:
Rodriguez MJ, Serodes JB, Morin M. Estimation of water utility compliance with River Kennet, UK. J Hydrol 2006;330:204–20.
trihalomethane regulations using a modelling approach. J Water Supply Res Wilhelm S, Adrian R. Impact of summer warming on the thermal characteristics of a
Technol-Aqua 2000;49:57–73. polymictic lake and consequences for oxygen, nutrients and phytoplankton. Freshw
Rodriguez E, Sordo A, Metcalf JS, Acero JL. Kinetics of the oxidation of cylindosper- Biol 2008;53:226–37.
mopsin and anatoxin-a with chlorine, monochloramine and permanganate. Water Williams DT, Lebel GL, Benoit FM. Disinfection by-products in Canadian drinking water.
Res 2007;41:2048–56. Chemosphere 1997;34:299–316.
Rook JJ. Formation of haloforms during chlorination of natural waters. Water Treat Worrall F, Burt T. Trends in DOC concentration in Great Britain. J Hydrol 2007;346:
Exam 1974;23:234–43. 81–92.
Rosario-Ortiz FL, Snyder SA, Suffet IH. Characterization of dissolved organic matter in Worrall F, Burt T, Adamson J. Can climate change explain increases in DOC flux from
drinking water sources impacted by multiple tributaries. Water Res 2007;41: upland peat catchments? Sci Total Environ 2004;326:95–112.
4115–28. Yang X, Shang C, Westerhoff P. Factors affecting formation of haloacetonitriles, haloketones,
Rositano J, Nicholson BC, Pieronne P. Destruction of cyanobacterial toxins by ozone. chloropicrin and cyanogens halides during chloramination. Water Res 2007;41:
Ozone-Sci Eng 1998;20:223–38. 1193–200.
Rositano J, Newcombe G, Nicholson B, Sztajnbok P. Ozonation of NOM and algal toxins in Yu L, Fink G, Wintgens T, Melin T, Ternes TA. Sorption behaviour of potential organic
four treated waters. Water Res 2001;35:23–32. wastewater indicators with soils. Water Res 2009;43:951–60.
Rothwell JJ, Evans MG, Daniels SM, Allott TEH. Baseflow and stormflow metal Zhang X, Echigo S, Lei H, Smith ME, Minear RA, Talley JW. Effects of temperature and
concentrations in streams draining contaminated peat moorlands in the Peak chemical addition on the formation of bromoorganic DBPs during ozonation. Water
District National Park (UK). J Hydrol 2007;341:90–104. Res 2005;39:423–35.
Sardans J, Peñuelas J, Estiarte M. Changes in soil enzymes related to C and N cycle and in Zhu Z, Arp PA, Mazumder A, Meng F, Bourque CPA, Foster NW. Modeling stream water
soil C and N content under prolonged warming and drought in a Mediterranean nutrient concentrations and loadings in response to weather condition and forest
shrubland. Appl Soil Ecol 2008;39:223–35. harvesting. Ecol Model 2005;185:231–43.
Shawwa AR, Smith DW. Kinetics of microcystin-LR oxidation by ozone. Ozone-Sci Eng Zwolsman JJG, van Bokhoven AJ. Impact of summer droughts on water quality of the
2001;23:161–70. Rhine River—a preview of climate change? Water Sci Technol 2007;56:44–55.
Shin JY, Spinette RF, O'melia CR. Stoichiometry of coagulation revisited. Environ Sci
Technol 2008;42:2582–9.
Chemosphere 80 (2010) 228–234
Chemosphere
journal homepage: www.elsevier.com/locate/chemosphere
Interaction of pyrene fluoroprobe with natural and synthetic

humic substances: Examining the local molecular organization
from photophysical and interfacial processes
A.-V. Jung a,*, C. Frochot b, F. Villieras c, B.S. Lartiges c, S. Parant d, M.-L. Viriot b, J.-L. Bersillon c
a
Ecole des Métiers de l’Environnement (EME), Campus de Ker Lann, Avenue Schuman, 35170 Bruz, France
b
Nancy-University CNRS-INPL-ENSIC UMR 7630, Département de Chimie Physique des Réactions (DCPR), 1 rue Grandville, BP 20451, 54000 Nancy, France
c
Nancy-University CNRS-INPL-ENSG UMR 7569, Laboratoire Environnement et Minéralurgie (LEM), 15 avenue du Charmois, BP 40, 54501 Vandœuvre-lès-Nancy, France
d
Nancy-University CNRS-UHP UMR 7565, Laboratoire de Structure et Réactivité des Systèmes Moléculaires Complexes (SRSMC), 54500 Vandœuvre-lès-Nancy, France
Article history: The direct and indirect interaction mechanisms of pyrene with: (i) various molecular weight fractions of a
Received 4 November 2009 synthetic humic-like substance (SyHA) and (ii) extracts of natural humic acids (NHA) from Moselle River
Received in revised form 11 April 2010 suspended matter were investigated using quenching fluorescence and surface tension measurements.
Accepted 12 April 2010
Humic materials were characterized in a previous study. The Stern–Volmer associative constants were
Available online 7 May 2010
determined from the quenching technique. Surface tension measurements revealed an increase in surface
activity as a function of concentration for each humic fraction independently of the pyrene presence in
Keywords:
solution, even during the formation of humic micelles. The results obtained suggest the possibility of spe-
Natural
Synthetic humic substances
cific intermolecular interactions occurring during pyrene entrapment within humic acids. In addition, we
Fluorescence quenching show that molecular weight, aliphatic chains (especially those containing nitrogen groups) and number
Pyrene of acidic groups are determinant characteristics for pollutant entrapment capacity at concentrations
Surface tension below the critical micellar concentration (CMC) of humic substances.
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction occurring during pyrene entrapment within humic acids. In addi-

tion, we show that molecular weight, aliphatic chains (especially
Understanding the fate and the transport of anthropic contam- those containing nitrogen groups) and number of acidic groups
inants in natural aquatic systems is of a primary importance for are determinant characteristics for pollutant entrapment capacity
water resources management and treatment optimization. Among at concentrations below the critical micellar concentration (CMC)
these contaminants, polycyclic aromatic hydrocarbons (PAHs) are of humic substances.
ubiquitous in the environment and have been identified in a num- PAHs also occur naturally, high levels of those molecules in the
ber of soils, effluents, river sediments, as well as in the colloidal environment are usually attributed to point and non-point pollu-
compartment of surface waters (Bouloubassi and Saliot, 1991; tion sources such as oil spills, automobile exhaust fumes, and coal
Countway et al., 2003; Agence de l’Eau Rhin-Meuse, 2005). refining processes (Short et al., 2007; Achten and Hofmann, 2009).
Although the direct and indirect interaction mechanisms of pyrene According to eco- and human toxicity exposure tests (Jeffrey et al.,
with: (i) various molecular weight fractions of a synthetic humic- 1977; Hua et al., 2009), mutagenic or carcinogenic effects of PAHs
like substance (SyHA) and (ii) extracts of natural humic acids can be expected in the natural environment.
(NHA) from Moselle River suspended matter, were investigated It has long been recognized that natural organic matter (NOM)
using quenching fluorescence and surface tension measurements. has an indisputable role in the sorption and leaching of hazardous
Humic materials were characterized in a previous study. The organic chemicals such as PAHs in the environment (Gauthier
Stern–Volmer associative constants were determined from the et al., 1987; Chin et al., 1997; Peuravuori, 2001; Countway et al.,
quenching technique. Surface tension measurements revealed an 2003). Various techniques have been used to examine the binding
increase in surface activity as a function of concentration for each capacities of NOM towards PAHs such as adsorption, nuclear mag-
humic fraction independently of the pyrene presence in solution, netic resonance (NMR), chromatography, and fluorescence meth-
even during the formation of humic micelles. The results obtained ods (Ghosh et al., 2000; Pan et al., 2006; Nakashima et al., 2007;
suggest the possibility of specific intermolecular interactions Hur et al., 2009). Most of these classical approaches require the
determination of free pollutant concentrations, before and after
* Corresponding author. Tel.: +33 2 99 05 88 00; fax: +33 2 99 05 88 09. their binding with dissolved natural organic matter. However,
E-mail address: av2jung@live.fr (A.-V. Jung). the necessary separation steps often lead to a poor quantitative
0045-6535/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.chemosphere.2010.04.035
A.-V. Jung et al. / Chemosphere 80 (2010) 228–234 229
estimate of binding capacity. On the other hand, a fluorescence two different nitrogen sources in distinct syntheses. Two molecu-
quenching technique does not require such a separation step. lar weight synthetic humic-like substance (SyHA) fractions
Therefore, fluorescence spectroscopy offers unique advantages in (1–3.5 kDa and >8 kDa, respectively CTHA1 and CTHA2) were ob-
the study of contaminant binding to NOM which is either dis- tained for the catechol/triglycine synthesis. A greater than 8 kDa
solved, suspended or within sediment matter. catechol/glycine synthetic humic-like substance (SyHA) fraction
Humic substances represent a predominant component of nat- was also collected (CGHA). Andreux et al. (1980) polymerization
ural organic matter in freshwater (Zumstein and Buffle, 1989). protocol was used. Catechol (6.6 g) was dissolved in 500 mL of a
They are commonly described as amphiphilic heterogeneous com- solution of sodium phosphate buffer NaH2PO4/Na2HPO4 at pH 8
pounds of about 1–2 nm in size (Sielechi et al., 2008). The organi- to obtain a 0.03 M catechol solution. This was then mixed with
zation of humic substances, either viewed through polymeric or an equal volume of buffered glycine or triglycine 0.03 M solution.
supramolecular assembly, remains a debated issue (Conte and The resulting mixture was then allowed to react in a 2 L flask in
Piccolo, 1999; Schulten and Gleixner, 1999; Piccolo, 2001, 2002; the dark at 25 °C and under pure oxygen constant pressure
Hetkorn et al., 2007). Their molecular size has been shown to (0.9 bar) for 5 d. A first dialysis step, using Spectra/Por 7 mem-
strongly influence the binding capacity and the environmental brane- 1 kDa molecular weight cut-off (Roth-Sochiel Lauterbourg,
transport of both non-polar (Chiou et al., 1987; Chin and Weber, France), was performed in order to eliminate the lowest molecu-
1989) and polar organic contaminants such as pesticides (Piccolo lar weight compounds (<1 kDa). The retentate then underwent a
et al., 1996, 1998). The polarity of humic material could also affect second dialysis step (Spectra/Por 7 membrane- 3.5 kDa molecular
pollution transport. Previous reports have suggested either a parti- weight cut-off, Roth-Sochiel Lauterbourg, France). The permeate
tion of contaminants between water and micelle-like structures of of this second dialysis step was our 1–3.5 kDa SyHA (referred to
humic substances and/or binding to well-defined sorption sites of as CTHA 1). The retentate was again allowed to undergo a third
the humic substances (Engebretson and von Wandruska, 1994; dialysis step (Spectra/Por 7 membrane- 8 kDa molecular weight
Sierra et al., 2005). Therefore, a better understanding of the humic cut-off, Roth-Sochiel Lauterbourg, France) to obtain the final
substance building blocks and structure can certainly be attained retentate (refered to as CTHA2). For CGHA, one single dialysis at
by relating their sorption properties with their chemical character- 8 kDa was performed to obtain that fraction. All the synthetic
istics. One approach is to correlate the ‘‘binding” coefficients materials in a sodium form were freeze-dried and stored at 4 °C
determined by quenching measurements with others hydrophobic in the dark. Monitoring of the synthesis has been described else-
experimentally determined parameters. Some studies (Engebretson where (Jung et al., 2005).
and von Wandruska, 1994; Sierra et al., 2005) have showed that hu- Natural organic matter (NOM) designates natural humic acids
mic polymers in aqueous solution could coil up and ‘‘squeeze” out extracted from suspended matter taken from the Moselle River at
the solvating water. This last point has often been offered in expla- Messein (France). The extraction was performed according to the
nation of the well-known cloud-point behaviour of non-ionic sur- International Humic Substances Society (IHSS) standard procedure
factants (Myers, 1992). (Thurman and Malcolm, 1981). Sodium saturated humic sub-
Recent research (Quagliotto et al., 2006) tends to build on such stances having a greater than 8 kDa molecular weight were dia-
concepts and views natural organic matter (NOM) as ‘‘a competi- lyzed to remove excess salt and then freeze-dried. All natural
tive low-cost bio-surfactant”. From process cost and environmen- and synthetic humic substances were collected under their sodium
tal impact points of view, the bio-surfactant has the advantages form. Two natural humic extracts corresponding to two different
of operating at much lower additive concentrations and should field sampling campaigns were obtained. Natural humic acids
be available at significantly lower cost than the synthetic surfac- (NHA1 and NHA2) extractions were performed during the summer
tants. However, the mechanistic basis for identifying the interac- and autumn seasons, respectively.
tions between polycyclic aromatic hydrocarbons (PAHs) and
natural organic matter (NOM) is not yet sufficiently clarified to
predict sorption coefficients. One approach to circumvent the 2.2. Preparation of pyrene–humic substances solutions
intrinsic complexity of humic substances consists in synthesizing
functionally well-defined polymers which integrate the main char- Pyrene (99+%) was purchased from Sigma–Aldrich Chemicals
acteristics of amino-phenolic humic-like molecules (Jung et al., (Lyon, France) and carefully stored at 4 °C in the dark. Crystals
2005). A comparison of the interaction of pyrene with: (i) synthetic were dissolved in ethanol (99.5+% spectrograde, Carlo Erba) to
humic-like substances (SyHA) of different molecular weights made make a stock solution of 103 M.
from the oxidation of catechol/glycine (CGHA) or catechol/trigly- Solutions of 10, 5 and 0.1 g L1 of synthetic humic-like sub-
cine (CTHA) oxidation, (ii) natural humic extracts from suspended stance (SyHA) and natural humic acids (NHA) were first prepared
matter centrifuged from Moselle River water (France), was ex- in deionized ultra-pure Milli-Q water (Millipore). Dilutions of these
plored in the present paper. The amount of pollutant bound to each solutions were made to obtain humic substances (HS) concentra-
of the synthetic and natural humic acid fractions was correlated tion ranging from 0.1 to 30 mg L1 for fluorescence measurements
with their chemical characteristics. and from 0.1 to 1 g L1 for surface tension measurements. Twenty
This paper attempts to clarify the formation of micellar aggre- milliliter aliquots of synthetic humic-like substance (SyHA) and
gates between pyrene and humic acids. The aim is to detect NHA solutions were then placed in glass beakers and mixed with
whether an interfacial modification could explain the pyrene the pyrene stock solution (103 M in ethanol) to obtain final con-
entrapment in micellar-like domains, and finally whether the pyr- centrations of 106 M of pyrene. The volumetric content of ethanol
ene has an effect on the formation of micelle-like humic substances. introduced in the samples for fluorescence measurements did not
exceed 0.1% and was thus considered negligible with respect to
the solubility of pyrene in the samples. The pyrene–synthetic
2. Experimental section humic-like substance (SyHA) and pyrene–natural humic acids
(NHA) solutions were then adjusted to pH 8 under stirring with
2.1. Humic substances dropwise addition of 0.1 M HCl. The pyrene–SyHA and
pyrene–NHA solutions were allowed to equilibrate for 1 h. Prior
Synthetic humic-like substances (SyHA) were prepared with to fluorescence measurements, the samples were homogenized
catechol and glycine or triglycine as reactants, thus representing by stirring.
230 A.-V. Jung et al. / Chemosphere 80 (2010) 228–234
2.3. Surface tension measurements place in the excited state of the fluorescence compound and may
be collisional in nature, in which case it is referred to as ‘‘dynamic
Surface tension measurements were carried out with solutions quenching”. The excitation energy transferred to the quencher, is
prepared similarly as for the fluorescence measurements. The sur- usually dissipated non-radiatively. Alternatively, the fluorophore
face tensions (cs) were determined at pH 8 using the Wilhelmy and the quencher may associate in the ground state, thus leading
plate technique both for humic materials alone and in presence to species that lack the original fluorescence characteristics. In that
of pyrene at 106 M. The measurements were performed with glass case, it is termed ‘‘static quenching”.
plates carefully cleaned with sulfochromic acid, rinsed with Milli-Q Quenching interactions are described by the Stern–Volmer
ultra-pure water, and then heated in a Bunsen burner flame to equation (Gauthier et al., 1986). The equation holds for both dy-
avoid organic contaminant traces. cs values were obtained at namic and static quenching, provided that only one type of
25 ± 0.1 °C, using a KSV Sigma-70 tensiomat model (Cahn, Texas, quenching is operative and that the fluorophore is present as a sin-
USA) and performed with a constant solution volume of 20 mL. gle population. A linearisable Stern–Volmer plot (I0/I versus [Q]0) is
The accuracy of measurements was ±0.1 mN m1. The surface ten- then obtained, where I0 and I are respectively the fluorescence
sion of each solution was calculated from the average of two intensity in the absence and in the presence of quencher, KSV the
immersion/emersion cycles using the linear intercept at the origin Stern–Volmer quenching constant (slope), and [Q]0 the quencher
of the receding branch. The average standard errors between mea- initial concentration.
surements for synthetic humic-like substance (SyHA) fractions and The pyrene fluorescence spectrum exhibits five main emission
natural extracts were 0.2 mN m1 and 0.35 mN m1, respectively. vibronic bands called I1, I2, I3, I4 and I5 respectively at 372, 377,
All experimental errors were found below 1 mN m1 (the maxi- 382, 390 and 395 nm in ethanol after excitation at 332 nm (Kalya-
mum difference between two immersion/emersion cycles mea- nasudaram and Thomas, 1977). For quenching measurements, the
surements and taking into account the experimental deviation). I0/I2 ratio was calculated, with I2 and I0 designating the pyrene
Critical micellar concentration (CMC) of humic substances have fluorescence intensities at 377 nm in emission for a 332 nm excita-
been determined from the break points of surface tension versus tion wavelength with and without humic substances, respectively.
logarithm of humic acid concentration straight lines. The second Three replicates were run for each sample and for each quench-
horizontal line was plotted using the first c stabilization value, that ing measurements to determine average binding constants.
is to say when the difference between two c mean measurements
was at maximum three times the average standard error deter-
mined for the fractions. 3. Results and discussion
Maximal surface adsorption of the solute Cmax (mol cm2) and
the minimal cross-sectional Amin (cm2 molecule1) was calculated 3.1. Synthetic humic-like substance (SyHA) and natural humic acids
using the following Gibss adsorption equations (Taubman, 1956) (NHA) characterizations
at the maximal value of dc/d ln c:
The main characteristics of both synthetic and natural humic
1ðdcÞ 1 2 1 materials have been presented in a previous paper using a multi-
C¼ mol cm2 and Amin ¼ cm molecule
RTðd ln cÞT Cmax N analytical technique approach (Jung et al., 2005). A summary of
those properties (elemental analysis, hydrophobic indices) is given
where R = gas constant, T = absolute temperature (RT at 25 °C = in Table 1. The nitrogen contents for synthetic humic-like sub-
2.47 1010 erg mol1), C = surface tension (erg cm2 or dyne stance (SyHA) were found to be approximatively three times high-
cm1), and C = concentration of solute in percent (weight/volume) er in CTHA (1–3.5 kDa and >8 kDa, respectively CTHA1 and CTHA2
N = Avogadro’s number (6.02 1023 molecules mol1). obtained with catechol/triglycine) than in CGHA (>8 kDa with cat-
Cmax were calculated by the lest square method from linearisa- echol/glycine). The higher nitrogen content of NHA1 (natural hu-
ble portions (10–31, 6 mg L1 of humic substances) of c versus log mic acid summer extract) compared with NHA2 (autumn extract)
[HA] plots. may be explained by microbial decomposition by-products. H/C
atomic ratio is a useful qualitative parameter when comparing
2.4. Fluorescence measurements the aromaticity and the hydrophobicity of humic substances (Chen
et al., 1977; Thurman, 1985). The aromaticity increases as H/C de-
Steady-state fluorescence measurements were performed on a creases, and thus, CGHA is more aromatic than CTHA2 and eventu-
Spex Fluorolog-3 spectrofluorometer (Jobin–Yvon, Stanmore, UK) ally than CTHA1. Natural humic acids (NHA) were found more
using 0.5 mm slits for the excitation and emission monochroma- aliphatic than synthetic humic-like substance (SyHA) based on ele-
tors. The instrument was equipped with a 450 W Xenon lamp mental analysis and on Fourier Transform InfraRed (FTIR) spectros-
(Jobin–Yvon). The sample temperature was maintained to copy results (Jung et al., 2005). CTHA fractions are also presumed
25.0 ± 0.1 °C by a LTD6G temperature control unit (Grant, Sarose, to possess more ‘‘labile” peptidic bonds and to present a lower aro-
UK). In order to compare the fluorescence intensities of all fluores- maticity degree than that of catechol–glycine humic acid. Finally, a
cence spectra and to avoid time apparatus drift, the fluorescence great difference between natural and synthetic humic-like
was normalized with a 0.16 mg L1 fluorescein solution buffered
at pH 10. Auger’s correction (Auger et al., 1995) was used to correct
Table 1
the DOM absorption. Absorbance was measured by means of a Per- Humic substances under investigation (Jung et al., 2005 excepted for).
kin Elmer (Boston, USA) Lambda-2 spectrophotometer. The syn-
Humic acid O/Ca H/Oa (O + N)a/Ca H/Ca Total acidity HIb
thetic humic-like substance (SyHA) and natural humic acids
fraction (meq g1)
(NHA) intrinsic fluorescence were corrected by subtracting the
CGHA 0.73 0.032 1.49 0.023 3.7 84.6
emission intensities of SyHA and NHA solutions without pyrene.
CTHA2 0.77 0.048 3.87 0.037 5.6 82.8
CTHA1 1.40 0.048 5.61 0.067 5.4 76.6
2.4.1. Determination of binding constants by fluorescence NHA1 0.40 0.224 2.45 0.089 13.1 51.1
Fluorescence quenching identifies an interaction between the NHA2 0.56 0.128 2.08 0.072 18.0 52.2
fluorophore and another species leading to a diminution or even a

Atomic ratio.
b
an extinction of the fluorescence signal. This interaction may take HI: hydrophobic indices determined by RP-HPLC (see Jung et al. (2005)).
substance concerning both aromatic and peptidic structures is re- To determine Cmax for our different HS fractions, the equations
vealed by the elemental composition. were (a) y = 3.34x + 72.4 (R2 = 0.905) (b) y = 10.02x + 71.7 (R2 =
0.943) (c) y = 1.97x + 71.8 (R2 = 0.681) (d) y = 7.98x + 69.9
(R2 = 0.989) (e) y = 12.99x + 68.9 (R2 = 0.948) respectively for
3.2. Surface tension measurements at the air/water interface CTHA, CTHA2, CGHA, NHA1 and NHA2. The corresponding Cmax
and Amin values are reported in Table 2. According to Chen and
As illustrated in Fig. 1, the surface tensions of all humic sub- Schnitzer (1978), the crowding of molecules at the liquid–gas
stances decrease with an increase in humic substances (HS) con- interface could be partially explained by a combination of two
centration. No significant differences between solutions of humic parameters: the molecular weights and the charge density (more
acid alone and humic acid and pyrene could be detected (data functional groups per unit area) of the substances. These authors
not shown). have for instance determined (Cmax, Amin) values of (1.722
Catechol triglycine low-molecular-weight humic acid (CTHA1) 1010 mol cm2, 96.5 Å2 molecule1) and (4.107 1010 mol cm2,
and catechol–glycine humic acids (CGHA) solutions present a de- 40.4 Å2 molecule1) respectively for a fulvic and humic acid frac-
crease of surface tension as a function of the logarithm of the con- tion. Considering our results, the maximal surface adsorption in-
centration, that is reminiscent of those of natural soil and leaf crease from CGHA, CTHA1 to CTHA2. However, CTHA2 and CGHA
humic acids extracts (Chin et al., 1996; Tuckermann and have the same dialysis molecular weight cut-off. Besides, this
Cammenga, 2004; Quagliotto et al., 2006). On the other hand, the parameter neither correlate with the number of acidic groups nor
two natural humic acids extracts (NHA1 and NHA2) and catechol the hydrophobic indice of the molecules. Without taking account
triglycine high-molecular-weight humic acid (CTHA2) show hyper- of the poor linearisability of the results, we can suggest, as Chen
bolic curves that resemble some determined in the literature and and Schnitzer (1978), that other parameters, such as particle shape
reported in Fig. 1 (Quagliotto et al., 2006). These different behav- or size should be considered to explain the adsorption capacity of
iours suggest that natural humic acids and CTHA higher molecular the HS.
weight fraction (CTHA2) likely present a molecular crowding at the
interface, whereas synthetic CGHA and CTHA1 seem to behave like
less charged entities concerning the number of acidic groups point 3.3. Stern–Volmer plots
of view: they can deliver less protons in the solution than the other
fractions (Table 1). The critical micellar concentration (CMC) values The I0/I2 versus synthetic humic-like substance (SyHA) and nat-
are reported in Table 2. The CMC of our natural extracts are twice ural humic acid (NHA) concentration plots are shown in Fig. 2. For
lower than those determined in the literature for humic substances all humic fractions, the quenching plots present a linearisable evo-
extracted from marine sediments (typically close to 1 g L1 accord- lution in the range of concentrations investigated. Moreover, the
ing to Shinozuka and Lee, 1991) but close to those of compost hu- evolution of the quenched fluorescence in solution is very similar
mic acid-like (typically 403 mg L1 according to Quagliotto et al., to other authors studies, especially for CTHA1 fraction (Peuravuori,
2006). The two natural extracts (summer, NHA1 and autumnal, 2001, presented in Fig. 2). The slopes of Stern–Volmer plots de-
NHA2) possess a similar critical micellar concentration (CMC) crease from synthetic catechol–glycine humic acid (CGHA), the
(with respect to experimental accuracy). Among synthetic frac- two catechol–triglycine humic acid (CTHA2 and CTHA1) to natural
tions, CTHA2 is the only material that shows a true CMC. It can also NHA1 and NHA2.
be hypothesized that to allow the formation of a micellar phase, It has been established (Gauthier et al., 1987; Lakowicz, 1999)
the humic substances should present a good balance between that two distinct kinds of fluorophore can exist in solution, i.e.,
functional groups (especially protein-like, when comparing results some molecules are bound to the humic-like material whereas oth-
from CTHA to CGHA) and high-molecular weight (comparing ers remain free in solution. The first group has a totally quenched
CTHA2 with CTHA1). fluorescence (this mechanism is usually called ‘‘static quenching”)
Fig. 1. Surface tension versus humic acid (HA) concentration (mg L1). Experimental errors: not represented (smaller than marks).
Table 2
Stern–Volmer association constants, critical micellar concentrations, maximal surface adsorption and minimal cross-sectional area values for humic substances.
Humic acid fraction KSVa (rc = ±0.15 104 mL g1) CMCb (rc = ±50 mg L1) Cmaxd (mol cm2) Amind (Å2 molecule1)
4 1
CGHA 5.50 10 mL g 0.797 208.3
CTHA2 4.50 104 mL g1 320 mg L1 4.056 40.9
CTHA1 1.51 104 mL g1 1.352 122.9
NHA1 0.52 104 mL g1 500 mg L1 3.230 51.4
NHA2 0.31 104 mL g1 450 mg L1 5.259 31.6
a
Stern–Volmer constants with 106 M pyrene, [HA] 1–30 mg L1, pH 8.
b
Critical micellar concentration values determined as HA first concentration for c stabilization value (no stabilization value obtained).
c
Maxima standard deviations values for all fractions.
d
Values in the tables have been multiplied by 1010.
Fig. 2. Synthetical and natural Stern–Volmer plots ([HA] in mg L1). Experimental errors: not represented (smaller than marks).
whereas the second group shows no interaction with the humic the Stern–Volmer constants determined (Table 2). These results
matrix, but remains sensitive to other types of quenching (this suggest that the more hydrophobic the humic substances at the lo-
mechanism is usually called ‘‘dynamic quenching”). Both quench- cal scale, the more effective the quenching fluorescence of pyrene.
ings (static and dynamic) can be observed by linearisable Stern– A previous work (Jung et al., 2005) has shown that catechol–gly-
Volmer plots at concentrations generally below 20–30 mg L1. cine humic acid (CGHA) presents the most condensed aromatic
Since no significant change in interfacial tension was observed structure. It can then be concluded that the more condensed the
when adding the pyrene to the suspension of humic material, we humic substance, the more efficiently quenched pyrene fluores-
can assume that the obtained values are mainly due to humic cence. Moreover, the binding constants could be discussed with
materials. Besides, all combined data suggest a close physical asso- the total number of acidic groups determined in Table 1: the more
ciation between the fluorophore and the quencher attributed to an the humic substances are acid, the less pyrene could be associated
association of the former into hydrophobic pockets. with, whatever kind of acidic group is concerned. Indeed, a previ-
The binding constants (KSV) obtained for our systems (Table 2) ous study (Jung et al., 2005) showed that carboxylic groups are
are in good agreement with the range found in the literature for more abundant in the synthetic polymers whereas phenolic groups
different polycyclic aromatic hydrocarbons (PAHs) complexed with are ten times more numerous in natural substances. These results
natural organic matter (Gauthier et al., 1986; Borisover et al., 2006; suggest that either (1) phenolic or carboxylic groups are repulsive
Peuravuori, 2001). For both natural and synthetic humic-like sub- for pyrene association with humic substances or (2) a higher num-
stance, we can conclude that pyrene is incorporated into the ber in acidic groups, i.e., a higher number in negative charges im-
hydrophobic sites of humic substances. plies an improved hydratation. The second hypothesis (2), which
The much higher KSV values for catechol–glycine humic acid seems to be the more realistic, suggests a change in conformation
(CGHA) may be explained by the differences in elemental analysis of the HA molecules due to the electrostatic repulsion between
reported in Table 1. The H/C atomic ratio, which could provide an multiple negative charged moieties. As a result, a potentially
estimation of the aromaticity degree (Chen et al., 1977; Thurman, hydrophobic interior of HA could be suppressed, and the driving
1985), is substantially lower (greater aromaticity) for this humic force for association with pyrene would then be eliminated.
substance. The greater aromaticity could also be responsible for The very slight difference in the Stern–Volmer slopes relative to
an increase of van der Waals forces due to p bonding interactions catechol triglycine higher molecular-weight humic acid (CTHA2)
between the aromatic structure of the pyrene and the aromatic and catechol–glycine humic acid (CGHA) suggests that peptidic
moieties of the synthetic humic-like substance (SyHA) and natural bonds do not significantly contribute during the first step of the
humic acid (NHA) (Chin et al., 1997). Besides, the hydrophobic interaction mechanism (static quenching). Other molecular parts
indices (HI in Table 1) calculated by reversed-phase high liquid of natural humic acid (summer extract, NHA1) not identified in this
chromatography in a water/acetonitrile elution gradient (Jung study, could be responsible for other complex photophysical elec-
et al., 2005) for the different fractions show good correlation with tron charge transfer, which should then lead to higher I0/I2 values.
Moreover, our results showed that KSV values obtained for syn- Acknowledgments
thetical humic acids are at least three times higher than those for
natural ones. We can assume that synthetic humic-like substance This work was carried out within the frame of the programme
(SyHA) hydrophobic domains are more tightly structured than ‘‘Zone Atelier du bassin de la Moselle” with Grants from Région
those of natural humic acids (NHA). Pyrene also shows more signif- Lorraine and CNRS. The field-centrifuge provided by Agence de
icant diffusional freedom with NHA and a lesser tendency to be l’Eau Rhin-Meuse for natural extracts sampling is gratefully
captured by humic acids. acknowledged.
Summarizing our results, there are two main domains for
pyrene–humic acids (HA) associations with increasing HA Appendix A
concentrations:
(i) a low-concentration region in the 0–30 mg L1 range, where Amin minimal cross-sectional
molecular photophysical ‘‘weak-adsorption” interactions cs surface tension
between the fluoroprobe ‘‘pyrene” and the most hydropho- [Q]0 quencher initial concentration
bic HA sites take place. This region allows to determine the CGHA catechol/glycine humic acid
Stern–Volmer binding constants between pyrene and HA CMC critical micellar concentration
molecules; CTHA catechol/triglycine humic acid
(ii) a high-concentration region at least greater than 250 mg L1 CTHA1 CTHA having a 1–3.5 kDa mean molecular weight
corresponding to a ‘‘pseudo-micellar” organization phase at CTHA2 CTHA having a >8 kDa mean molecular weight
the air/water interface. This latter region reveals that organic FTIR Fourier Transform InfraRed
matter micellar-like behaviour is independent from pyrene HI hydrophobic indice
entrapment capacity. In accordance with previous literature HS humic substances
(Peuravuori, 2001; Borisover et al., 2006; Nakashima et al., I fluorescence intensity in the presence of quencher
2007), these results confirm that a true entrapment in a I0 fluorescence intensity in the absence of quencher
micellar phase is not a prerequisite to the quenching pro- I1, I2, pyrene fluorescence intensity respectively at 372,
cess. Moreover, the presence of pyrene seems to have no I3, 377, 382, 390 and 395 nm (in ethanol after excitation
effect on a humic micelle formation. Considering our aque- I4, I5 at 332 nm)
ous pyrene concentration (0.2 mg L1) and the mean binding IHSS International Humic Substances Society
constant of 104 determined in this study, we can expect the KSV Stern–Volmer quenching constant
pyrene concentration in a micelle about 2000 mg kg1. Such NHA natural humic acids
result is particularly meaningful for drinking water treat- NHA1 NHA extracted during the summer season
ment, since pyrene could be entrapped in organic micelles NHA2 NHA extracted during the autumn season
contained in the raw surface water to be treated. Due to NOM natural organic matter
the coagulation process (such as charge destabilization), PAH polycyclic aromatic hydrocarbon
the pyrene could then be released at high coagulant concen- SyHA synthetic humic-like substance
tration in the treated water (Kazpard et al., 2006). Cmax maximal surface adsorption
4. Conclusion
Hydrophobic characteristics responsible for pyrene entrapment

References
in humic structures are mainly due to polarity and polarizability,
both correlated with the aromaticity and the presence of aliphatic Achten, C., Hofmann, T., 2009. Native polycyclic aromatic hydrocarbons (PAH) in
chains in humic acids, molecular weight, nitrogen nature and/or coals – a hardly recognized source of environment contamination. Sci. Total
Environ. 407, 2461–2473.
content and finally the number of acidic groups of the substances.
Agence de l’Eau Rhin-Meuse, 2005. <http://www.eau-rhin-meuse.fr/Berm/htm>.
The comparison of natural and synthetic humic acid interfacial Andreux, F., Golebiowska, D., Metche, M., 1980. Oxydative polymerization of O-
and Stern–Volmer analysis suggests that the hydrophobic interac- diphenols in presence or absence of amino-acids. Topics on (catechol–
tions are responsible for the organization of the aggregates, regard- glycocolle) and (catecholdiglycylglycine). General Assembly of Polyphenols
Group Report, Logrono, Spain, Bulletin 9, pp. 178–188.
less of HA origin. Therefore, even if the hydrophobic characteristics Auger, R.L., Jacobson, A.M., Domach, M.M., 1995. Aqueous phase fluorescence
of humic acid are well correlated with the pyrene quenching quenching technique for measuring naphthalene partition coefficients in
capacity, the specific interactions with aromatic molecular sites nonionic surfactant Micelles. Environ. Sci. Technol. 29, 1273–1278.
Borisover, M., Laor, Y., Bukhanovsky, N., Saadi, I., 2006. Fluorescence-based evidence
depend on the whole tridimensional photophysical environment for adsorptive binding of pyrene to effluent dissolved organic matter.
of humic acids in solution. Chemosphere 65, 1925–1934.
Surface tension measurements revealed an increase in surface Bouloubassi, I., Saliot, A., 1991. Composition and sources of dissolved and
particulate PAH in surface waters from the Rhone Delta (NW Mediterranean).
activity as a function of concentration for each humic fraction inde- Mar. Pollut. Bull. 22, 588–594.
pendently of pyrene presence in solution. Those results confirm the Chen, Y., Schnitzer, M., 1978. The surface tension of aqueous solutions of soil humic
possibility of specific intermolecular interactions occurring during substances. Soil Sci. 125, 7–15.
Chen, Y., Senesi, N., Schnitzer, M., 1977. Information provided on humic substances
pyrene entrapment within humic acids. From the comparison of by E4/46 ratios. Soil Sci. Soc. Am. J. 41, 352–358.
SyHA and NHA, we showed that the presence of pyrene had no ef- Chin, Y.P., Aiken, G.R., Danielsen, K.M., 1997. Binding of pyrene to aquatic and
fect on humic micelle formation. The pyrene fluorescence quench- commercial humic substances: the role of molecular weight and aromaticity.
Environ. Sci. Technol. 31, 1630–1635.
ing study pointed out the capacity of HA to incorporate large
Chin, Y.P., Buterbaugh, J.M., Gustafson, T.J., Traina, S.J., Danielsen, K.M., 1996.
hydrophobic molecules into its micellar core, without change of Response to comment on ‘‘solubility enhancement and fluorescence quenching
its CMC value determined in the absence of other organic mole- of pyrene by humic substances”. Environ. Sci. Technol. 30, 1409–1411.
cules. Other parameters, like size and shape of the humic sub- Chin, Y.P., Weber Jr., W.J., 1989. Estimating the effects of dispersed organic
polymers on the sorption of contaminants by natural solids I: a predictive
stances should be considered in the future to explain the thermodynamic humic acid–organic solute interaction model. Environ. Sci.
adsorption capacity at the air/water interface. Technol. 23, 978–985.
Chiou, C.T., Kile, D.E., Brinton, T.I., Malcolm, R.L., Leenheer, J.A., Mac Carthy, P., 1987. Nakashima, K., Maki, M., Ishikawa, F., Yoshikawa, T., Gong, Y.-K., Miyajima, T., 2007.
A comparison of water solubility enhancements of organic solutes by aquatic Fluorescence studies on binding of pyrene and its derivatives to humic acid.
humic materials and commercial humic acids. Environ. Sci. Technol. 21, 1231– Spectrochim. Acta, Part A 67, 930–935.
1234. Pan, B., Ying, B.S., Liu, W.X., Tao, S., Lin, X.M., Zhang, X.X.M., Zhang, Y.X., Xiao, Y., Dai,
Conte, P., Piccolo, A., 1999. Conformational arrangement of dissolved humic H.C., Yuan, H.S., 2006. Distribution of sorbed phenanthrene and pyrene in
substances. Influence of solution composition on association of humic different humic fractions of soils and importance of humin. Environ. Pollut. 143,
molecules. Environ. Sci. Technol. 33, 1682–1690. 24–33.
Countway, R.E., Dickhut, R.M., Canuel, E.A., 2003. Polycyclic aromatic hydrocarbon Peuravuori, J., 2001. Partition coefficients of pyrene to lake aquatic humic matter
(PAH) distributions and associations with organic matter in surface waters of determined by fluorescence quenching and solubility enhancement. Anal. Chim.
the York River, VA Estuary. Org. Geochem. 34, 209–224. Acta 429, 65–73.
Engebretson, R.R., von Wandruska, R., 1994. Microorganization in dissolved humic Piccolo, A., 2001. The supramolecular structure of humic substances. Soil Sci. 166,
acids. Environ. Sci. Technol. 28, 1934–1941. 810–832.
Gauthier, T.D., Seitz, W.R., Grant, C.L., 1987. Effects of structural and compositional Piccolo, A., 2002. The supramolecular structure of humic substances: a novel
variations of dissolved humic materials on pyrene K sub(oc) values. Environ. Sci. understanding of humus chemistry and implications in soil science. Adv. Agron.
Technol. 21, 243–248. 75, 57–134.
Gauthier, T.D., Syhane, E.C., Guerin, W.R., Seitz, W.R., Grant, C.L., 1986. Fluorescence Piccolo, A., Celano, G., Conte, P.J., 1996. Adsorption of glyphosate by humic
quenching method for determining equilibrium constants for polycyclic substances. J. Agric. Food Chem. 44, 2442–2446.
aromatic hydrocarbons binding to dissolved humic materials. Environ. Sci. Piccolo, A., Conte, P., Scheunert, I., Paci, M., 1998. Atrazine interactions with soil
Technol. 20, 1162–1166. humic substances of different molecular structure. J. Environ. Qual. 27, 1324–
Ghosh, U., Gillette, J.S., Richard, G.L., Zare, R.N., 2000. Microscale location 1333.
characterization, and association of polycyclic aromatic hydrocarbons on Quagliotto, P., Montoneri, E., Tambone, F., Fabrizio, A., Gobetto, R., Viscardi, G., 2006.
harbour sediment particles. Environ. Sci. Technol. 34, 1729–1736. Chemicals from wastes: compost-derived humic acid-like matter as surfactant.
Hetkorn, N., Ruecker, C., Meringer, M., Gugisch, R., 2007. High-frequency Environ. Sci. Technol. 40, 1686–1692.
measurements: indispensable tools at the core of the molecular-level analysis Schulten, H.R., Gleixner, G., 1999. Analytical pyrolysis of humic substances and
of complex systems. Anal. Bioanal. Chem. 389, 1311–1327. dissolved organic matter in aquatic systems: structure and origin. Water Res.
Hua, G., Lyons, B., Killham, K., Singleton, I., 2009. Potential use of DNA adducts to 33, 2489–2498.
detect mutagenic compounds in soil. Environ. Pollut. 157, 916–921. Shinozuka, N., Lee, C., 1991. Aggregate formation of humic acids from marine
Hur, J., Lee, D.-H., Shin, H.-S., 2009. Comparison of the structural, spectroscopic and sediments. Mar. Chem. 33, 229–241.
phenanthrene binding characteristics of humic acids from soils and lake Short, J.W., Irvine, G.V., Mann, D.H., Maselko, J.M., Pella, J.J., Lindeberg, M.R., 2007.
sediments. Org. Geochem. 40, 1091–1099. Slightly weathered Exxon Valdez oil persists in Gulf of Alaska beach sediments
Jeffrey, A.M., Weinstein, I.B., Jennette, K., Grzeskowiak, K., Nakanishi, K., Harvey, after 16 years. Environ. Sci. Technol. 41, 1245–1250.
R.G., Autrup, H., Harris, C., 1977. Structures of benzo[a]pyrene nucleic acid Sielechi, J.-M., Lartiges, B.S., Kayem, G.J., Hupont, S., Frochot, C., Thieme, J., Ghanbaja,
adducts formed in human and bovine bronchial explants. Nature (London) 269, J., d’Espinose de la Caillerie, J.B., Barres, O., Kamga, R., Levitz, P., Michot, L.J.,
348–350. 2008. Changes in humic acid conformation during coagulation with ferric
Jung, A.-V., Frochot, C., Parant, S., Lartiges, B.S., Selve, C., Viriot, M.-L., Bersillon, J.-L., chloride: implications for drinking water treatment. Water Res. 42, 2111–2123.
2005. Synthesis of amino-phenolic humic-like substances and comparison with Sierra, M.M.D., Rauen, T.G., Tormen, L., Debacher, N.A., Soriano-Sierra, E.J., 2005.
natural aquatic humic acids: a multi-analytical techniques approach. Org. Evidence from surface tension and fluorescence data of a pyrene-assisted
Geochem. 36, 1252–1271. micelle-like assemblage of humic substances. Water Res. 39, 3811–3818.
Kalyanasudaram, K., Thomas, J.K., 1977. Environmental effects on vibronic band Taubman, A.V., 1956. The structure of adsorption layers in dilute solution. Zh. Fis.
intensities in pyrene monomer fluorescence and their application in studies of Khim. 26, 389–398.
micellar systems. J. Am. Chem. Soc. 99, 2039–2044. Thurman, E.M., Malcolm, R.L., 1981. Preparative isolation of aquatic humic
Kazpard, V., Lartiges, B.S., Frochot, C., De la Caillerie, J.B., d’Espinose, Viriot, M.L., substances. J. Am. Chem. Soc. 4, 463–466.
Portal, J.M., Goerner, T., Bersillon, J.L., 2006. Fate of coagulant species and Thurman, E.M., 1985. Organic Geochemistry of Natural Waters. Martinus Nijhoff/
conformational effects during the aggregation of a model of a humic substance Junk Publishers, Dordrecht, The Netherlands.
with Al sub(13) polycations. Water Res. 40, 1965–1974. Tuckermann, R., Cammenga, H.K., 2004. The surface tension of aqueous solutions of
Lakowicz, J.R., 1999. Principles of Fluorescence Spectroscopy, second ed. Kluwer some atmospheric water-soluble organic compounds. Atmos. Environ. 38,
Academic/Plenum, New York. 6135–6138.
Myers, D., 1992. Surfactants Science and Technology, second ed. Wiley-VCH, Zumstein, J., Buffle, J., 1989. Circulation of pedogenic and aquagenic organic matter
Weinheim. in an eutrophic lake. Water Res. 23, 229–239.
Science of the Total Environment 409 (2011) 1683–1688
Science of the Total Environment

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / s c i t o t e n v
Impacts of rainfall events on runoff water quality in an agricultural environment in

temperate areas
Ianis Delpla a, Estelle Baurès a, Aude-Valérie Jung a,b, Olivier Thomas a,⁎
a
Environment and Health Research Laboratory, School of Advanced Studies in Public Health, 35043 Rennes, France
b
School of Environmental Engineering, Campus de Ker-Lann, Avenue Robert Schumann, 35170 Bruz, France
Article history: Since a rise in dissolved organic carbon (DOC) concentrations has been observed for surface waters at least
Received 1 September 2010 over the last two decades, a change in weather conditions (temperature and precipitations) has been
Received in revised form 8 January 2011 proposed to partly explain this increase. While the majority of DOC delivery from soils to stream occurs during
Accepted 18 January 2011
rainfall events, a better understanding of the rainfall influence on DOC release is needed. This study has been
Available online 12 February 2011
conducted in Brittany, western France, on agricultural experimental plots receiving either cattle manure (CM)
Keywords:
or pig slurry (PS) as fertilizers in accordance with local practices. Each plot was instrumented with a flow
Rainfall events meter and an auto sampler for runoff measurements. The results show that export of DOC during high
Runoff intensity events is higher than during lower intensity rainfalls. Fertilization has a noticeable impact on total
Organic carbon organic carbon (TOC) fluxes with an increase of five to seven folds for PS and CM respectively. If TOC shock
Fertilizers load occurs shortly after the rainfall peak, DOC maximum appears with the first flush of the event. Organic
Water quality carbon (OC) is mainly under colloidal (41.2%) and soluble (23.9%) forms during the first stage of a rainfall
event and a control of rainfall intensity on OC colloidal transport is suggested. These findings highlight the
potential risk of receiving water quality degradation due to the increase of heavier rainfall events with climate
change in temperate areas.
© 2011 Elsevier B.V. All rights reserved.
1. Introduction land use and agricultural practices have also been proposed as drivers of
rising DOC trends, with for example the increasing manure spreading on
The occurrence of extreme events such as droughts and heavy soils (Morel et al., 2009). Intensive agriculture often generates large
rainfalls is expected to increase with climate changes. For countries in quantities of animal manure spread onto agricultural land, because it is
the temperate zone, a decrease in the number of rainy days but an considered as a valuable source of macro- and micronutrients (N, P, K,
increase in the average volume of each rainfall event is predicted (Bates Ca, and Mg) for crop growth, and a low-cost alternative to mineral
et al., 2008). In the past decades, European countries' major river fertilizers (Sharlpley and Smith, 1995). However, over-application of
catchments have experienced several flood episodes, and a higher-than- animal manure to soils can generate an excess of nitrate and phosphate
average rate of global sea level rise (of approximately 3.1 mm/year) has that can ultimately contaminate both ground- and river waters. Manure
been estimated (WHO, 2010). Consequently, a change in climatic factors over-application can also strongly increase the amount of soil–water
may also alter the transport of organic matter through runoff to surface extractable organic matter, thereby leading to an increase of organic
waters. An increasing awareness among scientists concerning an matter fluxes (Chantigny, 2003). Some undesirable organic compounds
enhancement of the carbon cycle with climate changes is emerging. In (such as veterinary/pharmaceuticals compounds) or pathogenic micro-
fact, several recent studies highlight an increase in Dissolved Organic organisms can be present in manure (e.g., Hoese et al., 2009), and thus
Carbon (DOC) in rivers and lakes over the last two decades in Europe could migrate in the surface water after spreading.
and North America (Evans et al., 2005; Hejzlar et al., 2003; Hongve et al., High organic matter (and nitrate) concentrations in surface waters
2004; Monteith and Evans, 2005; Skjelkvåle et al., 2005; Worrall et al., are of important concern because of the capacity limitation of small
2004). Various explanations have been proposed as changes in air water supply systems leading sometimes to the noncompliance with
temperature, increased precipitations (Evans et al., 2005), increasing drinking water standards (WHO, 2010). It could also constitute a
atmospheric CO2 concentration (Porcal et al., 2009) and changes in substrate for bacterial regrowth within the distribution system and an
atmospheric deposition chemistry (Monteith et al., 2007). Changes in adsorption support for some micropollutants such as pesticides,
pharmaceuticals and metals (Morel et al., 2009). As a consequence, an
increase in hazardous substances (pesticides, pharmaceuticals and
⁎ Corresponding author. Tel.: +33 2 99 02 29 20; fax: +33 02 99 2 29 29. other micropollutants) in the water resources following extreme events,
E-mail address: olivier.thomas@ehesp.fr (O. Thomas). could lead to a potential threat for human health (Delpla et al., 2009).
0048-9697/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.scitotenv.2011.01.033
1684 I. Delpla et al. / Science of the Total Environment 409 (2011) 1683–1688
Hydrologic flow paths are an important control on surface water temperature and relative humidity by a Campbell Scientific CS215
dissolved organic matter concentrations during high flows (Fellman probe. Measurements of additional parameters such as wind speed and
et al., 2009). Several studies have investigated the effects of various direction (Young Wind sentry 03002-5) and soil temperature were also
hydrological conditions on water quality in small catchments and made. A data logger (Campbell scientific CR800) continuously moni-
experimental plots. It was found that most of the DOC is exported tored the data and transmitted it via a cellular modem. A message was
from catchments during storm events (Hope et al., 1994). According send on mobile phones each time a threshold of 3 mm/h was reached
to Clark et al. (2007), 36% to more than 50% of the annual DOC export which allowed a fast sampling intervention. For each rainy event, all
occurs during short-time high-intensity rainfall events. Similar results data were downloaded from the data logger at the end. Finally, soil
have been obtained by studies conducted in forested catchments moisture was measured monthly by soil sampling and gravimetry.
(Frank et al., 2000; Inamdar et al., 2006; Kawasaki et al., 2008),
grassland catchments (Frank et al., 2000) and peatlands (Clark et al., 2.3. Sampling and analysis
2007). In agricultural environment, many studies also focused on the
erosion impact of rainfalls at the plot scale (Assouline and Ben-Hur, A runoff water sampling device was designed and installed on each
2006; Le Bissonnais et al., 1998; Raclot et al, 2009), or on management of the three plots. For each plot, a drain pipe placed at the lowest point
practices impact on nutrient (Kato et al., 2009) and organic matter led the run off to a triangular profile spillway retainer for flow
(Veum et al., 2009) export at the watershed scale. measurement. The flow meter (ISCO 4230) was programmed to
Jacinthe et al. (2004) studied the influence of various soil ensure a sample collection for a predetermined volume (0.6 or 1.2 L)
management practices in small watersheds on labile carbon transport with a 24 bottle auto sampler (ISCO 3700). In addition, a final
and found that soil organic carbon losses were proportional to runoff. collection tank was placed at the outlet of the installation. Total runoff
Since most of the DOC is exported from soils to streams during storm volume was calculated from the sample volume and one of the
events, there is a need to improve knowledge on carbon losses at the collection tanks.
plot scale namely for intensive agricultural areas. This need is linked As the runoff regime is non-permanent, runoff response to a
on the existing monitoring practices collecting water samples daily, rainfall is variable from few minutes for the heaviest events to few
weekly, or monthly, depending on the objectives of researches. This hours for the less intense ones. On-site measurements were
sampling frequency, longer than the time lag of a rainy event performed with Neotek Ponsell probe for pH and with Hach Lange
(few hours to 1 or 2 days) is thus not adapted and consequently, probe for conductivity. Samples collected were stored at 4 °C until
the impact of rainfall events on water quality degradation could be analysis immediately after collection.
underestimated (Clark et al., 2007). Thus, a focus has to be made on Total and dissolved organic carbon (TOC and DOC) were
the whole rainfall event to assess its impact on runoff water quality. determined by thermal oxidation coupled with infrared detection
This study has been conducted in the Brittany region, western (Multi N/C 2100 Analytik Jena) following acidification with HCl (NF
France, with a land use dominated by intensive farming (Gascuel-Odoux EN 1484, 1997). Particulate organic carbon (POC) was calculated by
et al., 2009) and a resulting high production of animal waste. The main subtracting DOC to TOC. Kjeldhal nitrogen (KN) was analyzed
objective of this work is to study the runoff water fluxes during various following mineralization with H2SO4 (Block digester 435 Büchi) and
rainfall events (storm and shower events) by an adapted monitoring. distillation (distillation unit 360 Büchi) by titration (Metrohm Titrino)
according to NF EN 25663 (1994). Turbidity was measured by
2. Materials and methods nephelometry (NF EN ISO 7027, 2000) and UV spectra were acquired
with a UV–vis spectrometer PerkinElmer Lambda 35, with a 10 mm
2.1. Experimental field optical path length quartz cell (acquisition step: 1 nm, slit width:
2 nm, and scan speed: 1920 nm/min). Dissolved organic carbon and
The study was carried out in an experimental site (48°07 N and 1°43 UV spectra acquisition were carried out after a filtration through
O) managed by the Institut National de la Recherche Agronomique 0.45 μm HAWP nitrocellulose filter papers (Millipore).
(INRA Rennes, France) since 1993. The site is at 40 m above sea level, Finally, fractionations were realized for some samples at specific
and subjected to temperate climatic conditions with a mean annual times (first flush, stationary flow and last runoff) for one of the events.
temperature of 12.5 °C and a mean annual rainfall of 750 mm. 70% of the Three fractions were considered: soluble (b0.1 μm), colloidal (0.1–
rainfall falls from September to March (Morvan et al., 2007). The 1.2 μm) and particulate (N1.2 μm). Samples were filtered with 1.2 μm
experimental field was divided in 15 plots of 150 m2 continuously maize AP40 Millipore glass fiber filters and 0.1 μm VVLP Millipore poly-
cropped. The average slope is 3% and the soil was characterized as a vinylidene fluoride filters. UV spectra and organic carbon were
leached Aeolian loam lying on schist material and referenced as a Luvisol measured on each fraction.
Redoxisol (Morvan et al., 2007) with the following composition: silt
71%, clay 15%, and sand 14%. This experimental work was conducted 2.4. Data handling
from August 2009 to March 2010 on three plots. According to local
animal waste management practices, two plots amended respectively The estimation of the fluxes (expressed in g/ha) was calculated for
with cattle manure (CM, 50 t/ha, every 2 years) and pig slurry (PS, each of the three plots and of the events as the product of
40 m3/ha, each year) and one control (C) without spreading were concentration by the corresponding runoff volume divided by the
selected. After the ploughing, cattle manure was applied on April 3, 2009 plot surface. Rainfall maximum intensity was expressed for a 5 minute
and pig slurry on June 3, 2009, the soil being harrowed superficially step in mm/h. Correlations were studied using non parametric
(10 cm) after fertilizer application on the soil. Maize was sown on each Spearman correlation test as water quality parameters and flow did
plot on April 23, 2009 and harvested on September 9, 2009. In order to not show a normal distribution.
study the impact of winter spreading on water quality, plots were
fertilized again on January 21, 2010, for PS plot and January 28, 2010 for 3. Results
CM plot at the same rate of application than the previous one.
3.1. Fluxes variation
2.2. Weather data acquisition
A total of fifteen rainfall events have been monitored among which
A weather station was installed close to the site (150 m). Precipita- eight main events have been exploited considering the rainfall
tions were measured by a tipping bucket rain gauge (ARG 100), air intensity and runoff volume. It should be noticed that no record was
I. Delpla et al. / Science of the Total Environment 409 (2011) 1683–1688 1685
Table 1
Rainfall characteristics and fluxes of DOC, POC, TOC and KN by campaign.
Rainfall event 21/09/2009 01/11/2009 02/11/2009 08/11/2009 15/11/2009 08/02/2010 25/02/2010 27–28/02/2010
Precipitation (mm) 6.6 15.4 10.6 20.0 22.6 10.4 14.2 23.8
Duration (h) 2.2 7.0 22.3 15.4 25.1 33.5 31.6 14.8
Mean intensity (mm/h) 3.0 2.2 0.5 1.3 0.9 0.3 0.4 1.6
Max intensity (mm/h) 38.4 12.0 7.2 28.8 26.4 9.6 9.6 7.2
DOC (g/ha) Plot C 1.0 8.5 6.3 15.5 9.2 6.1 8.5 7.4
Plot CM 8.1 9.6 10.1 16.4 17.2 14.8 46.2 n.d.
Plot PS 2.2 11.7 11.5 20.1 13.8 8.3 16.6 21.9
TOC (g/ha) Plot C 1.8 17.5 8.0 28.6 75.6 8.6 17.7 17.1
Plot CM 10.5 16.5 12.5 26.8 73.8 105.7 330.3 n.d.
Plot PS 2.6 19.7 13.4 23.4 30.3 37.8 125.2 157.7
POC (g/ha) Plot C 0.8 9.0 1.6 13.1 66.3 2.5 9.2 9.7
Plot CM 2.4 6.8 2.4 10.6 56.7 90.9 284.0 n.d.
Plot PS 0.4 8.0 1.9 3.3 16.5 29.5 108.5 135.8
KN (g/ha) Plot C 0.8 4.5 11.4 11.5 9.4 2.1 4.0 4.9
Plot CM 5.9 5.4 2.0 10.0 13.4 15.7 45.7 n.d.
Plot PS 0.5 4.6 2.6 5.9 6.0 4.9 18.0 30.0
C: control, CM: cattle manure, PS: pig slurry.

n.d.: nondetermined.
available on the CM plot for the 27–28th February 2010 event due to a considering another soil types, management practices and rainfall
flow meter breakdown. Rainfall characteristics as duration (D), characteristics.
precipitation (P), mean intensity (Imean) and maximum intensity
(Imax) and fluxes of POC, TOC, DOC and KN are presented in Table 1. 3.2. Study of typical rainfall events
Concerning the variation of parameters with rain characteristics, a
strong difference was observed between control and fertilized plots, The results presented here focused on two main events in autumn
these ones showing generally higher values for highest intensity and winter, characterized by a strong shower and a heavy continuous
events, particularly after manure or slurry spreading. rainfall. Table 1 gives an expected linear relation between the rainfall
Low values of POC, TOC, DOC and KN were exported during the duration and the mean intensity (I mean = − 0.083 ⁎ D + 2.86,
heaviest event (September 21, 2009) and were explained by the very R2 = 0.92), whereas no other trends can be drawn between the rain
low runoff volume collected on all plots due to the short duration of volume, duration or intensities. According to the maximum intensity,
the rainfall (2.2 h). a shower can be identified for Imax values greater than 20 mm/h.
Besides this, some strong relations exist between water quality Notice that the beginning of each event (t0) is set up with the first
parameters. A relationship between POC and TOC is found for all plots tipping bucket of the rain gauge separated from the following by less
(control: POC = 0.92 ⁎ TOC − 6.1, R 2 = 0.97; cattle manure: than 30 min.
POC = 0.89 ⁎ TOC − 8.3, R2 = 0.99; pig slurry: POC = 0.93 ⁎ TOC − 9.5,
R2 = 0.99). These results reveal that most of the carbon is exported 3.3. Shower
under particulate form. As expected, DOC shows a positive relation-
ship with TOC and POC for amended and control plots, with higher Two main convective type rainfall events were monitored during
values after spreading. autumn (November 8, 2009 and November 15, 2009). These two
On the other hand, KN shows a positive relationship with TOC and showers were characterized by two main high intensity short
POC for amended plots, but not for control. While a slight relation is duration peaks (sub showers), and one lower intensity and longer
found between DOC and KN for control plot, these parameters are duration peak. However, the selected shower, in November 8, 2009,
strongly linked for CM and PS plots. For this last plot, two different had a higher mean intensity (Table 1) and was the first heavy rainfall
trends are noticed before and after spreading, with a higher slope after after a dry period which occurred from August to October. Variations
spreading. of DOC, TOC, POC and NK for this event are presented in Fig. 1. The
Due to plant uptake and mineralization, no clear impact of spring results obtained on pig slurry plot are presented hereafter.
2009 spreading could be noticed for autumn events, except for DOC The first sample (point 1) was collected a few times before
with higher values for fertilized plots. The impact of 2010 early (t = 34 min) the maximum rainfall intensity which occurred at the
fertilization, applied just before February events, was clearly noticed beginning of the shower (28.2 mm/h, t = 40 min). DOC concentration
on mean values for TOC fluxes (+678% for CM, +498% for PS), DOC was decreasing from point 1 (10.3 mg/L) to the end of the peak (point
fluxes (+ 148% for CM, +32% for PS) and for NK fluxes (+ 231% for 2, 4.8 mg/L) meanwhile TOC maximum was measured for the
CM, +350% for PS). It should be specified that TOC and NK fluxes from maximum intensity (peak, 13.7 mg/L) thus with a time delay of
the control plot were two fold lower after spreading. Furthermore, it 6 min with regard to the maximum DOC value. A slight increase in
could be noticed that TOC and NK export after spreading was higher DOC and TOC concentrations was observed between the end of the
for runoff waters originating from CM plot than from PS plot. two sub showers (point 3) to the maximum of the last sub event
Finally, the rainfall characteristics differed before and after winter (point 4) with values (around 6 mg/L) close to the initial ones. Similar
spreading. While rainfall depth remains unchanged, a decrease in results on DOC concentrations were obtained for the November 15,
maximum intensities and an increase in duration after spreading are 2009, rainfall event with a maximum measured at the beginning of
noticed. Mean soil and air temperatures also vary respectively from the runoff whereas the rainfall peak intensity occurred at the end of
13.5 °C before spreading to 7.2 °C after, and from 13.1 °C before the rainfall event. As for organic carbon (OC) measurements, UV
spreading to 8.1 °C after. Noticeably, an increase in soil moisture spectra revealed higher absorbance values for the first flush and the
occurred at the end of October, leading to a field capacity exceedance. end of the rainfall event. Furthermore, the dilution effect observed on
Conversely, air relative humidity didn't show strong variations (90.7% DOC measurements (points 2 and 3 of Fig. 1) is clearly noticed on
before to 93.4% after spreading). All these results have to be confirmed spectra area with a twofold decrease in absorbance values (data not
1 2 3
Flow rate (m3/h) Rainfall (mm/h)

1 2 3 4
Flow rate (m3/h) Rainfall (mm/h)
30 9
20 6
10 3
0
0 0.03
0.09
0.06 0.02
0.01
0.03
0.00
0.00
TOC 120 TOC 30
DOC
12 DOC 12
TOC (mg/L)
DOC (mg/l)
10 10
TOC (mg/L)
DOC (mg/L)
80 20
8 8
6 6
40 10
4 4
2 2
0 0 0 0
120 40
8 8 POC
POC
KN KN
6 6 90 30
POC (mg/L)
POC (mg/L)
KN (mg/L)
KN (mg/L)
4 4 60 20
2 2 30 10
0 0 0 0
0 90 180 270 360 450 0 120 240 360 480 600
Time (min) Time (min)
Fig. 1. Variation of flow and concentrations of TOC, DOC, POC and NK during the rainfall Fig. 2. Variation of flow and concentrations of TOC, DOC, POC and NK during the rainfall
event of November 8, 2009 on pig slurry plot (n = 16 samples). event of February 27–28, 2010 on pig slurry plot (n = 24 samples).
shown). KN and POC concentrations followed a rather different pattern the maximum was observed for the first flush of the runoff (around
than DOC and TOC namely at the beginning with a rapid increase in 10 mg/L, point 1). At the end of the rainfall event, corresponding to
concentration from very low values to a maximum (POC= 7.2 mg/L; the heaviest intensity stage (point 3), DOC values increased up to the
KN= 7.0 mg/L) during the highest rainfall intensity (peak). initial concentration values. Contrary to the shower, DOC did not
Spearman test shows positive correlations between DOC and TOC increase during the peak (point 2). A comparison between TOC and
(R = 0.75, p b 0.05), DOC and KN (R = 0.55, p b 0.05), TOC and KN DOC concentrations showed that a majority of organic carbon was
(R = 0.81, p b 0.05). Flow rate was positively correlated with DOC under particulate form. POC variation followed the same pattern than
(R = 0.55, p b 0.05), TOC (R = 0.71, p b 0.05) and KN (R = 0.82, p b 0.05). the TOC one, as well as KN variation, with however more slight
variations. For both, the maximum was reached for the intensity and
3.4. Continuous rainfall flow rate peak (point 2). As for shower events, UV spectra gave
consistent information with regard to those given by the DOC
Two main rainfall events were monitored during winter (February measurements (data not shown).
25, 2010, and February 27–28, 2010). These rainfalls presented different Parameters linked to flow rate were total organic carbon (R = 0.71,
characteristics relatively to autumn events, with a longer duration and p b 0.05), Kjeldhal nitrogen (R = 0.70, p b 0.05), but also total phos-
smaller rain intensity values (mean and maximum). The rainfall on 27– phorus (R = 0.79, p b 0.05), turbidity (R = 0.89, p b 0.05) and conduc-
28th February, which was the most important of the studied period by tivity (R = −0.80, p b 0.05). Moreover, TOC was correlated with KN
its duration and depth, corresponded to the tail of the storm Xynthia, (R = 0.61, p b 0.05) and with DOC (R = −0.44, p b 0.05) and pH (R =
which was a frontal precipitation type. This rainfall event was −0.70, p b 0.05). The effect of particle transport, washing and dilution
characterized by two different periods, a rather long period of by rain could explain these results.
continuous rain (6 h — 1.9 mm/h), followed by two shorter and heavier
intensity periods (70 min — 3.4 mm/h; 50 min — 5.1 mm/h) (Fig. 2). 3.5. Fractionation experiment
Thus, contrary to the previous case, a more important delay was
observed between the beginning of the rainfall and the one of the runoff. Colloidal materials in waters are of great importance to study as
This can be explained by relatively low intensities and runoff flows. they are a significant vector of adsorbed and particle bound
Several observations can be highlighted concerning the para- contaminants (Gimbert and Worsfold, 2009). In order to study the
meters variation. For TOC, the concentration values were very high repartition of organic carbon in runoff waters, fractionation experi-
(notice that spreading occurred few weeks ago) and regularly ments were conducted on the February 27–28, 2010, rainfall event for
increasing up to a maximum (around 115 mg/L, point 2) the PS plot runoff waters (Fig. 3). Three fractions were considered:
corresponding to the intensity and runoff flow peak. Then, TOC soluble (b0.1 μm), colloidal (0.1–1.2 μm) and particulate (N1.2 μm).
quickly decreased to increase again with the following peak. On the Remind that DOC corresponds to the OC of the 0.45 μm filtrate,
other hand, as for shower, DOC concentrations were rather low and including thus the soluble fraction and the small colloids.
I. Delpla et al. / Science of the Total Environment 409 (2011) 1683–1688 1687
Rainfall (mm/h) 8 Northern Europe being potentially waterlogged during the rainy late
6 autumn and winter period, the runoff volume produced during a
4 rainfall event tends to increase with a risk of pollutant transport. For a
2 storm, no similar experiments have been reported to our knowledge.
A comparison between November and February rainfalls shows a
0
100 strong impact of January spreading on TOC concentrations and to a
lesser extent for KN. Globally, TOC and KN concentrations are
correlated with flow rate for both rainfall types. While DOC maximum
80
Organic carbon (%)
loads are measured at the beginning of a shower event, this parameter

seems to evolve differently during a storm event, with a slow dilution
60 during the continuous stage of the rainfall followed by an increase
after a rise in rainfall intensity leading to the mobilization of small
40 colloids. The existence of the surface crust seemed to limit the DOC
export, the surface being washed at the beginning of the event. By
comparing these two rainfall events, some assumptions about their
20
expected impact on surface water quality could be made. On one
hand, a strong DOC pulse, induced by a shower event and
0
independent of the occurrence of the rainfall peak intensity, could
140 182 365 500 518
lead to fast surface water degradation depending obviously on the
Time (min) watershed characteristics and land-uses. On the other hand, storm
Particulate (>1.2µm) Colloids (1.2-0.1µm) Soluble (<0.1µm) events have the potential to lead to a strong export of organic carbon
into receiving waters, especially under colloidal form if the rainfall
Fig. 3. Comparison of total organic carbon distribution with rainfall intensity for each
fraction.
intensity is high and the event occur few times after a spreading.
Finally, for the fractionation experiment, the findings are consis-
tent with the results of Gimbert and Worsfold (2009) who found more
Organic matter fractionation gave more detailed information colloidal material contribution at the peak of the storm and a high
about the repartition of organic carbon in the three fractions and abundance of colloids for the first sample collected. The results can be
reveals that soluble organic carbon rate was at maximum during the explained by considering that for soils particularly prone to crusting,
first flush of the rainfall event (23.9%) and decreased continuously fragments resulting from raindrop detachment are generally elemen-
afterwards, whatever the rainfall intensity. Furthermore, carbon was tary particles or small micro aggregates lower than 100 μm which are
mainly under colloidal form during highest intensity stages of the easily transported by runoff (Le Bissonnais, 1996). Furthermore, a
storm (t = 365 min: 52.1%; t = 518 min: 72.7%). However, the first short time dry period before the event (36 h) coupled with positive air
sample (t = 140 min) showed that a great proportion of carbon temperatures (mean: 8.4 °C) and windy conditions leading to an
(41.2%) was transported under colloidal form during the first flush of abundance of colloids easily mobilized by overland flow could also
the runoff even if the intensity was low. explain these results. Globally, while most of the soluble OC (fraction
b0.1 μm) was flushed at the beginning of the event, the progressive
4. Discussion increase in intensity seemed to have mobilized more small colloidal
material (between 0.1 and 0.45 μm), which could explain the relative
Concerning the flux variation, the main differences between the rise in DOC concentrations at the end of the rainfall event.
studied rainfall events are globally explained by the slurry and cattle
manure application on soils which increases the soil organic matter 5. Conclusion
and N content (Aust et al., 2009; Morvan et al., 2007). Highest
contents of carbon and nitrogen supplied to amended plots (carbon: To our knowledge, this work is the first dealing with runoff water
57.3 kg for CM against 17.3 kg for PS; nitrogen: 2.3 kg for CM against quality variation during rainy events for fertilized agricultural plots,
1.0 kg for PS) can also explain these discrepancies. Moreover, a lowest with a high frequency monitoring. Thanks to a dedicated equipment
C/N ratio for PS fertilizer than for CM fertilizer also leads to a faster for runoff monitoring and sampling, it was possible to follow with a
degradation of organic materials after spreading. For all fertilized short time interval the impact of the rainfall on the variation of the
plots, the variation of the C/N ratio (here assessed by TOC/NK in runoff main water quality parameters such as organic carbon (total,
samples) indicates a spreading effect with an increase (from 4 to 7) dissolved and particulate) and Kjeldahl nitrogen during the whole
reflecting a higher fresh organic matter content in soils. rainy event. This work thus provides interesting results about the
Concerning the impact of a shower, the results of this study differ impacts of rainfall events on water quality in agricultural plots
significantly from the ones of Frank et al. (2000) founding higher DOC fertilized according to common practices (cattle manure CM, and pig
concentrations on the descending than on the rising limb collected slurry PS) in an experimental area (Brittany, France).
runoff in a small grassland catchment located in Switzerland. These The study of the runoff water quality variation between August
discrepancies are probably due to the smaller scale of our study and 2009 and March 2010 showed that two important factors must be
the difference between land uses and vegetation cover. Our results taken into account. Winter spreading on CM and PS plots had the most
could be explained by an accumulation of DOC in the upper soil which important impact on runoff water quality degradation with major
was flushed at the beginning of the runoff (Hornberger et al., 1994). increase for TOC fluxes and a lower one for KN and DOC fluxes.
However, they are comparable with those obtained on impervious Furthermore, highest intensive rainfall events mainly occurring
surfaces in urban runoff studies by Li et al. (2007) which found that during autumn, led to the highest DOC export.
the first 30% of the runoff volume carry 46 to 72% of the pollutant load. These findings can help understanding how climatic change
In comparison, the soils of our experimental site were particularly consequences (increase in intensity and duration of extreme rainfall
prone to crusting. Clod disaggregation under raindrop impacts result events) can influence the organic carbon and nutrients transport in
in a formation of surface crusts (Le Bissonnais et al., 1998), reducing agricultural areas of temperate countries, leading to the increase of at
the soil infiltration rate and inducing erosion by increasing runoff (Le risk situations in receiving waters. Such results are of interest
Bissonnais et al., 1995). A significant part of cultivated areas in especially for small scale water supply services that are particularly
vulnerable to short term transient water quality degradation. Gimbert LJ, Worsfold PJ. Temporal variability of colloidal material in agricultural storm
runoff from managed grassland using flow field-flow fractionation. J Chromatogr A
Appropriate sampling strategies and monitoring tools have to be 2009;1216:9120–4.
implemented to assess the water quality degradation during extreme Hejzlar J, Dubrovsky M, Buchtele J, Ružička M. The apparent and potential effects of
events, especially during the flow rising stage. climate change on the inferred concentration of dissolved organic matter in a
temperate stream (the Malše River, South Bohemia). Sci Total Environ 2003;310:
Furthermore, there is a need to trace the fate of manure-derived 143–52.
organic compounds once manure has been spread onto agricultural Hoese A, Clay SA, Clay DE, Oswald J, Trooien T, Thaler R, et al. Chlortetracycline and
soils, in order to detect any eventual contamination of soils, or of tylosin runoff from soils treated with antimicrobial containing manure. J Environ
Sci Heal B 2009;44:371–8.
groundwater and surface waters. One way to monitor the fate of these Hongve D, Riise G, Kristiansen JF. Increased colour and organic acid concentrations in
compounds is to perform molecular characterization and to follow Norwegian forest lakes and drinking water — a result of increased precipitation?
anthropogenic inputs of the different kinds of manure recycled by Aquat Sci 2004;66(2):231–8.
Hope D, Billett MF, Cresser MS. A review of the export of carbon in river water: fluxes
agriculture, with the aim of detecting some specific compounds to
and processes. Environ Pollut 1994;84:301–24.
trace each manure type in soils and waters. So, this work could be Hornberger GM, Bencala KE, McKnight DM. Hydrological controls on dissolved organic
complementary to the use of isotopic analysis (Diebel and Vander carbon during snowmelt in the Snake River near Montezuma, Colorado.
Zanden, 2009). There is also a need for accurate models to predict DOC Biogeochemistry 1994;25:147–65.
Inamdar SP, O'Leary N, Mitchell MJ, Riley JT. The impact of storm events on solute
concentrations from different soil types and management practices in exports from a glaciated forested watershed in western New York, USA. Hydrol
relation to rainfall characteristics. Reducing the carbon and nitrogen Process 2006;20:3423–39.
contents of fertilizers is also of importance for a better management of Jacinthe P-A, Lal R, Owens LB, Hothem DL. Transport of labile carbon in runoff as
affected by land use and rainfall characteristics. Soil Till Res 2004;77:111–23.
fertilizers applied to soils. Some studies have been carried out with Kato T, Kuroda H, Nakasone H. Runoff characteristics of nutrients from an agricultural
the aim to reduce nutrient excretion by adapting the composition of watershed with intensive livestock production. J Hydrol 2009;368:79–87.
animal diets and consequently reducing environmental consequences Kawasaki M, Ohte N, Kabeya N, Katsuyama M. Hydrological control of dissolved organic
carbon dynamics in a forested headwater catchment, Kiryu Experimental
of spreading (Toor et al., 2005). Finally, if all these findings are of Watershed, Japan. Hydrol Process 2008;22:429–42.
particular importance for the understanding of short term water Le Bissonnais Y. Aggregate stability and assessment of soil crustability and erodibility:
quality variation, they are also very useful for the implementation of I. Theory and methodology. Eur J Soil Sci 1996;47:425–31.
Le Bissonnais Y, Renaux B, Delouche H. Interactions between soil properties and
adapted actions needed for the production of a safer drinking water moisture content in crust formation, runoff and interrill erosion from tilled loess
supply after heavy rainy events. soils. Catena 1995;25:33–46.
Le Bissonnais Y, Benkhadra H, Chaplot V, Fox D, King D, Daroussin J. Crusting, runoff and
sheet erosion on silty loamy soils at various scales and upscaling from m2 to small
Acknowledgments
catchments. Soil Till Res 1998;46:69–80.
Li L-Q, Yin C-G, He Q-C, Kong L-L. First flush of storm runoff pollution from an urban
This work was funded by the French National Agency of Sanitary catchment in China. J Environ Sci 2007;19:295–9.
Security (ANSES) under the first ERA-EnvHealth call (FP7-ENV-2007- Monteith DT, Evans CD. The United Kingdom Acid Waters Monitoring Network: a
review of the first 15 years and introduction to the special issue. Environ Pollut
CSA-1.2.3-01). The authors sincerely thank Thierry Morvan and the 2005;137:3-13.
Institut National de la Recherche Agronomique (INRA Rennes) for the Monteith DT, Stoddard JL, Evans CD, de Wit HA, Forsius M, Høgåsen T, et al. Dissolved
use of their experimental site. The authors are indebted to Modibo organic carbon trends resulting from changes in atmospheric deposition chemistry.
Nature 2007;450:537–40.
Moungengui Diallo and the technical team of the Environment and Morel B, Durand P, Jaffrezic A, Gruau G, Molenat J. Sources of dissolved organic carbon
Health Research Laboratory for the site instrumentation and sample during stormflow in a headwater agricultural catchment. Hydrol Process 2009;23:
analysis. 2888–901.
Morvan T, Ruiz L, Viaud V. Cumulative effects of applications of organic fertilizers on soil
organic matter dynamics. In: De Neve S, Salomez J, Van Den Bossche A, Haneklaus S,
References Van Cleemput O, Hofman G, Schnug E, editors. Mineral Versus Organic Fertilization:
Conflict Or Synergism? Proceedings of the 16th International Symposium of the
Assouline S, Ben-Hur M. Effects of rainfall intensity and slope gradient on the dynamics International Scientific Centre of Fertilizers. 2007, Sept 16-19, Gand, Belgium; 2007.
of interrill erosion during soil surface sealing. Catena 2006;66:211–20. p. 362–70.
Aust M-O, Thiele-Bruhn S, Eckhardt K-U, Leinweber P. Composition of organic matter in NF EN 1484. Juillet 1997. Analyse de l'eau. Lignes directrices pour le dosage du Carbone
particle size fractionated pig slurry. Bioresour Technol 2009;100:5736–43. Organique Total et du Carbone Organique Dissous.
Bates BC, Kundzewicz ZW, Wu S, Palutikof JP, Bates BC, Kundzewicz ZW, et al. Climate NF EN 25663. Janvier 1994. Qualité de l'eau. Dosage de l'azote Kjeldhal. Méthode après
change and water. Technical Paper of the Intergovernmental Panel on Climate minéralisation au sélénium.
Change, Geneva, Switzerland; 2008. 210 pp. NF EN ISO 7027. Mars 2000. Détermination de la turbidité.
Chantigny MH. Dissolved and water-extractable organic matter in soils: a review on the Porcal P, Koprivnjak J-F, Molot LA, Dillon PJ. Humic substances—part 7: the
influence of land use and management practices. Geoderma 2003;113:357–80. biogeochemistry of dissolved organic carbon and its interactions with climate
Clark JM, Lane SN, Chapman PJ, Adamson JK. Export of dissolved organic carbon from an change. Environ Sci Pollut R 2009;16:714–26.
upland peatland during storm events: implications for flux estimates. J Hydrol Raclot D, Le Bissonnais Y, Louchart X, Andrieux P, Moussa R, Voltz M. Soil tillage and
2007;347:438–47. scale effects on erosion from fields to catchment in a Mediterranean vineyard area.
Delpla I, Jung AV, Baures E, Clement M, Thomas O. Impacts of climate change on surface Agr Ecosyst Environ 2009;134:201–10.
water quality in relation to drinking water production. Environ Int 2009;35(8): Sharlpley AN, Smith SJ. Nitrogen and phosphorus forms in soils receiving manure. Soil
1225–33. Sci 1995;159:253–8.
Diebel MW, Vander Zanden MJ. Nitrogen stable isotopes in streams: effects of Skjelkvåle BL, Stoddard JL, Jeffries DS, Tørseth K, Høgåsen T, Bowman J, et al. Regional
agricultural sources and transformations. Ecol Appl 2009;19:1127–34. scale evidence for improvements in surface water chemistry 1990–2001. Environ
Evans CD, Monteith DT, Cooper DM. Long-term increases in surface water dissolved Pollut 2005;137:165–76.
organic carbon: observations, possible causes and environmental impacts. Environ Toor GS, Sims JT, Dou Z. Short communication: reducing phosphorus in dairy diets
Pollut 2005;137:55–71. improves farm nutrient balances and decreases the risk of nonpoint pollution of
Fellman JB, Hood E, Edwards RT, D'Amore DV. Changes in the concentration, surface and ground waters. Agr Ecosyst Environ 2005;105:401–11.
biodegradability, and fluorescent properties of dissolved organic matter during Veum KS, Goyne KW, Motavalli PP, Udawatta RP. Runoff and dissolved organic carbon
stormflows in coastal temperate watersheds. J Geophys Res 2009;114:G01021. doi: loss from a paired-watershed study of three adjacent agricultural watersheds. Agr
10.1029/2008JG000790. Ecosyst Environ 2009;130(3–4):115–22.
Frank H, Patrick S, Peter W, Hannes F. Export of dissolved organic carbon and nitrogen World Health Organization (WHO). Guidance on water supply and sanitation in
from Gleysol dominated catchments — the significance of water flow paths. extreme weather events. In: Sinisi L, Aertgeerts R, editors. World Health
Biogeochemistry 2000;50:137–61. Organization, Geneva, Switzerland; 2010, 105 pp.
Gascuel-Odoux C, Aurousseau P, Durand P, Ruiz L, Molenat J. The role of climate on Worrall F, Harriman R, Evans CD, Watts CD, Adamson J, Neal C, et al. Trends in dissolved
inter-annual variation in stream nitrate fluxes and concentrations. Sci Total Environ organic carbon in UK rivers and lakes. Biogeochemistry 2004;70:369–402.
2009. doi:10.1016/j.scitotenv.2009.05.003.
Chapter
RAINFALL AND WATER QUALITY
Benoit Roig1,2,*, Estelle Baures1,2, Aude Valérie Jung3,

Ianis Delpla1,2 and Olivier Thomas1,2
1
Advanced School of Public Health (EHESP), LERES, Avenue du
Professeur Léon Bernard - CS 74312, Rennes Cedex, France
2
INSERM U1085, LERES, Rennes Cedex, France
3
School of Environmental Engineering (EME), Campus de Ker Lann,
Avenue Robert Schumann, Bruz, France
1. INTRODUCTION
The consideration of rainfalls impact on water management is not new and
started with the history of sanitation. At the beginning of the 20th century
sewer networks in urban area were designed both for sewage and for
stormwater management. In the 70’s the impact of urban runoff on water
quality was already studied (Pitt et al. 1977) and more recently the question of
climate change impacts on water quality was considered with the effect of
heavier rainy events. Commonly, categories of rain are defined by their
intensity: heavy, moderate and light rain being characterised by the amount of
mm of water/hour (respectively 2.5, 2.5 to 7.5, >7.5). According to the
duration of a given rainfall, the same intensity can be observed during a period
(shower) or a longer one (storm). Consequences of these precipitations can
*
E-mail: benoit.roig@ehesp.fr.
2 Benoit Roig, Estelle Baures, Aude Valérie Junge et al.
result in so called extreme rain or flood respectively at local or regional scale.

As reported by the 2007 IPCC report, it is likely that the frequency of heavy
precipitation events has increased over most areas, particularly in Europe and
North America, and that available research suggests a significant future
increase in heavy rainfall events in many regions. Such phenomena, increasing
in frequency and intensity, may have a strong impact on the quality of the
waters. Indeed, resulting changes in flow regimes will influence the chemistry,
hydromorphology and ecology of water bodies at the environmental level as
well as, in urban area, the wastewater quality, the sewage treatment efficiency
and the corresponding pollutants load. The effect of rainy events can also
reach drinking water production system, impairing the quality of the
distributed water (Figure 1)
This chapter aims to make an overview of the impact of rainfall on water
quality, by considering, within the water life cycle, chemical, biological and
physico-chemical parameters of water quality.
Wastewater Drinking water

Treatment production
Human activities Water Ressource Distribution
(wastewater) (surface and groundwater) (drinking water)
Figure 1. Impact of heavy events on water life cycle.

Rainfall and Water Quality 3
2. WATER QUALITY
Water bodies are characterised regarding their chemical and biological
status. A number of parameters are used to analyse water quality. European
water regulations (Drinking-water Directive (DWD) and the Water Framework
Directive (WFD)) have been established, since the end of the 1990s, to ensure
public health security for water intended for human consumption by promoting
a policy for the management and protection of water resources at European
level. More recently, a new preventive risk management approach was
initiated by WHO and water suppliers, via the implementation of Water Safety
Plans (Bartram et al., 2009), the aim of which is to guarantee the safety of a
drinking-water supply through the use of a comprehensive risk assessment and
risk management approach. Despite this regulatory framework, numerous
water bodies are often of impaired quality and non-compliant with European
regulations (Bloech, 2011).
In general, raw water quality is influenced by both 'natural' and
anthropogenic driving forces. The former include spatial and temporal
variation in water availability, wildlife, topography, geology, vegetation and
the future impact of climate change, particularly with respect to the frequency
and severity of droughts or rainfalls. Anthropogenic driving forces come from
each of the key sectors that use water (industry, energy production, public
water supply and agriculture) and include point sources (e.g. wastewater
discharges) and non-point sources (e.g. surface runoff). In general, water
quality degradation is directly linked to anthropogenic pressure, and climate
change acts as an indirect aggravating factor for these pressures on the
environment. In particular, the main determinants of climate change having a
direct or indirect impact on water quality are air temperature and extreme
water events (flood, drought) (Delpla et al., 2009). Resource availability is
linked to these parameters, and it becomes important to consider extreme
events in assessing both evolution of water stress and efficiency of the
treatment processes (sanitation and sewage treatment).
3. IMPACT OF RAINFALL ON URBAN WATER CYCLE

(FROM WASTEWATER TO THE ENVIRONMENT)
The historical question of rainfall impact on urban water cycle is linked to

urban runoff, well studied since the 70’s. For instance, Pitt et al. (1977)
studied the water-quality effects from urban runoff for a hypothetical city of
100,000 people. Total solids concentration and oxygen consumption were
calculated for receiving waters using various storms simulated in the study.
and more research on the long-term toxic effects of storm flow were suggested
to help solve urban runoff problems.
Cities are becoming increasingly vulnerable to flooding because of rapid
urbanization, installation of complex infrastructure and changes in the
precipitation patterns caused by anthropogenic climate change (Willems et al;
2012). For more than a century, sewer systems have been constructed at large
scale across cities worldwide. These sewer systems could make the cities more
vulnerable to extreme rainfall, partly due to the lack of consideration to what
occurs when the design criteria are exceeded. Next to this increase in the
vulnerability, there is strong evidence that, due to global warming, the
probabilities and risk to sewer surcharge and flooding are under discussion.
But impacts of climate change on urban drainage are still weakly investigated
(Willems et al., 2012).
In AMWA report (2007), the different impacts of climate changes on
water suppliers are identified and described as follow:
i. direct impacts resulting from the effects of climate change on water

utility functions and operations,
ii. indirect impact resulting from the effects of climate change on the
baseline environment in which water utility functions and operations
are carried out and
iii. compound impact resulting from the cumulative effect of the direct
and indirect impacts which affect the same natural systems or utility
systems.
Among the direct impacts, more frequent and intense rainfall events are
intended to increase turbidity, sedimentation, risk of direct flood and
consequently damage water utility facilities: for example, loss of reservoir
capacity storage (shallower, warmer water, increased evaporation and
eutrophication and potential conflicts with flood control objectives) with
consequences on water treatment (AMWA, 2007). However for urban
watercourses, there is often “flashy” hydraulic regimes and poor water quality,
making them potential hazards.
Furthermore, the knowledge of impact of intense rainfall events on urban

river water discharge and, in particular, fine particle transport is limited (Old et
al., 2003). The washed off pollutants such as suspended solids increases the
turbidity of the receiving water, thereby reducing the penetration of light,
resulting in decreased activity and growth of photosynthetic organisms.
Clogging of fish gills has also been attributed to the presence of suspended
solids. These ones settle in receiving water pose long-term threats resulting
from increased oxygen demand and gradual accumulation of toxic pollutants.
Such detrimental effects have made suspended solid concentration one of the
pollutant measurement indices (Aryal, 2009).
Moreover, Old et al. (2003) demonstrated that this extremely large and
urban flux of fine sediment is likely to have a major downstream impact on
water quality in the river.
Brodie and Egodawatta (2011) have shown that suspended particles in
stormwater runoff generated from road surfaces loads increase linearly with
average rainfall intensity. Above a threshold intensity, there is evidence to
suggest a constant or plateau particle load is reached.
Cheng et al (2009) have studied the impact of four different intensity
rainfall (light rain, moderate rain, heavy rain and storm) on water quality of
urban combined sewerage system (CSS) in highly urbanized region and the
correlations of parameters with both rainfall intensity and runoff graph. Basic
parameters and metals concentrations were monitored and shown good
relation between them and runoff.
Temporal evolution of microbiological, physical, and chemical quality of
stormwater runoff from a stormwater drain was studied by He et al (2010) in
an urban residential area in Calgary, Canada. Intensive rainfall events
produced highly polluted stormwater runoff when pollutant source limitation
did not occur. Inconsistent event-based correlations between total suspended
solids (TSS) concentrations and water quality parameters were observed.
During storms, the loading of TSS exhibited a flow-dependent nature, whereas
microorganism discharge appeared to be governed by a flow-independent
mechanism. No strong first-flush effect was observed in either TSS or
microorganisms, on average. No correlations of first-flush loads of TSS with
rainfall characteristics were identified.
Rainfall events can also play an important role regarding micropollutants
load at sewage treatment works (STWs). In general, urban runoff tends to
originate from five main categories: transportation (i.e. roads and vehicles),
roofing materials, construction activities, vegetation and soil. There may be
many variables within a given source, e.g. runoff from roads and associated
traffic emissions may contain vehicle lubricants, exhaust emissions,

degradation from tyres and brake linings, road maintenance and surface
degradation, accidental spillages and road salting in winter. The extent of
urban runoff is dependent on levels of contaminants deposited on urban
surfaces through wet and dry deposition, the length of the preceding dry spell
and the characteristics and size of the urban catchment. Assessing the effects
of rainfall STW influent has produced results that suggest that with short
rainfall incidences the level of most micropollutants will increase due to ‘first
flushing’ events, whereas continuous high levels of rainfall will result in a
dilution effect on micropollutant levels (Rowsell et al., 2010).
Another consequence of runoff on water quality may be ecotoxilogical
impact linked to the presence of contaminants. Meland et al. (2010), exposed
brown trout (Salmo trutta L.) to highway runoff samples during 24 h. The
metabolism of the fish through reduced condition factor was affected and the
observed effects were likely caused by multiple stressors and not by a single
contaminant. A sedimentation pond clearly reduced the toxicity of the
highway runoff. But even in the least polluted exposure condition signs of
physiological disturbances were evident.
Considering the threats originating from wastewater utilities, the type of
wastewater system is of crucial importance. Moreoften a combined wastewater
system is prevalent in urban area, with domestic wastewater and storm water
runoff from impervious surfaces (streets, roofs etc.) collected in the same
sewer system.
During rain events, the wastewater treatment plants as well as the sewer
systems have to cope with a high load of combined wastewater that may
exceed the systems’ capacity. Thus, certain volumes of wastewater must be
stored or discharged. This may be achieved by storm water overflow basins or
storm water tanks (SWT), with both retention and treatment functions (Heins,
2009). Combined sewage overflows which were previously designed to
directly discharge the exceeding flow tend to be removed from wastewater
networks.
In case of heavy rainfall, the wastewater treatment plant temporarily
receives low concentrated sewage and high hydraulic loading. Prolonged
exposure to storm water conditions will negatively affect all treatment steps by
decreasing the residential time and particularly the settling processes (primary
and secondary) as well as the biological part of the treatment with activated
sludge, and possibly denitrification and phosphorus removal processes. For
this last step, it is still not clear whether this is due to inhibition of
phosphorous-removing bacteria (PRB) or that the low COD concentrations in
the influent are causing the problem. It has been reported that, for example, the
deterioration of PRB efficiency regularly occurred at some wastewater
treatment plants (WWTPs) after heavy rainfall. The phenomenon was
attributed to low plant loading that took place during such events (Brdjanovic
et al., 1998).
An observatory of urban pollutants was created in France for the purpose
of assessing the dynamic of wastewater and wet weather flow (WWF). The
magnitude of WWF pollutant loads characterized by physical-chemical
parameters: total suspended solid, DOC, heavy metals (Cu, Zn), polycyclic
aromatic hydrocarbons, and their subsequent acute impact on receiving waters
have been extensively studied since 1970 (Suarez and Puertas, 2005; Gasperi
et al., 2010). Some studies have observed the becoming of specific pollutants
concentrations during wet weather flows in wastewater. Gilbert et al., (2011)
for instance showed that alkylphenols (AP) and polybromodiphenylether
(PBDE) concentrations are 1.5 to 5 fold higher under wet weather flows.
The removal rates of several antiphlogistics and lipid regulating agents
was investigated during a rainfall event (Ternes, 1998). The elimination rates
of drugs like bezafibrate, diclofenac, naproxen and clofibric acid were
significantly reduced on the rainfall day. For bezafibrate a reduced elimination
rate (<5%) was observed the last day of the investigation period. These results
indicate that the rainfall event may be presumably responsible for the
decreased drug elimination rates by the sewage treatment plant probably due
to a reduced microbial activity and/or altered sorption and flocculation
conditions within this rain period (Ternes, 1998).
4. RAINFALL AND WATER RESOURCE

(QUALITY MODIFICATION OF NATURAL WATER)
Water quality depends on stressors among which climate-related
parameters such as water temperature and river flows (Murdoch et al., 2000).
According to these authors, changes in water quality is related to hydrological
factors (e.g. limited dilution of point sources during low river flows, or
increased runoff from agricultural area during rainstorms), terrestrial factors
(e.g. changes in vegetation and soil structure) and resource-use factors (e.g.
increased water use, increased demand for cooling capacity of industrial
processes). More recently, bibliographic reviews and case studies (Delpla et
al., 2009, Zwolsman and van Bokhoven, 2007, Van Vliet and Zwolsman,
2008) concluded that climate change has a negative impact on water quality.
Depending on the season, the same impact (increase of water pollution) may
be related to different conditions. During winter rainfall, water and air
temperatures are low, river flows generally high and of soil leaching and
runoff increase), contrary to summertime where very high air temperature may
lead to drought (e.g. Krysanova et al., 2005). Both cases may give a high
pollution level either by runoff (heavy rain in winter) or by physical
concentration (drought in summer).
According to Bates et al. (2008) and Brunetti et al. (2001), climate change
in temperate area will lead to a decrease of the number of rainy days, with
however a significant increase of the mean volume for each event.
Consequently, the drought and re-humidification cycles for the soils will allow
the decomposition and leaching of the organic matter to the surface waters and
thus impact water quality as a whole (Evans et al., 2005). Heavy rainfall
events could lead to an increase in dissolved organic carbon concentration
(Evans et al., 2005 ; Fellman et al., 2009) which is known to play an important
role concerning the transport and release of pollutants (Lennartz and Louchart,
2007; Pédrot et al., 2008).
Nutrients concentrations (nitrates, phosphates) and obviously particulate
and colloidal suspended matter are directly correlated with rainy events
(Prathumratana et al., 2008; Bhat et al., 2007; Drewry et al., 2009) and could
consequently exceed water quality regulated limits.
For heavy rainy events leading to an increase of river flows, the major risk
is also certainly the transport of pollutants by run-off (such as pesticides and
pharmaceuticals). It was often acknowledged that the timing and magnitude of
rainfall events are significant on pesticide transport through rapid flow
processes (Leonard, 1990, Flury, 1996; Kladivko et al., 2001, McGrath et al.,
2010) with in general an increase of pesticide leaching with increasing rainfall
amount (Tiktak et al. 2004, Jiang et al., 2012). Pharmaceuticals considered as
« emerging substances » are of a special importance because they are not yet
regulated. Some studies specifically show the variations of some of these
compounds (carbamazepine, monensin, clofibric acid, iopromide,
sulfamethoxazole, trimethoprim, flumequine) with flow rates variation during
rainfall events (Oppel et al., 2004; Lissemore et al., 2006; Tamtam et al.,
2008).
Variations of temperature, pH and water composition occurring during
rainfall events could also have an influence on contaminants by impacting
their sorption on mineral phases. Moreover, the influence of soils properties
has to be taken into account for the leaching behaviour assessment of

anthropogenic compounds (Oppel et al., 2004; Yu et al., 2009).
Neal et al (2004), studied the water quality of rainfall and runoff for two
catchments of different characteristics. Average water quality concentrations
and fluxes vary across the sites, typically by about 30% but stream chemistry
is much less variable due to water coming from aquifer sources of high
storage. Moreover, the catchments appear to be retain both P and N coming
from agricultural practices and sewage inputs to the streams.
Some studies on coastal water quality impacts of rainfall like Coulliette et
al. (2008) have shown the effect of non-point source contamination linked to
stormwater. Fecal coliforms concentrations increased with rainfall intensity
but higher than expected values existed during conditions of negligible rainfall
(< 0.25 cm), indicating a possible reservoir population in the sediment.
Liu et al (2010), worked on the occurrence and distribution of three typical
endocrine-disrupting chemicals (EDCs), nonylphenol mixture (NPs),
bisphenol A (BPA), and 17-ethynilestrdiol (EE2), in the coastal water,
suspended solid, and sediment. With the increasing rainfall, the concentrations
of target EDCs decreased in water and sediment and increased in suspended
solid.
The genotoxic impact on fish of heavy rainfall following herbicide
application has been investigated by Polard et al. (2011). For three
hydrological conditions (basal flow, winter flood, and spring flood), chemical
analysis of the water samples confirmed the higher contamination of the spring
flood water, mainly explained by a peak of metolachlor. Genotoxicity was
evaluated on Crucian carp and exposure to spring flood water resulted in the
highest damage induction. Further experiments on pesticides mixture adde on
water flood revealed a mutagenic impact of water contamination during the
spring flood, emphasizing the need to consider these transient events in water
quality monitoring programs
Concerning pathogens, floods frequently lead to a faecal contamination of
surface and deep waters and also to waterborne diseases (Beaudeau et al.,
2010). As an example, very elevated levels of bacteria E. Coli (2000 – 7000
CFU/mL) have been measured downstream Milwaukee river basin after
rainfall events (Patz et al., 2008). Curriero et al., (2001) have shown that half
of the waterborne diseases outbreaks in United States during the second half of
the 20th century followed a period of extreme rainfall (defined as the excess of
the 90th percentile based on 50-year distribution).
Finally changes in rainfalls patterns and temperature could also have an
effect on blooms occurrence of cyanobacteria. Intense rainfalls could interrupt
bloom development in a water body by flushing and mixing at the event scale
but increase the nutrient load and the rate of eutrophication which favour
cyanobacteria at a longer time scale (Reichwaldt and Ghadouani, 2011).
5. IMPACT ON DRINKING WATER PRODUCTION

Rainfalls events could lead to elevated levels of turbidity and organic
matter in river waters which in turn could cause a decline in treatment
performance. Mineral particles leaching could lead to high concentrations in
natural waters, having a direct impact on coagulant demand during water
treatment (Shin et al., 2008) and on Disinfection By-Products (DBPs)
formation. However, this decrease in efficiency could be also due to a
combination of a change in the nature and increased concentrations of Natural
Organic Matter (NOM) and lower water temperatures in the natural water
(Hurst et al., 2004). This could explain why seasonal differences have a
significant impact on process robustness which is independent of the raw
water turbidity. Moreover, in the distribution system, the highest spatial
variation in Trihalomethans concentration (from 2 to 4 times) is measured for
elevated water temperatures (>18°C) typically during the summer season,
whereas this variation is small for water temperature lower than 15°C during
the spring or fall seasons (Rodriguez and Serodes, 2001).
Although very few studies reported the relationship between algae and
DBPs precursors, some authors pointed out the algal contribution to some
DPBs formation such as Haloacetic Acids (Chen and Zhang, 2008).
Heavy rainfall and associated floods could flush biological contaminants
into waterways and aquifers, for example with combined sewer systems
overflowing and agricultural lands leaching. Greater water flow contributes to
increase the pathogen load and the penetration speed of pathogens into
resource waters and the drinking-water supply (Unc and Goss, 2003). A
significant number of waterborne disease outbreaks have involved the
transport of bacteria, viruses, or small parasites into water systems or well-
heads during an heavy rainfall event (Auld et al., 2004; Thomas et al., 2006).
As an example, heavy rainfalls contributed to the Milwaukee (United States)
outbreak of Cryptosporidium in 1993 (54 deaths, 403000 cases) and the
Walkerton (Canada) E. Coli O157:H7 outbreak in 2000 (7 deaths, 4300 cases).
Monitoring the change of the biological charge with adequate and rapid tools
is very important to ensure maximum efficiency of the treatment and
protection of the population after rainfall events. This is of primary importance

regarding the fact that routine monitoring programs mainly use techniques that
are time consuming and not allow a representative picture of the wide range of
biological contaminants that could be present in water.
The main recent studies dealing with the occurrence and fate of
micropollutants with respect to drinking water treatment are related to
pharmaceuticals. Actually, the majority of studies on pharmaceuticals focused
on wastewater treatment, for which the insuficient removal efficiency may
affect the quality of water resources downstream a treated effluent discharge.
Although they are partially removed, residual quantities may remain in treated
water and contaminate drinking water resources (Al-Ahmad et al., 1999;
Hernando et al., 2006). Actually, more than 20 substances (parent compound
and residues) have been found in drinking water (Mompelat et al., 2009).
The efficiency of pharmaceuticals removal varies with treatment processes
and also with temperature and weather (Choi et al., 2008). For instance,
diclofenac showed largely different elimination rates whose range, comprised
between 17% (Heberer et al., 2002), 69% (Ternes et al., 1998), and 100%
(Thomas and Foster, 2004) depends on these two last parameters.
Pharmaceuticals removal could only be achieved with advanced processes as
ozonation, activated carbon adsorption, membrane filtration (Ternes et al.,
2002) or UV treatment (Canonica et al., 2008). Finally, only some lipophilic
pesticides are removed adequately in conventional physico-chemical drinking
water treatment processes, such as flocculation, sedimentation, filtration, or
lime softening (Baldauf, 2006).
Concerning natural micropollutants, mainly represented by cyanotoxins,
chlorination, micro/ultra-filtration and especially ozonation are the most
effective water treatment procedures in destroying cyanobacteria and in
removing microcystins (Hitzfeld et al., 2000). Lots of studies concerning
removal of cyanobacterial toxins from water showed that the effectiveness of
the oxidation process is not only dependant on the reactant concentration, but
also on temperature, pH, ionic composition (Rositano et al., 1998; Shawwa
and Smith, 2001) and NOM concentration (Al Momani et al., 2008).
Moreover, some studies tried to improve by-products identification and
toxicity assessment for cyanotoxins (Rodriguez et al., 2007; Mérel et al.,
2009). Although DBPs formation has largely not been investigated, the few
studies conducted have concluded to a decrease in cyanotoxins toxicity
(microcystin, cylindrospermopsin, saxitoxin) after chlorination (Mérel et al.,
2010).
Extreme weather events could affect the efficiency of drinking water

treatment processes, leading to a possible degradation of quality in distributed
drinking water (Sinisi and Aertgeerts, 2010). These concerns are of primary
importance for small-scale water suppliers (SSWSs), private, community-
based or public, which usually prevail in rural areas. A significant part of the
population living in these areas (16 million people in Europe (WHO, 2010) do
not have access to improved drinking water resources. This vulnerability of
SSWSs is due to an insufficient sanitary protection of resources, agricultural
pressures, limited water treatment technologies, relatively greater capital costs
of technical installations and larger per unit costs of materials and
construction. By considering these existing pressures, and the indirect risks
posed by climate change, SSWSs are facing several challenges, including their
regulatory environment, administration, management, operation, and their
available technical, human and financial resources (Sinisi and Aertgeerts,
2010).
CONCLUSION
Rainfall can be a significant source of variation in water quality. In urban
area, depending on its intensity, rainfall can have an effect on utility functions
and operation as well as water discharge due to runoff (road, roof, …). The
relation between water quality and runoff is in particular well known since the
70’s as an increase of turbidity, sedimentation, transport of more or less fine
particles and pollutants (conventional and/or emerging). The main outcomes
were the improvement of sewer and rainstorm networks design with in situ
settling systems for TSS removal and flow storage. More recently the effect of
rainfall on soil erosion has been also highlighted, in particular the suspended
solid exportation and increase of nutrient in surface water.
Water resource is also stressed by rainfall. Changes in water quality rely
in particular to hydrological, terrestrial and resource-use factors as well as
climate. Climate changes are more and more studied with consequences on
micropollutants transportation and cyanobacteria crisis
One important point of the relation between water quality and rainfall is
the evolution of transient effect (in particular toxicity and ecotoxicity) and the
effect on sanitary water.
REFERENCES
Al-Ahmad, A; Daschner, FD; Kummerer, K. Biodegradability of cefotiam,
ciprofloxacin, meropenem, penicillin G, and sulfamethoxazole and
inhibition of waste water bacteria. Arch Environ Contam Toxicol 1999,
37,158–63.
Al Momani, F; Smith, DW; El-Din, MG. Degradation of cyanobacteria toxin
by advanced oxidation process. J Hazard Mater 2008, 150, 238–49.
AMWA (Association of metropolitan water agencies), Implications of climate
change for urban water utilities, Washington DC, December 2007, 18 p.
Auld, H; MacIver, D; Klaassen, J. Heavy rainfall and waterborne disease
outbreaks: the Walkerton example. J. Toxicol. Environ. Health A 2004,
67, 1879-1887.
Aryal, RK; Lee, B.-K. Characteristics of Suspended Solids and
Micropollutants in First-Flush Highway Runoff. Water Air Soil Pollut:
Focus 2009, 9, 339–346.
Baldauf, G. Removal of pesticides in drinking water treatment. Acta
Hydrochim Hydrobiol 2006, 21, 203–8.
Bartram, J; Corrales, L; Davison, A; Deere, D; Drury, D; Gordon, B; Howard,
G; Rinehold, A; Stevens, M; Water Safety Plan Manual: Step-By-Step
Risk Management for Drinking-Water Suppliers, World Health
Organization, Geneva, Switzerland, 108p, 2009.
Bates, B; Kundzewickz, ZW; Wu, S; Palutikof, J. (Editors) Climate Change
and Water, Technical Paper of the Intergovernmental Panel on Climate
Change, IPPC Secretariat, Geneva, Switzerland, 210 p., 2008.
Beaudeau, P; Valdes, D; Mouly, D; Stempfelet, M; Seux, R. Natural and
technical factors in faecal contamination incidents of driniking water in
small distribution networks, France, 2003-2004: a geographical study. J.
Water Health 2010, 08.1, 20-34.
Bhat, S; Hatfield, K; Jacobs, JM; Lowrance, R; Williams, R. Surface runoff
contribution of nitrogen during storm events in a forested watershed.
Biogeochemistry 2007, 85, 253-262.
Bloech, H. Summary Record of the Meeting of the Committee under Article 12
of Directive 98/83/EC, DG ENV, 22/02/2011.
Brdjanovic, D; Slamet, A; Van Loosdercht, MCM; Hooijmans, CM; Alaerts,
GJ; Heijnen, JJ. Impact of excessive aeration on biological phosphorus
removal from wastewater, Water Research 1998, 32, 200-208.
Brodie, IM; Egodawatta, P. Relationships between rainfall intensity, duration

and suspended particle washoff from an urban road surface, Hydrology
Research 2011, 42, 239-249.
Brunetti, M; Maugeri, M; Nanni, T. Changes in total precipitation rainy days
and extreme events in Northeastern Italy. Int. J. Climatol. 2001, 21, 861-
871.
Canonica, S; Meunier, L; von Gunten, V. Phototransformation of selected
pharmaceuticals during UV treatment of drinking water. Water Res. 2008,
42, 121-128.
Chen, C; Zhang, X-J; Zhu, L-x; Liu, J; He, W-j; Han, H-d. Disinfection by-
products and their precursors in awater treatment plant in North China:
seasonal changes and fraction analysis. Sci Total Environ 2008, 397, 140–
7.
Cheng, J; Yang, K; Huang, X-F; Lue, Y-P. Hydrology and Water Quality of
Rainfall-Runoff in Combined Sewerage System Along Suzhou Creek in
Central Shanghai. Environmental Science 2009, 30, 1893-1900.
Choi, K; Kima, Y; Park, J; Park, CK; Kim, M; Kim, HS; Kim, P. Seasonal
variations of several pharmaceutical residues in surface water and sewage
treatment plants of Han River. Sci Total Environ 2008, 405, 102–28.
Coulliette, AD; Noble, R T. Impacts of rainfall on the water quality of the
Newport River Estuary (Eastern North Carolina, USA). Journal of Water
and Health 2008, 6, 473-482.
Curriero, FC; Patz, JA; Rose, JB; Lele, S. The association between extreme
precipitation and waterborne disease outbreaks in the United States, 1948-
1994. Am. J. Public Health, 2001, 91, 1194-1199.
Delpla, I; Jung, A-V; Baures, E.; Clement, M; Thomas, O. Impacts of climate
change on surface water quality in relation to drinking water production.
Environment International 2009, 35, 1225-1233.
Drewry, JJ; Newham, LTH; Croke, BFW. Suspended sediment, nitrogen and
phosphorus concentrations and exports during storm-events to the Tuross
estuary, Australia. J. Environ. Manag. 2009, 90, 879-887.
Evans, CD; Monteith, DT; Cooper, DM. Long-term increases in surface water
dissolved organic carbon: observations, possible causes and environmental
impacts. Environment Pollution 2005, 137, 55–71.
Fellman, B; Hood, E; Edwards, RT; D’Amore, DV. Changes in the
concentration, biodegradability, and fluorescent properties of dissolved
organic matter during stormflows in coastal temperate watersheds.
Journal of Geophysical Research, 2009, 114, G01021, doi:10.1029/
2008JG000790.
Flury, M. Experimental evidence of transport of pesticides through field soils

— a review. J. Environ. Qual. 1996, 25, 25–45.
Gasperi, J; Gromaire, MC; Kafi, M; Moilleron, R; Chebbo, G. Contributions
of watewater, runoff and sewer deposit erosion to wet weather pollutant
load in combined sewer systems, Water Research 2010, 44, 5875-5886.
Gilbert, DS; Gasperi, J; Rocher, V; Lorgeoux, C; Chebbo, G. Efficiency of the
seine centre wastewater treatment plant during wet weather periods
regarding conventional wastewater parameters, alkylphenols and
polybromodiphenylethers, Techniques Sciences et Méthodes 2011, 4, 63-
72.
He, J.; Valeo, C.; Chu, A.; Neumann, N. F. Characteristics of Suspended
Solids, Microorganisms, and Chemical Water Quality in Event-Based
Stormwater Runoff from an Urban Residential Area. Water Environment
Research 82. 12 (Dec 2010): 2333-2345.
Heberer, T; Reddersen, K; Mechlinski, A. From municipal sewage to drinking
water: fate and removal of pharmaceutical residues in the aquatic
environment in urban areas. Water Sci Technol 2002, 46, 81–88.
Heinz, B; Birk, S; Liedl, R; Geyer, T; Straub, KL; Andresen, J; Bester, K;
Kappler, A. Water quality deterioration at a karst spring (Gallusquelle,
Germany) due to combined sewer overflow: evidence of bacterial and
micro-pollutant contamination. Environmental Geology 2009, 57, 797-
808.
Hernando, MD; Mezcua, M; Fernandez-Alba, AR;, Barceló, D. Environmental
risk assessment of pharmaceutical residues in wastewater effluents,
surface waters and sediments. Talanta 2006, 69, 334–342.
Hitzfeld, BC; Höger, SJ; Dietrich, DR. Cyanobacterial toxins removal during
drinking water treatment and human risk assessment. Environ Health
Perspect 2000, 108, 113–122.
Hurst, AM; Edwards, MJ; Chipps, M, Jefferson, B; Parsons, SA. The impact
of rainstorm events on coagulation and clarifier performance in potable
water treatment. Sci Total Environ 2004, 321, 219–230.
Jiang, W; Haver, D; Rust, M; Gan, J. Runoff of pyrethroid insecticides from
concrete surfaces following simulated and natural rainfalls. Water Res.
2012, 46, 645-52.
Kladivko, EJ; Brown, LC; Baker, JL. Pesticide transport to subsurface tile
drains in humid regions of north America. Crit. Rev. Environ. Sci. Tech.
2001, 31, 1–62.
Krysanova, V; Hattermann, F., Habeck, A. Expected changes in water

resources availability and water quality with respect to climate change in
the Elbe River basin (Germany). Nordic Hydrology 2005, 36, 321-333.
Lennartz, B; Louchart, X. Effect of drying on the desorption of diuron and
terbuthylazine from natural soils. Environmental Pollution 2007, 146,
180-187.
Leonard, R., 1990. Movement of pesticides into surface waters. In: Cheng,
H.H. (Ed.), Pesticides in the Soil Environment. Soil Science Society of
America, Madison, USA, pp. 303–349.
Lissemore, L; Hao, C; Yang, P; Sibley, PK; Mabury, S; Solomon, KR. An
exposure assessment for selected pharmaceuticals within a watershed in
Southern Ontario. Chemosphere 2006, 64, 717–729.
Liu, Y.; Guan, Y.; Tam, NFY; Mizuno, T.; Tsuno, H.; Zhu, W. Influence of
Rainfall and Basic Water Quality Parameters on the Distribution of
Endocrine-Disrupting Chemicals in Coastal Area. Water, Air, & Soil
Pollution 2010, 209 333-343.
Meland, S; Salbu, B; Rosseland, BO. Ecotoxicological impact of highway
runoff using brown trout (Salmo trutta L.) as an indicator model. Journal
of Environmental Monitoring 2010, 12, 654-664.
Mérel, S; Lebot, B; Clement, M; Seux, R; Thomas, O. MS identification of
microcystin-LR chlorination by-products. Chemosphere 2009, 74, 832–9.
McGrath, G; Hinz, C; Sivapalan, M. Assessing the impact of regional rainfall
variability on rapid pesticide leaching potential. J Contam Hydrol. 2010,
113, 56-65.
Mérel, S; Clément, M; Thomas, O. State of the art on cyanotoxins in water and
their behaviour towards chlorine. Toxicon 2010, 55, 677-691.
Mompelat, S; Le Bot, B; Thomas, O. Occurrence and fate of pharmaceutical
products and by-products, from resource to drinking water. Environ Int.
2009, 35, 803-814.
Murdoch, PS; Baron, JS; Miller, TL. Potential effects of climate chance on
surface water quality in North America. Journal of the American Water
Resources Association 2000, 36, 347-366.
Neal, C; Skeffington, R; Neal, M; Wyatt, R; Wickham, H; et al. Rainfall and
runoff water quality of the Pang and Lambourn, tributaries of the River
Thames, south-eastern England. Hydrology and Earth System Sciences
2004, 8, 601-613.
Old, GH; Leeks, GJL; Packman, JC; Smith, BPG; Lewis, S; Hewitt, EJ;
Homles, M; Young, A. The impact of convectional summer rainfall event
on river flow on fine sediment transport in a highly urbanized catchment:
Bradford, West Yorkshire, The Science of the Total Environment 2003,

314-316, 495-512.
Oppel, J; Broll, G; Löffler, D; Meller, M; Römbke, J; Ternes, T. Leaching
behaviour of pharmaceuticals in soil testing systems: a part of an
environmental risk assessment for groundwater contamination. Sci Total
Environ 2004, 328, 265–273.
Patz, JA; Vavrus, SJ; Uejio, CK; McLellan, SL. Climate change and
waterborne disease risk in the Great Lakes region of the US. Am J Prev
Med 2008, 35, 451–458.
Pédrot, M; Dia, A; Davranche, M; Bouhnik-Le Coz, M; Henin, O; Gruau, G.
Insights into colloid-mediated trace element release at the soil/water
interface. J Colloid Interface Science 2008, 325, 187–197.
Pitt, R; Field, R. Water-Quality Effects from Urban Runoff. J. Amer Water
Works Assoc. 1977, 69, 432-436.
Polard, T; Jean, S; Gauthier, L; Laplanche, C; Merlina, G; et al. Mutagenic
impact on fish of runoff events in agricultural areas in south-west France.
Aquatic Toxicology 2011, 101, 126-134.
Prathumratana, L; Sthiannopkao, S; Kim, KW. The relationship of climatic
and hydrological parameters to surface water quality in the lower Mekong
River. Environ. Int. 2008, 34, 860-866.
Reichwaldt, ES; Ghadouani, A. Effects of rainfall patterns on toxic
cyanobacterial blooms in a changing climate: Between simplistic
scenarios and complex dynamics. Water Research 2012, 46, 1372-1393.
Rodriguez, MJ; Serodes, JB. Spatial and temporal evolution of
trihalomethanes in three water distribution systems. Water Res 2001, 35,
1572–1586.
Rodriguez, E; Sordo, A; Metcalf, JS; Acero, JL. Kinetics of the oxidation of
cylindospermopsin and anatoxin-a with chlorine, monochloramine and
permanganate. Water Res 2007, 41, 2048–2056.
Rositano, J; Nicholson, BC; Pieronne, P. Destruction of cyanobacterial toxins
by ozone. Ozone-Sci Eng 1998, 20, 223–38.
Rowsell, VF; Tangney, P; Hunt, C; Voulvoulis, N. Estimating levels of
micropollutants in municipal wastewater, water, air and soil pollution
2010, 206, 357-368.
Shawwa, AR; Smith, DW. Kinetics of microcystin-LR oxidation by ozone.
Ozone-Sci Eng 2001, 23, 161–170.
Shin, JY; Spinette, RF; O'melia, CR. Stoichiometry of coagulation revisited.
Environ Sci Technol 2008, 42, 2582–2589.
Sinisi, L; Aertgeerts, R. Guidance on Water Supply and Sanitation in Extreme

Weather Events, World Health Organization, Geneva, Switzerland, 2010.
Suarez, J; Puertas, J. Determination of COD, BOD and suspended solids loads
during combined sewer overflow (CSO) event in some combined
catchments in Spain, ecological engineering 2005, 24, 199-217.
Tamtam, F; Mercier, F; Le Bot, B; Eurin, J; Tuc Dinh, Q; Clément, M;
Chevreuil, M. Occurrence and fate of antibiotics in the Seine River in
various hydrological conditions. Sci Tot Env 2008, 393, 84 – 95.
Ternes, TA. Occurrence of drugs in German sewage treatment plant and rivers,
Water Research 1998, 32, 3245-3260.
Ternes, T; Hirsch, R; Mueller, J; Haberer, K. Methods for the determination of
neutral drugs as well as betablockers and alpha2-sympathomimetics in
aqueous matrices using GC/MS and LC/MS/MS. Fresen J Anal Chem
1998, 362, 329–40.
Ternes, T; Meisenheimer, M; McDowell, D; Sacher, F; Brauch, H-J; Haist-
Glude, B; Preuss, G; Wilme, U; Zulei-Seibert, N. Removal of
pharmaceuticals during drinking water treatment. Environ Sci Technol.
2002, 36, 3855-3863.
Thomas, KM; Charron, DF; Waltner-Toews, D; Schuster, C; Maarouf, AR;
Holt, JD. A role of high impact weather events in waterborne disease
outbreaks in Canada, 1975 – 2001. Int. J. Environ. Health Res. 2006, 16,
167-180.
Thomas, PM; Foster, GD. Determination of nonsteroidal anti-inflammatory
drugs, caffeine, and triclosan in wastewater by gas chromatography-mass
spectrometry. J Environ Sci Health A 2004, 39, 1969–1678.
Tiktak, A; de Nie, DS; Piñeros Garcet, J; Jones, A; Vanclooster, M.
Assessment of the pesticide leaching risk at the pan-European level: the
EuroPEARL approach. J. Hydrol. 2004, 289, 222–238.
Unc, A; Goss, M. Movement of Faecal Bacteria through the Vadose Zone.
Water Air Soil Pollut. 2003, 149, 327-337.
Van Vliet, MTH; Zwolsman, JJG. Impact of Summer Droughts on the Water
Quality of the Meuse River. Journal of Hydrology, 2008, 353, 1-17.
Willems, P; Arnbjerg-Nielsen, K; Olsson, J; Nguyen, VTV. Climate change
impact assessment on urban rainfall extremes and urban drainage:
methods and shortcomings, Atmosperic Research 2012, 103, 106-118
World Health Organization (WHO), Vision 2030 – The Resilience of Water
Supply and Sanitation in the Face of Climate Change, World Health
Organization/Department for International Development, Geneva,
Switzerland, 2010 (http://www.who.int/water_sanitation_health/ publi-

cations/9789241598422_cdrom/en/index.html, accessed 5 January 2011).
Yu, L ; Fink, G ; Wintgens, T ; Melin, T ; Ternes, TA. Sorption behaviour of
potential organic wastewater indicators with soils. Water Res 2009, 43,
951–60.
Zwolsman, JJG; Van Bokhoven, A. Impact of summer droughts on the water
quality of the Rhine River - a preview of climate change? Water Science
& Technology, 2007, 56, 45–55.
Trends in Analytical Chemistry, Vol. 30, No. 8, 2011 Trends
Analytical issues in monitoring

drinking-water contamination
related to short-term,
heavy rainfall events
B. Roig, I. Delpla, E. Baurès, A.V. Jung, O. Thomas
Heavy rainfall events, increasing in frequency and intensity with climate change, impact on the quality of the water resource used
for drinking-water production. Small-scale water suppliers are particularly sensitive because of their management and the related
difficulties of adapting treatment to variations. Decision-support systems, based on monitoring and analytical tools, need to be
developed to improve crisis-management procedures related to such events. After presenting the issues related to heavy rainfall
events, the article summarizes the tools currently used for quality control of drinking water within this framework, the need for
developments and other requirements.
ª 2011 Elsevier Ltd. All rights reserved.
Keywords: Analytical tool; Climate change; Contamination; Decision-support system; Drinking water; Monitoring; Quality control; Rainfall;
Small-scale water supplier; Water treatment
1. Introduction with rapid changes in source-water qual-

B. Roig*, I. Delpla, E. Baurès, ity. This point is very important, because
A.V. Jung, O. Thomas
For half-a-century, worldwide political freshwater resources have the potential to
Environment and Health
Research laboratory (LERES),
awareness of river-water quality issues has be strongly impacted by climate change,
Advanced School of Public been increasing substantially. For exam- with wide-ranging consequences for
Health (EHESP), Avenue du ple, the Water Framework Directive (WFD) human societies [4]. According to the
Professeur Léon Bernard – CS asks European Union (EU) member states 2007 IPCC report [5], it is likely that the
74312, 35043 Rennes Cedex,
to ensure the necessary protection for the frequency of heavy precipitation events
France
A.V. Jung
bodies of water identified with the aim of has increased over most areas, particu-
School of Environmental avoiding deterioration in their quality in larly in Europe and North America, and
Engineering (EME), order to reduce the level of purification that available research suggests a signifi-
Campus de Ker Lann, treatment required in the production of cant future increase in heavy rainfall
Avenue Robert Schumann,
drinking water [1]. Moreover, within the events in many regions. Regarding the
35170 Bruz,
France
framework of the WFD Common Imple- change in rainfall patterns, the increase in
mentation Strategy, EU member states frequency of intense rainfalls of short
may establish safeguard zones for these duration is already being observed. Fig. 1
water bodies by following the guidance on presents a synthesis of the different types
Groundwater in Drinking Water Protected of events, from light rain to flash and plain
Areas [2]. In this context, water-safety floods on a local or regional scale.
plans (WSPs) aim to ensure the safety of a Resulting changes in flow regimes will
drinking-water supply through the use of a influence the chemistry, hydromorpholo-
comprehensive risk-assessment and risk- gy and ecology of water bodies [6]. In
management approach [3]. particular, associated with the type of
The first step of this approach is identi- heavy rainfall events, the impacts on the
fication of hazards and hazardous events, quality of water resources can be severe
*
Corresponding author. the first of which comprises the meteorol- for surface water and karstic ground-
Tel.: +33 4 66 27 95 71; ogy and the weather patterns associated water, through run-off transportation of
E-mail: benoit.roig@ehesp.fr
0165-9936/$ - see front matter ª 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.trac.2011.04.008 1243
Trends Trends in Analytical Chemistry, Vol. 30, No. 8, 2011
local “ regional”
Space
Flash Flood
Extreme rain Plain Flood
Intensity / vol of water
Heavy rain Permeability

High intensity Land use
Slope
Storm
Soil moisture
Moderate rain
Shower
Low intensity
Storm
Light rain
Duration
Figure 1. Types of events with regard to intensity and duration of rain. Categories of rain are defined by their intensity. Commonly, heavy, mod-
erate and light rain are characterized respectively by 2.5 mm, 2.5–7.5 mm and >7.5 mm of water/hour. Showers and low-intensity or high-inten-
sity storms can also be distinguished according to the duration of such rainfall. The consequences also depend on several parameters (soil
moisture, permeability, land use and field slope) and can result in extreme rain or flooding, respectively, at local or regional scales.
suspended solids, chemical substances (natural and in sorption and leaching of hazardous organic chemicals
anthropogenic) [7,8] and pathogens [9]. The quality of in the environment [19,20]. Refractory organic sub-
the drinking water may also be degraded in the event of stances are also found in drinking water, and are
heavy rainfall or floods because of water-treatment fail- important precursors, along with NOM, of the formation
ures (from overloading to flooding) leading to outbreaks of haloforms [also known as trihalomethanes (THMs)]
of water-borne disease [4,10]. At the very least, an and further disinfection by-products (DBPs) during
uncontrolled overload of microbial and chemical pollu- application of chlorine [21,22].
tants can be difficult to handle through the use of con- As stated in a recent WHO report [17], extreme
ventional drinking-water-treatment (DWT) processes. weather events could affect the efficiency of DWT
The diversity of native flora and wildlife generally also processes, leading to a possible degradation of quality in
decreases, while tolerant species become more abundant distributed drinking water. Water suppliers can prepare
[11–13]. As an example, some studies showed a higher to minimize the impact of extreme events on the service
presence of enteric pathogens resistant to chlorination provided in a number of ways (e.g., implementation of
(e.g., Cryptosporidium) during the rainy season [14,15]. pro-active measures to identify changes in quantity and
Moreover, the increased load of suspended particles can quality of water resources).
jeopardize the efficiency of water-filtration and water- Finally, these concerns are of primary importance for
treatment systems, increasing the risk of contamination small-scale water suppliers (SSWSs), private, commu-
of the drinking-water supply [16]. nity-based or public, which usually prevail in rural
Short-term, intense rainfall events could lead to a areas. Considering that part of the population living in
strong increase in dissolved organic carbon (DOC) in rural areas does not have access to improved drinking-
surface waters. It was found that most DOC is exported water sources {e.g., 16 million people in Europe [23]},
from catchments during storm events [17]; meanwhile, SSWSs are facing several challenges, including their
a DOC increase in surface waters has already been regulatory environment, administration, management,
observed for the past two decades due to reduction in operation, and their available technical, human and
acidic deposition [18]. Among the constituents influ- financial resources [16].
encing the transport and the mobility of pollutants, In this context, simple tools are needed in various
natural organic matter (NOM) has an indisputable role areas {e.g., for the rapid assessment of the vulnerability
1244 http://www.elsevier.com/locate/trac
of water utilities to climate change [23] or the adequate frequency is useful to characterize baseflow
improvement of compliance monitoring, validating water quality, it is not adapted to the context of extreme
effectiveness and monitoring performance against water- rainfall events. Guidance has been proposed to assist in
quality limits, particularly during or after heavy rainfall designing appropriate sampling strategies; in particular,
events}. several general recommendations have been made,
This article summarizes the analytical issues and including the use of flow-interval or variable time-interval
currently available tools required to produce water at a sampling, and also composite sampling to limit the
constant quantity and quality in situations of extreme number of samples collected [26,27]. Grab sampling has
climatic events (e.g., heavy rain). It also aims to been judged a relatively inefficient methodology for cap-
provide an overview of suitability, needs, and future turing mean concentrations for water subjected to highly
developments for drinking-water-safety policy required variable loads [28]. Moreover, when comparing grab,
to meet the special conditions of such events. integrated, and automated storm-sampling techniques,
the frequency and the duration of sampling, on the one
hand, and the location of sample collection, on the other,
2. Needs and available solutions have to be taken into account. In particular, reduction in
the uncertainty of measured concentrations can be con-
Within the implementation of the water-safety plan, sidered by using automated sampling, multiple integrated
water-quality monitoring is a key factor for all hazards samples throughout the duration of the event, or, if
related to potential deterioration of water bodies, namely integrated or automated sampling cannot be carried out,
surface and karstic waters [3,4]. In this context, drink- multiple grab samples from various locations throughout
ing-water monitoring during short-term, intense rainfall a cross-section [29].
events involves careful consideration of the whole ana- Automatic samplers are designed to obtain a more
lytical process, from sampling design to analysis and representative picture of water quality. These devices
delivery of results. Indeed, such a climatic event is could be used with peristaltic and vacuum sampling lines.
mainly characterized by strong increase in water flow, In the event of heavy rainfall events, vacuum samplers
leaching, and run-off, all resulting in a change in the seem to be more suitable, considering recommendations
transport of contaminants and likely alteration to the of the field investigation section of the US Environmental
quality of raw water (in particular, surface and karstic Protection Agency (EPA) [30]. More representative
waters) used for drinking-water production. Moreover, samples are obtained, particularly when turbidity and
the modification to flows can disrupt the performance of particles are present in the water flow, as is the case in
the water-supply plants, particularly if they are small rainfall events. Moreover, in these conditions, absence of
scale. Thus, it is important to implement analytical a pump seems to be an advantage, because it ensures the
methods adapted to these changes in order to anticipate continuous correct functioning of the sampler.
any modification or disruption affecting the production Furthermore, two different collection regimes, either
of drinking water. at regular time intervals (time-controlled) or in response
to discharge changes (flow-controlled) can be proposed.
2.1. Sampling Flow-controlled configuration, in rainfall conditions, is
Sampling is a critical step in the monitoring procedure more viable for detecting rapid changes in raw-water
because it could represent the main contribution to error quality, in particular because it can adapt the sample
in the whole analytical process [24]. Thus, defining drawn to the flow rate [24] and then collect flow-
temporally accurate sampling strategies during fast connected fractions. More generally, automatic samplers
variations in flow is essential for accurate assessment of are adaptable to suit the type of collection required for
the degradation in quality of raw water and drinking analytical purposes (e.g., glass or polyethylene) and
water. The WFD has set up guidelines for designing permit sample identification (e.g., sampling time and
monitoring programs that include the objectives of location of sampling).
supporting risk-assessment procedures and appraising
long-term changes whose causes are both natural and 2.2. Analysis
anthropogenic [25]. This involves consideration of fast Conventional laboratory methods are often too time
degradation in water quality during rainfall events. As a consuming for adequate in situ operational response in
consequence, we need appropriate sampling tools and the case of fast degradation of raw-water quality, and do
methods, easy to use and reliable, taking into account not make it possible to protect human health in real time
flow variations and allowing an adequate number of [31]. Other rapid tools can optimize the time necessary
samples to be taken. for parameter determination and reduce sample alter-
Several studies have compared the efficiency of grab ation due to transport and storage. These techniques are
and composite samples during fluctuations in water vol- required by water utilities, in particular SSWSs, which
ume. While periodic manual grab-sample collection with are more vulnerable to any changes in influent quality.
http://www.elsevier.com/locate/trac 1245
Three main field methods are commonly differentiated: ular, the high amount of suspended solids and organic
in situ; on-line; and, off-line [32]: matter, can lead to clogging or high interference with
(1) in situ techniques require the use of sensors placed sensors, and consequently decrease their performance.
in the water body with no sampling step. To prevent clogging, some techniques already exist (e.g.,
(2) on-line methods involve a sampling step, followed implementation of ultrasonic cleaning and the develop-
generally by pre-treatment and measurement based ment of inductive probes to reduce interference). To
on optical techniques; all these steps could be auto- prevent interference, optical techniques (e.g., UV-visible
mated. absorptiometry) could require a dilution stage to ensure
(3) off-line methods generally require a conventional a quantifiable measurement.
spot-sampling step, followed by an analytical Generally, one common solution for on-line systems is
method that is faster and/or less expensive than to add a dilution loop to prevent interference [38]
commonly-used laboratory methods. although this could limit sensitivity.
These tools, which were developed for surface waters
and wastewaters [33], allow fast diagnosis of water- 2.4. Chemical micropollutants
quality degradation, so they present a clear potential for Micropollutants are not systematically monitored in raw
implementation in SSWSs during fast variations in flow. and treated waters. Chromatographic techniques
It should be noted that these alternative monitoring tools coupled with mass spectrometry have already been used
are not fit for purpose in terms of complying with reg- on-line to measure micropollutants {e.g., THM [38–40]
ulations (e.g., WFD), but they can be used as to com- and pesticides [41]}, and can provide quasi-real-time
plement standardized laboratory methods. However, field information [31]. However, in order to prevent device
and on-line methods used for chemical monitoring pro- clogging by heavily loaded waters, these systems could
grams should, for example, have to comply with the be equipped with a filtering unit [42] or other
requirements of Directive 2009/90/EC, which set mini- pre-treatment [32].
mum performance criteria (uncertainty of measurement Immunoassay-based tests have been widely used in
and limit of quantification). Directive 2009/90/EC also detection of micropollutants, in particular in environ-
requires that methods are validated and documented in mental waters. They could make fast analysis possible,
accordance with the EN ISO/IEC-17025 standard or and comparison with reference methods shows similar
other equivalent standards accepted at the international results. As an example, the on-line RIANA system was
level [34]. able to measure on-line the presence of atrazine and
isoproturon in a DWT plant [43]. Enzyme-linked
immune sorbent assay (ELISA) tests have proved suitable
2.3. Physico-chemical parameters for measuring atrazine, glyphosate, alachlor and phe-
Water utilities already use sensors for process control and nylureas in raw waters during precipitations/run-off
regulatory compliance {e.g., contaminant detection [32] events [44]. Nevertheless, it was found that these tests
throughout the DWT process and the distribution have a significant bias towards finished (drinking)
system}. In drinking-water plants, sensors can be used for waters [44] and could also cross-react with other
in situ measurement of temperature, pH, redox potential, molecules with similar structures [37].
conductivity, dissolved oxygen, free chlorine, total chlo- The effect of the presence of micropollutants can also
rine, total organic carbon, nitrates, nitrites, ammonia, be assessed by in situ biological systems (e.g., Daphnia
turbidity and particle counts [34]. These sensors are taximeter, ToxControl Musselmonitor and Truitosem).
mainly based on electrochemical [35] or spectroscopic These early warning systems (EWSs) are integrated
techniques [32]. Acquisition of ultraviolet (UV)-visible monitoring systems that can analyze and interpret
spectra is on-line or off-line for measuring turbidity and results in real time [31]. EWSs are based on living
nitrates/nitrites, and assessing chemical oxygen demand, organisms and monitor overall water quality by mea-
biological oxygen demand, total organic carbon, total suring changes in species behavior. These systems were
suspended solids, and aromatic compounds [31,36]. initially developed to protect water supplies against
Laboratory-derived techniques (e.g., on-line DOC accidental contamination, but they could also be used to
analyzers) could also represent an efficient method, monitor short-term degradation of raw-water quality
already implemented in a DWT plant [37], even if this after rainfall events and to prevent delivery of water
technique is not commonly used at present. unsuitable for drinking purposes. Such EWSs could
Finally, colorimetric test kits could also be useful to constitute a viable option for remote monitoring and
determine quickly nutrient concentrations (nitrates, control of small drinking-water facilities. However, in
nitrites, ammonium and phosphates) in raw water, spite of their high sensitivity, such devices find their limit
despite their low sensitivity [37]. in treated waters through their poor selectivity and the
However, whatever the techniques, the increase in sensitivity to chlorine of the living organisms chosen
water-pollution charge during rainfall events, in partic- [31].
Finally, micropollutant analysis needs to face new long time. But, because there are questions about their
challenges, including the presence of emerging con- ability to predict the likely presence of pathogens, other
taminants (e.g., human and veterinary pharmaceuticals, indicators have been proposed (e.g., bacteroides, RNA
steroids and hormones, prescription and non-prescrip- phage and host-specific viruses) [55]. However, they
tion drugs, and personal-care products). It is well known remain difficult to implement directly in the field because
that they are present in the different environmental they require DNA amplification as a principal step of the
compartments (soil, water, sediments) and their analytical process. Moreover, one additional difficulty is
behavior can depend on environmental parameters [45]. the preconcentration of microorganisms, in particular
In particular, intense rainfalls and associated run-off viruses, as there is no universal method for good
modify their transport and mobility from soil to water recovery of the majority of these [54].
[46–48], increasing the contamination levels of surface Here, therefore, is a very important challenge: to
waters and, potentially, drinking water. As they are develop new methods for the rapid characterization of
suspected to have an impact on both the environment biological contamination. Currently, characterization of
and human health [49], mainly through chronic expo- the global bacteria charge is often rapidly assessed by ATP
sure to low concentrations (ng/L–lg/L), there is a clear measurement [56,57] but without any selectivity. Rapid
need for quick detection of these substances as well as tests are commercially available [e.g., QGA from Lumin-
detection of by-products generated by disinfection pro- ultra (www.luminultra.com) or Dendridiag SW from
cesses. Indeed, it has been shown that chlorinated GLBiocontrol (www.gl-biocontrol.com)]. With regard to
by-products could lead to higher genotoxicity or viruses, rapid methods are scarce, particularly because of
carcinogenicity than their mother molecules [38,50]. the difficulty of quantitatively concentrating them from
water. Nevertheless, methods [e.g., reverse-transcription
2.5. Biological contaminants polymerase chain reaction (RT-PCR)] have shown satis-
As for sewage, fertilizers and other organic wastes and factory efficiency in detecting viruses (e.g., hepatitis A) in
chemicals, heavy rainfall and associated floods, if tap water in the laboratory [58], but, so far, the PCR stage
occurring, can flush microorganisms into waterways has been difficult to implement in the field.
and aquifers. Greater water flow contributes to increas-
ing the pathogen load and the penetration speed of
pathogens into resource waters and the drinking-water 3. Trends
supply [51]. Levels of biological contaminants tend to
increase during such events. As an example, a signifi- As previously shown, short-term, intense rainfall events
cant number of outbreaks of waterborne disease have are characterized by high volumes of water, high loads of
involved the transport of bacteria, viruses, or small suspended particulate matter, increases in organic
parasites into water systems or well-heads {[52] and matter transported by run-off and increased concentra-
references therein; [53] and references therein}. Moni- tion of microorganisms (some of them being pathogens).
toring the increase of this biological charge is very Large drinking-water units can absorb this kind of
important to ensure maximum efficiency of the treat- variation and are not really disrupted by such events,
ment and protection of the population. except in particularly extreme conditions. Conversely,
The main laboratory technique currently used to SSWSs, fed by surface water, are not well adapted to deal
detect the presence of pathogenic strains in drinking- with such events and will face severe difficulties. This is
water and/or water resources is based on microbiological confirmed by a very recent decision of the Committee of
culture. Although reliable under normal conditions, it is the European Drinking Water Directive (DWD), which
not suited to times of crisis (e.g., periods of intense concluded that increased implementation and enforce-
rainfall). Indeed, the tedious, time-consuming process of ment efforts are required to ensure safe drinking water,
culture represents a huge drawback with regard to rapid particularly in smaller supplies, and that no revision of
quality control (QC). Moreover, only cultivable bacteria legislation is required for the DWD [59]. As an example,
(in appropriate culture media) can be detected, while increased pH and turbidity (high load of organic matter)
there is a need for wider bacterial screening. As for can modify the efficiency of coagulation-sedimentation
viruses, most species of fecal origin cannot be detected treatment [60]. Higher TOC and turbidity can also inter-
with conventional cell-culture methods [54]. fere in chlorination processes by reducing the free-
Molecular techniques have been developed in the chlorine residual and creating chlorine demand [61].
laboratory for more sensitive, rapid characterization of Consequently, processes need to be adapted to consider
biological contamination. These techniques are used for such variation, and there is an important need to help
the rapid detection of specific species (bacteria and water managers to take the appropriate decisions;
viruses) as indicators of human fecal contamination. monitoring is a key element. These efforts will also help to
Traditional fecal indicator bacteria (e.g., fecal coliforms, improve the treatment efficiency of smaller supplies even
Escherichia coli and enterococci) have been used for a outside rainy periods.
Rapid tools for water-quality monitoring are required biological methods [e.g., PCR, quantitative PCR (QPCR)
in this case, not only systems giving a quantitative and microarray] have been developed in some drinking-
response (e.g., concentration of contaminant), but also water monitoring applications [63–65]. They are based
systems providing information on changes in water on the detection of specific gene(s) after amplification.
quality (e.g., indicator and index). In the case of an However, these methods are currently difficult to
unusual event, a follow-up procedure using comparisons implement on-site because of the lack of miniaturized
with regular conditions, based on EWSs, could often be PCR devices. A recent study tested the efficiency of a new
sufficient to manage and to anticipate a potential crisis. portable real-time RT-PCR (rRT-PCR) unit (RAPID,
Thus, implementation of indicators of the global physico- Idaho Technologies, Salt Lake City, UT, USA) for in-field
chemical, chemical and biological quality should be detection of avian influenza virus in wild birds, and
promoted. The first of these (e.g., probes and sensors) are compared it with virus isolation combined with rRT-PCR
already available, even if in some cases their sensitivity conducted in a laboratory. Results indicated that the
can be questionable, as explained previously. portable rRT-PCR unit had equivalent specificity to virus
Concerning chemical (micropollutant) and biological isolation with no false positives, but sensitivity was
(pathogen) indicators, there are very few adequate compromised at low viral titers [66]. Another portable
solutions, and here there is a major challenge. Indeed, PCR is available {[67], Ahram biosystem http://
the strategy used so far, involving development of one www.ahrambio.com/product.html}, but still needs some
system for each substance, is limited because of the validation for its application in the context of climate
number of substances to monitor, the cost, and the time change and water quality.
required to perform a total analysis. A strategy for each Detection of viral particles also suffers, due to their
family or group of compounds must be favored. Sensors relatively low density in aquatic environments that
should be developed to give access to quantitative makes representative, reliable measurement particularly
information (concentration or representative parameter) difficult to perform without preliminary concentration.
and, at the same time, qualitative information (effect or Different solutions are available for concentrating
representative parameter). This can be achieved by viruses from water, but their recovery rate does not
molecular biodiagnostic tools, which couple a molecular exceed 60% [68], except in particular cases [54]. Poly-
recognition element (e.g., enzyme, antibody, peptide, mers using biomaterial (e.g., chitosane and lysine) can
receptor, nucleic acid) with a physico-chemical trans- be used due to their affinity [e.g., weak interaction (e.g.,
ducer (e.g., optical, electrochemical, and physical) [62]. electrostatic, hydrogen bond)] with viral particles (con-
These transducer technologies can be miniaturized and taining electrostatic charges and hydrophobic groups
allow compact, portable devices to be made available. expressed variably as a function of pH, chemical and
Furthermore, these tools require improvements in organic components of the water). To be compatible with
rapid sampling and on-site extraction procedures in rapid, on-site applications, they must be packaged in an
order to achieve increased sensitivity and purity of appropriate form (i.e. powder, magnetic beads, hollow
samples. (Micro)extraction using solid phases in associ- fibers, or foam). This application has already been
ation with portable devices needs to be developed. In- developed for medical diagnostics [e.g., Ademtech
deed, this technique has been successfully developed as a (http://www.ademtech.com/) and ApoH technologies
pre-treatment for chromatography (liquid and/or gas) (http://www.apohtech.com/en/home.html)], and is to be
but now needs to be adapted to on-line or on-site devices. extended to environmental applications.
Very few models are currently available. Finally, as for resource and water management, the
The Hocer Aquapod SPE50 (http://hocer.images- development of decision-support systems (DSSs) could be
creations.fr/content.cfm?id=129) couples a solid-phase particularly advantageous in cases of extreme rainfall
extraction (SPE) module with a UV detector and is used to events. According to Power et al. [69], any system
detect pesticides, hydrocarbons and other industrial sub- supporting decision-making (e.g., executive information
stances. Improvements are planned for this system, with systems, executive support systems, geographic infor-
remote sensing and assistance, and better sensitivity. Note mation systems, and on-line analytical processing and
that passive samplers, which are one of the main sampling software agents) may be called a DSS. Generally, DSSs
methods of the future for micropollutants in water [32], involve the development of models to support decision
could also be suggested for use in heavy rainfall events, making for water (and wastewater) preventive man-
but they would need to be adapted to an on-line sampling agement systems. Different criteria can be used (e.g.,
belt or connected directly to a sensor. Unfortunately, there health, environment, economy, social, or technology),
has been no such development to date. each requiring a set of indicators {i.e. quantitative or
The rapid assessment of pathogenic contamination is qualitative data, including satellite imaging [70–72]}, to
even more challenging. Indeed, the culture method still assess comparatively the different options in the deci-
remains the method principally used in all circumstances sion-making process, taking into account the different
in spite of its drawbacks. For several years, molecular criteria [73].
As a recent example, GESCAL, a DSS for water-quality quality of the associated drinking water may also be
issues, simulates conductivity, phosphorous, carbona- degraded because of a lack of adaptability or failures in
ceous organic matter, dissolved oxygen, organic nitro- water treatment, in particular in small-scale supply
gen, ammonia, and nitrates to provide an integrated services.
model of water quality for whole water-resource systems, While Water Safety Plans are increasingly being
including reservoirs and rivers [74]. The practice of integrated into water regulations with the implementa-
modeling is not recent, as shown by the modeling tion of hazard assessment for risk management, drink-
support systems (WODA) developed to register biological ing-water quality compliance in small-scale water
oxygen demand (BOD) and dissolved oxygen (DO) con- treatment should be improved. These actions do not
centrations in a stretch of river under different hydro- necessarily fit in with the need for better understanding
logical and thermal conditions [75] or the generic DSS of the impact of short-term, heavy rainfall events, based
(Aquatool) designed for the planning and operational on more frequent, relevant measurements in order to
stages of decision making associated with complex river monitor short-term variations in water quality. But for-
basins [76]. But advances in computer technology, getting the impacts of climate change will cancel out
availability of real-time hydroclimatic data, and these efforts, so, at least for the next decade, development
improvements in the ability to develop user-friendly and implementation of suitable monitoring tools and
graphical model interfaces have made possible develop- procedures must be planned with a view to a better
ment and application of DSS for drinking-water-resource knowledge of water-quality variation for a complete risk
systems. EPAÕs National Homeland Security Research assessment and management program. In particular,
Center (NHSRC) has developed the Water Systems field systems (on-line, on-site) need to be employed.
Materials Disposal Decision Support Tool (DST), a Web- Sampling procedures should be adapted to meet the
based platform, to assist and to help incident responders special nature of such events and the associated water
in emergency response to biological or chemical con- quality. Concerning analytical issues, global parameters
tamination of water systems (DWT or wastewater could be measured in real time with available tools.
treatment plant) [77]. A real-time DSS for adaptive daily However, these are lacking for chemical and biological
and weekly management of a reservoir system that micropollutants, as such analytical developments are
provides drinking-water has also been described [78]. By very challenging. Some methods are described, but are
integrating watershed models, reservoir hydraulic not widely used.
models, and a reservoir water-quality model, it offers the Finally, to respond and to react adequately to the
ability to optimize reservoir operations. As shown in crisis, analytical data need to be transferred to, and
these examples, DSS represents an interesting way of translated for, water managers for relevant, accurate
controlling water quality, in particular during strong decision making. DSSs represent the solution of the
and rapid modifications of water-system conditions. It future, and are beginning to be employed in the
should allow improved reactivity during a period of drinking-water supply sector.
crisis, with the selection of the best scenario(s) from the
rapid assessment of the degradation of the water system.
Although currently limited to a Web interface, the References
[1] European Commission, Directive 2000/60/EC of the European
concept is intended to be developed through the devel-
Parliament and of the Council of 23 October 2000 establishing a
opment of real-time data transmission (e.g., phone and framework for Community action in the field of water policy, Off. J.
Wi-Fi). Finally, DSS should be implemented by both Eur. Comm. L327 (22.12.2000).
drinking-water managers and producers and sanitary [2] European Commission, Common Implementation Strategy for the
authorities. This is of primary importance for SSWSs, Water Framework Directive (2000/60/EC) Guidance Document
No. 16, 2007 (http://circa.europa.eu/Public/irc/env/wfd/librar-
with regard to efforts planned for the improvement of
y?l=/framework_directive/guidance_documents&vm=detailed&sb
drinking-water quality compliance in one-third of SSWSs =Title).
in Europe and the implementation of a risk-based [3] J. Bartram, L. Corrales, A. Davison, D. Deere, D. Drury, B. Gordon,
approach for more effective QC [59]. G. Howard, A. Rinehold, M. Stevens, Water Safety Plan Manual:
Step-By-Step Risk Management for Drinking-Water Suppliers,
World Health Organization, Geneva, Switzerland, 2009.
[4] B. Bates, Z.W. Kundzewicz, S. Wu, J. Palutikof (Editors), Climate
4. Conclusion Change and Water, Technical Paper of the Intergovernmental
Panel on Climate Change, IPCC Secretariat, Geneva, Switzerland,
Heavy rainfall events are characterized by high vol- 2008, 210 pp.
umes of water, increased flow rates, and, consequently, [5] M.L. Parry, O.F. Canziani, J.P. Palutikof, P.J. van der Linden, C.E.
Hanson (Editors), Contribution of Working Group II to the Fourth
more significant run-off transportation of suspended
Assessment Report of the Intergovernmental Panel on Climate
solids and possibly chemical and biological substances. Change, Cambridge University Press, Cambridge, UK, 2007.
The quality of water resources (surface water and [6] P.G. Whitehead, R.J. Wilby, R.W. Battarbee, M. Kernan, A.J.
karstic groundwater) can be severely impacted, and the Wade, Hydrol. Sci. J. 54 (2009) 101.
[7] J.H. Pardue, W.M. Moe, D. McInnis, L.J. Thibodeaux, K.T. [34] Y.J. Yang, R.C. Haught, J.A. Goodrich, J. Environ. Manage. 90
Valsaraj, E. Maciasz, I. van Heerden, N. Korevec, Q.Z. Yuan, (2009) 2494.
Environ. Sci. Technol. 39 (2005) 8591. [35] S. Capelo, F. Mira, A.M. de Bettencourt, Talanta 71 (2007) 1166.
[8] I. Delpla, A.V. Jung, E. Baures, M. Clement, O. Thomas, Environ. [36] C. Gonzalez, E. Prichard, S. Spinelli, J. Gille, E. Touraud, Trends
Int. 35 (2009) 1225. Anal. Chem. 26 (2007) 315.
[9] F.C. Curriero, J.A. Patz, J.B. Rose, S. Lele, Am. J. Public Health 91 [37] B. Roig, Techniques de lÕingénieur, P 3900 (2007).
(2001) 1194. [38] M.A. Brown, G.L. Emmert, Anal. Chim. Acta 555 (2006) 75.
[10] T. Kistemann, T. Claßen, C. Koch, F. Dangendorf, R. Fischeder, J. [39] M.A. Brown, S. Miller, G.L. Emmert, Anal. Chim. Acta 592 (2007)
Gebel, V. Vacata, M. Exner, Appl. Environ. Microbiol. 68 (2002) 154.
2188. [40] T.C. Chen, G.R. Her, J. Chromatogr., A 927 (2001) 229.
[11] C.J. Walsh, A.H. Roy, J.W. Feminella, P.D. Cottingham, P.M. [41] H. Bagheri, E.R. Brouwer, R.T. Ghijsen, U.A.Th. Brinkman, J.
Groffman, R.P. Morgan, J. North Am. Benthol. Soc. 24 (2005) Chromatogr., A 647 (1993) 121.
706. [42] S. Lacorte, J.J. Vreuls, J.S. Salau, F. Ventura, D. Barceló, J.
[12] M.L. McKinney, Biol. Conserv. 127 (2006) 247. Chromatogr., A 795 (1998) 71.
[13] K. Vermonden, G. van der Velde, R.S.E.W. Leuven, Resources, [43] S. Rodriguez-Mozaz, M.J. López de Alda, D. Barceló, Talanta 69
Conserv. Recycling (2011) (doi: 10.1016/j.resconrec.2011.01. (2006) 377.
003). [44] N. Graziano, M.J. McGuire, A. Roberson, C. Adams, H. Jiang, N.
[14] M. Nchito, P. Kelly, S. Sianongo, N.P. Luo, R. Feldman, M. Blute, Environ. Sci. Technol. 40 (2006) 1163.
Farthing, K.S. Baboo, Am. J. Trop. Med. Hyg. 59 (1998) 435. [45] B. Roig, Pharmaceuticals in the Environment: Current Knowledge
[15] G. Kang, B.S. Ramakrishna, J. Daniel, M. Mathan, V. Mathan, and Need Assessment to Reduce Presence and Impact, IWA
Epidemiol. Infect. 127 (2001) 107. Publishing, London, 2010.
[16] L. Sinisi, R. Aertgeerts, Guidance on Water Supply and Sanitation [46] P. Drillia, K. Stamatelatou, G. Lyberatos, Chemosphere 60 (2005)
in Extreme Weather Events, World Health Organization, Geneva, 1034.
Switzerland, 2010. [47] L. Barron, E. Nesterenko, K. Hart, E. Power, B. Quinn, B. Kelleher,
[17] J.B. Fellman, E. Hood, R.T. Edwards, D.V. DÕAmore, J. Geophys. B. Paull, Anal. Bioanal. Chem. 397 (2010) 287.
Res. 114 (2009) G01021. [48] S.C. Kim, J.G. Davis, C.C. Truman, J.C. Ascough, K. Carlson, J.
[18] C.D. Evans, D.T. Monteith, D.M. Cooper, Environ. Pollut. 137 Hazard. Mater. 175 (2010) 836.
(2005) 55. [49] M. Schriks, M.B. Heringa, M.M.E. van der Kooi, P. de Voogt, A.P.
[19] B. Lennartz, X. Louchart, Environ. Pollut. 146 (2007) 180. van Wezel, Water Res. 44 (2010) 461.
[20] M. Pédrot, A. Dia, M. Davranche, M. Bouhnik-Le Coz, O. Henin, G. [50] S.D. Richardson, M.J. Plewa, E.D. Wagner, R. Schoeny, D.M.
Gruau, J. Colloid. Interf. Sci. 325 (2008) 187. DeMarini, Mutat. Res. 636 (2007) 178.
[21] P.C. Singer, Water Sci. Technol. 40 (1999) 25. [51] A. Unc, M. Goss, Water Air Soil Pollut. 149 (2003) 327.
[22] M. Palacios, J.F. Pampillon, M.E. Rodriguez, Water Res 34 (2000) [52] H. Auld, D. MacIver, J. Klaassen, J. Toxicol. Environ. Health A 67
1002. (2004) 1879.
[23] World Health Organization (WHO), Vision 2030 – The Resilience [53] K.M. Thomas, D.F. Charron, D. Waltner-Toews, C. Schuster,
of Water Supply and Sanitation in the Face of Climate Change, A.R. Maarouf, J.D. Holt, Int. J. Environ. Health Res. 16
World Health Organization/Department for International Devel- (2006) 167.
opment, Geneva, Switzerland, 2010 (http://www.who.int/ [54] R. Girones, M.A. Ferrús, J.L. Alonso, J. Rodriguez-Manzano, B.
water_sanitation_health/publications/9789241598422_cdrom/ Calgua, A. de Abreu Corrêa, A. Hundesa, A. Carratala, S. Bofill-
en/index.html, accessed 5 January 2011). Mas, Water Res. 44 (2010) 4325.
[24] Y. Madrid, Z.P. Zayas, Trends Anal. Chem. 26 (2007) 293. [55] W. Ahmed, A. Goonetilleke, T. Gardner, J. Aust. Water Waste-
[25] European Commission, Water Framework Directive, Common water Assoc. (2008) 38.
Implementation Strategy Working Group 2.7. Monitoring: Guid- [56] H. Tanaka, T. Shinji (Iwano), K. Sawada, Y. Monji, S. Seto, M.
ance on Monitoring for the Water Framework Directive, Final Yajima, O. Yagi, Water Res. 31 (1997) 1913.
version, 23 January 2003 (http://www.uicnmed.org/web2007/ [57] D.J. Squirrell, R.L. Price, M.J. Murphy, Anal. Chim. Acta 457
cdflow/conten/5/pdf/5_1_2/Monitoring/guidan-Jan03.pdf). (2002) 109.
[26] R.D. Harmel, K.W. King, R.M. Slade, Appl. Eng. Agric. 19 (2003) [58] E. Dubois, C. Hennechart, G. Merle, C. Burger, N. Hmila, S. Ruelle,
667. S. Perelle, V. Ferré, Int. J. Food Microbiol. 117 (2007) 141.
[27] R.D. Harmel, K.W. King, B.E. Haggard, D.G. Wren, J.M. Sheridan, [59] H. Bloech, Summary Record of the Meeting of the Commit-
Trans. ASABE 49 (2006) 937. tee under Article 12 of Directive 98/83/EC, DG ENV, 22/02/
[28] A. Facchi, C. Gandolfi, M.J. Whelan, Chemosphere 66 (2007) 2011.
746. [60] G. Annadurai, S.S. Sung, D.J. Lee, Adv. Environ. Res. 8 (2004)
[29] R.D. Harmel, R.M. Slade Jr., R.L. Haney, J. Environ. Qual. 39 713.
(2010) 1734. [61] M.W. Le Chevallier, T.M. Evans, R.J. Seidler, Appl. Environ.
[30] Environmental Protection Agency (EPA), Report on Sampling Microbiol. 42 (1981) 159.
Equipment comparing Vacuum and Peristaltic Sampling Methods, [62] B. Roig, I. Bazin, S. Bayle, D. Habauzit, J. Chopineau, in: C.
1974, 80 pp. (Document PB259875, available from NTIS, Gonzalez, R. Greenwood, P. Quevauviller (Editors), Rapid Chem-
Alexandria, VA, USA). ical and Biological Techniques for Water Monitoring – Water
[31] M.V. Storey, B. van der Gaag, B.P. Burns, Water Res. 45 (2010) Quality Measurements Series, John Wiley & Son, Chichester, West
741. Sussex, UK, 2009, p. 175.
[32] R. Greenwood, G.A. Mills, B. Roig, Trends Anal. Chem. 26 (2007) [63] K.S. Reynolds, C.P. Gerba, I.L. Pepper, Water Sci. Technol. 35
263. (1997) 423.
[33] B. Roig, I. Allan, G.A. Mills, N. Guigues, R. Greenwood, C. [64] J.C. Vivier, M.M. Ehlers, W.O.K. Grabow, Water Res. 38 (2004)
Gonzalez, in: C. Gonzalez, R. Greenwood, P. Quevauviller (Editors), 2699.
Rapid Chemical and Biological Techniques for Water Monitoring – [65] P.T. Monis, C.P. Saint, Water Res. 35 (2001) 1641.
Water Quality Measurements Series, John Wiley & Son, Chiches- [66] J.Y. Takekawa, S.A. Iverson, A.K. Schultz, N.J. Hill, C.J. Cardona,
ter, West Sussex, UK, 2009, p. 39. W.M. Boyce, J.P. Dudley, J. Virol. Methods 166 (2010) 92.
[67] G. Maltezos, W. Chantratita, A. Gomez, F. Gomez, E. Kartalov, A. [73] P.-A. Malmqvist, H. Palmquist, Water Sci. Technol. 51 (2005) 41.
Scherer, IVD Technol. 16 (2010) No. 7. [74] J. Paredes, J. Andreu, A. Solera, Sci. Total Environ. 408 (2010)
[68] J.C. Block, L. Schwartzbrod, Viruses in Water Systems. Detection 2576.
and Identification, VCH Publishers, New York, USA, 1989. [75] A. Kraszewski, R. Soncini-Sessa, Environ. Software 1 (1986)
[69] D.J. Power. On-Line Exec. J. Data-Intensive Decision Support 3 90.
(1997) (http://dssresources.com/papers/whatisadss/index.html). [76] J. Andreu, J. Capilla, E. Sanchı́s, J. Hydrol. 177 (1996) 269.
[70] F.L. Hellweger, P. Schlosser, U. Lall, J.K. Weissel, Estuar. Coast [77] Environmental Protection Agency, National Homeland Security
Shelf S. 61 (2004) 437. Research Center, Technical brief, 2007 (available at www.
[71] V.M. Chowdary, R.V. Chandran, N. Neeti, R.V. Bothale, Y.K. michigan.gov/.../deq-wws-wb-Water_Systems_DST_Overview_
Srivastava, P. Ingle, D. Ramakrishnan, D. Dutta, A. Jeyaram, J.R. 265102_7.pdf).
Sharma, R. Singh, Agric. Water Manage. 95 (2008) 754. [78] K.S. Westphal, R.M. Vogel, P. Kirshen, S.C. Chapra, J. Water
[72] S. Chen, W. Huang, W. Chen, H. Wang, Ecol. Inform. 6 (2011) Resources Plan. Manage. 129 (2003) 165.
147.
Provided for non-commercial research and educational use only.
Not for reproduction or distribution or commercial use.
This article was originally published by IWA Publishing. IWA Publishing recognizes
the retention of the right by the author(s) to photocopy or make single electronic
copies of the paper for their own personal use, including for their own classroom use,
or the personal use of colleagues, provided the copies are not offered for sale and
are not distributed in a systematic way outside of their employing institution.
Please note that you are not permitted to post the IWA Publishing PDF version of
your paper on your own website or your institution’s website or repository.
Please direct any queries regarding use or permissions to wst@iwap.co.uk

227 & IWA Publishing 2011 Water Science & Technology 9 63.2 9 2011
Issues of drinking water quality of small scale water

services towards climate change
I. Delpla, E. Baures, A. V. Jung, M. Clement and O. Thomas
ABSTRACT
As climate change could impact water quantity and quality, important concerns are related to I. Delpla
E. Baures
water quality degradation in small scale water services (SSWS). SSWS using surface waters M. Clement
O. Thomas (corresponding author)
resources (rivers and lakes) for drinking water production are particularly vulnerable to short term School of Public Health (Ecole des Hautes Etudes
en Santé Publique),
transient events due to their low adaptation capacity and their lack of support and technical Laboratoire d’Etude et de Recherche en
knowledge compared to major centralized systems. Based on weather and water quality Environnement et Santé (LERES),
Avenue Professeur Léon Bernard,
databases, a case study was conducted on a SSWS in Brittany (France) pumping from surface 35000 Rennes,
France
water. Results show an important vulnerability in treatment efficiency related to the lowest and E-mail: olivier.thomas@ehesp.fr
highest river flows and provide first assumptions about the impacts of an increase in extreme A. V. Jung
weather events with climate change on drinking water quality. School of Environment (Ecole des Métiers de
l’Environnement),
Key words 9 climate change, drinking water, small scale water services, water quality, Campus de Ker Lann,
Avenue Robert Schumann,
water treatment 35170 Bruz,
France
INTRODUCTION
Global change affects water quality through several factors. treatment with the quality variation of the resource. Among
The most evident is climate change with namely global water quality degradation factors, chemical and physical
warming increasing the number of extreme meteorological (increase of concentrations of chemicals and suspended
events (heavy rain falls, floods or droughts). A recent review solids) and biological ones (pathogens occurrence) are often
(Delpla et al. 2009) underlines that the main consequences on correlated with extreme meteorological events like heavy
water quality is an increase of total and dissolved organic rainstorms or droughts. Moreover, as SSWS are dispersed
carbon (TOC and DOC), nutrients (nitrate namely) and in rural area, the impact of chemical risk associated with
possibly total suspended solids (TSS) and pathogens (Hunter some micropollutants or emerging substances (namely pesti-
2003; Zwolsman & van Bokhoven 2007; Van Vliet & Zwols- cides and pharmaceuticals) is potentially present. Starting
man 2008). Heatwaves also promotes harmful cyanobacteria from the French water quality database for SSWS, the che-
blooms (Jöhnk et al. 2008). A second factor is the evolution of mical evolution of water quality with extreme meteorological
land use with for example intensive agricultural practices and events (from lowest flows and droughts to highest flows and
soil waterproofing in urban area, increasing TSS, nutrients floods) is studied for one SSWS on a period of near 20 years.
and micropollutants concentration with run off. The third The evolution of water quality is examined from samples
factor is a consequence of industrial efforts by reducing water taken both from the resource (river water) and the distribu-
and air discharges with for example acid rain decrease also tion network.
leading to DOC increase in lake water in glaciated landscapes
(Monteith et al. 2007).
The main consequence of water quality degradation MATERIAL AND METHODS
under global changes concerns small scale water services
(SSWS) pumping in lakes, at streams surface or after bank In Brittany, 80% of drinking waters are supplied from surface
filtration, the larger waterworks being designed to adapt the waters. Moreover, surface water quality is affected by strong
doi: 10.2166/wst.2011.038
228 I. Delpla et al. 9 Water quality from small-scale water services under climate change Water Science & Technology 9 63.2 9 2011
agricultural pressures (organic matter production: 4.7 billions geneous series because sampling frequency is variable from
of tons/year) and, topographic and geologic conditions also year to year and may be low for some. Moreover, sampling
favour the leaching of pollutants from agricultural area to frequency varied also between resource and treated waters,
rivers (low slopes and impervious soils). from 4/year to 12/year respectively. On the other hand,
For this work, one SSWS was selected by taking into changes in parameters determination also occurred in this
consideration the amount of people supplied, the water period. Thus, in order to extend the dataset for TOC results,
treatment plant capacity and the origin of water (surface which was less frequently measured in the resource than
water). This SSWS is located in Brittany in a little river permanganate oxydability (OxA), a statistical relationship
basin of 468 km2 under strong agricultural and related indus- between these parameters was used (TOC ¼ 1.045*OxA þ
trial pressures. Relevant information about the studied water 0.34, R2 ¼ 0.83, n ¼ 59), to include estimated TOC data from
supply service and its resource is presented in Table 1. OxA. Finally the use of TOC instead of DOC is of less
The presence of a small dam allows drinking water importance, with very low suspended solids concentration
treatment and distribution for the whole population even in in the great majority of samples.
droughts period with river flow less than 50 L/s.
Data are provided from the database SISE-EAUX (Agence
Régionale de Santé Bretagne) for water quality (resource and RESULTS
drinking water network) and from the database HYDRO
(Ministère de l’Ecologie et du Développement Durable) for Starting from raw data (resource quality and drinking water
the river flows measure at a limnimetric station in front of the network, respectively referenced as RES and SSWS), the evolu-
treatment plant. The data acquisition covers the period 1993– tion of some main parameters is firstly studied both for droughts
2009. Air temperature is provided by the weather station of or low water levels and floods or high water level periods. In a
Rennes Saint-Jacques, located at 18 km from the site. second time, the frequency of water quality limits exceedances
The choice of water quality parameters to be included is given for these events, namely for TOC.
was carried out with regard to the aim of this study. Para- Before studying the evolution of parameters during drou-
meters relevant with regard to drinking water quality evalua- ghts and floods let’s show their yearly evolution (Figure 1).
tion (TOC, nitrates, turbidity) are considered in the data By comparing TOC concentration and flow rate evolution
analysis. Micropollutants (namely pesticides) were not during the period 1993–2009, it could be noticed that highest
included due to their scarce determination at the beginning TOC values are observed during wet periods (1993–1995 and
of the studied period. Concerning the choice of environ- 1998–2001). Air temperature didn’t show significant evolu-
mental parameters external to water quality, river flow was tion during the period considered, even if an increase in
only considered, taking into account the lack of historical warm days and mean temperature have been observed in
meteorological data like rain volume or intensity, only known Brittany during the last sixty years (Tréguer et al. 2009).
from the weather station of Rennes Saint Jacques (see above). The first step of our study deals with the examination of
Before examining the main outcomes of this work, it has the impact of extreme meteorological events (from droughts
to be underlined that using water quality database provided to floods) on water quality of the resource (RES) and drinking
by the sanitary monitoring program may lead to hetero- water network (SSWS). Figure 2 presents the relation
Table 1 9 Main characteristics of the SSWS selected
Altitude Catchment area River Q (meana) River Q River Q

(mean dry periodb) (mean wet periodb)
35 m 468 km2 3.1 m3/s 1.9 m3/s 4.3 m3/s
Treatment process Mean daily volume Annual Volume Inhabitants supplied
Advanced treatment (Activated 1243 640000 5600

Carbon) þ Bleach/Ozone
aFor 43 years.
bCalculated for five-year dry and wet period values for the river studied.
RES
20 30
15
20
Flow (m3/s)
TOC (mg/l)
10
10
5
0 0
1993 1995 1997 1999 2001 2003 2005 2007 2009 1993 1995 1997 1999 2001 2003 2005 2007 2009
SSWS
6 40
30
Air temperature (°C)
4
TOC (mg/l)
20
10
2
0 -10
1993 1995 1997 1999 2001 2003 2005 2007 2009 1993 1995 1997 1999 2001 2003 2005 2007 2009
Figure 1 9 TOC evolution on resource and treated waters/ air temperature and river flow evolution.
between river flows and parameters values during droughts or particles from land to the resource. As for low water levels,
severe low water levels. no relation between air or water temperature and flows can
The main outcomes are first a slight concentration phe- be drawn.
nomenon for the TOC both in resource and drinking waters In order to clarify the nitrates evolution during dry and
(TOC values decreasing with flow). On the contrary, the wet period and extreme events (droughts and floods), a
evolution of nitrates concentration showed a very different relation with TOC is shown on Figure 4.
trend with the lowest values for the lowest flows. This typical The decrease in nitrates concentration for the dry period
evolution can be explained by the high mobility of this could be explained by the biomass consumption at low water
substance and consequently the reduced supply from soil velocity (high residential time), explaining the increase of
leaching under low hydrological conditions (Van Vliet & TOC. The decrease in nitrates concentration is also observed
Zwolsman 2008), but also by biomass nitrates consumption for high water levels (Q44.3 m3/s) with a lower slope and
favoured by very low water speed. No clear trend could be can be explained as discussed above by TOC and nitrates
noticed for turbidity. Finally, no relation between air or water behavior with higher flows.
temperature and flows can be drawn. After the examination of the impact of climate change on
For high water level or floods, the relations between river drinking water quality (SSWS) and resource (RES), the
flow and parameters values are displayed on Figure 3. second step of this study consists in a comparison of para-
Despite the number of data less numerous than for low meters data with quality limits or references of quality. These
water levels, the main outcomes are a slight dilution phenom- two thresholds give useful information with respect to regula-
enon for nitrates, and an increase in concentrations for TOC tion compliance and health risk assessment. Table 2 presents
and turbidity. TOC and turbidity show a distinct evolution the percentage of samples with higher TOC concentration
during high flow levels, probably related to an increase in run than the French limit of quality (QL) for catchment (10 mg/L)
off and consequently in transport of organic matter and and the reference of quality (QR) for the drinking water
RES SSWS
15 5
TOC (mg / L)
TOC (mg / L)
10
3
2
5
1
0 0
50 50
40 40
NO3- (mg / L)
NO3- (mg / L)
30 30
20 20
10 10
0 0
15 2.0
Turbidity (NTU)
Turbidity (NTU)
1.5
10
1.0
5
0.5
0 0.0
0.0 0.5 1.0 1.5 2.0 0.0 0.5 1.0 1.5 2.0
Flow (m3/s) Flow (m3/s)
Figure 2 9 Relation between river flow and parameters values during droughts or severe low water levels (resource-RES and network-SSWS).
network (2 mg/L) set by the ‘‘arrêté du 11 janvier 2007’’. DISCUSSION

It should be noticed that only the quality limit is imperative in
France. For SSWS, an international mobilisation is carried out in the
The main conclusions drawn from Table 2 are: frame of the implementation of WHO water and Health
Protocol, including water supply during extreme weather
– for low water levels or droughts period, an exceedance of
events. Even if experience sharing on good practices in the
5% and 35% respectively for resource and drinking water;
safe operation of water supply in rural areas subject to the
– for high water levels or floods period, an exceedance of
impact of short-term critical situations is planned (UNECE/
33% and 60% respectively for resource and drinking
WHO 2009), the regulatory trends is to simplify the sanitary
water.
control by introducing a risk assessment/risk management
– In between, an exceedance of 61% only for drinking water
approach, as in Europe with the future revision of the drink-
(with less data and values close to 2 mg/L
ing water directive.
Among the exceedances values in all classes for drinking Therefore, the requirement for operational monitoring
water around 50% of the exceeding values are lower than should be simplified (reduced) if there is evidence that a
2.5 mg/L and the more recent samples (since 2005) shows substance is not present or not used in the catchment. Unfor-
less excess than before. However efforts have to be made for tunately, emerging substances like pharmaceuticals are not yet
regulation compliance. However, the worst scenario is monitored in drinking water but can however be present
obviously for high water levels and floods with about a (Mompelat et al. 2009). In this case, new monitoring tools
third of samples exceed the limit of 10 mg/L for the TOC, and procedures for on site use and more appropriate labora-
needing a strong adaptation capacity of the treatment plant. tory monitoring methods must be proposed. New sampling
RES SSWS
20 5
15 4
TOC (mg / L)
TOC (mg/L)
3
10
2
5
1
0 0
60 60
NO3- (mg / L)
NO3- (mg/L)
40 40
20 20
0 0
150 5
Turbidity (NTU)
Turbidity (NTU)
4
100
3
2
50
1
0 0
4 12 20 28 36 44 52 60 4 12 20 28 36 44 52 60
Flow (m3/s) Flow (m3/s)
Figure 3 9 Relation between river flow and parameters values during floods or high water levels (resource-RES and network-SSWS).
strategies must also be designed to cover the water quality reduce trihalomethanes (THMs) precursors (Teksoy et al.
variability around extreme weather events. 2008). Finally, as recommended by the report on the work-
Besides water quality monitoring, decision support shop on safety of SSWS in the European region (UNECE/
system should also be proposed for safe water supply manage- WHO 2009), laboratories that specialize in emerging health
ment and risk prevention. For example, water treatment must risks, including cyanobacterial blooms, should be encouraged
be adapted to face with transient degradation of raw resource to work with small-scale water suppliers to assist and
in order not only to ensure safe water to end users but also to strengthen current health risk assessment.
60 Table 2 9 Frequency of quality standards exceedances for extreme conditions of river flow
Q < 1.9 m3/s (from droughts to floods) (n ¼ number of samples of the class)
Q > 4.3 m3/s

TOC max value (mg/L) 410 (QL) 42 (QR)
40
NO3- (mg / L)
Qa mean ¼ 3.1 m3/s Flow condition RES (%) SSWS (%)

(m3/s)
20 Droughts & o1.9 4.8 (n ¼ 84) 35 (n ¼ 60)

low water levels
Median water levels 1.9rQo4.3 0 (n ¼ 26) 61.5 (n ¼ 13)

0 High water QZ4.3 33.3 (n ¼ 33) 60 (n ¼ 25)
0 5 10 15 20
levels & floods
TOC (mg / L)
Figure 4 9 Relation between nitrates and TOC in the resource for dry () and wet (*) periods. aAverage value on 43 years.
CONCLUSION REFERENCES
The findings of the present study highlights expected results Delpla, I., Jung, A. V., Baures, E.; Clement, M. & Thomas, O. 2009
for the variation of the water quality of the resource: Impacts of climate change on surface water quality in relation to
drinking water production. Environ Int. 35, 1225–1233.
– for droughts periods, a decrease in nitrates concentration, Hunter, P. R. 2003 Climate change and waterborne and vector-borne
– for floods, an increase of TOC concentration and a disease. J Appl Microbiol. 94, 37S–46S.
decrease of nitrates levels. Jöhnk, K. D., Huisman, J., Sharples, J., Sommeijer, B., Visser, P. M. &
Stroom, J. M. 2008 Summer heatwaves promote blooms of harmful
For the quality of drinking water of the studied SWSS, the Cyanobacteria. Global Change Biol. 14, 495–512.
worst situation is related to high water levels and floods with Mompelat, S., Le Bot, B. & Thomas, O. 2009 Occurrence and fate of
pharmaceutical products and by-products, from resource to drink-
a rather high frequency of TOC quality reference excee-
ing water. Environment International 35, 803–814.
dances. These first conclusions are of great importance con- Monteith, D. T., Stoddard, J. L., Evans, C. D., de Wit, H. A., Forsius, M.,
sidering the possible increase in frequency and intensity of Hgåsen, T., Wilander, A., Skjelkvåle, B. L., Jeffries, D. S., Vuor-
droughts and floods in the future with climate change. enmaa, J., Keller, B., Kopácek, J. & Vesely, J. 2007 Dissolved
Another parameter such as DOC is essential considering its organic carbon trends resulting from changes in atmospheric
deposition chemistry. Nature 450, 537–541.
impact on quality of treated waters and has to be studied,
Teksoy, A., Alkan, U. & Baskaya, H. S. 2008 Influence of treatment
especially in SSWS. Further studies will also include other
process combinations on the formation of THM species in water.
water quality parameters like pesticides and other sites. Sep Purif Technol. 61, 447–454.
Tréguer, P., Lhuillery, M. & Viard, F. 2009 Changement climatique
et impacts sur les écosystèmes marins de l’ouest de la France.
ACKNOWLEDGMENTS Report of the Conseil scientifique de l’Environnement de
Bretagne, Rennes, France.
The authors sincerely thanks the Agence Régionale de Santé UNECE/WHO, 2009, http://www.unece.org/env/water/meetings/
(ARS) and the Ministère de l’Ecologie et du Développement wgwh/Secondmeeting_2009/ece_mp_wh_wg_1_2009_9.pdf,
accessed 02 July 2010.
Durable (MEDD) for the data they provided.
Van Vliet, M. T. H. & Zwolsman, J. J. G. 2008 Impact of summer
This work has been carried out in the frame of an ERA-Net droughts on the water quality of the Meuse river. J Hydrol. 353,
Environment and Health contract funded by the European 1–17.
commission and the French agency for environmental and Zwolsman, J. J. G. & van Bokhoven, A. J. 2007 Impact of summer
occupational health (AFSSET), the authors wish to sincerely droughts on water quality of the Rhine River - a preview of climate
change? Water Sci Technol. 56, 44–55.
thanks.
Association of iron oligomeric species with
natural organic matter: a combined EELS
and Mössbauer investigation
A.-V. Jung, V. Chanudet, B. S. Lartiges,

J. Ghanbaja, M. Abdelmoula & J.-
L. Bersillon
Aquatic Sciences
Research Across Boundaries
ISSN 1015-1621
Volume 74
Number 4
Aquat Sci (2012) 74:769-779

DOI 10.1007/s00027-012-0260-9
1 23
Your article is protected by copyright and all
rights are held exclusively by Springer Basel
AG. This e-offprint is for personal use only
and shall not be self-archived in electronic
repositories. If you wish to self-archive your
work, please use the accepted author’s
version for posting to your own website or
your institution’s repository. You may further
deposit the accepted author’s version on
a funder’s repository at a funder’s request,
provided it is not made publicly available until
12 months after publication.
1 23
Author's personal copy
Aquat Sci (2012) 74:769–779
DOI 10.1007/s00027-012-0260-9 Aquatic Sciences
RESEARCH ARTICLE
Association of iron oligomeric species with natural organic

matter: a combined EELS and Mössbauer investigation
A.-V. Jung • V. Chanudet • B. S. Lartiges •
J. Ghanbaja • M. Abdelmoula • J.-L. Bersillon
Received: 15 November 2011 / Accepted: 26 March 2012 / Published online: 27 April 2012
Ó Springer Basel AG 2012
Abstract The nature of the interaction between iron oxy- literature. The areas beneath the peaks (Fe L3 edge) and the
hydroxide compounds and natural organic matter (NOM) peak shapes of EELS spectra differ according to the origin
may take various forms and is still a matter of debate. It is an of the organic matter, suggesting that various types of Fe
important field to understand, especially for water treatment populations can be distinguished using the EELS technique.
applications and for the knowledge of iron transport in the Combining the selectivity of both Mössbauer spectroscopy
environment. The nature of association reached between iron for identifying trace, poorly crystalline Fe solids, oxidation
oligomeric species and NOM is here investigated using state and of EELS for being able to characterize the popu-
Mössbauer spectroscopy and electron energy-loss spectros- lation of Fe based on L-edge spectra, appears promising for
copy (EELS) at the Fe-L3 edge. Raw water NOM taken from characterizing Fe in systems containing NOM.
Moselle River (France), natural humic substances extracted
from the riverine suspended matter, and a synthesized Keywords EELS Fe-L3 edge Natural organic matter
humic-like substance, are coagulated with iron nitrate Humic colloids Mössbauer
according to a jar-test procedure. The results from Möss-
bauer spectroscopy indicate that Fe is present in an Abbreviations
octahedral coordination environment, which is consistent CS Center shift
with prior X-ray absorption spectroscopy reported in the EELS Electron energy-loss spectroscopy
ELNES Electron energy-loss near edge structure
A.-V. Jung (&) J. Ghanbaja

School of Environmental Engineering (EME), Service Commun de Microscopie Electronique par
Campus de Ker Lann, Avenue Robert Schumann, Transmission, University of Nancy, BP 239,
Bruz, France 54506 Vandœuvre-lès-Nancy, France
e-mail: av2jung@live.fr
M. Abdelmoula
A.-V. Jung Laboratoire de Chimie-Physique et Microbiologie de
Advanced School of Public Health (EHESP), LERES, l’Environnement (LCPME), UMR 7564, CNRS-Nancy
Avenue du Professeur Léon-Bernard-CS 74312, Université, 54 600 Villers-lès-Nancy, France
Rennes Cedex, France
J.-L. Bersillon
A.-V. Jung Laboratoire Environnement et Minéralurgie (LEM),
INSERM U1085, LERES, Rennes Cedex, France CNRS-INPL-ENSG UMR 7569, University of Nancy,
15 avenue du Charmois, BP 40, 54501 Vandœuvre-lès-Nancy,
V. Chanudet France
EDF-CIH Savoie Technolac, 73373 Le Bourget du Lac, France
B. S. Lartiges
Geosciences Environment Toulouse, UMR UPS-CNRS-IRD
5563, University of Toulouse (Paul Sabatier),
14 Avenue E. Belin, 31400 Toulouse, France
123
770 A.-V. Jung et al.
EFTEM Energy-filtered transmission electron groups attached to NOM aromatic moieties control the
microscopy (EFTEM) interactions with a-FeOOH goethite particles (Klaiser
EXAFS Extended X-ray absorption fine structure 2003), whereas Fourier transformed infra-red spectroscopy
FWHM Full with at half maximum (FTIR) emphasized the importance of NOM carboxylic
H Magnetic hyperfine field groups in the formation of stable complexes with Fe3?
FTIR Fourier transformed infra-red spectroscopy (Piccolo and Stevenson 1982; Senesi 1992; Gu et al. 1994).
NHS Natural humic substances (extracted from X-ray absorption spectroscopy (XAS) showed that poorly
suspended matters) polymerized iron species—monodentate and bidentate ferric
NOM Natural organic matter ions complexes, dimers and double corner trimers—are pre-
NMR Nuclear magnetic resonance spectroscopy dominantly identified in the presence of NOM colloids (Rose
OCC Optimum coagulation Concentration et al. 1998; Vilgé-Ritter et al. 1999; Olivie-Lauquet et al.
Py/GC–MS Pyrolysis gas chromatography–mass 2000; Senesi et al. 1977; Karlsson et al. 2008).
spectroscopy Electron paramagnetic resonance spectroscopy (EPR)
QS Quadrupole splitting has also been widely used to characterize complexes
SAXS Small-angle X-ray spectroscopy formed by NOM and iron. Applied to the suspended
SHS Synthesized humic-like substance material transported in tropical ecosystems, this technique
w Half width at half maximum of the confirmed that the iron species occurs mainly as FeIII both
Mössbauer peaks in oxyhydroxides and in complexes associated with NOM
XAS X-ray absorption spectroscopy (Olivie-Lauquet et al. 1999).
57
XRD X-ray diffraction Fe Mössbauer spectroscopy ideally complements the
other physico-chemical techniques since it determines
the iron oxidation states with high accuracy, it quantifies
the distribution of iron between the particles, and it pro-
Introduction vides information about crystallinity and particle size. It
has been shown particularly useful in systems where iron
Complexation of natural organic matter (NOM) with iron oxides may be either too low in concentration or poorly
oxyhydroxide mineral species plays a major role in the crystallised to be detected by X-ray diffraction (XRD), for
geocycling of carbon and the transport of metallic trace instance in soils, sediments, rust, natural water, and in
elements in terrestrial and aquatic environments (Salomons living organisms (e.g., Lakatos et al. 1977; Goodman 1988;
and Forstner 1984; Olivie-Lauquet et al. 2000; Pokrovsky Chevrier and Mathe 2007; Bender Koch et al. 2009; Ona-
and Schott 2002; Lyvén et al. 2003; Hassellöv and von der Nguema et al. 2009; Oshtrakh et al. 2009). EELS has
Krammer 2008). It is also of significant interest to the water recently been used to describe the carbon speciation in
treatment field, where an efficient removal of NOM by NOM and HS, to establish a fingerprint database at the
coagulation using iron hydrolyzed species remains a crucial nanoscale (Jung et al. 2005a). Such spectroscopy is also
issue (Vilgé-Ritter et al. 1999; Jung et al. 2005a, b; Sielechi well-adapted for studying the oxidation state of 3d transi-
et al. 2008). Most often, humic substances (HS) that account tion metals using L-edge spectra (Leapman et al. 1982;
for 30–50 % of aquatic NOM, are used to restrict the working Krivanek and Paterson 1990; van Aken et al. 1998; van
definition of NOM (Zumstein and Buffle 1989).The nature of Aken and Liebscher 2002; Schmid and Mader 2006; Ikeno
the interaction between iron oxyhydroxide compounds and et al. 2006). For instance, EELS enables the quantification
NOM may take various forms from a simple precipitation of iron valence in clay minerals (van Aken et al. 1998), iron
through complexation (e.g. Gu et al. 1994), to a drastic oxides phases such as FeO, Fe2O3 and Fe3O4, being used as
modification of the oxido-reduction state of environmental references to develop a deconvolution fitting procedure
nanocolloids (Lovley et al. 1996). In the latter case, NOM (Guillaume et al. 2001; Guillaume 2002). In Jung et al.
functional moieties specifically involved in iron reduction (2005a), thanks to a C-EELS identification method, we
are still a matter of debate (Chen et al. 2003). showed that iron-hydrolyzed species were associated with
Numerous techniques have been used to characterize the carboxylic groups of humic colloids at both pH 6 and 8
interaction of NOM with iron oxyhydroxide in natural during the coagulation stage of water treatment. C-EELS
waters. Adsorption isotherms of dissolved organic matter proved to be a very interesting technique to obtain nano-
onto the surface of various iron mineral phases revealed metric information on chemical organic bounds engaged
that NOM carboxylic groups are partially complexed with with the coagulant (iron salt). In this paper, following this
the hydroxyl groups of the mineral surface (e.g. Wu et al. previous work applied to NOM (Jung et al. 2005a), we
2008). Nuclear magnetic resonance (NMR) provided evi- focus on iron status. As mentioned above, the nature of the
dence that both the number and the position of acidic interaction of NOM with iron oxyhydroxide in natural
123
Association of iron oligomeric species with NOM 771
waters is still a matter of debate, therefore, we investigate Preparation of organic matter/iron oxyhydroxide
at the nanoscale the association between iron oligomeric precipitates
species and NOM combining Fe-EELS with Mössbauer
spectroscopy. Various synthetic waters were obtained by dissolving either
synthetic or natural humic material in Milli-Q ultra-pure water
(30 mg L21). NaHCO3 (4.10-3 mol L-1) (Sigma-Aldrich,
Materials and methods analytical grade) was also added to the suspension to provide a
carbonate alkalinity similar to that of european natural river
Humic substances waters (336 mg L-1) (GEMS 2011). The pH of synthetic
waters was adjusted to 8 with dropwise addition of 0.1 M HCl.
Natural humic acids (NHS) were extracted from suspended The surface water taken from the Moselle River, was used as
matter collected from the Moselle River. The sampling site collected (natural pH 7.4). It is named ‘‘raw water river NOM’’
was located at Richardmenil (France) in the vicinity of the in the following. The aggregation experiments between
pumping facility supplying the water treatment plant of organic matter and iron hydrolyzed species were conducted in
Greater Nancy Urban Community in surface water. 1 L baffled reactors of known power dissipation characteris-
Approximately 12 m3 of river water were pumped tics. The stirring rate was fixed at 100 rpm which corresponds
(1 m3 h-1) into a flow-through centrifuge operating at to a mean velocity gradient of 135 s-1. For all organic matter
17,000 rpm. The humic acids contained in the centrifuged samples, various iron (0.1 mol L-1 ferric nitrate, Fe(NO3)3,
solid suspended matter (supernatant not used) were then 9H2O, Sigma-Aldrich) concentrations were added under
extracted according to the IHSS standard procedure (Inter- agitation, and mixing was continued for 20 min. The sus-
national Humic Substances Society) (Thurman and Malcolm pension pH was not readjusted after coagulant addition. The
1981), adapted to yield the humic material in a sodium form. iron oxyhydroxide reference was obtained from a
Grab samples of the Moselle River water (9.3 mg L-1 total 5 9 10-4 mol L-1 iron solution hydrolyzed at pH 8 in the
organic carbon content) were also collected at the same absence of humic substances. For all samples, after 24 h of
location, thus providing raw water with unfractionated quiescent settling in graduated Imhoff cones, the residual
NOM. The synthetic humic-like substances (SHS) were turbidity of supernatants was measured with a HACH Ratio
prepared by oxidation of catechol with glycine (Andreux XR turbidimeter. All sediments were then collected and
1978; Andreux et al. 1980). Catechol (6.6 g) was dissolved in freeze-dried (Benchtop 3.3 EL105 Sentry).
500 mL of sodium phosphate buffer solution (NaH2PO4/
Na2HPO4) at pH 8 to yield a 0.03 mol L-1 catechol solution. Experimental techniques
An equal volume of buffered glycine 0.03 mol L-1 solution
was then added and the resulting mixture was allowed to Mössbauer spectra were recorded on powdered samples
react for 5 days at 25 °C in the dark and under a pure oxygen with a constant acceleration spectrometer using a 57Co
constant pressure (0.9 bar). After dialysis (Spectra/PorÒ 7 source embedded in a Rh matrix. The system was sealed in
membrane-8 kDa molecular weight cut-off; Roth-Sochiel a cryogenic workstation (Cryo Industries of AmericaÒ)
Lauterbourg, France), the synthetic material in a sodium under N2 gas to minimize re-oxidation of reduced samples.
form was freeze-dried and stored at 4 °C in the dark. Spectra were obtained at 77 and 298 K and analysed by
Extensive characterizations of SHS and relevance to v2 minimization using Lorentzian-shape lines with variable
natural humic substances have been discussed in previous widths for taking into account spectral asymmetry (Benali
publications (Jung et al. 2005b; Kazpard et al. 2006). et al. 2001). Metallic a-Fe was used for calibration.
Elemental analysis, FTIR and pyrolysis gas chromatogra- EELS investigations were carried out at 298 K using a
phy–mass spectroscopy (Py/GC–MS) investigations 666 GATAN parallel electron energy-loss spectrometer
revealed that synthetic and natural organic materials have adapted to a Phillips CM20 transmission electron micro-
similar molecular moieties. Potentiometric titration showed scope operated at 200 kV with an unsaturated LaB6
that the synthetic polymers feature the two main carboxylic cathode. A spectrometer entrance aperture of 2 mm, a
and phenolic acidities even though phenolic moieties collection half angle of 10 mrad, and an energy dispersion
contribute to a higher proportion in the total acidity for of 0.1 eV per channel were selected as experimental con-
natural materials. Moreover, the hydrophilic to hydropho- ditions. The energy resolution measured at the full width at
bic balance ratio was found lower for synthetic molecules half maximum (FWHM) of the zero-loss peak was 1.2 eV.
than for natural ones. The aromatic parts of the synthetic The freeze-dried powders were resuspended in ethanol
molecules (models) possess a higher polymerisation degree under ultrasonication, and a drop of suspension was
(higher aromatic contribution) than the natural compounds evaporated on a carbon-coated copper grid (Euromedex,
isolated in this work. Mundolsheim, France). For each iron concentration
123
investigated, 15 spectra on distinct iron–NOM aggregates treatments, the standard error for the peak area determi-
were acquired. nation was \1 %.
The spectra were analysed by means of Gatan EL/P 2.1
PEELS software. The low energy-loss spectrum was used
to remove the multiple-inelastic-scattering effect in the Results and discussion
core-loss region using a Fourier-ratio deconvolution
method (Egerton 1996). Each spectrum was recorded by The effect of iron (III) concentration on the residual turbidity
fixing the energy of the Fe-L3 edge maximum at 711.0 eV after settling of various organic matter suspensions investi-
(Mansot et al. 1994). The background intensity, extrapo- gated is presented in Fig. 2. All the turbidity curves display a
lated from the pre-edge region by an inverse power law, classical behaviour with an increase in turbidity at low iron
was subtracted, and the Fe-L3 edge was then fitted with concentrations that characterizes the formation of small
three bands, i.e. L3a, L3b and L3c, following a method aggregates, followed by a decrease in turbidity at higher
developed by Guillaume (2002) to investigate the local dosages that corresponds to the formation of larger settleable
chemistry of iron in clay minerals. According to that aggregates. For a given suspension, the point where the
study, the Fe-L3 edge shape was best fitted with a mixture extrapolated steep portion of the turbidity curve intersects the
of Gaussian (65 %) and Lorentzian (35 %) contributions. x-axis defines the optimum aggregation concentration (OCC),
L3a and L3b were then assigned to FeII and FeIII, respec- which is indicated with an arrow on the graph. At higher iron
tively. L3c was not assigned even though this band was concentrations, the residual turbidity remains low, except for
systematically observed at the same energy and with the NHS suspension where restabilization is induced.
almost the same relative abundance on the investigated Three iron concentrations: (1) below OCC, (2) at OCC,
iron oxides and clay samples (Guillaume 2002). The and (3) above OCC, were selected for EELS and Möss-
presence of this characteristic band was also evidenced by bauer examination (respectively for SHS and NHS, at pH
Leapman et al. (1982) but again not clearly assigned. The 8: 8.3 9 10-4 mol L-1 and 6.4 9 10-4 mol L-1; for
deconvolution/fitting method is illustrated in Fig. 1 in the NOM, at pH 7.4: 1.7 9 10-4 mol L-1). However, the
case of hydrolyzed iron in the absence of organic matter. amount of material recovered below OCC for raw Moselle
Fitting of the Fe-L3 band and calculation of peak areas, River water and for SHS was not sufficient to perform the
i.e. A3a, A3b, and A3c for L3a, L3b and L3c, respectively, Mössbauer experiments.
were performed in the 710.0–725.0 eV range using Opus As illustrated in Fig. 3, the Mössbauer spectra recorded
softwareÒ from Bruker (Billerica, MA, USA). A Leven- at 298 K for coagulated NHS present two quadrupole ferric
berg–Marquardt algorithm was applied, the band position, doublets (C1 and C2). At this temperature, all Mössbauer
shape (35 % Lorentzian–65 % Gaussian), and intensity, spectra are similar for all organic matter types and iron
being parameters fixed according to Guillaume et al. concentrations. Mössbauer parameters, C1 and C2 values,
(2001, 2003) and Guillaume (2002). For all deconvolution center shift (CS), quadrupole splitting (QS), and half width
of the corresponding peaks (w), are listed in Table 1. The
Mössbauer spectra can be fitted with two doublets C1 and
Experimental
ΣL3a , L3b, L3c
band fitting
decomposition
Intensity (cps)
Max L3b : 711.61 eV

A 3b : 56%
Max L3a : 710.10 eV

A3a : 41% Max L3c : 714.23 eV
A 3c : 3%
700 705 710 715 720
Energy loss (eV)

Fig. 2 Evolution of residual turbidity in NTU with coagulant
Fig. 1 Three-band fitting (65 % Gaussian–35 % Lorentzian) decom- concentration for NHS and SHS at pH 8 and NOM at natural pH
position on Fe L3 coagulant (hydrolysed ferric nitrate) electron- 7.4 (optimum coagulant concentration, OCC) mentioned by arrows
energy loss spectrum for each organic material)
123
100 is also in accordance with previous XAS results suggesting

Transmittance (%) that iron oligomeric species associated with natural organic
98
matter present the same oxidation state and coordination as
96 a-FeOOH goethite (Vilgé-Ritter et al. 1999). The presence
94 of the sextets at low temperature is typical for magnetically
ordered species (Hawthorne 1988). Their magnetic hyperfine
92
field value (H) is relatively low, which would indicate the
90 presence of poorly crystallized species (Goodman and
88 Cheshire 1979; Goodman et al. 1991). Such finding is con-
-3 -2 -1 0 1 2 3
sistent with previous EXAFS (extended X-ray absorption fine
-1
Velocity (mm.s ) structure) results showing that iron (III) hydrolyzed in the
presence of NOM is poorly polymerized and mainly found as
single-corner sharing trimers (Vilgé-Ritter et al. 1999).
100
In the case of hydrolyzed iron, no change in the Mössbauer
Transmittance (%)
98
parameters was observed between 298 and 77 K. The spec-
96 trum shape was similar to organic matter coagulated with iron
94 (Fig. 3). This suggests the existence of only one valence of
92 iron species in the precipitate since Mössbauer spectroscopy is
90
very sensitive to this parameter. The presence of two kinds of
NHS_2 doublets could indicate the existence of two populations for
88
iron (III) concerning quadrupole splitting distribution.
-3 -2 -1 0 1 2 3
-1
Figure 5 shows the EELS spectra obtained in the
Velocity (mm.s ) 700–725 eV range (Fe-L3 peak) for NHS, SHS, and NOM
aggregated with various concentrations of iron hydrolyzed
100 species. Peak areas are given in Table 1. The positions of
the maximum of L3a, L3b and L3c remain constant at about
98
Transmittance (%)
710, 711.5, and 714.0 ± 0.1 eV, respectively, regardless of

96
the nature of organic matter. The presence of three main
94 deconvolution peaks for Fe-L3 edge was also used by
92 Calvert et al. (2005) to quantify the iron valence ratio
P
90 (Fe3?/Fe2? and Fe3?/ Fe) in various organic and inor-
88
NHS_3 ganic materials. The standard error for peak position is
-3 -2 -1 0 1 2 3 lower than the resolution of the electron microscope. The
Velocity (mm.s ) -1 relative peak areas are significantly different, depending on
the origin of organic matter. For SHS, A3a is observed to
Fig. 3 Mössbauer spectra (298 K) of natural humic substances increase with iron concentration, whereas it decreases for
coagulated with three critical for water treatment iron concentrations NOM. In the case of NHS, no obvious trend can be rec-
for water treatment (1 below OCC, 2 at OCC, 3 above OCC)
ognized from the three iron concentrations investigated.
It should be noted that previous C-EELS and N-EELS
C2 which differ noticeably in their quadrupole splittings results revealed that the nature of the river NOM removed
(1.04 and 0.59 mm s-1), whereas the center shifts is by application of iron-hydrolyzed species varies as a
identical for the both components. This procedure used to function of iron concentration, HS being preferentially
fit broadened Mössbauer lines suggests the presence of a eliminated at low iron concentrations (Jung et al. 2005a).
slowly relaxing magnetic component. The existence of an Interestingly, the relative proportions of A3a and
additional component is confirmed by spectra recorded at (A3b ? A3c) obtained from fitting the EELS spectra of iron
77 K in the velocity range ±11 mm s-1; all three samples hydrolyzed species, SHS, and raw water NOM, seem to be
show magnetic sextets in addition to the paramagnetic related to the corresponding C1 and C2 Mössbauer areas at
doublets (Fig. 4). The hyperfine parameters for natural 298 K (Table 2). At 77 K, C1 seems to be over-estimated
organic matter examined at 77 K are listed in Table 1d. due to the contribution of magnetically ordered species
The typical values of CS and QS doublets at both tem- (sextets) which are not detected at 298 K. Garvie and
peratures (298 and 77 K) are consistent with paramagnetic Buseck (1998), Guillaume et al. (2001), Guillaume (2002),
parameters usually found for iron (III) in an octahedral Calvert et al. (2005) and more recently Pan et al. (2010)
coordination environment (Hawthorne 1988). Such finding focused on the shape of the Fe-L3 edge to display a
123
Table 1 EELS and Mössbauer parameters

Iron concentrationa CS QS w C1 C2 A3a A3b A3c A3b ? A3c
(a) Synthetic humic-like substance (SHS) at 298 K

1 –b –b –b –b –b 37.0 58.0 5.0 63.0
2 0.47 0.557 0.174 41.0 44.0 50.0 6.0 56.0
0.46 0.978 0.214 59.0
3 0.46 0.605 0.196 59.0 48.0 35.0 17.0 52.0
0.46 1.049 0.193 41.0
(b) Raw river water (NOM) at 298 K
1 –b –b –b –b –b 60.0 34.0 6.0 40.0
2 0.46 0.998 0.201 54.0 57.0 39.0 4.0 43.0
0.46 0.595 0.171 46.0
3 0.46 1.046 0.196 48.0 52.0 40.0 8.0 48.0
0.46 0.608 0.178 52.0
(c) Natural humic substance (NHS) at 298 K
1 0.47 1.039 0.317 29.0 32.0 60.0 8.0 68.0
0.46 0.618 0.372 71.0
2 0.47 1.026 0.356 39.0 24.0 63.0 13.0 76.0
0.47 0.587 0.362 61.0
3 0.47 1.031 0.353 38.0 31.0 59.0 10.0 69.0
0.47 0.598 0.356 62.0
Iron concentrationa CS QS w C1 C2 S H e
(d) Natural humic substance (NHS) at 77 K

1 0.4653 1.109 0.209 30.6
0.4593 0.634 0.235 63.0
0.5600 0.640 6.3 444 -0.14
2 0.4604 1.190 0.274 22.8
0.4551 0.685 0.725 65.6
0.5090 11.6 448 -0.11
3 0.4659 1.129 0.239 31.0
0.4596 0.647 0.261 61.7
0.5100 0.350 7.3 440 -0.05
a
1, 2, 3: Iron concentrations before, at and after the optimum concentration coagulation (Jung et al. 2005a)
b
Non-calculated parameters
CS center shift in mm s-1, QS quadrupole splitting in mm s-1, w half width at half maximum of the Mössbauer peaks in mm s-1, A3a, A3b, A3c
areas (expressed in % total area) issued from Fe-L3 three band (L3a, L3b, L3c) peak-fitting EELS decomposition according to Guillaume (2002), e
quadrupole shift in mm s-1, H magnetic hyperfine field in kOe, C1, C2 % areas from Mössbauer doublets decomposition according to Benali
et al. (2001), S areas from Mössbauer sextets decomposition with Lorentzian-shape lines according to Benali et al. (2001)
correlation between this shape and the valence of iron. possibility of sample degradation under beam irradiation
They modelled the evolution of the Fe-L3 edge energy as a has already been discussed in several studies (e.g. Garvie
function of this shape to quantify the chemical shift of the et al. 2004; Pan et al. 2010).
energy at maximum intensity of the Fe-L3 edge, as a When comparing EELS and ELNES results (Pan et al.
function of the valence of iron. Guillaume et al. (2003) 2010), changes in composition and oxidation state of
compared Fe3?/Fetotal calculated by transmission Möss- minerals with progressive irradiation have been evidenced.
bauer spectroscopy and by EELS using such deconvolution The electron beam had a significant effect, causing a
for montmorillonite in presence of metallic iron and found change of Fe3? co-ordination from octahedral to tetrahe-
a good correlation. For the highest Fe3?/Fetotal ratio, the dral. It also progressively reduced Fe3? to Fe2? as the
slight deviation observed has been explained by sample accumulated electron fluence increased. However, for low-
preparation or sample reduction under the beam. This K dielectric material samples, no evidence for chemical
123
The evolution of iron oxidation state in natural media is

7200000
still a matter of debate. Several parameters can interfere
7100000
upon reduction of Fe(III) to Fe(II) such as the presence of
7000000
oxygen depleted redox zones within sediments, or the
6900000
presence of microorganisms acting as electron acceptors.
Counts
6800000
The role of mineral phases allowing iron chelation by
6700000
organic matter has been discussed too (Hakala et al. 2009).
6600000
However, very few results exist on the influence on NOM
6500000 1 on reduction process. For instance, the humic fraction of
6400000
NOM in the absence of iron determines the reduction of
-15 -10 -5 0 5 10 15
organic molecules, but the Fe(II)–NOM complex is gen-
Velocity (mm.s-1)
erally viewed as a stronger reductant than either NOM or
11000000 Fe(II) alone (Hakala et al. 2007). Others authors revealed
that the effects of iron on NOM molecular weight, absor-
bance, and fluorescence properties depend upon the redox
10500000
state of iron and the composition of NOM (Pullin et al.
2007). Our results show that the aromatic content of humic
Counts
10000000 substances is not an impact parameter which will induce a

change in iron valence during coagulation. This result is
2
NHSLN_2 not contradictory with previous works (Murphy et al. 1990;
9500000
Bose and Reckhow 1998; Meier et al. 1999; Zhou et al.
2001) especially based on Freundlich and Langmuir or
-10 0 10
non-Langmuir adsorption isotherms on NOM and high
Velocity (mm.s-1)
pressure size exclusion chromatography results which
show that the more hydrophobic NOM parts are the more
6000000
abundant to be eliminated by coagulation.
Mössbauer spectroscopy is particularly sensitive to the
5800000 valence state of iron. Previous studies using Mössbauer
spectroscopy to examine the status of iron in the presence of
Counts
5600000 humic materials suggested various binding mechanisms, e.g.

iron (III) weakly adsorbed on humic colloids, iron strongly
5400000 bound and protected by tetrahedral and/or octahedral coor-
3
dination in NOM complexes (Senesi et al. 1977; Hansen and
5200000 Mosbaek 1970; Lakatos et al. 1977). The results presented
-10 0 10
here prove that the iron status does not evolve whatever the
Velocity (mm.s-1)
content in aromatic moieties. Concerning carboxylic moie-
Fig. 4 Mössbauer spectra (77 K) of natural humic substances ties, we know that these groups contribute to the coagulation
coagulated with three critical for water treatment iron concentrations mechanism (Gu et al. 1994; Jung et al. 2005a). The two iron
for water treatment (1 below OCC, 2 at OCC, 3 above OCC)
populations of iron (III) here observed could perhaps be
linked to (1) the existence of different kind of carboxylic
bound alteration by the TEM sample preparation process moieties (Hay and Myeni 2007; Deshmukh et al. 2007) or to
and analysis was found (Stegmann et al. 2008). In our study, (2) different polymerisation status of iron (III) or still to (3) a
we only observed one type of iron oxidation state, poorly reorganization of humic molecules from a stretched to a
crystallised or in an octahedral coordination environment coiled conformation upon interaction with coagulant species
(regarding Mössbauer results at 77 K). Other studies (Namjesnik-Dejanovic and Maurice 2000; Jung et al. 2005a;
(McCammon et al. 2004; Calvert et al. 2005) proved the Sielechi et al. 2008). In the case of (2) scenario, the two iron
method robust with an error in the range of\10 % on Fe3?/ populations observed here are consistent with previous
P
Fe ratio for samples with[2 atom% Fe, which is the case reports identifying two types of iron species associated with
of our samples (Jung et al. 2005a). We can also conclude NOM, i.e. mononuclear FeIII–NOM complex and polymeric
that our samples are relatively stable under the TEM beam FeIII (hydr)oxides (Karlsson et al. 2008; van Schaik et al.
and that the evolution of iron status during the experiments 2008; Karlsson and Persson 2010). In particular, it has been
is negligible. Therefore, in our study, L3a and L3b cannot be shown that the iron (III) oxy(hydr)oxides significantly con-
assigned to FeII and FeIII, respectively. tribute (more than 30 %) to the iron transport in tropical
123
Table 2 EELS and Mössbauer parameters for coagulant at 298 and 77 K

Temperature (K) CS QS w C1 C2 A3a A3b A3c A3b ? A3c
298 0.46 1.002 0.349 37.0 41.0 56.0 3.0 59.0

0.46 0.57 0.29 63.0
77 0.45 1.095 0.22 35.0 –a –a –a –a
0.45 0.582 0.25 65.0
a
Non-calculated parameters
Fig. 5 298 K Fe L3 EELS peak Natural humic extract Synthetic humic substance Raw-river NOM
fitting for organic matters (1)
before (2) at and (3) after iron
Normalized intensity (cps)

1
1 1

optimum coagulation
concentration 711.38
711.49
710.33
711.69
709.92 709.82
714.41 714.08
714.31
700 705 710 715 720 725 700 705 710 715 720 725 700 705 710 715 720 725
Energy loss (eV) Energy loss (eV) Energy loss (eV)


2 2 2
711.36 711.49
710.23
711.79
709.94
709.88
714.16 714.1
713.93 8
700 705 710 715 720 725 700 705 710 715 720 725 700 705 710 715 720 725

3 3
3
711.53
710.29
711.58 710.41
709.92 711.5
713.77
714.08
713.85
700 705 710 715 720 725 700 705 710 715 720 725 700 705 710 715 720 725
coloured waters (Olivie-Lauquet et al. 1999). In the case of hypotheses are to be considered with care since functional
(3) scenario, data from Deshmukh et al. (2007) indicate an group chemistry and molecular structure are two properties
important contribution of electron-withdrawing substituents interrelated and one influences the other (Myeni 2002). A
(CO2H, COR, COOR, CHO, and phenyl groups on the a–b– future quantitative use of the Mössbauer results could per-
c–C atoms) for carboxyl groups of natural soil humic sub- haps give more precise conclusions on this point considering
stances. These groups improve the ability of the substances the different proportion of these two acidities in both natural
to bind metals through chelation processes. All these and synthetic molecules.
123
Conclusion References
This study shows that Mössbauer spectroscopy combined Andreux F (1978) Etude des étapes initiales de la stabilisation
physico-chimique et biologique d’acides humiques modèles.
with EELS at the Fe-L3 edge provide a better insight into PhD thesis Université de Nancy I
the status of iron associated with natural organic matter. Andreux F, Golebiowska D, Metche M (1980) Oxydative polymer-
Contrary to Mössbauer spectroscopy, EELS could only be ization of O-diphenols in presence or absence of amino-acids.
used to relatively quickly identified iron valence. More- Topics on (catechol-glycocolle) and (catechol-diglycylglycine).
General Assembly of Polyphenols Group Report, Logrono
over, we showed that EELS could provide information (Spain), Bulletin 9:178–188
about the iron types engaged in various bonds with NOM. Benali O, Abdelmoula M, Refait P, Génin JM (2001) Effect of
Mössbauer spectroscopy however remains essential to orthophosphate on the oxidation products of Fe(II)-Fe(III)
determine the iron oxidation state with high accuracy and hydroxycarbonate: the transformation of green rust to ferrihy-
drite. Geochem Cosmochem Acta 65:1715–1726
to give an insight into crystallinity. Bender Koch C, Rasmussen HK, Mørup S (2009) Iron speciation in
Progressing on from the EEL fingerprint studies of natural hyperacid water investigated by Mössbauer spectroscopy.
NOM coagulated with iron, we demonstrate a method for Spectrochim Acta Part A 74:928–930
the qualification of iron species using the observed Fe-L3 Bose P, Reckhow DA (1998) Adsorption of natural organic matter on
preformed aluminium hydroxide flocs. J Environ Eng 803–811
EEL spectra, with no needs for measurement of standards. Calvert CC, Brown A, Brydson R (2005) Determination of the local
This spectroscopy can also be used as a first step—neither chemistry of iron in inorganic and organic materials. J Electron
time nor sample amount-consuming technique—to inves- Spectrosc Relat Phenom 143:173–187
tigate the reactivity of iron towards organic colloids at the Chen J, Gu B, Royer RA, Burgos WD (2003) The roles of natural
organic matter in chemical and microbial reduction of ferric iron.
nanoscale. In particular, the high-resolution of the method Sci Total Environ 307:167–178
(\100 nm) allows the characterization of natural matrices Chevrier V, Mathe PE (2007) Mineralogy and evolution of the surface
heterogeneity. The main advantage of the technique of Mars: a review. Planet Space Sci 55:289–314
remains the possibility of coupling Fe-L and C-K edges Deshmukh AP, Pacheco C, Hay MB, Myeni SCB (2007) Structural
environments of carboxyl groups in natural organic molecules
electron-energy loss spectroscopies to provide ultrastruc- from terrestrial systems. Part 2: 2D NMR spectroscopy.
tural scale overviews of iron–OM aggregates. Therefore, in Geochim Cosmochim Acta 71:3533–3544
the case of iron–NOM complexes, Fe-EELS should only be Egerton RF (1996) Electron energy-loss spectroscopy in the electron
used to discriminate iron ‘‘types’’ or ‘‘populations’’. microscope, 2nd edn. Plenum Press, New York
Garvie L, Buseck P (1998) Ratios of ferrous and ferric iron
Mössbauer results show that only one valence can be nanometer-sized areas in minerals. Nature 396:667–670
evidenced for iron in natural and synthetical substances Garvie LAJ, Zega TJ, Rez P P, Buseck R (2004) Nanometer-scale
coagulated with this metal. Iron is found to be poorly measurements of Fe3?/ Fe by electron energy-loss spectros-
crystallised or in an octahedral coordination environment. copy: a cautionary note. Am Mineral 89:1610–1616
GEMS—Global Environment Monitoring System (2011) http://www.
The iron status is independent of the aromatic content of gemswater.org/atlas-gwq/acid-e.html#2. Last consulted 13 April
organic molecules. EELS suggests the presence of two iron 2011
populations that could either be linked to the functional Goodman BA (1988) The characterization of iron complexes with soil
chemistry of organic molecules (presence of several car- organic matter. In: Stucki JW, Goodman BA, Schwertmann U
(eds) Iron in soils and clay minerals. Dordrecht, pp 677–687
boxylic moieties) or to the molecular reorganisation of Goodman BA, Cheshire MV (1979) A Mössbauer-effect study of the
humic network upon coagulation, or still to different effect of pH on the reaction between iron and humic acid in
polymerisation status of iron (III). Further work is still aqueous media. J Soil Sci 30:85–91
needed to define the meaning of the two iron populations Goodman BA, Cheshier MV, Chadwick J (1991) Characterization of
the Fe(III)-fulvic acid reaction by Mössbauer spectroscopy.
found associated with natural organic matter. In-solution J Soil Sci 42:25–38
atomic-force microscopy (Namjesnik-Dejanovic and Gu B, Schmitt J, Chen Z, Lian L, McCarthy JF (1994) Adsorption and
Maurice 2000) and perhaps the recent development of desorption of natural organic matter on iron oxide: mechanisms
quantitative core-loss energy-filtered transmission electron and models. Environ Sci Technol 28:38–46
Guillaume D (2002) Etude expérimentale du système fer-smectite en
microscopy (EFTEM) could be helpful for such purposes présence de solution à 80°C et 300°C, PhD thesis Université
(Jin-Philipp et al. 2011). Indeed, the merit of the latter Henri Poincaré
technique lies in its chemical selectivity, providing 3D Guillaume D, Pironon J, Ghanbaja J (2001) Valence determination of
information, not only about the size and shape, but also on iron in clays by electron energy loss spectroscopy, Berichte des
Deutschen mineralogischen Gesellshaft, Beihefte zum European
the composition of a nano-object. Journal of Mineralogy 13
Guillaume D, Neaman A, Cathelineau M, Mosser-Ruck R, Peiffert C,
Acknowledgments This work was achieved within the framework Abdelmoula M, Dubessy J, Villiéras F, Baronnet A, Michau N
of ‘‘Zone Atelier du bassin de la Moselle’’, with financial support (2003) Experimental synthesis of chlorite from smectite at
from Region Lorraine and CNRS (Contrat de Plan Etat, Région and 300°C in the presence of metallic Fe. Clay Miner 38:281–302
Programme ‘‘Environnement Vie et Société’’).
123
Hakala JA, Chin Y-P, Weber AJ (2007) Influence of dissolved Mansot JL, Leone P, Euzen P, Palvadeau P (1994) Valence of
organic matter and Fe(II) on the abiotic reduction of pentachlo- manganese, in a new oxybromide compound determined by
ronitrobenzene. Env Sci Technol 41:7337–7342 means of electron energy loss spectroscopy. Microsc Microanal
Hakala JA, Fimmen RL, Chin Y-P, Agrawal SG, Ward CP (2009) Microstruct 5:79–90
Mössbauer studies of an iron (III) fulvic acid complex. Acta McCammon CA, Lauterbach S, Seifert F, Langenhorst F, van Aken
Chem Scand 8:3983–3984 PA (2004) Iron oxidation state in lower mantle mineral
Hansen GH, Mosbaek LH (1970) Mössbauer studies of an iron (iii): assemblages I. Empirical relations derived from high-pressure
fulvic acid complex. Acta Chem Scand 8:3083–3884 experiments. Earth Planet Sci Lett 222:435–449
Hassellöv M, von der Krammer F (2008) Iron oxides as geochemical Meier M, Namjesnik-Dejanovic K, Maurice PA, Chin Y-P, Aiken GR
nanovectors for metal transport in soil–river systems. Elements (1999) Fractionation of aquatic natural organic matter upon
4:401–406 sorption to goethite and kaolinite. Chem Geol 157:275–284
Hawthorne FC (1988) Spectroscopic methods in mineralogy and Murphy EM, Zachara JM, Smith SC (1990) The role of sorbed humic
geology. In: Ribbe PH (ed) Reviews of mineralogy. Mineralog- substances on the distribution of organic and inorganic contam-
ical Society of America, Blacksburg, pp 255–340 inants in groundwater. Environ Sci Technol 28:1507–1516
Hay MB, Myeni SCB (2007) Structural environments of carboxyl Myeni SCB (2002) Soft X-ray spectroscopy and spectromicroscopy
groups in natural organic molecules from terrestrial systems. Part studies of organic molecules in the environment. Rev Min
1: Infrared spectroscopy. Geochim Cosmochim Acta 71:3518– Geochem 49:485–579
3532 Namjesnik-Dejanovic K, Maurice PA (2000) Conformations and
Ikeno H, Mizoguchi T, Kayama Y, Kumagai Y, Tanaka I (2006) First- aggregate structures of sorbed natural organic matter on muscovite
principles multi-electron calculations for L2,3 ELNES/XANES and hematite. Geochim Cosmochim Acta 65:1047–1057
of 3d transition metal monoxides. Ultramicroscopy 106:970–975 Olivie-Lauquet G, Allard T, Benedetti M, Muller JP (1999) Chemical
Jin-Philipp NY, Koch CT, van Aken PA (2011) 3D elemental distribution of trivalent iron in riverine material from a tropical
mapping on nanomaterials using core-loss EFTEM tomography. ecosystem: a quantitative EPR study. Water Res 33:2726–2734
Ultramicroscopy 111:1255–1261 Olivie-Lauquet G, Allard T, Berteaux J, Muller JP (2000) Crystal
Jung AV, Chanudet V, Ghanbaja J, Lartiges BS, Bersillon JL (2005a) chemistry of suspended matter in a tropical hydrosystem, Nyong
Coagulation of humic substances and dissolved organic matter basin (Cameroon, Africa). Chem Geol 170:113–131
with a ferric salt: an electron energy loss spectroscopy inves- Ona-Nguema G, Morin G, Wang Y, Menguy N, Juillot F, Olivi L,
tigation. Water Res 39:3849–3862 Aquilanti G, Abdelmoula M, Ruby C, Bargar JR, Guyot F, Calas
Jung AV, Frochot C, Parant S, Lartiges BS, Selve C, Viriot ML, Bersillon G, Brown GE (2009) Arsenite sequestration at the surface of
JL (2005b) Synthesis of amino-phenolic humic-like substances and nano-Fe(OH)2, ferrous-carbonate hydroxide, and green-rust after
comparison with natural aquatic humic acids: a multi-analytical bioreduction of arsenic-sorbed lepidocrocite by Shewanella
techniques approach. Org Geochem 36:1252–1271 putrefaciens. Geochim Cosmochim Acta 73:1359–1381
Karlsson T, Persson P (2010) Coordination chemistry and hydrolysis Oshtrakh MI, Semionkin VA, Milder OB, Novikov EG (2009)
of Fe(III) in a peat humic acid studied by X-Ray absorption Mössbauer spectroscopy with high velocity resolution: new
spectroscopy. Geochim Cosmochim Acta 74:30–40 possibilities in biomedical research. J Mol Struct 924–926:20–26
Karlsson T, Persson P, Skyllberg U, Mörth CM, Giesler R (2008) Pan YH, Vaughan G, Brydson R, Bleloch A, Gass M, Sader K, Brown
Characterization of iron (III) in organic soils using extended A (2010) Electron-beam-induced reduction of Fe3? in iron
X-ray absorption fine structure spectroscopy. Env Sci Technol phosphate dehydrate, ferrihydrite, haemosiderin and ferritin as
42:5449–5454 revealed by electron energy-loss spectroscopy, 2010. Ultrami-
Kazpard V, Lartiges BS, Frochot C, De la Caillerie d’Espinose JB, croscopy 110:1020–1032
Viriot ML, Portal JM, Goerner T, Bersillon JL (2006) Fate of Piccolo A, Stevenson FJ (1982) Infrared spectra of Cu2?, Pb2?, and
coagulant species and conformational effects during the aggre- Ca2? complexes of soil humic substances. Geoderma 27:195–208
gation of a model of a humic substance with Al sub(13) Pokrovsky OS, Schott J (2002) Iron colloids/organic matter associated
polycations. Water Res 40:1965–1974 transport of major and trace elements in small boreal rivers and
Klaiser K (2003) Sorption of natural organic matter fractions to their estuaries (NW Russia). Chem Geol 190:141–179
goethite (a-FeOOH): effect of chemical composition as revealed Pullin MJ, Anthony C, Maurice PA (2007) Effects of iron on the
by liquid-state 13C NMR and wet-chemical analysis. Org molecular weight distribution, light absorption, and fluorescence
Geochem 34:1569–1579 properties of natural organic matter. Env Eng Sci 24:987–997
Krivanek OL, Paterson JH (1990) ELNES of 3d transition-metal Rose J, Vilgé A, Olivie-Lauquet G, Masion A, Frechou C, Bottero JY
oxides. I. Variations across the periodic table. Ultramicroscopy (1998) Iron speciation in natural organic matter colloids.
32:313–318 Colloids Surf A 136:11–19
Lakatos B, Korecz L, Meisel J (1977) Comparative studies on the Salomons W, Forstner U (1984) Metals in the hydrosphere. Springer,
Mössbauer parameters of iron humates and polyuronates. Heidelberg
Geoderma 19:149–157 Schmid HK, Mader W (2006) Oxidation states of Mn and Fe in
Leapman RD, Grunes LA, Fejes PL (1982) Study of the L23 edges various compound oxide systems. Micron 37:426–432
in the 3d transition metals and their oxides by electron-energy- Senesi N (1992) Metal-humic substance complexes in the environ-
loss spectroscopy with comparisons to theory. Phys Rev B ment. Molecular and mechanistic aspects by multiple
26:614–635 spectroscopic approach. In: Adriano DC (ed) Biogeochemistry
Lovley DR, Coates JD, Blunt-Harris EL, Hayes LA, Philips EJP, of trace metals. Lewis Publishers, Boca Raton, pp 429–497
Coates JD (1996) Humic substances as electron acceptors for Senesi N, Griffith SM, Schnitzer M, Townsend MG (1977) Binding of
microbial respiration. Nature 382:445–448 Fe3? by humic materials. Geochem Cosmochem Acta 41:969–976
Lyvén B, Hassellöv M, Turner DR, Haraldsson C, Andersson K Sielechi JM, Lartiges BS, Kayem GJ, Hupont S, Frochot C, Thieme J,
(2003) Competition between iron- and carbon-based colloidal Ghanbaja J, d’Espinose de la Caillerie JB, Barres O, Kamga R,
carriers for trace metals in a fresh-water assessed using flow Levitz P, Michot LJ (2008) Changes in humic acid conformation
field-flow fractionation coupled to ICPMS. Geochim Cosmo- during coagulation with ferric chloride: implications for drinking
chim Acta 67:3791–3802 water treatment. Water Res 42:2111–2123
123
Stegmann H, Ozdol VB, Koch C, van Aken PA, Engelmann HJ, Vilgé-Ritter A, Rose J, Masion A, Bottero JY, Lainé JM (1999)
Potapov P, Zschech E (2008) Chemical bonds in damaged and Chemistry and structure of aggregates formed with Fe-salts and
pristine low-K materials: a comparative EELS study. Micro- natural organic matters. Colloids Surf 147:297–308
electron Eng 85:2169–2171 Wu H, Lin Y, Wu J, Zeng L, Zeng D, Du J (2008) Surface adsorption
Thurman EM, Malcolm RL (1981) Preparative isolation of aquatic of iron oxide minerals for phenol and dissolved organic matter.
humic substances. Env Sci Technol 15:463–466 Earth Sci Frontiers 15:133–140
van Aken PA, Liebscher B (2002) Quantification of ferrous/ferric Zhou Q, Maurice PA, Cabaniss SE (2001) Size fractionation upon
ratios in minerals: new evaluation schemes of Fe L2,3 electron adsorption of fulvic acid on goethite: equilibrium and kinetic
energy-loss near-edge spectra. Phys Chem Miner 29:188–200 studies. Geochim Cosmochim Acta 65:803–812
van Aken PA, Liebscher B, Styrsa VJ (1998) Quantitative determi- Zumstein J, Buffle J (1989) Circulation of pedogenic and aquagenic
nation of iron oxidation states in minerals using Fe L2,3-edge organic matter in an eutrophic lake. Water Res 23:229–239
electron energy-loss near-edge structure spectroscopy. Phys
Chem Miner 25:323–327
van Schaik JWJ, Persson J, Kleja DB, Gustafsson JP (2008) EXAFS
study on the reactions between iron and fulvic acid in acid
aqueous solutions. Env Sci Technol 42:2367–2373
123
This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/copyright
Journal of Hydrology 477 (2013) 86–93
Contents lists available at SciVerse ScienceDirect
Journal of Hydrology
journal homepage: www.elsevier.com/locate/jhydrol
Variation of organic carbon and nitrate with river flow within an oceanic regime
in a rural area and potential impacts for drinking water production
E. Baurès a,b,⇑, I. Delpla a,b, S. Merel c, M.-F. Thomas d, A.-V. Jung a,b,d, O. Thomas a,b
a
Environment and Health Research laboratory (LERES), EHESP School of Public Health, Avenue du Professeur Léon Bernard, CS 74312, 35043 Rennes Cedex, France
b
Inserm, U 1085 Institute of Research in Environmental and Occupational Health (IRSET), Avenue du Professeur Léon Bernard, CS 74312, 35043 Rennes Cedex, France
c
Department of Chemical and Environmental Engineering, University of Arizona, 1133 James E. Rogers Way, Tucson, AZ 85721, United States
d
School of Environmental Engineering (EME), Campus de Ker Lann, Avenue Robert Schumann, 35170 Bruz, France
a r t i c l e i n f o s u m m a r y
Article history: Over the last two decades, climate change has become a major environmental and public health concern
Received 21 April 2011 due to the increase of the mean temperature on the Earth and its consequences on extreme meteorolog-
Received in revised form 31 October 2012 ical events such as floods and droughts. These events induce very low or very high river flows that may
Accepted 2 November 2012
impair surface water quality, and therefore result in potential health impacts when used for drinking
Available online 10 November 2012
This manuscript was handled by Andras
water production. The present study aims at assessing the impact of hydrologic regime on surface water
Bardossy, Editor-in-Chief, with the quality with a particular emphasis on total organic carbon (TOC) and nitrate. Water quality data from
assistance of Sheng Yue, Associate Editor three French rivers acquired over a 27 years period, from January 1983 to December 2009, show the influ-
ence of extreme flows. Variation in TOC and nitrate concentrations showed opposite patterns for the
Keywords: whole range of flow rate (from less than 10% up to more than 100% of the mean flow). Regarding fluxes,
Resource water quality TOC increased continuously with flow rate while nitrate was stable for very high discharges. The C/N ratio
Data analysis expressed from TOC and nitrate concentrations showed high values for extreme flows and particularly for
Weather events very low flow rates, generally in summer, where nitrate is assimilated by biomass. Considering TOC and
Climate change nitrate fluxes, it is confirmed that the worst situations were encountered for very high flow rates, namely
for TOC exportation during surface runoff which was related to heavy rains or floods. These findings are of
great importance with regard to the adaptation for drinking water treatment in facing extreme hydrolog-
ical conditions, of which the frequency is increasing with climate change.
Ó 2012 Elsevier B.V. All rights reserved.
1. Introduction exploitation of thousands of data points that there are negative

nonlinear relationships between nitrate and organic carbon
At a river basin scale, any water management actions should concentrations for a hydrologic continuum from soils, streams,
take into account water quantity and quality variations. lakes, and oceans. Concerning rivers in rural areas, whose water
This is of primary importance within the frame of water safety can be treated for supply systems, mostly of small size, the knowl-
plans implementation when surface water is used for drinking edge of water quality variation should be documented in the risk
water supply after treatment. Regarding water quality variation, assessment step of water safety plans. Climate change impacts
a lot of data is available worldwide primarily for the main phys- with a possible increase in extreme weather events (floods and
ico-chemical parameters, nutrients and some micropollutants. droughts), may lead to the modification of hydrologic regimes of
While the majority of recent research has focused on the contam- rivers with very low or very high flow rates during droughts or
ination of emerging substances like pharmaceuticals, an under- floods (at least for heavy rainfalls) and lead to deterioration of
standing of the relationship between basic parameters of water water quality. This is particularly true for heavy rain impacts on
quality like organic carbon and nitrate is still relevant today small watershed where runoff is a nonlinear, complex procedure
(Taylor and Townsend, 2010). These authors showed from the applying the watershed’s geomorphologic properties, such as
topology, vegetation, soil type, and climatic factors i.e., precipita-
⇑ Corresponding author at: Environment and Health Research laboratory (LERES), tion, temperature, etc. (Wang et al., 2007; Jin et al., 2009).
EHESP School of Public Health, Avenue du Professeur Léon Bernard, CS 74312, One of the first studies performed in North American rivers
35043 Rennes Cedex, France. Tel.: + 33 2 99 02 29 19; fax: + 33 2 99 02 29 29. regarding climate change impacts (warming) and water quality,
E-mail addresses: estelle.baures@ehesp.fr (E. Baurès), ianis.delpla@ehesp.fr (I. anticipated a decreased oxygen-carrying capacity, a decreased
Delpla), sylvain.merel@gmail.com (S. Merel), marieflorencethomas@ecole-eme.com
(M.-F. Thomas), audevaleriejung@ecole-eme.com (A.-V. Jung), olivier.thomas@
volume for dilution of chemical inputs and the invasion by
ehesp.fr (O. Thomas). temperature-sensitive exotic species (Murdoch et al., 2000).
0022-1694/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jhydrol.2012.11.006
E. Baurès et al. / Journal of Hydrology 477 (2013) 86–93 87
Soil leaching following rainfall events in agricultural areas can used as a resource for drinking water production by a SSWS in Brit-
lead to high levels of nitrate in surface waters used for potable tany, France. The land use is mainly agricultural (86% of area) with
water production and consequently in distributed waters (WHO, a large number of animal farms (cattle, pig, and poultry) and some
2003). The main health risk associated with nitrate exposure is agro-food industries including four slaughterhouses and four dair-
the methaemoglobinaemia, the infants being the susceptible pop- ies. Over the five total counties located in the entire watershed,
ulation (WHO, 2003). Consequently, in order to protect the popula- three showed an excess of the regulatory N production of
tion, a water quality guideline for nitrate concentration has been 170 kg N ha1 (Agreste – DRAAF Bretagne, 2004, 2006; Council
set at 50 mg/L in France (Arrêté du 11 janvier, 2007). Furthermore, Directive, 1991) and three rural districts located at less than
high levels of dissolved organic matter, suspended solids and 20 km upstream of the abstraction point present the most impor-
pathogens are often measured in streams following rainfall events tant percentage of agricultural lands (from 97% to 100%) and could
(Delpla et al., 2009; Hunter, 2003; van Vliet and Zwolsman, 2008; potentially present the greatest risk to water contamination
Zwolsman and van Bokhoven, 2007; Worrall and Burt, 2007). Most (Corine Land Cover, 2010). The human density is approximately
of the pathogens associated with suspended solids are responsible 50 inhabitants km2. There are a few villages, and the main one
for water-borne diseases such as gastroenteritis (Beaudeau et al., owns a SSWS serving approximately 6000 people. Water samples
2001, 2010). The water quality reference for TOC in France is set were collected from the main sampling site of this SSWS which
at 10 mg/L in raw waters both to protect human health and ensure is the pump inlet of the treatment plant. The treatment process
treatment efficiency (Arrêté du 11 janvier, 2007). is after a conventional coagulation-settling step, followed by sand
The C/N (here defined as total organic carbon/nitrogen-nitrate, filtration, adsorption on activated carbon, and chlorine disinfec-
TOC/N-NO3) ratio is a parameter commonly used to characterize soil tion. This site is also characterized as a schist bedrock and soil with
organic matter and microbial activity which is less used for stream high silt content (67%), inducing strong hydrological variations and
water (Meybeck, 1982; Taylor and Townsend, 2010). However, it high soil sensibility to erosion.
could be used as an indicator to determine organic matter origin. The second site (site B) is located approximately a third of the
This study will mainly focus on the effects of flows on total or- way downstream of a 101 km length river associated with a
ganic carbon (TOC) and nitrate concentrations, including C/N ratio, 710 km2 catchment area and characterized by a 7.1 m3 s1 mean
in regards to health risk assessment and SSWS adaptation. It is flow (calculated over 43 years). However, the discharge ranges
based on the interpretation of water quality data and flow rate from less than 0.9 m3 s1 during droughts, to more than
measurements acquired from three small rivers in Brittany, France 28.8 m3 s1 during floods. This watershed is characterized by an
over a 27 years period. The three rivers considered have a rather alternation of schist and granitic bedrock and pedology differs
stretched basin of low slope and a pluvial oceanic regime. Samples from the first site because soils contain a higher sand content
were collected from monitoring stations for small-scale water ser- (26%) and a lower silt content (56%) than site A. This site also faces
vices (SSWSs) with a treatment plant drawing water directly from strong agricultural pressures with 90% of the watershed covered by
the river. Relationships between parameters are studied for the agricultural area and two counties located near the source among
three sites. the seven covering the watershed, showing a N production exceed-
ing the regulatory threshold of 170 kg N ha1. Cultivated areas rep-
resent approximately 65% of catchment area with 2700 farms
2. Material and methods mostly dedicated to cattle farming (116 km2), pigs (184 km2)
and poultry (2902 km2). Despite the minimal data available, it ap-
2.1. Description of the experimental field pears that there are lower industrial pressures at this site com-
pared to the first site. However, at site B drinking water
The characteristics of the three catchments (catchments A, B treatment is more complex as it includes ozonation and nitrate re-
and C) are described in Table 1. moval in addition to conventional treatment processes.
The site A is located downstream on a small river, approxi- The third site (site C) is located approximately at the mid-point
mately at a third of the way down its length (87 km). The river is of a 41 km length river associated with a 147 km2 catchment area.
Table 1
Characteristics of the sites studied.
Site A Site B Site C

River length (km) 87 101 41
Upstream catchment area (km2) 468 710 147
Total catchment area (km2) 815 1220 170
Altitude (m) 35 10 96
Farms 1300 2760 310
Cattle (number km2) 100.4 116.0 199.1
Pigs (number km2) 384.4 183.7
Poultry (number km2) 10124.1 2902.1 1996.6
Natural/semi-natural areaa (km2) 10% 5% 1%
Industries (km2) 16 food industries
Geology Schist bedrock Schist and granitic bedrock Granitic bedrock
Mean soil particles compositionb Silt: 67%, clay: 17%, sand: 16% Silt: 56%, clay: 18%, sand: 26% Silt: 69%, clay: 14%, sand: 17%
Hydrologyc (m3 s1)
Mean flow (m3.s1) 3.1 7.1 1.9
Very Low Flow (m3.s1) Q < 0.4 Q < 0.9 Q < 0.3
Low Flow (m3.s1) 0.4 < Q < 1.9 0.9 < Q < 4.7 0.3 < Q < 1.5
Intermediate Flow (m3.s1) 1.9 < Q < 4.3 4.7 < Q < 9.6 1.5 < Q < 2.6
High Flow (m3.s1) 4.3 < Q < 12.9 9.6 < Q < 28.8 2.6 < Q < 7.8
Very High Flow (m3.s1) Q > 12.9 Q > 28.8 Q > 7.8
a
Corine Land Cover database (Corine Land Cover, 2010), administrated by the French ministry of environment.
b
French database of soil analysis (BDAT, 2011) administrated by the French National Institute for Agricultural Research (INRA).
c
French hydrological database (Banque hydro, 2010) administrated by the French ministry of environment.
88 E. Baurès et al. / Journal of Hydrology 477 (2013) 86–93
With a discharge ranging from 0.3 m3 s1 or less to 7.8 m3 s1 or thus assessed by using a statistical relationship (determined from
more, a mean flow of 1.9 m3 s1 was calculated for 37 years of data. data during the period were both tests were being conducted) be-
Unlike the other two sites, this site is characterized as a granitic tween these two parameters (TOC = OxA/1.045, R2 = 0.83, n = 59).
bedrock. Soil composition is similar to the first site (silt: 69%, clay: Oxydability was analyzed by manual titration with permanganate
14%, sand: 17%). This site is under substantial agricultural pres- (NF EN ISO 8467, 1995). TOC was determined by thermal oxidation
sures with 310 farms, intensive livestock farming and an exceed- coupled with infrared detection (Multi N/C 2100 Analytik Jena) fol-
ance of the regulatory threshold of 170 kg N ha1 (Council lowing acidification with HCl (NF EN 1484, 1997). Nitrate concen-
Directive, 1991) in the two counties covering the watershed. Like tration was expressed in mg L1 of N-NO3. Nitrate was determined
the first site, the drinking water treatment processes include con- after a filtration through 0.45 lm HAWP nitrocellulose filter papers
ventional coagulation-settling, sand filtration, and chlorination. (Millipore) with a continuous flux analyzer according to standard
River hydrology for the three sites is presented in Table 1. The method (NF EN ISO 13395, 1996). For these analyses, the samples
data was collected for a period of 43 years, for sites A and B, and were stored at 4 °C. In the present study, the mean flow calculated
37 years for site C. The dry discharge corresponds to the annual from the dataset was different from the one given by the French
mean flow ðQ m43y Þ that has 4/5 probability of being exceeded each hydrological database. The daily flow rates (Q) were shown to obey
year. In the same way, the wet discharge corresponds to the annual a log-normal distribution. Consequently, in order to avoid a huge
mean flow that has 1/5 probability of being exceeded each year. dispersion of plots and make graphs readable, the normalized flow
These values were used to sort out the daily flow rate (Q) into five rates (Q/Qm) and their logarithmic values log (Q/Qm) were also con-
different classes: (i) very low flow, VLF (under 20% of dry dis- sidered. Moreover, normalized fluxes were calculated as the prod-
charge), (ii) low flow, LF (between VLF and dry discharge), (iii) uct of the concentrations and the corresponding normalized flow
intermediate flow, IF (between LF and wet discharge), (iv) high rates (Q/Qm). Correlation matrix calculations were created using
flow, HF (between IF and three times wet discharge) and (v) very the Pearson correlation test. Finally, correlation and linear and
high flow, VHF (greater than three times wet discharge). non-linear adjustments were calculated with Statistica V6, from
StatSoft Inc.
2.2. Raw data
Water quality data were obtained from the French Ministry of 3. Results and discussion
Health through the national database SISE-EAUX. This database
gathers the results of water quality analysis performed on drinking The first step was to compare the parameters through the calcu-
water resources and drinking water supply networks as part of the lation of correlation coefficients (Table 2).
mandatory sanitary control program. Since no data was available The first observation is that more than 3/4 of the correlation
prior to 1983, the present study covers the period ranging from coefficients were statistically significant, considering the related
January 1983 to December 2009 (27 years) for the three sites. p-values. All parameters show fair correlations with log of normal-
Parameters monitored were water temperature, pH, conductivity, ized values of river discharges, and lower correlations with nor-
turbidity, oxidability, TOC and nitrate. Considering that on one malized flow rate, except for turbidity and TOC which have a
hand the monitoring frequency (over 27 years) varies with the size similar behavior and are linked together and to normalized flow
of SSWS, and on the other hand, data set presents some missing rate. Conductivity and pH are linked to other parameters except
values, the number of sampling vary from 150 to 266, depending turbidity for conductivity and TOC for pH. The highest correlations
on the site to be considered. Notice that the sampling was not de- with flow rate are observed for conductivity and pH and are plot-
signed to take into account the river hydrology, but was large en- ted in Fig. 1.
ough to include storm and base flow conditions. More precisely, at Thus TOC values are not well correlated with water tempera-
each site 2/3 of the samples were taken on the falling limb of the ture in this case, which is contrary to previous studies showing
hydrograph and 1/3 were taken on the rising limb. Overall, the that TOC load of river is not only a rainfall driven but also a tem-
greatest amount of data was related to site A, with 58 samples ana- perature driven biological process (Ouyang, 2003). Notice that cor-
lyzed for turbidity and 230 for nitrate. relation coefficients are either positive or negative depending on
the parameters. Considering the importance of TOC and nitrate
2.3. Data pre-treatment concentrations for the assessment of water quality and their vari-
ation with flowrate particularly in the context of heavy rain or
Data conversions were needed for the coherence of the results droughts (Delpla et al., 2011a) it was decided to focus on the im-
of organic matter. Results were expressed only by permanganate pact of river flow rate on water quality parameters. TOC and nitrate
oxidability in acidic conditions (OxA) at the beginning of the mon- were also chosen for the high concentrations found locally (GIP
itoring from 1983 to 1993, coupled with TOC after this date, and Bretagne Environnement, 2011) and their potential health risks
measured only as TOC after 1995. TOC values before 1993 were (Bull et al., 1995; Katsoyiannis and Samara, 2007).
Table 2
Pearson correlation between parameters for the three sites (number of samples is in brackets).
Log(Q/Qm) Q/Qm Water temperature Turbidity (NFU) Conductivity at pH TOC (mg L1)
(°C) 20 °C (lS cm1)
Q/Qm 0.67 (n = 579)
Water temperature (°C) 0.52 (n = 340) 0.30 (n = 340)
Turbidity (NFU) 0.35 (n = 142) 0.47 (n = 142) 0.13 (n = 157)
Conductivity at 20 °C (lS cm1) 0.70 (n = 215) 0.28 (n = 215) 0.23 (n = 257) 0.07 (n = 105)
pH 0.65 (n = 354) 0.30 (n = 354) 0.47 (n = 402) 0.04 (n = 161) 0.57 (n = 270)
TOC (mg L1) 0.01 (n = 291) 0.37 (n = 291) 0.15 (n = 318) 0.41 (n = 147) 0.27 (n = 215) 0.07 (n = 330)
N-NO3 (mg L1) 0.50 (n = 512) 0.09 (n = 512) 0.39 (n = 390) 0.09 (n = 157) 0.48 (n = 265) 0.37 (n = 409) 0.42 (n = 319)
In bold: statistically significant values (p < 0.05).

3.1. Variation of organic carbon and nitrate with river flow to March). However, these seasonal trends should be viewed with
caution as these are linked to high pressure from agriculture, with
Firstly, the hydrology follows a seasonal behavior with higher the application of manure or slurries to the fields each year during
flows from December to February and lower flows from July to mid-winter and the start of summer from the year 2000. This may
September consistent with previous studies conducted in temper- have contributed to the elevated nitrate and TOC concentrations in
ate areas (Attrill and Power, 2000). The seasonal values of TOC and the river water.
nitrate (N-NO3), calculated over the 27-year period are reported in
Fig. 2. 3.2. Distribution of organic carbon and nitrate according to river flow
As expected, after the above correlation study, nitrate concen-
tration and the flow rate of the rivers, followed the same yearly The relationship between TOC and nitrate concentration with
pattern, with the lowest values occurring during summer (from flow rate (log(Q/Qm)) is shown in Fig. 3, which also considers the
July to September) and highest values during winter (from January French limits of quality for drinking water resources, 10 mg L1
10 2000
Conductivity (µS/cm)
9 1500
pH
8 1000
7 500
6 0
-3 -2 -1 0 1 2 -3 -2 -1 0 1 2
Log (Q/Qm) Log (Q/Qm)
Fig. 1. Variation of pH and conductivity with log of normalized flow rate for the three sites. Pearson correlation coefficients (r) are given for each plot.
20 20
15 15
N-NO3 (mg.L-1)
TOC (mg.L-1)
10 10
5 5
0 0
Jan-Mar Apr-Jun Jul-Sep Oct-Dec Jan-Mar Apr-Jun Jul-Sep Oct-Dec
Months Months
Fig. 2. Seasonal distribution of TOC and nitrate concentrations from 1983 to 2009 for the three sites. Boundaries of boxes indicate interquartile range (25th and 75th
percentiles) and median values (midline), whiskers indicate 5th and 95th percentiles and symbols indicate maximum values.
20 20
15 15
N-NO3 (mg.L-1)
TOC (mg.L-1)
10 10
5 5
0 0
-3 -2 -1 0 1 2 -3 -2 -1 0 1 2
Log (Q/Qm) Log (Q/Qm)
Fig. 3. Relation between TOC and nitrate concentrations with flow rate (log(Q/Qm)) for the three sites. Pearson correlation coefficients (r) are given for each plot.
for TOC and 11.3 mg L1 N-NO3 (i.e. 50 mg NO3 L1). Two evolution In order to clearly reveal the potential impact of extreme flows,
schemes are observed for TOC and nitrate concentrations. TOC con- Fig. 6 displays the evolution of the C/N (TOC/N-NO3) concentra-
centration tends to decrease when log(Q/Qm) increases from 2 to tions ratios, with river flows represented by the inverse of log(Q/
0 (corresponding to Q ranging from 1% to 100% of Qm) and increase Qm). This choice, which places very low or very high flow rates near
for log(Q/Qm) greater than 0 (corresponding to Q higher than 100% the origin, shows that the corresponding C/N ratios increase
of Qm). Contrary to TOC, nitrate concentration rises from 0 to more strongly when 1/log(Q/Qm) tends towards 0. More precisely, this
than 11.3 mg L1 for log(Q/Qm) ranging from 2 to 0 and decrease is observed for 1 < 1/log(Q/Qm) < 1 corresponding to Q < 0.1Qm
for higher values. Even if the majority of the points stay below or Q > 10Qm. However, there is a difference between very low flows
quality limit for nitrate, the highest values (18.9 mg L1) are corresponding to droughts for which C/N values rise up to 50, and
encountered for mean flow rates. Concerning instances of TOC con- very high flows corresponding to floods with C/N values lower than
centrations exceeding 10 mg L1, some samples correspond to low 10. Between these values of extreme flow rates, i.e. for values
flows, but for the highest values of TOC the samples correspond to around the mean flow rate, the C/N ratio is very low and close to
flows upper than the mean flow rate. Overall, the relationship be- one.
tween nitrate and flow rate (R = 0.53) was stronger than TOC and
flow rate (R = 0.12, see Table 2). 4. Discussion
The wide variation in TOC and nitrate concentrations in relation
to river flow make it difficult to quickly identify the hydrological 4.1. C/N variation
conditions that would cause a increase in the number of times that
the water quality guidelines would be exceeded. TOC and nitrate Among the above results, the variation of C/N (TOC/N-NO3) with
results were sorted into five main flow rate classes, previously de- river flows (Fig. 6) must be highlighted. For the majority of low
scribed in Table 1. This shows the conditions for a higher vulnera- flows, C/N is quite low, around 1–1.3, in accordance with Kim
bility of surface water quality with respect to health risk et al. (2007). For extreme flows, the C/N ratio increases to reach
assessment (Fig. 4). The box plots TOC graph highlights that there very high values (around 50), with a difference between the max-
is a greater risk of exceeding the French limit of quality for drinking ima of C/N corresponding to low and high flows, the latter giving
water resources when the flow rate is high (floods). For nitrate, the lower maxima (less than 10). Similarly, Wysocki et al. (2006)
risk of exceeding the limit seems to be lower. Overall, the highest showed that C/N ratios are higher during low flows than during
probability for nitrate exceedances occurs at flow rates greater high flows. The data analysis shows that C/N increase is observed
than the mean value with the risk decreasing for higher flows. for flows lower or greater than 10% or 1000% respectively of the
mean flow. Considering these results and the ones of Delpla et al.
3.3. Variation of TOC and nitrate fluxes with flow rate (2011a), the following explanations can be proposed: (i) For low
flows, occurring mostly in summer, nitrate is assimilated by the
The evolution of fluxes is quite different between TOC and ni- biomass (phytoplankton and macrophytes) (van Vliet and Zwols-
trate. TOC flux (FTOC) increases regularly, while nitrate flux (FN- man, 2008) and moreover supply from soil leaching is reduced dur-
NO3) remains relatively constant for very high flows (Fig. 5a and b). ing the dry season (Delpla et al., 2011a; van Vliet and Zwolsman,
Polynomial adjustment lead to good correlations (R2 = 0.94 for 2008). The biomass growth leads to an increase in TOC concentra-
FTOC and 0.83 for FN-NO3), but variation of both fluxes can be con- tion observed for low flows. (ii) For high flows, generally in winter,
sidered as partly linear for flow rates less than 3 Q/Qm (see zoom in C/N is controlled by surface runoff and dilution. The importance of
Fig. 5b). organic carbon as soluble, colloidal, and particulate forms exporta-
tion by surface runoff during rainy events, is depending on phys-
3.4. Variation of organic carbon-nitrate relationship with river flow ico-chemical parameters of soil (nature, slope, moisture, etc.),
rain (intensity, duration) (Delpla et al., 2011b; Jacinthe et al.,
TOC and nitrate concentrations evolution with log of normal- 2004) and land use (Tejada and Gonzalez, 2008). For nitrate con-
ized flow rate is quite different (Fig. 3). Consequently, the study centration, the value remains always not negligible despite dilu-
of C/N ratio calculated from these parameters could provide some tion because an higher supply by soil leaching in winter (at least
interesting findings about the origin of organic matter (Meybeck, for the area studied) (Fig. 2). Finally, these C/N values are close
1982) and therefore on its treatability. Moreover, according to to that of organic soils reflecting a high supply of exogenous organ-
Taylor and Townsend (2010), there exist some relationships ic constituents (Meybeck, 1982).
between nitrate and organic carbon (dissolved or particulate) In case of agricultural area amended by animal waste or bioso-
along a hydrologic continuum from soils to streams and lakes. lids as fertilizers, organic carbon in runoff is mostly related to
20 20
15 15
N-NO3 (mg.L-1)
TOC (mg.L-1)
10 10
5 5
0 0
VLF LF IF HF VHF VLF LF IF HF VHF
Flowrate class Flowrate class
Fig. 4. Distribution of TOC and nitrate concentrations by flow rate classes for the three sites. Boundaries of boxes indicate interquartile range (25th and 75th percentiles) and
median values (midline), whiskers indicate 5th and 95th percentiles and symbols indicate maximum values. n corresponds to the number of samples for each flowrate class.
(a) (b)
60
TOC fluxes (mg.L-1)
200
40
100
20
0 0
150
60
N-NO3 fluxes (mg.L-1)
100
40
50
20
0 0
0 5 10 15 0 1 2 3 4 5 6
Q/Qm Q/Qm
Fig. 5. (a) Impact of normalized flow rate on TOC and N-NO3 fluxes for the three sites and (b) zoom.
for concentration, fluxes and C/N variations when the relationship

between TOC and nitrate with flow rates were examined. However,
because the dynamic of flows were less important at sites B and C
as compared to site A, a decrease in nitrate concentration or fluxes
was not observed. Flow rates of rivers located in granitic bedrocks
catchments generally undergo less variability than those located
on schist bedrock (Mérot, 2003) and could explain the lowest var-
iability in flow rates for site C. Despite this, the relationship be-
tween TOC and nitrate fluxes and flow rates are linear for the
three sites, with strong correlation coefficients (Table 3).
For nitrate flux, the slope values are very close for the three sites
(respectively 8.92, 9.59 and 9.26 for A, B and C). These trends could
be strengthened by considering land use practices, which at pres-
ent include two amendment periods (during the end of winter
and summer beginning) corresponding to nitrate supply. More
Fig. 6. Impact of discharge on C/N ratio variation for the three sites. generally, these similarities could be explained by the controlled
agricultural practices all year long, which results in a dynamic pro-
duction and release of nitrate nearly constant and limited (Oeurng
manure or slurry, particularly after spreading (Haynes and Naidu, et al., 2010).
1998; Schulten and Leinweber, 1991). Nitrate flux depends on
different mechanisms and mainly the transfer from the aqueous
form or groundwater if any and a dilution with rain.
Finally, studies undertaken on larger rivers indicate that lower Table 3
nitrate concentrations are associated with lower river flows (Attrill Relations between TOC and nitrate normalized fluxes (y = ax + b) with flow rates for
and Power, 2000; van Vliet and Zwolsman, 2008). Thus the inter- the three sites.
pretation of nitrate sources based on discharge and nitrate concen- Site Flux a b R n
tration relationship is still uncertain because of the difficulty in
Aa TOC 9.01 0.96 0.90 164
interpreting nitrate sources.
Nitrate 8.92 0.54 0.91 233
B TOC 6.44 1.07 0.91 84
4.2. Towards a generalization
Nitrate 9.59 0.18 0.90 183
C TOC 4.88 0.65 0.90 34
After the theory of C/N variation with extreme flow rates is dis-
Nitrate 9.26 0.21 0.95 92
cussed, it is important to compare the results between the three
a
sites studied (A, B and C). All the sites follow the same patterns For site A, fluxes are calculated for flow rates < 500% of Qm.
For TOC flux, the slope values are different for the three sites, tries and nature also explains sediment deposition/transportation
(respectively 9.01, 6.44 and 4.88 for A, B and C). These differences with current (Mianaei and Keshavarzi, 2010). Depending of the
could be explained by several parameters which influence organic development of river, water quality may vary with the presence
matter concentration, which include the geology and pedology. For of a dam for example, reducing both sediments transportation
example, clod disaggregation under raindrop impacts result in a and fauna distribution of derived zooplankton (Takao et al.,
formation of surface crusts (Le Bissonnais et al., 1998), reducing 2008). River branching related to watershed topography, and its
the soil infiltration rate and inducing erosion by increasing runoff consequences on flowrate (Yousefi and Ghiassi, 2011) can also be
(Le Bissonnais et al., 1995) and so increasing organic matter flux. considered local factor of water quality variation. Finally, floods
In fact, the soils on the site A watershed could be particularly prone frequency increase is a strong factor of water quality variation with
to crusting (Delpla et al., 2011b). Moreover, these observations pollutants and solids transportation and modification of flow rout-
might be consistent with the presence of two pools of organic mat- ing (He et al., 2007).
ter: allochtonous large-size colloids formed by lixiviation from
upper soil horizons (Kim et al., 2007) and autochthonous (aquatic)
5. Conclusion
small molecular-size substances, probably linked to bacterial and
phytoplankton exudates. The proportion of both possible pools is
To date, except a couple of studies (Delpla et al., 2009; van Vliet
highly dependent on the season. Allochtonous input of organic car-
and Zwolsman, 2008), very few papers describe the potential ef-
bon strongly increases during the spring flood, whereas the bacte-
fects of climate change and particularly the hydrologic regime of
rial/plankton exudates of the second pool of organic matter are
river on water quality. The present study provides original results
seasonally variable reflecting both light and temperature changes
on the variation of organic carbon - nitrate relationship with this
in photosynthesis/respiration.
hydrologic regime, covering a relatively long period ranging from
Finally, our findings must be considered as a first step in simple
January 1983 to December 2009 (27 years). This study shows the
modelling for TOC and nitrate fluxes, which are complementary to
importance of the extreme hydrologic regime particularly for the
powerful tools like deterministic models (Wriedt and Rode, 2006),
very low (Q < 10% of Qm) flows on C/N (TOC/N-NO3) ratio. For
which certainly must be checked at other sites with different envi-
the dry period, especially in summer, with low water levels, the
ronmental and socio-economic conditions.
C/N ratio increase could be explained mainly by the decrease in
nitrate concentration. This decrease is mainly justified by the
4.3. Impact on small scale water services in-stream biomass consumption, its reduced supply due to the
high mobility of this substance and plant mobilization during dry
Floods and droughts, which are likely to increase with climate periods. However, a public health risk assessment is more sensible
change, may impact the quality of water resources used for drink- if the water quality is considered in term of flux rather than
ing water production by direct effects of dilution or concentration concentration. This study shows that, for very high flow
of dissolved substances (Delpla et al., 2009). Therefore, as stated in (Q > 500% of Qm), the TOC flux increase and the nitrate flux
a recent World Health Organisation (WHO) report (WHO, 2010), decrease. Thus the TOC flux becomes a major concern for the SSWS
these extreme weather events could affect the efficiency of drink- without adapted waterworks. Such findings should be useful for
ing water treatment processes and the stability of drinking water the administrations in charge of sanitary control but also for those
quality in distribution. These concerns are important for small- in charge of drinking water production. In fact, the present
scale water services in rural areas, which are particularly sensitive conclusions promote the adaptation of drinking water treatment
because of difficulties to adapt treatment to such variations. Con- based on the hydrological conditions. However, further research
sidering that a large part of the population living in rural areas should be undertaken to validate and check our findings for
do not have access to improved drinking-water sources, e.g. 16 other conditions, taking into account the local variability of water
million people in Europe (WHO and DFID, 20101), small-scale water quality in river sections, and to integrate other water quality
supplies producing tap water from surface resources are facing sev- parameters like microbiological contamination or concentration
eral challenges including their regulatory environment, administra- of micropollutants and organic carbon under soluble, colloidal
tion, management, operational or available techniques, as well as and particulate forms into an accurate statistical model for water
dwindling personnel and financial resources (WHO, 2010). quality prediction and health risk assessment.
4.4. Hydrological limits for monitoring relevance Acknowledgements
The previous results suppose homogeneous conditions in the The authors sincerely thanks the French Agence Régionale de
section of rivers were samples were taken in order to be able to Santé (ARS) of Brittany and the Ministère de l’Ecologie, du Dével-
monitor a representative state of water quality. Among factors of oppement Durable, des Transports et du Logement (MEDDTL) for
variability, small rivers like the ones considered in the present the data provided and are grateful to Darryl Jones, Ph.D. student
studies are characterized by various hydromorphological charac- at the University of Arizona (USA), for his assistance in editing. This
teristics (slope, banks, bed, etc.) and the presence of several tribu- work was partially funded by the French National Agency of Sani-
taries on the related watersheds may induce a local distribution of tary Security (ANSES) under the first ERA-EnvHealth call (FP7-ENV-
pollutants as nitrates and organic carbon. For example, the influ- 2007-CSA-1.2.3-01).
ence of the bottom topography (Thomas and Linden, 1998;
Cenedese and Linden, 2002) or the lateral mixing of pollutants References
when the tributary channel is shallower than the main channel
(Biron et al., 2004) may lead to a sectional gradient of water quality Agreste – DRAAF Bretagne, 2004. Fichiers administratifs PAC et BDNI, recensement
avicole.
particularly in the downstream vicinity of a confluence. Moreover,
Agreste – DRAAF Bretagne, 2006. Fichiers administratifs PAC et BDNI, recensement
flow mixing rates downstream confluences is depending on bed des sites porcins.
roughness and nature (Parsons et al., 2007), and bed form geome- Arrêté du 11 janvier 2007 relatif aux limites et références de qualité des eaux brutes
et des eaux destinées à la consommation humaine mentionnées aux articles R.
1321–2, R. 1321–3, R. 1321–7 et R. 1321–38 du code de la santé publique.
1
Department For International Development. Journal Officiel, n°31 du 6 février 2007, page 2180, texte no. 17.
Attrill, M.J., Power, M., 2000. Modelling the effect of drought on estuarine water Mianaei, S.J., Keshavarzi, A.R., 2010. Study of near bed stochastic turbulence and
quality. Water Res. 34, 1584–1594. sediment entrainment over the ripples at the bed of open channel using image
Banque hydro, 2010. Online French hydrological database accessible at <http:// processing technique. Stoch. Environ. Res. Risk Assess. 24, 591–598.
www.hydro.eaufrance.fr/> (accessed 10.12.10). Murdoch, P.S., Baron, J.S., Miller, T.L., 2000. Potential effects of climate change on
BDAT, 2011. Online French Soil analysis database accessible at http:// surface-water quality in North America. J. Am. Water Resour. Assoc. 36, 347–
acklins.orleans.inra.fr/geosol/index.php (accessed 03.02.11). 366.
Beaudeau, P., Leboulanger, T., Lacroix, M., Hanneton, S., Wang, H.Q., 2001. NF EN 1484, Juillet 1997. Analyse de l’eau. Lignes directrices pour le dosage du
Forecasting of turbid floods in a coastal, chalk karstic drain using an artificial Carbone Organique Total et du Carbone Organique Dissous.
neural network. Ground Water 39 (1), 109–118. NF EN ISO 13395, Octobre 1996. Détermination de l’azote nitreux et de l’azote
Beaudeau, P., Valdes, D., Mouly, D., Stempfelet, M., Seux, R., 2010. Natural and nitrique et de la somme des deux par analyse en flux continu et détection
technical factors in faecal contamination incidents of drinking water in small spectrométrique.
distribution networks, France, 2003–2004: a geographical study. J. Water NF EN ISO 8467, Juillet 1995. Qualité de l’eau. Détermination de l’indice
Health 08 (1), 20–34. permanganate.
Biron, P.M., Ramamurthy, A.S., Han, S., 2004. Three-dimensional numerical Oeurng, C., Sauvage, S., Sánchez-Pérez, J.-M., 2010. Temporal variability of nitrate
modeling of mixing at river confluences. J. Hydraul. Eng. 130 (3), 243–253. transport through hydrological response during flood events within a
Bull, R.J., Birnbaum, L.S., Cantor, K.P., Rose, J.B., Butterworth, B.B., Pegram, R., large agricultural catchment in south-west France. Sci. Total Environ. 409,
Tuomisto, J., 1995. Water chlorination: essential process or cancer hazard? 140–149.
Fund. Appl. Toxicol. 28, 155–166. Ouyang, Y., 2003. Simulating dynamic load of naturally occurring TOC from
Cenedese, C., Linden, P.F., 2002. Stability of a buoyancy-driven coastal current at the watershed into a river. Water Res. 37, 823–832.
shelf break. J. Fluid Mech. 452, 97–121. Parsons, D.R., Best, J.L., Lane, S.N., Orfeo, O., Hardy, R.J., Kostaschuk, R., 2007. Form
Corine Land Cover, 2010. Online French geographical database accessible at <http:// roughness and the absence of secondary flow in a large confluence-diffluence,
www.stats.environnement.developpement-durable.gouv.fr/bases-de-donnees/ Rio Paraná, Argentina. Earth Surf. Process. Landforms 32, 155–162.
occupation-des-sols-corine-land-cover> (accessed 10.12.10). Schulten, H.-R., Leinweber, P., 1991. Influence of long-term fertilization with
Council Directive 91/676/EEC of 12 December 1991 Concerning the Protection of farmyard manure on soil organic matter: characteristics of particle-size
Waters against Pollution caused by Nitrates from Agricultural Sources. fractions. Biol. Fert. Soils 12, 81–88.
Delpla, I., Jung, A.V., Baurès, E., Clément, M., Thomas, O., 2009. Impacts of climate Takao, A., Kawaguchi, Y., Minagawa, T., Kayaba, Y., Morimoto, Y., 2008. The
change on surface water quality in relation to drinking water production. relationships between benthic macroinvertebrates and biotic and abiotic
Environ. Int. 35, 1225–1233. environmental characteristics downstream of the Yahagi dam, central japan,
Delpla, I., Baurès, E., Jung, A.V., Clément, M., Thomas, O., 2011a. Issues of drinking and the state change caused by inflow from a tributary. River Res. Applic. 24 (5),
water quality of small scale water services towards climate change. Water Sci. 580–597.
Technol. 63, 227–232. Taylor, P.G., Townsend, A.R., 2010. Stoichiometric control of organic carbon–nitrate
Delpla, I., Baurès, E., Jung, A.-V., Thomas, O., 2011b. Impacts of rainfall events on relationships from soils to the sea. Nature 464, 1178–1181.
runoff water quality in an agricultural environment in temperate areas. Sci. Tejada, M., Gonzalez, J.L., 2008. Influence of two organic amendments on the soil
Total Environ. 409, 1683–1688. physical properties, soil losses, sediments and runoff water quality. Geoderma
GIP (Groupement d’Interêt Public). Bretagne environnement, accessible at <http:// 145, 325–334.
www.observatoire-eau-bretagne.fr> (accessed 01.10.11). Thomas, P.J., Linden, P.F., 1998. A bi-modal structure imposed on gravity driven
Haynes, R.J., Naidu, R., 1998. Influence of lime. Fertilizer and manure applications on boundary currents in rotating systems by effects of the bottom topography. Exp.
soil organic matter content and soil physical conditions: a review. Nutr. Cycl. Fluids 25, 388–391.
Agroecosyst. 51, 123–137. van Vliet, M.T.H., Zwolsman, J.J.G., 2008. Impact of summer droughts on the water
He, H.M., Zhou, J., Yu, Q., Tian, Y.Q., Chen, R.F., 2007. Flood frequency and routing quality of the Meuse river. J. Hydrol. 353, 1–17.
processes at a confluence of the Middle yellow river in China. River Res. Applic. Wang, X.K., Lu, W.Z., Cao, S.Y., Fang, D., 2007. Using time delay neural network
23 (4), 407–427. combined with genetic algorithm to predict runoff level of Linshan watershed,
Hunter, P.R., 2003. Climate change and waterborne and vector-borne disease. J. Sichuan, China. J. Hydrol. Eng.-ASCE 12 (2), 231–236.
Appl. Microbiol. 94, 37S–46S. World Health Organization (WHO), 2003. Nitrate and nitrite in drinking-water.
Jacinthe, P.-A., Lal, R., Owens, L.B., Hothem, D.L., 2004. Transport of labile carbon in Background Document for Preparation of WHO Guidelines for Drinking-Water
runoff as affected by land use and rainfall characteristics. Soil Till Res. 77, 111– Quality. Geneva, World Health, Organization (WHO/SDE/WSH/03.04/56).
123. World Health Organization (WHO), 2010. Guidance on water supply and sanitation
Jin, K., Cornelis, W.M., Gabriels, D., Baert, M., Wu, H.J., Schiettecatte, W., Cai, D.X., De in extreme weather events. In: Sinisi, L., Aertgeerts, R. (Eds.), World Health
Neve, S., Jin, J.Y., Hartmann, R., Hofman, G., 2009. Residue cover and rainfall Organization, Geneva, Switzerland, 105 pp.
intensity effects on runoff soil organic carbon losses. Catena 78, 81–86. World Health Organization (WHO) and Department For International Development
Katsoyiannis, A., Samara, C., 2007. The fate of dissolved organic carbon (DOC) in the (DFID), 2010. Vision 2030: the resilience of water supply and sanitation in the
wastewater treatment process and its importance in the removal of wastewater face of climate change at <http://www.who.int/water_sanitation_health/
contaminants. Environ. Sci. Pollut. Res. Int. 14 (5), 284–292. publications/9789241598422_cdrom/en/index.html> (accessed 05.01.11).
Kim, J.-H., Ludwig, W., Schouten, S., Kerherve, P., Herfort, L., Bonnin, J., Sinninghe Worrall, F., Burt, T.P., 2007. Trends in DOC concentration in Great Britain. J. Hydrol.
Damste, J.S., 2007. Impact of flood events on the transport of terrestrial organic 346, 81–92.
matter to the ocean: a study of the Têt River (SW France) using the BIT index. Wriedt, G., Rode, M., 2006. Modelling nitrate transport and turnover in a lowland
Org. Geochem. 38, 1593–1606. catchment system. J. Hydrol. 328, 157–176.
Le Bissonnais, Y., Renaux, B., Delouche, H., 1995. Interactions between soil Wysocki, L.A., Bianchi, T.S., Powell, R.T., Reuss, N., 2006. Spatial variability in the
properties and moisture content in crust formation, runoff and interrill coupling of organic carbon, nutrients, and phytoplankton pigments in surface
erosion from tilled loess soils. Catena 25, 33–46. waters and sediments of the Mississippi River plume. Estuar. Coast. Shelf Sci. 69,
Le Bissonnais, Y., Benkhadra, H., Chaplot, V., Fox, D., King, D., Daroussin, J., 1998. 47–63.
Crusting, runoff and sheet erosion on silty loamy soils at various scales and Yousefi, S., Ghiassi, R., 2011. Modeling and analyzing flow diversion in branching
upscaling from m2 to small catchments. Soil Till. Res. 46, 69–80. channels with symmetric geometry. River Res. Applic. 27 (7), 805–813.
Mérot, P., 2003. Small rural catchment functioning in the framework of floods and Zwolsman, J.J.G., van Bokhoven, A.J., 2007. Impact of summer droughts on water
rainfall–runoff events. La Houille Blanche 6, 1–9. quality of the Rhine River – a preview of climate change? Water Sci. Technol. 56,
Meybeck, M., 1982. Carbon, nitrogen and phosphorus transport by world rivers. Am. 44–55.
J. Sci. 282, 401–450.
Chemosphere 107 (2014) 115–120
Chemosphere
journal homepage: www.elsevier.com/locate/chemosphere
Revealing organic carbon–nitrate linear relationship from UV spectra

of freshwaters in agricultural environment
O. Thomas a,b,d,⇑, A.V. Jung a,b,c,d, J. Causse a,b,d, M.V. Louyer a,b,d, S. Piel e, E. Baurès a,b,d, M.F. Thomas c,d
a
EHESP Rennes, Sorbonne Paris Cité – CS 74312, 35043 Rennes Cedex, France
b
Inserm, U 1085 (IRSET), LERES – CS 74312, 35043 Rennes Cedex, France
c
School of Environmental Engineering (EME), Campus de Ker Lann, 35170 Bruz, France
d
European University of Brittany (UEB), France
e
Saur R&D – Les Cyclades, 1 rue Antoine Lavoisier, 78280 Guyancourt, France
h i g h l i g h t s
Normalization of UV spectra of freshwater samples to reveal hidden isosbestic point.

Simple linear model between nitrate and DOC, for agricultural environments.
Validation on more than 150 samples in rivers and lakes (Brittany, France).
Seasonal evolution of nitrate–carbon relationship.
Article history: A strong non linear relationship between nitrate and organic matter (assessed by dissolved organic
Received 21 August 2013 carbon, DOC) has been recently demonstrated by Taylor and Townsend (2010), namely for freshwaters.
Received in revised form 14 February 2014 In this context, our study explores this relation from the behavior of sets of normalized UV spectra (same
Accepted 1 March 2014
area under each spectrum) of different water samples showing a hidden isosbestic point (HIP) around
225 nm. This HIP is linked to the existence of a simple relation between nitrate and DOC, the proportions
Handling Editor: X. Cao of which vary according to the sampling location and environmental factors. In a second step, a simple
linear model is proposed for nitrate–DOC relationship (aNO3 + bDOC = 1) and a validation is proposed
Keywords: for more than 150 samples of different Brittany rivers and lakes. For samples of the largest watershed,
Water quality a complementary exploitation from data acquired during the different campaigns confirmed the seasonal
Agricultural environment evolution between spring (high nitrate/low DOC) and autumn (high DOC/low nitrate). Further investiga-
UV spectrophotometry tion on other freshwater samples is needed in order to improve the limits of this linear model.
Normalization Ó 2014 Elsevier Ltd. All rights reserved.
Model
1. Introduction and anthropogenic, the latter being related to point source (e.g.
treated wastewater discharge) and non-point source (e.g. animal
For more than one century, the characterization of organic matter waste spreading) pollution. NOM is known to be closely linked to
(OM) in natural waters has been and is still an important issue. From biogeochemical cycling of nutrients such as nitrogen forms (Woh-
BOD (biochemical oxygen demand) to COD (chemical oxygen lers-Zoellner et al., 2011; Fork and Heffernan, 2013). Decomposi-
demand) and TOC/DOC (total/dissolved organic carbon) measure- tion of NOM supplies both nutrients and energy for plants and
ment, these aggregate parameters have been completed by fraction- microbial organisms. Two models can be considered to explain
ation methods and optical methods such as UV spectrophotometry the growth and fate biomass in the environment (Allen and Gillo-
and fluorimetry for a better understanding of organic matter. oly, 2009): the metabolic theory of ecology (MTE) and the ecolog-
Dissolved organic matter (DOM) is the principal form of OM in ical stoichiometry theory (EST). Body size and temperature are the
surface waters. DOM is a mixture of natural organic matter (NOM) two main parameters governing MTE, whereas environmental fac-
tors and nutrient balance explain EST. Increasing nitrate concentra-
tions in rivers is one of the most current and challenging
⇑ Corresponding author at: EHESP Rennes, Sorbonne Paris Cité – Av. prof. environmental concern, in both developing and developed coun-
Leon-Brenard, CS 74312, 35043 Rennes Cedex, France. Tel.: +33 (0) 299 022 920. tries (Galloway et al., 2008). Nitrogen is a key nutrient driving
E-mail address: olivier.thomas@ehesp.fr (O. Thomas).
http://dx.doi.org/10.1016/j.chemosphere.2014.03.034
0045-6535/Ó 2014 Elsevier Ltd. All rights reserved.
116 O. Thomas et al. / Chemosphere 107 (2014) 115–120
eutrophication processes in surface waters: excess of nitrogen is temperature catalytic oxidation (NF EN 1484) and nitrate by flow
the main factor in the development of green algae in marine envi- analysis (CFA) and spectrometric detection (NF EN ISO 13395).
ronments and blue-green algae (cyanobacteria) in freshwaters
(Heisler et al., 2008). Regarding these two main components of
2.2. UV spectra acquisition
freshwaters -OM and nitrate- Taylor and Townsend (2010) have
recently shown consistent and negative nonlinear relationships
All UV spectra were acquired between 200 and 350 nm with a
between nitrate and DOC occurring along a hydrological contin-
10 mm SuprasilÒ quartz cell. The spectrophotometer used was a
uum from soils to water bodies (e.g. streams and lakes). In a recent
Lambda 35 Perkin–Elmer with a bandwidth of 2 nm, a scan speed
study focusing on the evolution of these parameters with the
around 2000 nm min1 and an acquisition step of 1 nm. When
hydrological regime of rivers, Baurès et al. (2013) demonstrated
signal saturated (absorbance above 2.5 u.a.), a dilution step was
opposite behavior of organic carbon and nitrate fluxes for high
carried out with water of ultra high quality (R > 18 MX), and the
flowrates (OC increased and nitrate decreased), and a strong
measured absorbance was corrected by the dilution factor.
nitrate uptake for low flows (in-stream biomass consumption).
Nutrient analysis has been undertaken for decades mainly for
eutrophication monitoring, with a great emphasis on nitrate mea- 2.3. Spectra normalization
surement. For the latter, the colorimetric method is often replaced
by ionic chromatography or even UV spectrophotometry for rapid All UV spectra acquired are normalized. The normalization step
assessment. consists in giving the same area to each spectrum (Vaillant et al.,
In this context UV spectrophotometry is commonly used for the 2002). Thus each absorbance value is corrected (or normalized) fol-
assessment of DOC and nitrate concentrations as well as water lowing equation:
quality variation (Thomas and Theraulaz, 2007) and should offer
a way to characterize the relationship between both parameters
A ðki Þ ¼ Aðki Þ F UVj ð1Þ
from the UV absorption spectrum shape of water samples.
where A⁄(ki) is the normalized absorbance at the wavelength ki, A(ki)
Although Thomas and Cerda (2007) described patterns in UV
is the measured absorbance at the wavelength ki, FUVj is the factor of
spectra for water and wastewaters, the proof of existing relation-
normalization of the jth UV spectrum (for a 10 mm pathlength cell
ships between OM and nitrate has not yet been demonstrated from
and an acquisition step of 1 nm) calculated from the following
UV spectra. , In this context, the aim of this study is to explore this
equation:
relation from the examination of sets of normalized UV spectra
(same area under each spectrum) of different water samples. N
F UVj ¼ ð2Þ
ASj
with F is the ratio between N, N the arbitrary area (norm) after nor-
2. Materials and methods
malization (e.g. 100) and ASj is the area under spectrum j in the UV
region. AS is the sum of absorbance values in the UV range:
2.1. Sampling and analysis P350
i¼200 Aj ðki Þ.
Several rivers and lakes characterized by an oceanic climate and

granite bedrock in Brittany (France) were sampled several times 2.4. Hidden isosbestic point
between 2009 and 2013 (see Fig. a as Supplementary information
and Table 1). As intensive agriculture (animal farming, cereal pro- In some cases, several UV spectra of a given set of samples can
duction, legumes and fodder crops) is the main economic activity cross together at a given wavelength. The presence of such a par-
in mainland Brittany, surface waters are known to be highly con- ticular point, called isosbestic point (Fig. 1), means that there is a
centrated in nitrate. Regarding water quality the mean concentra- fixed linear relation between the concentrations of the two absorb-
tions of DOC varied from 4 to 7.4 mgC L1 with a low coefficient of ing components or pseudo-components of the form: a1C1 + a2C2 = 1
variation except for the small rivers studied (Ic and Frémur). The (Pouet et al., 2004).
mean values of nitrate were high particularly for Frémur with high In other cases, an isosbestic point may appear after a normaliza-
coefficients of variation. tion step (Vaillant et al., 2002; Thomas et al., 2005). Nitrate can be
For each river, between 4 and 16 sampling sites were defined one of the components and DOM which is in fact a complex mix-
from their source to the mouth and between 6 and 9 sampling ture, can be considered as a pseudo component. In this case, the
campaigns were carried out. For lakes, samples were taken twice composition of all water samples presenting a HIP is characterized
during the period at their discharge point. Each sample was stored by the presence of the same components, the proportion of which
at low temperature after spot sampling, and analyzed the same varies according to the sampling location and environmental fac-
day. The water quality was assessed by the acquisition of UV spec- tors. On the contrary, if no HIP can be revealed, the composition
trum (see hereafter). For some samples, DOC and nitrate were of water samples is likely depending on a complex mixture of more
determined from standard methods. DOC was analyzed by high than two components.
Table 1
Brittany rivers and lakes studied: characteristics and water quality.
River Length, (Qmean m3 s1) Watershed (km2) Number of samples (stations) DOC mgC L1 mean (CV%) Nitrate mgNO3 L1 mean (CV%)
La Vilaine 225 km, (80) 11 000 55 (12) 6.1 (20) 14.0 (75)
L’Oust 145 km, (27) 3630 18 (4) 4.6 (21) 22.0 (56)
Le Meu 85 km, (3.1) 468 13 (3) 6.8 (25) 16.3 (81)
L’Ic 17 km (0.6) 86 116 (16) 4.0 (52) 40.5 (31)
Le Frémur 19 km (0.3) 77 176 (16) 7.4 (62) 44.0 (42)
Lakes – – 40 (26) 6.7 (35) 14.3 (66)
O. Thomas et al. / Chemosphere 107 (2014) 115–120 117
6 5
raw spectra 4
normalized spectra
4
a b
3
a.u.
a.u.
2
2
1
0 0
200 250 300 350 200 250 300 350
1.0 1.5
zoom a zoom b
0.8
a.u.
a.u.
0.6 1.0
HIP
0.4
0.2 0.5
220 230 240 250 220 225 230 235
Fig. 1. Raw and normalized UV spectra of some samples of the River Vilaine and zoom showing a hidden isosbestic point (HIP).
3. Results and discussion organic carbon concentration corresponding to the normalized

spectrum.
3.1. Highlighting organic matter–nitrate relationship from UV spectra For the jth sample, the concentration corresponding to the nor-
of freshwaters malized spectra (Cnj) is calculated from the following equation:
Cnj ¼ C j F UVj ð4Þ

UV spectra of samples of four surface rivers were normalized
and displayed in Fig. 2. All sets of spectra present a hidden isos- where Cj is the measured concentration, and FUVj is the factor of nor-
bestic point at the same wavelength (225 ± 1 nm). malization of the jth UV spectrum (for a 10 mm pathlength cell and
As previously explained the presence of a HIP is associated to a an acquisition step of 1 nm) calculated from Eq. (2).
simple relation between two absorbing component(s) or pseudo This linear model of Eq. (3) could be considered, if validated, as
component(s) (mixture), nitrate and DOC. The wavelength corre- a simplification of the main outcome of the work of Taylor and
sponding to the HIP (225 ± 1 nm) can be explained by the presence Townsend (2010), showing the existence of a negative nonlinear
of an inflexion point linked to the shape of nitrate spectrum and a correlation, namely for freshwaters.
model of DOM spectrum (Thomas and Theraulaz, 2007). Nitrate
spectrum shows a Gaussian shape centered on 205 nm and DOM
3.3. Model validation
presents a monotonous decreasing absorption on the whole spec-
tral range from 200 to 350 nm. In terms of concentration, nitrate
In order to check the relevance of the above linear model, a val-
and DOM (DOC) can be considered as the dominant components
idation was carried out from a set of 55 samples, from samples of
in freshwater, compared to other absorbing compounds such as or-
River Vilaine. Fig. 3 shows the UV spectra set before (3a) and after
ganic micropollutants. Thus, the principle UV absorbing interfer-
normalization (3b). The relation between nitrate and DOC is dis-
ences in river waters can be suspended solids and colloids in
played in Fig. 3c and d for the measured and normalized concentra-
case of heavy rain or flood. However their corresponding spectra
tions respectively. As previously shown in Fig. 1, a HIP occurs for
being very close to the one of DOM as studied by Berho et al.
wavelength around 225 nm (Fig. 3b) and a linear relation between
(2004), the general pattern with HIP around 225 nm is conserved.
nitrate and DOC normalized concentrations can be shown (Fig. 3d).
From the equation of the regression line (NO3 = 1.45DOC + 31.1),
3.2. Towards a simple description of organic carbon–nitrate the coefficients a and b of Eq. (3) can be easily calculated
relationship for freshwaters (a = 0.0321 and b = 0.0466).
Without normalisation, it can be observed that the scatter
Assessing that only DOM (DOC) and nitrate are responsible for graph of the relation between measured concentration of nitrate
UV spectra evolution of freshwaters if a hidden isosbestic point oc- and DOC (Fig. 3c) is comparable to the results shown by Taylor
curs around 225 nm (for a set of normalized spectra), a simple rela- and Townsend (2010), in the range of DOC concentrations between
tionship (Eq. (3)) can be used, as proposed by Pouet et al. (2004): 5 and 10 mgC L1, confirming the non linear relationship.
a NO3n þ b DOC n ¼ 1 ð3Þ 3.4. Discussion
where NO3n is the nitrate concentration corresponding to the nor- A first point is the generalization of the approach of spectra
malized spectrum (e.g. in mgN-NO3 L1) and DOCn is the dissolved normalization (Vaillant et al., 2002; Pouet et al., 2004) and its
Meu River Oust River

5 5
4 4
3 3
a.u.
a.u.
2 2
1 1
0 0
200 250 300 350 200 250 300 350
Fremur River Ic River

6 6
4 4
a.u.
a.u.
2 2
0 0
200 250 300 350 200 250 300 350
Fig. 2. Normalized UV spectra set of samples of four Britanny rivers, showing a hidden isosbestic point (HIP).
10 5
a b
8 4
6 3
a.u.
a.u.
4 2
2 1
0 0
200 250 300 350 200 250 300 350
40 30
c d
30
NO3n (mg/L)
NO3 (mg/L)
20
20
10
10
0 0
0 5 10 15 0 5 10 15 20 25
DOC (mg/L) DOCn (mg/L)
Fig. 3. Raw (a) and normalized (b) UV spectra of samples of River Vilaine and relation between nitrate and DOC concentrations before (c) and after (d) normalization of
concentrations. After normalization, a linear relation is obtained (R2 = 0.936, n = 55).
applications. Among them, a qualitative exploitation of UV spectra A second point to be discussed is UV window and norm value of
of water and wastewater in a non parametric way, without mea- the factor of normalization (Eq. (2)). For UV window, the lower lim-
suring water quality parameters such as DOC directly (Thomas it is imposed by spectrophotometer performances and oxygen
et al., 2005), and the characterization of non domestic load in absorption around 190 nm. The upper limit, 350 nm is commonly
urban wastewater networks (Baurès et al., 2007) were proposed. used for UV spectra, between organic matter absorption shoulder
O. Thomas et al. / Chemosphere 107 (2014) 115–120 119
Table 2
Values of nitrate and DOC coefficients of equations 3, 5 and 6.
Water bodies Nb of samples Equation (3) Equation (5) Equation 6

2 2
a (10 ) b (10 ) a 0
b 0
a (102) b (102)
River La Vilaine 43 3.21 4.66 2.68 36.4 1.46 12.80
River Le Meu 10 3.14 4.87 2.29 45.9 1.85 10.61
River L’Oust 15 2.97 5.43 2.48 35.7 1.37 15.21
River Le Fremur 20 2.73 7.62 2.03 77.2 0.91 9.87
River L’Ic 25 2.82 7.21 2.05 61.7 1.25 11.69
Brittany lakes 40 2.86 5.02 1.77 48.9 2.23 10.27
around 270–280 nm and the visible region (from 380 nm). Regard- being normally distributed (D’Agostino and Pearson test,
ing the norm value, chosen as an arbitrary value of 100, it deter- a = 0.05), the mean values of residues are not statistically different
mines the value of the factors of normalization for each spectra from zero. Thus the linear model can be considered as validated for
of a given waterbody and thus, of coefficients of the linear model. rivers and lakes in Brittany in intensive agricultural environment
It also determines intercept and abscissa as, for data of Fig. 3d, a should be verified for other freshwaters in different environments
concentration of nitrate of 31.15 mgNO3 L1 (=1/0.0321) when (e.g. natural with forests, peatland, etc.). In these environments,
DOC concentration is equal to zero, and a DOC value of the absence of allochtonous inputs of nitrate should likely state
21.46 mgC L1 (=1/0.0466) supposing that no nitrate is present in the limit of the model with nitrate concentrations lower than the
water. These values are different from those of Fig. 3c) and the quantification limit.
range of concentrations after normalization is doubled for DOCn Besides the need for in depth explanations of the coefficients of
while nitrate is rather constant (Fig. 3c and d). This can be ex- the linear model from one water body to another, a tentative
plained from the relation between the area under spectra (AS) understanding of the behavior of the nitrate-carbon relationship
and the corresponding values of DOC and nitrate concentrations in response to seasonality (including hydro-climatic factors) can
(see Fig. b as Supplementary information), showing a strong corre- be proposed. Starting from the five sampling campaigns carried
lation for nitrate. This response is linked to the high absorption of out between 2009 and 2012 on the River Vilaine, at different peri-
nitrate, compared to DOC for concentrations in freshwaters in agri- ods, Fig. 5 shows the mean values of nitrate and DOC normalized
cultural area and is not affected by the value of the norm. The rela- concentrations for the different campaigns. Independently from
tion is: the hydrologic conditions (wet weather for April and October cam-
0 paigns), the variation with seasonality follows the same pattern as
AS ¼ a0 NO3 þ b ð5Þ observed by Aubert et al. (2013) from the pluri-annual monitoring
Starting from Eq. (5), the factor of normalization (F = N/AS) can of water quality of a small watershed in the same environment as
be expressed as a function of nitrate concentration. Thus, it is pos- this study. The existence of an intra annual inverse variation of ni-
sible to express a linear model between DOC and nitrate from Eqs. trate and DOC concentrations with discharge and air temperature,
(3) and (5), without reference to the normalization factor: with relatively low concentrations of nitrate in autumn and DOC in
spring and reciprocally, is in close agreement with our findings and
a NO3 þ b DOC ¼ 1 ð6Þ emphasizes the importance of seasonality.
with a = (aN a0 )/b0 and b = bN/b0 . The above results can be interpreted with regard to the different
The key point of the discussion is the feasibility of the transpo- mechanisms described by Taylor and Townsend (2010), explaining
sition of the linear model for other freshwaters. For the purpose, the importance of the carbon:nitrate ratio in the accumulation of
the exploitation of nitrate, DOC and UV spectra was carried out nitrate, independently from allochtonous inputs of DOC from
for the 5 rivers and the 26 lakes (Table 1). The values of the differ- wastewater discharges (even usually treated) in urban area and ni-
ent coefficients of Eqs. (3), (5), and (6), corresponding to linear trate with fertilizers (organics and mineral) from agricultural
regressions with coefficients of determination equal or greater practices. For Taylor and Townsend, the ratio DOC/NO3 regulates
than 0.85, are reported in Table 2. The linear model may be vali- the microbial processes such as nitrification and denitrification.
dated from the examination of residuals between the result of Thus, for the spring campaigns (March and April) influenced by
the linear combination of nitrate and DOC concentrations and 1 agricultural activity with fertilizers spreading, mineral or organic
(Eq. (6)). Fig. 4 shows the distribution of residuals for the different (manure and slurry), the low level of carbon and high level of
waterbodies studied from box and whiskers plots. All data sets
30
0.50 March
Mean_NO3n (mg/L)
20
April
0.25
June
residuals
0.00 October
10
August-Sept
-0.25
0
0 5 10 15 20 25
-0.50
Vilaine Meu Oust Fremur Ic Lakes Mean_DOCn (mg/L)
Fig. 4. Box and whiskers plots of residues (r) of the linear model: aNO3 + Fig. 5. Relation between the mean values of normalized nitrate and DOC concen-
bDOC = 1 + r (box: 25%, median and 75%, whiskers: 5% and 95%, +: mean), for trations for the five sampling campaigns on River Vilaine between 2008 and 2012
Brittany freshwaters (France). (data from Fig. 3, n = 55).
nitrate (leading to low C/NO3, 0.2 for March and 0.3 for April) changed during rainy events. In the same way, the other forms of
would led to nitrate accrual and denitrification process trying to nitrogen (Norg and NH4 mainly) but also phosphorus are also
limit the increase of nitrate concentration. For the summer interested to consider, in order to improve the interactions under-
campaign (August–September) the samples are characterized by standing between nutrients cycles, by establishing nutrients
relative high DOC concentrations and low nitrate values corre- balances.
sponding to high C/NO3 (2 for August–September), given that
nitrate discharged by treated wastewaters is partially eliminated Acknowledgements
by nitrification/denitrification processes in treatment plants.
Another mechanism explaining the strong decrease of nitrate in This work was partially funded by the French National Agency
the summer is the greater biomass consumption due to more of Sanitary Security (ANSES) under the ERA-EnvHealth call
favorable the hydrometeorological conditions (nutrient, high water (FP7-ENV-2007-CSA-1.2.3-01), by the French Agency of Technological
temperature and high radiation). Besides allochtonous inputs of Research (ANRT) for PhD grants for S. Piel and J. Causse, and by the
nitrate in spring (and end of winter) and DOC in autumn (with soil SAUR and CDFO partners. The authors would like to thanks Ianis
and vegetal wastes runoffs) the ratio C/NO3 corresponds probably Delpla and all the staff of the LERES.
to the stochiometric demand of biomass as suggested by Taylor
and Townsend (2010) which could explain the significant decrease Appendix A. Supplementary material
of nitrates observed between March and September.
Finally the method described in this work should be tested in Supplementary data associated with this article can be found, in
other environments and for other water bodies in view of the the online version, at http://dx.doi.org/10.1016/j.chemosphere.
generalization of the findings. 2014.03.034.
4. Conclusion References
This work has been realized, in order to contribute to the under- Allen, A.P., Gillooly, J.F., 2009. Towards an integration of ecological stoichiometry
and the metabolic theory of ecology to better understand nutrient cycling. Ecol.
standing of the relationship between organic matter and nitrate in Lett. 12 (5), 369–384.
freshwaters in intensive agricultural environments where algae Aubert, A.H., Gascuel-Odoux, C., Merot, P., 2013. Annual hysteresis of water quality:
blooms may occur in waterbodies (rivers, lakes and coastal a method to analyse the effect of intra- and inter-annual climatic conditions. J.
Hydrol. 478, 29–39.
waters). If nowadays, different approaches are proposed to better Baurès, E., Hélias, E., Junqua, G., Thomas, O., 2007. Fast characterization of non
understand nutrient cycling, namely based on ecological theory domestic load in urban wastewater networks by UV spectrophotometry. J.
and stoichiometry, further studies are needed for improving nitrate Environ. Monit. 9 (9), 959–965.
Baurès, E., Delpla, I., Mérel, S., Jung, A.-V., Pétavy, F., Thomas, O., 2013. Variation
behavior in freshwaters with a methodology adapted to water of organic carbon and nitrate with river flow within an oceanic regime in a
quality management. In this context, the results presented in this rural area and potential impacts for drinking water production. J. Hydrol.
paper are firstly comparable to Taylor and Townsend’s recent 477, 86–93.
Berho, C., Pouet, M.-F., Bayle, S., Azéma, N., Thomas, O., 2004. Study of UV–Vis
findings, showing a consistent and negative nonlinear relation-
responses of mineral suspensions in water. Colloids Surf., A 248, 9–16.
ships between nitrate and DOC concentrations occurring along a Delpla, I., Baures, E., Jung, A.-V., Thomas, O., 2011. Impacts of rainfall events on
hydrological continuum from soils to water bodies (e.g. streams runoff water quality in an agricultural environment in temperate areas. Sci.
Total Environ. 409, 1683–1688.
and lakes). On the other hand, the significant contribution of this
Fork, M.L., Heffernan, J.B., 2013. Direct and indirect effects of dissolved organic
work is the demonstration of a linear model between nitrate and matter source and concentration on denitrification in northern Florida rivers.
DOC, thanks to the exploitation of UV spectra showing an isos- Ecosystems. http://dx.doi.org/10.1007/s10021-013-9705-9.
bestic point after normalization. The linear model is simple Galloway, J.N., Townsend, A.R., Erisman, J.W., Bekunda, M., Cai, Z., Freney, J.R.,
Martinelli, L.A., Seitzinger, S.P., Sutton, M.A., 2008. Transformation of the
(aNO3 + bDOC = 1) and has been validated on more than 150 sam- nitrogen cycle: recent trends questions and potential solutions. Science (New
ples of different Brittany rivers and lakes. Further investigation of York, N.Y.) 320, 889–892.
other freshwater samples is needed in order to determine more Heisler, J., Glibert, P.M., Burkholder, J.M., Anderson, D.M., Cochlan, W., Dennison,
W.C., Dortch, Q., Gobler, C.J., Heil, C.A., Humphries, E., Lewitus, A., Magnien, R.,
accurate limits of this linear model, particularly for natural envi- Marshall, H.G., Sellner, K., Stockwell, D.A., Stoecker, D.K., Suddleson, M., 2008.
ronments with no allochtonous nitrate inputs (forest, peatland, Eutrophication and harmful algal blooms: a scientific consensus. Harmful Algae
etc.). This important finding could help to predict the concentra- 8, 3–13.
Pouet, M.-F., Baures, E., Vaillant, S., Thomas, O., 2004. Hidden isosbestic point(s) in
tion of the two main components (nitrate and DOC) from the ultraviolet spectra. Appl. Spectrosc. 58 (4), 486–490.
exploitation of UV spectra that may constitute a potential tool Taylor, P.G., Townsend, A.R., 2010. Stoichiometric control of organic carbon–nitrate
for the survey of water quality. Another result is the normalized relationships from soils to the sea. Nature 464 (7292), 1178–1181.
Thomas, O., Baures, E., Pouet, M.-F., 2005. UV spectrophotometry as a non-
factor is linked to nitrate for freshwaters in agricultural areas, con-
parametric measurement of water and wastewater quality variability. Water
trary to DOC. Further investigation of the normalization factor but Qual. Res. J. Can. 40 (1), 51–58.
also of the C/NO3 ratio (often used) are necessary to explain more Thomas, O., Cerda, V., 2007. From spectra to qualitative and quantitative results. In:
Thomas, O., Burgess, C. (Eds.), UV–Visible Spectrophotometry for Water and
precisely the evolution of nitrate and DOC observed during seasons
Wastewater. Elsevier, Amsterdam, pp. 21–46.
and along a river from the source to the discharge, especially in Thomas, O., Theraulaz, F., 2007. Aggregate organic constituents. In: Thomas, O.,
agricultural environments. The variation of DOC/NO3 suggests the Burgess, C. (Eds.), UV–Visible Spectrophotometry for Water and Wastewater.
importance of environmental factors (like climate change) in addi- Elsevier, Amsterdam, pp. 89–114.
Vaillant, S., Pouet, M.F., Thomas, O., 2002. Basic handling of UV spectra for urban
tion to stoechiometric control, and organic matter origin. Finally water quality monitoring. Urban Water 4 (3), 273–281.
the different forms of organic carbon may be particularly impor- Wohlers-Zoellner, J., Breithaupt, P., Walther, K., Juergens, K., Riebesell, U., 2011.
tant to take into consideration since previous works (Delpla Temperature and nutrient stoichiometry interactively modulate organic matter
cycling in a pelagi algal–bacterial community. Limnol. Oceanogr. 56 (2), 599–
et al., 2011) have shown that DOC (dissolved), COC (colloidal) 610.
and POC (particulate) could see their distribution and repartition
Int. J. Environ. Res. Public Health 2014, 11, 7354-7375; doi:10.3390/ijerph110707354
OPEN ACCESS
International Journal of
Environmental Research and
Public Health
ISSN 1660-4601
www.mdpi.com/journal/ijerph
Article
A Decision Support System for Drinking Water Production

Integrating Health Risks Assessment
Ianis Delpla 1,2,3,4, Donald T. Monteith 5, Chris Freeman 6, Joris Haftka 7, Joop Hermens 7,
Timothy G. Jones 6, Estelle Baurès 1,2,3, Aude-Valérie Jung 1,2,3,8 and Olivier Thomas 1,2,3,*
1
Advanced School of Public Health, Sorbonne Paris Cité, Avenue du Professeur Léon Bernard,
Rennes Cedex CS 74312, France; E-Mails: ianis.delpla@crad.ulaval.ca (I.D.);
estelle.baures@ehesp.fr (E.B.); audevaleriejung@ecole-eme.fr (A.-V.J.)
2
Inserm, U 1085-Institute of Research in Environmental and Occupational Health (IRSET), LERES,
Rennes 1 University, Avenue du Professeur Léon Bernard, Rennes Cedex CS 74312, France
3
European University of Brittany, Boulevard Laënnec, Rennes 35000, France
4
École Supérieure d’aménagement du territoire et de Développement Régional, Laval University,
Pavillon F.A. Savard, Ste-Foy, QC G1K 7P4, Canada
5
Centre for Ecology and Hydrology, Lancaster University, Lancaster LA1 4AP, UK;
E-Mail: donm@ceh.ac.uk
6
Wolfson Carbon Capture Laboratory, School of Biological Sciences, Bangor University,
Deiniol Road, Bangor, Gwynedd LL57 2UW, UK; E-Mails: c.freeman@bangor.ac.uk (C.F.);
t.jones@bangor.ac.uk (T.G.J.)
7
Toxicology Division, Institute for Risk Assessment Sciences, Utrecht University, Utrecht 3508 TD,
The Netherlands; E-Mails: j.haftka@uu.nl (J.H.); j.hermens@uu.nl (J.H.)
8
School of Environmental Engineering, Campus de Ker Lann, Avenue Robert Schumann,
Bruz 35170, France
* Author to whom correspondence should be addressed; E-Mail: olivier.thomas@ehesp.fr;

Tel.: +33-2-99-02-29-20; fax: +33-2-99-02-29-29.
Received: 28 January 2014; in revised form: 7 July 2014 / Accepted: 8 July 2014 /
Published: 18 July 2014
Abstract: The issue of drinking water quality compliance in small and medium scale water
services is of paramount importance in relation to the 98/83/CE European Drinking Water
Directive (DWD). Additionally, concerns are being expressed over the implementation of
the DWD with respect to possible impacts on water quality from forecast changes in
European climate with global warming and further anticipated reductions in north European
Int. J. Environ. Res. Public Health 2014, 11 7355
acid emissions. Consequently, we have developed a decision support system (DSS) named
ARTEM-WQ (AwaReness Tool for the Evaluation and Mitigation of drinking
Water Quality issues resulting from environmental changes) to support decision making by
small and medium plant operators and other water stakeholders. ARTEM-WQ is based on
a sequential risk analysis approach that includes consideration of catchment characteristics,
climatic conditions and treatment operations. It provides a holistic evaluation of the water
system, while also assessing human health risks of organic contaminants potentially present
in treated waters (steroids, pharmaceuticals, pesticides, bisphenol-a, polychlorobiphenyls,
polycyclic aromatic hydrocarbons, petrochemical hydrocarbons and disinfection by-products;
n = 109). Moreover, the system provides recommendations for improvement while
supporting decision making in its widest context. The tool has been tested on various
European catchments and shows a promising potential to inform water managers of risks
and appropriate mitigative actions. Further improvements should include toxicological
knowledge advancement, environmental background pollutant concentrations and the
assessment of the impact of distribution systems on water quality variation.
Keywords: decision support system; drinking water; small and medium scale
water services; health risks assessment; climate change; organic carbon
1. Introduction
Drinking water systems supplied by surface waters are vulnerable to short-term variation in inputs
of organic matter that affect raw water quality. Increases in rainfall intensity and the frequency of
heavy rainfall events and droughts are predicted for the end of the century [1]. Such changes could
strongly affect water quality and treatment operations [1]. Rainfall events, for example, can cause
rapid degradation of water quality, including elevated levels of Total Suspended Solids,
Dissolved Organic Carbon (DOC) and Total Organic Carbon (TOC), nutrients, some micropollutants
with high partition coefficients, and microbiological parameters such as bacteria, viruses or protozoa [2].
Conversely, drought events and their aftermath can lead to algal blooms and associated increases in
cyanotoxins, sediment micropollutants release, diminution of dissolved oxygen, and of the river
dilution capacity for nutrients (ammonium, orthophosphates) or heavy metals [3–6]. To complicate
matters further, DOC concentrations have been rising steadily across many areas of northern and
central Europe and North America in recent decades, primarily in response to the influence of
declining acid deposition on organic matter solubility [7–11], and are of great concern for drinking
water producers. Changes in land-use practice (e.g., [12]) and climate [7,13–15] have also been
identified as potential drivers of DOC increases.
Future hydrological changes coupled with a long-term rise in organic matter solubility may
therefore increase loads of organic matter and associated micropollutants in ―raw water‖,
impairing water treatment efficiency and leading to a rise in disinfection by-products [16].
Large water supplies have generally robust and adaptive water treatment processes but there is
a concern about the adaptability of small water supplies.
Small and medium scale water services supplying water to fewer than 10,000 inhabitants are
commonplace in Europe and supply almost one third of the population, especially located in rural
areas [17,18]. In Europe, it was shown that more than one third of the small water supply systems
(SSWS) delivered water not complying with the values set in the 98/83/CE European Drinking Water
Directive (DWD) [19]. Technical and management difficulties, combined with a lack of financial
resources were identified as the main limitations to ensuring good water quality distributed by
SSWS [19,20]. Moreover, small supplies are often located in isolated areas, and operators lacking easy
access to expert assistance were shown to be highly vulnerable to sudden changes in raw water quality.
The DWD revision process, started in 2003, was concluded in 2011 with the statements that no
revision of legislation was required but that increased implementation and enforcement efforts using a
risk-based approach would be necessary to ensure safe drinking water in smaller supplies [19].
In this context, decision support systems (DSS) have the potential to provide valuable tools to enable
improved and informed management of SSWSs.
DSSs are defined as any system supporting decision making and include executive information
systems, executive support systems, geographic information systems, and on-line analytical processing
and software agents, and are normally underpinned by computer-based models [21]. During the last
few decades, many DSSs have been developed to inform water resource management issues such as
prevention of water shortages (droughts), surpluses (floods), and water quality impairment (pollution) [22].
Some DSSs have been designed to inform management of large rivers basins such as Elbe [23,24],
or big cities [25], while others have focused on tackling issues inherent to the high degree of
uncertainty existing in water systems due to their complexity [26–28].
Water quality-based DSSs are increasingly being developed to underpin the management and
protection of water resources under the Water Framework Directive [22,29–34]. Less attention has
been given to DSS development with respect to informing implementation of the DWD although there
is a clear need to establish holistic approaches in this field linking resources to the consumer.
However, some decision support tools and technical guides, principally based on a semi-quantitative
approach (Table 1), have recently been developed to assist managers of SSWSs in the assessment of
risks to their water systems [26,35–39].
These various tools or guides are mainly based on the water safety plan (WSP) approach defined by
the World Health Organization (WHO) and adopted worldwide. In its third edition of the guidelines
for drinking water quality, the WHO also concluded that a comprehensive risk management approach
is the most effective way of ensuring the safety of drinking water supplies [40]. The WSP approach
includes hazard ranking for qualitative (or semi-quantitative) risk assessments (RA). This is especially
relevant for small supplies, as the frequency of monitoring is usually low. Due to the potentially threat
to health posed by waterborne disease following the spread of pathogens within a distribution system,
and concerns over the ability to detect such outbreaks, these tools are mainly focused on
microbiological risk assessment and management. However, chemical risks posed by cyanotoxins,
nitrates and disinfection by-products may also be high. Moreover, new issues related to
micropollutants, emerging substances or environmental changes are not taken into account by these
tools/guides. Finally, the impact of further anticipated changes in DOC loads and concentrations on the
viability of a water treatment system need to be factored in. As concerns about health impacts related
to water consumption become increasingly societally important, there is a clear need for a management
tool capable of integrating this information.
Table 1. Summary of the decision support systems (DSS) designed for studies of small
water supply systems (SSWS).
Leading Main Water Supply Application
Year Name of The Tool Outputs
Organization Characteristics Size Range Environment
Agricultural
Guidance Risk scoring
WHO/IWA 2009 Water Safety Plan Manual 2500–8,500,000 (extensive)
document Control measures
Urban
Handbook on
implementation for Water
Semi Agricultural
Ireland (EPA) 2010 Services Authorities 2–5000 Risk scoring
quantitative RA Urban pressure
for private water
supplies—Section 10
All pressures
(wild life,
United
agriculture,
Kingdom Private water supplies : Semi 1–50 Risk scoring
2003 forestry
(Scottish Technical Manual quantitative RA 50–1000 Recommendations
industry,
executive)
wastewater,
sludge, landfill
Ogeris, aide à l’évaluation
des risques
Recommendations
France microbiologiques dans les Vulnerability
2009 <5000 All types and priorization of
(ASTEE) petites unités de assessment
actions
production/distribution
d’eau potable
Technical note for
guideline W 1001.
Germany DVGW Rules, security of Guidance Risk scoring
2008
(DVGW) water supply risk document Control measures
management during
normal operation
HACCP Strategies for
Large supplies
Distribution System Guidance Risk analysis and
USA (EPA) 2007 (1 example Urban pressure
Monitoring, Hazard document recommendations
770,000)
Assessment and Control
Identification of
University of Fuzzy logic and which failure
Fuzzy-Logic Modeling of Small water
Guelph 2009 fault tree All types contributes to the
Risk Assessment supplies
(Canada) methodology high potential risk
Recommendations
The ARTEM-WQ decision support system (AwaReness Tool for the Evaluation and Mitigation of
drinking Water Quality issues resulting from environmental changes) has been designed in light of
DWD requirements for SSWS, forecast changes in the European climate in response to global
warming and further anticipated reductions in north European acid emissions and resulting
DOC changes. Its main purpose is to provide water stakeholders (sanitary authorities, water treatment
operators and other stakeholders) with a holistic tool for identifying and assessing the risks posed by
the complex range of pressures (agricultural, industrial, climatic, etc.) on water resources.
Moreover, the tool should enable operators to specify plant characteristics in a consistent manner,
and inform on appropriate mitigation measures for enabling improvement of drinking water safety in
small and medium supplies. Finally, the tool is designed to assess the viability of a water treatment
system in the context of further anticipated changes in DOC loads and concentrations.
ARTEM-WQ is targeted primarily at water system managers and operators of small and medium
scale treatment plants, and particularly those responsible for plant construction or renovation.
However, it should also be of value in informing sanitary authorities responsible for sanitary quality
monitoring and river basin managers (e.g., Agence de l’eau, DDTM, ONEMA, DREAL).
2. Description of the Tool
2.1. General Architecture
The general architecture takes the following sequential approach:

1. Water resources risk analysis based on a score determined from data on catchment land-use and
land management.
2. Water treatment risk analysis.
3. Risk reduction recommendations to support decision making.
4. Health risk assessment.
From the outset, it was necessary to distinguish between systems dominated by waters draining
unimproved upland environments where riverine organic matter is chiefly derived from terrestrial
unfertilized primary production, and more anthropogenically impacted catchments, i.e., those draining
land supporting improved or intensive agriculture, urban settlement and industry.
Vulnerability is defined as the degree to which a system is susceptible to, and unable to cope with,
adverse effects of a hazard, such as climate change [1]. The tool requires completion of a sequence of
five mandatory questionnaires that together describe drinking water production system vulnerability.
The questionnaires are based primarily on hazard parameters derived from a review of existing
guidance documents used for risk assessment and risk management of water supplies (cf. Table 1).
These have been discussed and refined through meetings with project partners and local stakeholders.
A semi-quantitative risk scoring approach is defined as a combination of the likelihood of a hazard
and the severity of its consequences [38,39]. This approach is used to assess the overall risks resulting
from the presence and relative magnitude of the listed hazards, with higher scores indicating greater
potential severity of consequences for the safety of water supply. This allows identification and
assessment of all risks in the catchment and the treatment plant that may result in a risk to health
and/or a breach of the required standard [39]. Risks are accorded positive scores while risk reducing
measures such as treatment management actions or water quality improvement actions plans are
treated as negative.
Moreover, the DSS includes the capacity to predict DOC concentration for unimproved upland
catchments. The future target date chosen is the medium term 2050 horizon. This DOC model may be
used to predict average DOC concentration in unimproved upland catchments that are largely free of
soil erosion after completion of ―Natural catchment‖ and ―Treatment‖ questionnaires if the user
determines that the catchment is subject to only slight human impact (by selecting the appropriate box
in the ―catchment type‖ tab).
Others tool components include:
1. Graphical representation of scores.
2. Recommendations for catchment and treatment management improvement.
3. Estimates of likely pollutant concentration following treatment and associated Health Risk
Assessment (HRA).
Figure 1. General architecture of the ARTEM-WQ Decision Support System.
General informations
Catchment type
Anthropogenically Unimproved upland

Catchment
impacted areas areas
type
Catchment
characteristics «Anthropogenic » «Natural »
module Catchment Catchment
System
Treatment Assessment
Treatment
module module
Outputs:
- Contaminants selection
- System assessment
- Recommendations
DOC DOC N
data available data available
Y Y ―Current‖ DOC
modelling
(DOCn, 1)
Health Risks
DOC conversion ―Future‖ DOC Assessment
(DOCa,2=k*DOCa,1) modelling module
(DOCn, 2)
Final contaminants Guidelines

concentrations comparison
Health Risk
Assessment
The HRA focuses on potentially hazardous substances predicted from the pressures identified for the
catchment in question, the potential influence of raw water quality by considering DOC concentration,
and the impact of treatment on substances removal and disinfection by-products formation.
Chemical contaminants are generally present in very low concentrations in drinking water resources.
The effects of chronic exposure to small quantities of contaminants in drinking water have been the most
studied endpoint these last decades and lead to a consequent production of toxicological literature [41]
that is useable for conducting a human health risk assessment. Consequently, we decided to conduct
the HRA for carcinogenic (―non-threshold‖) effects following ingestion.
The DSS architecture is detailed in Figure 1.
2.2. Parameters
Dissolved Organic Carbon, the primary focus of this DSS, is a key water quality parameter for
drinking water production as its concentration and quality influence micropollutants adsorption,
Disinfection By-Product (DBP) formation and biological regrowth within the distribution system.
High levels of organic matter in raw waters sources are known to reduce clarifier performance,
and can also significantly increase drinking water treatment costs due to a higher chlorine and
coagulant demand [16].
Overall, a total of 109 organic contaminants were considered for inclusion in the DSS. We included
several frequently detected micropollutants of surface waters such as steroids (estradiol, estrone),
pharmaceuticals (carbamazepine, diazepam and triclosan), pesticides (e.g., atrazine, bromophos
methyl, cypermethrin, deltamethrin, permethrin, aldrin, dieldrin, DDD, DDE, HCH), bisphenol-a,
polychlorobiphenyls (e.g., PCB 77, 118, 126, 156, 169, 170, 180), polycyclic aromatic hydrocarbons
(e.g., phenanthrene, pyrene, benz(a)anthracene) and petrochemical hydrocarbons
(e.g., 7,12-dimethylbenz(a)anthracene, nonane). Disinfection by-products such as the four
THM species (chloroform, bromoform, dichlorobromomethane and chlorodibromomethane)
and two haloacetic acids (dichloroacetate, trichloroacetate) were also included.
Where no DOC measurements were available for waters draining unimproved upland habitats that
are not subject to significant soil erosion, the DSS allows estimation of average concentrations using a
simple statistical model based on the relationship between long-term mean DOC concentration and
physico-geographic data representing mean catchment altitude, percentage peatland extent,
and effective precipitation (i.e., precipitation minus evaporation) according to [42].
2.3. Main Functionalities
2.3.1. Catchment Type Module
This questionnaire is used to classify the river basin according to dominant environmental pressures
into either anthropogenically impacted or anthropogenic and ―natural‖; on the basis of land use.
This simple river basin characterization system represents the first step in the tool, directing the user
toward further questionnaires.
2.3.2. Catchment Characteristics Module
Completion of this questionnaire generates an assessment of drinking water catchment area

vulnerability. The questionnaire content was adapted from [35]. It requires data on the type of land
use, animal density, agricultural practices in the catchment (farm waste spreading/storage, etc.),
urban and industrial pressures (industries, hospitals, tourism infrastructures, WWTP, landfill sites,
storm overflows, etc.), catchment characteristics (slopes) and management measures. The questionnaire’s
contents are generally similar for both major catchment types, but with additional data required for the
impacted systems. A description of the considered catchment characteristics with their corresponding
scores is given in Supplementary Information.
2.3.3. Treatment Module
Here, the efficiency of the drinking water treatment plant (DWTP) system is evaluated on the basis
of data on monitoring measures, treatment performance and routine plant exploitation
(inspections, control manuals, bypass). The questionnaire content was adapted from [35]. A total of
fifteen of the most commonly applied treatments steps are included, e.g., sand filtration, coagulation,
activated carbon (granular/powder), membrane filtration, disinfection (chlorination, UV, ozonation).
In this questionnaire, treatment steps and performance, appropriate monitoring measures, and adequate
plant management are considered as risk-reducing measures and therefore generate negative scores.
A description of the treatment questionnaire with its corresponding scores is given in
Supplementary Information (Tables S1–S3).
2.3.4. System Assessment Module
Scores generated by the ―Catchment characteristics‖ and ―Treatment‖ questionnaires are summed,
as it was commonly done in other DSS [35,36]. Firstly, the net score is graphically presented,
associated with a recommendation according to the net score value. The following algorithm is used
for water quality assessment (Figure 2).
Figure 2. Water quality assessment tree used in ARTEM-WQ showing the three different
system assessment scores’ thresholds (20, 40 and 60).
>40
Y N
>60 >20
Y N Y N
Poor Vulnerable Acceptable Good
The thresholds were defined following a sensitivity analysis of the sum of risk scoring provided by
the DSS by testing the tool in two extremes situations (pristine catchments with adequate water
treatment and management; substantially polluted catchments with insufficient water treatment and
management). The scores range from 0–94. Data for a considerably wider range of sites will be
necessary to provide a robust calibration of the risk score (see Discussion), but the example in Figure 2
is based on the assumption that it is proportional to the risk of delivering an unsafe drinking water to
the consumer. Consequently, in this example the score range has been divided equally into four water
quality classes (0–20, 21–40, 41–60 and >60).
Figure 3. Screenshot of the System Assessment tab.

An example of the System Assessment output is provided in Figure 3. The bars in the charts convey
scores for chief sub-categories determined by the Catchment and Treatment questionnaires.
Positives scores are marked in orange, while scores of zero or below are provided numerically.
This presentation allows the user to rapidly assess the key potential pressures and a range of water
treatment operations that might be improved (Figure 3).
A corrective action is proposed for each potential hazard identified on the water system
(i.e., for positive risk scores), such as high animal density, discharges from wastewater treatment
plants, or inadequate water treatment plant management. Corrective actions cover the following fields:
river basin management, water treatment process design, maintenance, performance and monitoring.
These corrective measures include on one hand primary recommendations which are strongly
suggested to be put in place, and on the other hand secondary recommendations which could be put in
place further or are not mandatory by the regulation. For instance, if there is evidence of storm water
overflows previously occurring in the catchment, the primary recommendation given by the software is
to implement a buffer tank and the secondary recommendation is to implement at the intake point an
alert system based on online measurement for global parameters of water quality (Figure 4).
Figure 4. Screenshot of the Recommendations tab.
2.3.5. Health Risk Assessment Module
The objective of this module is to highlight the substances that could present a health risk.
Contaminant concentrations are calculated and a Health Risk Assessment is performed for each
substance. The module requires input data on DOC concentrations that are derived either directly from
measurements or by modeling. Then, the calculation of organic contaminants’ removal following

water treatment is achieved. Finally, the final concentrations obtained are used as inputs to calculate
individual estimates of risks for each substance following an exposure by ingestion.
2.3.5.1. Contaminant Selection
Potentially harmful contaminants are selected according to the pressures identified on the river
basin in the ―Catchment‖ questionnaire. Each pressure is linked with one or more groups of substances
(Ex: PAH substances with ―Presence of industries‖). If the score associated with one pressure is strictly
positive, all substances potentially present and associated to this pressure are taken into account.
2.3.5.2. Dissolved Organic Carbon Concentration Calculation
This module is intended to allow the estimation of contemporary DOC concentration in unimproved
upland areas where DOC data are absent, in addition to providing estimates of likely future levels on
the basis of IPCC climate change scenarios.
In the case of using the DSS for unimproved upland environments (―Natural catchments‖ in Figure 1)
and when DOC data are not available, a dialog box allows the estimation of DOC concentrations from
geo-physical data (DOC model). The current DOC model provides an estimate of contemporary
DOC concentrations (i.e., at the time of the DSS use) but is also being developed to predict long term
trends according to current climate change scenarios (2050) in natural environments in the UK region.
Input variables are based on parameters identified as dominating spatial variation in
DOC concentrations variations in UK upland catchments according to [42].
1. Mean catchment altitude.
2. Percentage of peat or peaty gley soil in the catchment.
3. Annual effective rainfall (i.e., rain minus evaporation).
When using the DSS for directly anthropogenically impacted environments (―Anthropogenically
impacted catchment‖ in Figure 1), the user needs to provide DOC data for pursuing the Health Risk
Assessment. In order to provide a potential future scenario DOC concentration, we assumed a worst
case scenario stating a 50% increase in contemporary levels, on the assumption that trends experienced
in Brittany [43] over the last 30 years will continue into the future at the current rate. Consequently,
for long term trends’ concentrations (2050), the following equation was used: DOCa,2 = k × DOCa,1,
(with k = 1.5, DOCa,1: DOCconcentration provided by the user).
2.3.5.3. Calculation of Organic Contaminants Removal Following Water Treatment
In order to remove dissolved (and colloidal) organic carbon (DOC), coagulation and subsequent
flocculation precedes most water treatment operations. Hydrophobic organic contaminants sorbed to
DOC will also be removed from solution by coagulation/flocculation processes [44]. Organic chemicals
that are strongly sorbed to the hydrophobic fraction of DOC will therefore be more efficiently removed
by coagulation/flocculation compared to hydrophilic compounds that are weakly sorbed to DOC.The
removal efficiency of organic contaminants will also be dependent on the hydrophobicity of the
compound with higher removal efficiencies observed for chemicals with higher hydrophobicity [45]. It
is assumed in this work that the organic contaminant is solely sorbed to the hydrophobic fraction of
DOC and there is no appreciable sorption of freely dissolved chemicals to the precipitated flocs or to
the residual inorganic coagulants.
The removal of an organic contaminant sorbed to DOC was subsequently calculated with the
following equation:
𝐾DOC ∗ 𝐷𝑂𝐶
𝑐res ,x = 𝑐i,x 1 − Re (1)
1 + 𝐾DOC ∗ 𝐷𝑂𝐶 DOC
where cres,x is the residual contaminant concentration after coagulation of compound x, ci,x is the initial
contaminant concentration in the raw water, KDOC is the DOC to water partition coefficient,
DOC is the DOC concentration of raw water and ReDOC is the removal efficiency of DOC
following coagulation.
Other water treatment steps can subsequently remove the residual contaminant concentration after
the coagulation/flocculation process. Water treatment can include e.g., sand filtration, (pre-)disinfection
with either UV/H2O2, O3, Cl2 or NaClO2, filtration with granular activated carbon and finally
(post-) disinfection. Different removal efficiencies (e.g., for treatment a and b following
coagulation/flocculation) are combined, according to the following equation:
𝐾DOC ∗ DOC
𝑐f,x = 𝑐i,x 1 − Re 1 − Rea 1 − Reb (2)
1 + 𝐾DOC ∗ 𝐷𝑂𝐶 DOC
where cf,x is the final contaminant concentration in finished drinking water, and Rea,b is the removal
efficiency of a specific organic contaminant and water treatment.
Figure 5. Influence of Dissolved Organic Carbon on the calculation of initial and final
concentrations during preparation of drinking water via removal of (1) DOC sorbed and
residual contaminant concentration; (2) freely dissolved contaminant concentration;
and (3) total contaminant concentration.
Initial concentration,
Ci,x
Removal of DOC
sorbed concentration
No
Residual total
concentration,Cres,x coagulation
Removal of residual Removal of freely Removal of total

total concentration dissolved concentration concentration
Final concentration
Cf,x
The removal of an organic contaminant following a sequence of different water treatment processes
is schematically shown in Figure 5, where the removal via coagulation and any other water treatment
is shown in the left of the figure.
In case there is no coagulation/flocculation step at a water treatment plant, organic contaminants may
also be removed via the freely dissolved or the total concentration, respectively (Figure 5).
The final concentration after removal of the freely dissolved concentration from solution can theoretically
be calculated but there is no literature data available on the removal efficiencies of freely dissolved
contaminants. For this reason, removal of freely dissolved concentration was not considered here.
This step has required an extensive dataset constitution for grouping relevant data on environmental
background chemicals concentrations, organic contaminant sorption to DOC (KDOC) and treatment
efficiency (removal rates). KDOC values were obtained following sorption experiments conducted by
the Utrecht University (data not shown). Data on background concentrations and removal rates by
treatment was obtained following a large literature review [46–64].
When a group of substances is suspected to be found in raw waters, the DSS synthesize all initial
concentration values (Ci,x), corresponding values of KDOC (KDOC,x), DOC concentration in treated
waters (DOCtheo2, theoretical or measured, after validation of ―Treatment‖ questionnaire)
and the removal coefficient(s) corresponding to the water treatment(s) selected by the user
(―Treatment‖ questionnaire), in order to calculate final concentration values (after treatment).
2.3.5.4. Calculation of Disinfection By-Product Concentrations
The calculation of DBPs concentration (4 THM, 2 HAA) is realized by compound-specific equations:

For THM, the following equation is used [65]:
TTHM  1.392(DOC)1.092 (pH)0.531 (T)0.255 (3)
where TTHM = Total Trihalomethanes (µg/L), DOC = Dissolved Organic Carbon concentration (mg/L),
T = Mean Water Temperature at Montfort DWTP (°C) = 12.8 °C, pH = Mean pH value at
Montfort DWTP = 7.8.
A partition coefficient is also applied for the calculation of each THM substance. These coefficients
are taken from the results of the PollProx study (219 samples in Rennes area) [66].
In this survey, the four THM are distributed as follows:
Chloroform: 23.5%
Bromoform: 28.2%
Dibromochloromethane: 30.2%
Bromodichloromethane: 18.4%
Hence, the formula becomes:
Chloroform = 1.6 × (DOCtreated water)1.092
Bromoform = 1.92 × (DOCtreated water)1.092
Dibromochloromethane = 2.057 × (DOCtreated water)1.092
Bromodichloromethane = 1.253 × (DOCtreated water)1.092
For the HAA calculation, the following formula is used [67]:

Dichloroacetate = 1.534 + 0.566 × (Bromodichloromethane) − 0.258 × (Bromoform)
Trichloroacetate = 1.269 + 0.375 × (Chloroform)
2.3.5.5. Individual Excess Risks Calculation (Non-Thresholds Effects)
The last step involves the calculation of Individual Excess Risks (IER) for each substance for
carcinogenic (―non-threshold‖) effects for an adult following a lifetime exposure by ingestion.
Since the selected substances are very different in structure and toxicity, the quantitative health risk
assessment is performed for each one separately. Relevant toxicological information was collected by
consulting ten different toxicological databases (Integrated Risk Information System-USEPA,
Agency for Toxic Substances and Disease Registry, Health Canada, International Toxicity Estimates
for Risk-TERA, Office of Environmental Health Hazard Assessment, Dutch National Institute for
Public Health and the Environment-RIVM, Risk Assessment Information System, Hazard Substances
DataBase, Ineris and Furetox-InVS).
The following equation, derived from [68], is used for the IER calculation:
𝐶f,x ∗ 𝑉𝑤𝑎𝑡𝑒𝑟 ∗ 𝑇𝑝 ∗ 𝑇𝑅𝑉
𝐼𝐸𝑅 = (4)
𝑀∗𝑇
where Cf,x = Substance final concentration (THM, other, …) (mg/L), TRV = Toxicological Reference
Value (mg/kg/d)−1, Vwater = Daily water consumption, set to 2 L/day, M = Mean adult body weight,
set to 70 kg, T = Lifetime, set to 70 years, Tp = Exposure duration (ingestion), set to 70 years.
A mean exposure to 2 L/day of water during the lifetime (70 years) was considered.
Figure 6. Screenshot of the Health Risk Assessment tab.

Results of the quantitative health risk assessment are presented for individual substances.
The software highlights within a new tab risk values higher than 10−5 in red and the values lower than
10−5 in green. This threshold corresponds to the guideline set by WHO for defining risk as
―acceptable‖ [69]. These substances are ordered in IER descending order.
A screenshot of the Health Risk Assessment tab is presented in Figure 6.
3. DSS Validation
The methodology used in the DSS was validated by applying it to eight water systems located in
Brittany (France). Sites were selected to span a range of key pressures and on the basis of available
data. The systems covered a wide range of river basins environments ranging from natural forested
ecosystems to highly intensive agricultural zones and urbanized catchment. Among the catchments,
six are used as a water resource to produce drinking water.
The proportion of agricultural land varies from 16.1% for the less human-impacted catchment to
96.1% for the site with the most intensive agricultural practices. The water treatments are relatively
complex for all the sites and all include different steps such as coagulation, chlorination, ozonation and
granular activated carbon filtration and/or sand filtration. Data on water treatments was obtained for
five sites, corresponding to the most anthropized environments.
Questionnaires scoring and substances selection were validated separately.
Firstly, the risk analysis was conducted on the eight test catchments and compared with mean
DOC concentrations obtained from source waters data released by environmental and sanitary
monitoring networks between 1993 and 2011. The DOC parameter was chosen as it could be
considered as a proxy of raw sources water quality (Figure 7). While there is no significant linear
correlation, higher DOC concentrations are generally associated with higher risk scores.
Figure 7. Relationship between score (catchment characteristics questionnaire) and DOC

(n = 8, R2 = 0.28).
10
8
DOC (mg/L)
0
0 20 40 60 80
Score (Catchment)
Secondly, for the purpose of validating the assessment of risk posed by selected substances,
the data were compared with field data collected by the environment and health authorities.
The majority of substances selected based on existing pressures on the catchment were detected in
source waters. Moreover, the substances that are not selected by the DSS are effectively not present
in raw waters, proving that the contaminant selection procedure does not omit potentially harmful
contaminants (Table 2).
Table 2. Comparison between contaminants selected by the DSS and those currently
detected in waters.
Number of Contaminants Number of Contaminants
Sampling Contaminants Contaminants
Selected by the DSS Non Selected by the DSS
Site Detected * (%) Detected * (%)
and Analyzed * and Analyzed *
1 11 45.5 0 -
2 12 91.7 0 -
3 12 66.7 0 -
4 0 - 20 0.0
5 12 91.7 7 0.0
6 0 - 20 10.0
7 12 16.7 14 0.0
8 11 9.1 12 0.0
9 14 100 6 0.0
Note: * during the field sampling campaigns.
4. Discussion
By assessing the vulnerability of a water system using a semi-quantitative approach combined with
substances identification and health risk assessment, our decision support system identifies potential
problems and attests to the systems suitability for assessing the status of water quality.
Moreover, the statistically-based DOC model provides a useful guide to key factors and processes
determining DOC loadings to water treatment plants draining natural upland catchments.
Future iterations should enable responses to future environmental change to be assessed,
thus providing significant economic benefits to the water industry by informing planning procedures.
A number of issues require further development and testing before the DSS is suitable for
widespread use by water treatment managers. First, the scoring system used to rate individual criteria
stipulated in the catchment questionnaire and drawn largely from the existing literature requires more
methodical ground-truthing to ensure that appropriate weighting has been applied.
Second, while the statistical modeling of spatial variation in DOC for natural environments appears
to indicate significant roles for both temperature and precipitation, it is still unclear to what extent
these factors will influence trends at individual sites over time. Further work is ongoing to determine
their importance, and quantification of these effects will be necessary before a dynamic climate change
element can be incorporated.
Third, the calculation of contaminant concentrations in treated waters is based on removal rates which
are determined following a literature review of existing knowledge, some values then could be modified
in the near future. Moreover, this calculation is also based on environmental concentrations available
in the literature and could not reflect the existing background concentrations for a given catchment.
Fourth, the DSS is limited to specific substances for which toxicological data exist.
An improvement in knowledge for emerging substances is needed to complete the database,
such as for instance, data for antibiotics and anticancer drugs. The DSS will be improved as new
toxicological information will be released.
Finally, the physical structure of the water distribution network is likely to exert a significant
influence on variation in water quality. Residence times and material contact could lead to an increase
of disinfection by-products formation potential, especially for high water temperatures.
It has been shown that THM concentrations could significantly vary from 2–4 times within the
distribution system [70]. The residence time could also play a role on biofilm and microorganism’s
development [71]. Furthermore, the material from which water pipes are constructed should be
considered with respect to the potential to the release of endocrine disrupting substances such as
organotins [72] and carcinogenic substances such as vinyl chloride monomer [73].
5. Conclusions
The issue of drinking water quality compliance in small and medium scale water services is of
paramount importance in relation to the requirements of the DWD and the existing difficulties in its
implementation within such distribution systems. Various incentives such as free training, advice and
tools play a key role for successful DWD implementation in SSWS. In order to help water authorities
and managers tackle issues regarding the microbiological, chemical and climate risks in their water
treatment plants, a decision support system has been defined and tested on different catchments
representing a wide range of environments. ARTEM-WQ has been developed using a sequential
approach based on a risk analysis in conjunction with catchment characteristics and treatment
operations. The DSS allows operators to evaluate information on the current global status of the water
system, while also assessing human health risks of substances potentially present in finished waters.
In combination, this information allows formulation of recommendations for improvement while
supporting decision making in its widest context. Initial testing of the tool in various catchments shows
promising potential to inform water managers of risks and appropriate mitigative actions.
Further improvements are now needed, however, including advancement of toxicological knowledge,
environmental background pollutant concentrations and their respective removal rates.
Moreover, the impact of distribution systems on water quality variation should also be included
in the next version of the DSS.
Acknowledgments
This work was funded by the French National Agency of Sanitary Security (ANSES),
UK Natural Environment Research Council (NERC) and VROM in the Netherlands under the
first ERA-EnvHealth call (FP7-ENV-2007- CSA-1.2.3-01). The ERA EnvHealth project aims to
improve knowledge on the implications of recent rising trends in Dissolved Organic Carbon (DOC)
in surface water resources for human health and to propose a decision support system. The authors also
wish to thank Sylvain D’agata for his fruitful participation in this work.
Author Contributions
Ianis Delpla participates to the definition, implementation and testing of the DSS and to the writing
of the paper. Don T. Monteith construct the DOC model, and participates to the definition of the DSS
architecture and the writing of the paper. Tim G. Jones and Joris Haftka participates to the definition of
the DSS architecture and the writing of the paper. Olivier Thomas, Chris Freeman, Joop Hermens,
Estelle Baurès and Aude-Valérie Jung participates to the definition of the DSS architecture.
Conflicts of Interest
The authors declare no conflict of interest.
References
1. Climate Change and Water, Technical Paper of the Intergovernmental Panel on Climate Change;
Bates, B., Kundzewicz, Z.W., Wu, S., Palutikof, J., Eds.; IPCC Secretariat: Geneva, Switzerland,
2008; p. 210.
2. Schijven, J.; Bouwknegt, M.; de Roda Husman, A.M.; Rutje, S.; Sudr, B.; Suk, J.E.; Semenza, J.C.
A decision support tool to compare waterborne and foodborne infection and/or illness risks
associated with climate change. Risk Anal. 2013, 33, 2154–2167.
3. Curriero, F.C.; Patz, J.A.; Rose, J.B.; Lele, S. The association between extreme precipitation and
waterborne disease outbreaks in the United States, 1948–1994. Amer. J. Public Health 2001, 91,
1194–1199.
4. Delpla, I.; Jung, A.-V.; Baurès, E.; Clément, M.; Thomas, O. Impacts of climate change on
surface water quality in relation to drinking water production. Environ. Int. 2009, 35, 1225–1233.
5. Gallina, N.; Anneville, O.; Beniston, M. Impacts of extreme air temperatures on cyanobacteria
in five deep peri-Alpine lakes. J. Limnol. 2011, 70, 186–196.
6. Van Vliet, M.T.H.; Zwolsman, J.J.G. Impact of summer droughts on the water quality of the
Meuse river. J. Hydrol. 2008, 353, 1–17.
7. Freeman, C.; Evans, C.D.; Monteith, D.T.; Reynolds, B.; Fenner, N. Export of organic carbon
from peat soils. Nature 2001, 412, doi:10.1038/35090628.
8. Worrall, F.; Burt, T.; Adamson, J. Can climate change explain increases in DOC flux from upland
peat catchments? Sci. Total Environ. 2004, 326, 95–112.
9. Monteith, D.T.; Stoddard, J.L.; Evans, C.D.; de Wit, H.A.; Forsius, M.; Høgåsen, T.; Wilander, A.;
Skjelkvåle, B.L.; Jeffries, D.S.; Vuorenmaa, J.; et al. Dissolved organic carbon trends resulting
from changes in atmospheric deposition chemistry. Nature 2007, 450, 537–541.
10. Oulehle, F.; Hruška, J. Rising trends of dissolved organic matter in drinking-water reservoirs as a
result of recovery from acidification in the Ore Mts., Czech Republic. Environ. Pollut. 2009, 157,
3433–3439.
11. Evans, C.D.; Jones, T.G.; Burden, A.; Ostle, N.J.; Zieliński, P.; Cooper, M.D.A.; Peacock, M.;
Clark, J.M.; Oulehle, F.; Cooper, D.; et al. Acidity controls on dissolved organic carbon mobility
in organic soils. Glob. Chang. Biol. 2012, 18, 3317–3331.
12. Gough, R.; Holliman, P.J.; Willis, N.; Jones, T.G.; Freeman, C. Influence of habitat on the
quantity and composition of leachable carbon in the O2 horizon: Potential implications for
potable water treatment. Lake Reserv. Manag. 2012, 28, 282–292.
13. Freeman, C.; Fenner, N.; Ostle, N.J.; Kang, H.; Dowrick, D.J.; Reynolds, B.; Lock, M.A.; Sleep, D.;
Hughes, S.; Hudson, J. Dissolved organic carbon export from peatlands under elevated carbon
dioxide levels. Nature 2004, 430, 195–198.
14. Fenner, N.; Freeman, C. Drought-induced carbon loss in peatlands. Nat. Geosci. 2011, 4, 895–900.
15. Tang, R.; Clark, J.M.; Bond, T.; Graham, N.; Hughes, D.; Freeman, C. Assessment of potential
climate change impacts on peatland dissolved organic carbon release and drinking water treatment
from laboratory experiments. Environ. Pollut. 2013, 173, 270–277.
16. Matilainen, A.; Sillanpää, M. Removal of natural organic matter from drinking water by advanced
oxidation processes. Chemosphere 2010, 80, 351–365.
17. Report on Water Quality in France 2005–2006; DGS (French Health Directorate): Paris, France,
2008; p. 66.
18. Small-Scale Water Supplies in the pan-European Region: Background—Challenges—
Improvements; WHO-United Nations Economic Commission for Europe (UNECE): Copenhagen,
Denmark, 2011; p. 54.
19. Bloech, H. Summary Record of the Meeting of the Committee under Article 12 of Directive
98/83/EC; Directorate-General for the Environment, Brussels, Belgium, 2011.
20. Guidelines for Drinking-Water Quality, 3rd ed.; Incorporating The First and Second Addenda;
World Health Organization: Geneva, Switzerland, 2008; Volume 1, p. 668.
21. Roig, B.; Delpla, I.; Baurès, E.; Jung, A.V.; Thomas, O. Analytical issues for drinking water
contamination monitoring related to short term intense rainy events. Trends Anal. Chem. 2011,
30, 1243–1251.
22. Giupponi, C. Decision support systems for implementing the european water framework
directive: The MULINO approach. Environ. Model. Softw. 2007, 22, 248–258.
23. Berlekamp, J.; Lautenbach, S.; Graf, N.; Reimer, S.; Matthies, M. Integration of MONERIS and
GREAT-ER in the decision support system for the German Elbe river basin. Environ. Model. Softw.
2007, 22, 239–247.
24. Lautenbach, S.; Berlekamp, J.; Graf, N.; Seppelt, R.; Matthies, M. Scenario analysis and management
options for sustainable river basin management: Application of the Elbe DSS. Environ. Model. Softw.
2009, 24, 26–43.
25. Zeng, Y.; Cai, Y.; Jia, P.; Jee, H. Development of a web-based decision support system for
supporting integrated water resources management in Daegu city, South Korea. Expert Syst. Appl.
2012, 39, 10091–10102.
26. Lee, S.; Ni-Mesister, W.; Toll, D.; Nigro, J.; Gutierrez-Magness, A.L.; Engman, T. Assessing the
hydrologic performance of the EPA’s nonpoint source water quality assessment decision support
tool using North American Land Data Assimilation System (NLDAS) products. J. Hydrol. 2010,
387, 212–220.
27. Xu, Y.-P.; Tung, Y.-K. Decision rules for water resources management under uncertainty.
J. Water Resour. Plan. Man-Asce. 2009, 135, 149–159.
28. Zhang, X.; Huang, G.H.; Nie, X.; Lin, Q. Model-based decision support system for water quality
management under hybrid uncertainty. Expert Syst. Appl. 2011, 38, 2809–2816.
29. Arnold, J.G.; Fohrer, N. SWAT2000: current capabilities and research opportunities in applied
watershed modelling. Hydrol. Process. 2005, 19, 563–572.
30. Even, S.; Billen, G.; Bacq, N.; Théry, S.; Ruelland, D.; Garnier, J.; Cugier, P.; Poulin, M.; Blanc, S.;
Lamy, F.; et al. New tools for modelling water quality of hydrosystems: An application in the
Seine River basin in the frame of the Water Framework Directive. Sci. Total Environ. 2007, 375,
274–291.
31. Mirauda, D.; Ostoich, M. Surface water vulnerability assessment applying the integrity model as a
decision support system for quality improvement. Environ. Impact Assess. 2011, 31, 161–171.
32. Mysiak, J.; Giupponi, C.; Rosato, P. Towards the development of a decision support system for
water resource management. Environ. Model. Softw. 2005, 20, 203–214.
33. Paredes, J.; Andreu, J.; Solera, A. A decision support system for water quality issues in the
Manzanares river (Madrid, Spain). Sci. Total Environ. 2010, 408, 2576–2589.
34. Sieker, H.; Bandermann, S.; Schröter, K.; Ostrowski, M.; Leichtfuss, A.; Schmidt, W.; Thiel, E.;
Peters, C.; Mühleck, R. Development of a decision support system for integrated water
resources management in intensively used small watersheds. Water Pract. Technol. 2006, 1,
doi:10.2166/wpt.2006.004.
35. Handbook on Implementation for Water Services Authorities for Private Water Supplies—
Appendix 1; EPA Ireland: Wexford, Ireland, 2010; p. 63.
36. Scottish Executive. Private Water Supplies: Technical Manual; Scottish Executive: Edinburgh,
UK, 2006; p. 32.
37. Ogeris. Outil D’aide a L’évaluation des Risques Microbiologiques dans les Petites Unités de
Production/Distribution d’eau Potable; Association Scientifique et Technique pour l’Eau et
l’Environnement, Nanterre, Paris, 2009. (in French)
38. HACCP Strategies for Distribution System Monitoring, Hazard Assessment and Control; U.S.
Environmental Protection Agency: Washington, DC, USA, 2007; p. 81.
39. Bartram, J.; Corrales, L.; Davison, A.; Deere, D.; Drury, D.; Gordon, B.; Howard, G.; Rinehold, A.;
Stevens, M. Water Safety Plan Manual: Step-By-Step Risk Management for Drinking-Water
Suppliers; World Health Organization: Geneva, Switzerland, 2009; p. 108 .
40. Guidelines for Drinking Water Quality, 3rd ed.; World Health Organization: Geneva, Switzerland,
2004; Volume 1, p. 540.
41. Calderon, R.L. The epidemiology of chemical contaminants of drinking water. Food Chem. Toxicol.
2000, 38, S13–S20.
42. Monteith, D.T.; Henrys, P.A.; Evans, C.D.; Malcolm, I.; Shilland, E.M.; Pereira, M.G.; Spatial
controls on dissolved organic carbon in upland waters inferred from a simple statistical model. In
review.
43. Gruau, G.; Birgand, F.; Novince, E.; Jarde, E. Pollution des Captages d’eau Brute de Bretagne
par les Matières Organiques; Agence de l’eau Loire Bretagne: Rennes, France, 2004. (in French)
44. Laor, Y.; Rebhun, M. Complexation-flocculation: A new method to determine binding coefficients of
organic contaminants to dissolved humic substances. Environ. Sci. Technol. 1997, 31, 3558–3564.
45. Thuy, P.T.; Moons, K.; van Dijk, J.C.; Anh, N.V.; van der Bruggen, B. To what extent are
pesticides removed from surface water during coagulation-flocculation? Water Environ. J. 2008,
22, 217–223.
46. Association of River Water Supply Companies. Annual Report 2010. Available online:
http://www.riwa-rijn.org/uploads/tx_deriwa/RIWA_Jaarrapport_2010_internet.pdf (accessed on
6 May 2014).
47. Fernandez, M.A.; Alonso, C.; Gonzalez, M.J.; Hernandez, L.M. Occurrence of organochlorine
insecticides, PCBs and PCB congeners in water and sediments of the Ebro river (Spain).
Chemosphere 1999, 38, 33–43.
48. Kolpin, D.W.; Furlong, E.T.; Meyer, M.T.; Thurman, E.M.; Zaugg, S.D.; Barber, L.B.; Buxton, H.T.
Pharmaceuticals, hormones, and other organic wastewater contaminants in U.S. streams, 1999–2000:
A national reconnaissance. Environ. Sci. Technol. 2002, 36, 1202–1211.
49. Manoli, E.; Samara, C. Polycyclic aromatic hydrocarbons in natural waters: Sources, occurrence
and analysis. Trends Anal. Chem. 1999, 18, 417–428.
50. Gomez-Belinchon, J.I.; Grimalt, J.O.; Albaiges, J. Volatile organic compounds in two polluted
rivers in Barcelona (Catalonia, Spain). Water Res. 1991, 25, 577–589.
51. Murray, K.E.; Thomas, S.M.; Bodour, A.A. Prioritizing research for trace pollutants and
emerging contaminants in the freshwater environment. Environ. Pollut. 2010, 158, 3462–3471.
52. Vulliet, E.; Cren-Olivé, C. Screening of pharmaceuticals and hormones at the regional scale,
in surface and groundwaters intended to human consumption. Environ. Pollut. 2011, 159, 2929–2934.
53. Lapworth, D.J.; Baran, N.; Stuart, M.E.; Ward, R.S. Emerging organic contaminants in
groundwater: A review of sources, fate and occurrence. Environ. Pollut. 2012, 163, 287–303.
54. Huerta-Fontela, M.; Galceran, M.T.; Ventura, F. Occurrence and removal of pharmaceuticals and
hormones through drinking water treatment. Water Res. 2011, 45, 1432–1442.
55. Roberts, P.H.; Thomas, K.V. The occurrence of selected pharmaceuticals in wastewater effluent
and surface waters of the lower Tyne catchment. Sci. Total Environ. 2006, 356, 143–153.
56. Verliefde, A.; Cornelissen, E.; Amy, G.; van der Bruggen, B.; van Dijk, H. Priority organic
micropollutants in water sources in Flanders and the Netherlands and assessment of removal
possibilities with nanofiltration. Environ. Pollut. 2007, 146, 281–289.
57. Weston, D.P.; Holmes, R.W.; Lydy, M.J. Residential runoff as a source of pyrethroid pesticides to
urban creeks. Environ. Pollut. 2009, 157, 287–294.
58. Balmer, M.E.; Buser, H.R.; Muller, M.D.; Poiger, T. Occurrence of some organic UV filters
in wastewater, in surface waters, and in fish from Swiss lakes. Environ. Sci. Technol. 2005, 39,
953–962.
59. Westerhoff, P.W.; Yoon, Y.N.; Snyder, S.; Wert, E. Fate of endocrine-disruptor, pharmaceutical,
and personal care product chemicals during simulated drinking water treatment processes.
Environ. Sci. Technol. 2005, 39, 6649–6663.
60. Stackelberg, P.E.; Gibs, J.; Furlong, E.T.; Meyer, M.T.; Zaugg, S.D.; Lippincott, R.L.
Efficiency of conventional drinking-water-treatment processes in removal of pharmaceuticals and
other organic compounds. Sci. Total Environ. 2007, 377, 255–272.
61. Nghiem, L.D.; Schäfer, A.I.; Elimelech, M. Removal of natural hormones by nanofiltration
membranes: Measurement, modeling, and mechanisms. Environ. Sci. Technol. 2004, 38,
1888–1896.
62. Kimura, K.; Amy, G.; Drewes, J.E.; Heberer, T.; Kim, T.-U.; Watanabe, Y. Rejection of organic
micropollutants (disinfection by-products, endocrine disrupting compounds, and pharmaceutically
active compounds) by NF/RO membranes. J. Membr. Sci. 2003, 227, 113–121.
63. Huber, M.M.; Canonica, S.; Park, G.-Y.; von Gunten, U. Oxidation of pharmaceuticals during
ozonation and advanced oxidation processes. Environ. Sci. Technol. 2003, 37, 1016–1024.
64. Rosenfeldt, E.J.; Linden, K.G. Degradation of endocrine disrupting chemicals bisphenol A,
ethinyl estradiol, and estradiol during UV photolysis and advanced oxidation processes.
Environ. Sci. Technol. 2004, 38, 5476–5483.
65. Rodriguez, M.J.; Serodes, J.; Morin, M. Estimation of water utility compliance with trihalomethane
regulations using modeling approach. J. Water Supply Res. Technol.—Aq. 2000, 49, 57–73.
66. Clément, M.; Thomas, O. Détermination de la Contamination de l’air Intérieur des Habitats par
les Trihalométhanes, en vue d’une Évaluation de L’exposition à ces Substances par Inhalation;
PRIMEQUAL 2 – PREDIT report, 2009. (in french)
67. Villanueva, C.M.; Kogevinas, M.; Grimalt, J.O. Research note: Haloacetic acids and trihalomethanes
in finished drinking waters from heterogeneous sources. Water Res. 2003, 37, 953–958.
68. Human health evaluation manual (part A). In Risk Assessment Guidance for Superfund; Office of
Emergency and Remedial Response, U.S. Environmental Protection Agency: Washington, DC,
USA, 1989; Volume 1.
69. Guidelines for Drinking-Water Quality; World Health Organization: Geneva, Switzerland, 1993;
Volume 1.
70. Rodriguez, M.J.; Serodes J.B. Spatial and temporal evolution of trihalomethanes in three water
distribution systems. Water Res. 2001, 35, 1572–1586.
71. Lehtola, M.J.; Miettinen, I.T.; Keinänen, M.M.; Kekki, T.K.; Laine, O.; Hirvonen, A.; Vartiainen, T.;
Martikainen, P.J. Microbiology, chemistry and biofilm development in a pilot drinking water
distribution system with copper and plastic pipes. Water Res. 2004, 38, 3769–3779.
72. Sadiki, A.-I.; Williams, D.T. A study on organotin levels in Canadian drinking water distributed
through PVC pipes. Chemosphere 1999, 38, 1541–1548.
73. Beardsley, M.; Adams, C.D. Modeling and control of vinyl chloride in drinking water distribution
systems. J. Environ. Eng. 2003, 129, 844–851.
© 2014 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article
distributed under the terms and conditions of the Creative Commons Attribution license
(http://creativecommons.org/licenses/by/3.0/
Int. J. Environ. Res. Public Health 2014, 11, 4292-4310; doi:10.3390/ijerph110404292
OPEN ACCESS
International Journal of
Environmental Research and
Public Health
ISSN 1660-4601
www.mdpi.com/journal/ijerph
Review
Microbial Contamination Detection in Water Resources:

Interest of Current Optical Methods, Trends and Needs in
the Context of Climate Change
Aude-Valérie Jung 1,2,3,4,*, Pierre Le Cann 2,3,4, Benoit Roig 2,3,4,5, Olivier Thomas 2,3,4,
Estelle Baurès 2,3,4 and Marie-Florence Thomas 1,4
1
School of Environmental Engineering (EME), Campus de Ker Lann, Avenue Robert Schuman, Bruz
35170, France; E-Mail: marieflorencethomas@ecole-eme.fr
2
EHESP Rennes, Sorbonne Paris Cité, Avenue du Professeur Léon Bernard-CS 74312, Rennes Cedex
35043, France; E-Mails: pierre.lecann@ehesp.fr (P.L.C.); benoit.roig@unimes.fr (B.R.);
olivier.thomas@ehesp.fr (O.T.); estelle.baures@ehesp.fr (E.B.)
3
INSERM, UMR IRSET Institut de Recherche sur la Santé l’Environnement et le Travail - 1085,
LERES, Rennes 35043, France
4
European University of Brittany (UEB), 5 Boulevard Laënnec, Rennes 35000, France
5
Deparment of Sciences, Nîmes University, Rue du docteur George Salan, Nîmes 30000, France
* Author to whom correspondence should be addressed; E-Mail: audevaleriejung@ecole-eme.fr;

Tel.: +33-2-9905-8800; Fax: +33-2-9905-8809.
Received: 21 November 2013; in revised form: 10 April 2014 / Accepted: 11 April 2014 /
Published: 17 April 2014
Abstract: Microbial pollution in aquatic environments is one of the crucial issues with
regard to the sanitary state of water bodies used for drinking water supply, recreational
activities and harvesting seafood due to a potential contamination by pathogenic bacteria,
protozoa or viruses. To address this risk, microbial contamination monitoring is usually
assessed by turbidity measurements performed at drinking water plants. Some recent studies
have shown significant correlations of microbial contamination with the risk of endemic
gastroenteresis. However the relevance of turbidimetry may be limited since the presence of
colloids in water creates interferences with the nephelometric response. Thus there is a need
for a more relevant, simple and fast indicator for microbial contamination detection in water,
especially in the perspective of climate change with the increase of heavy rainfall events.
This review focuses on the one hand on sources, fate and behavior of microorganisms in
water and factors influencing pathogens’ presence, transportation and mobilization, and on
the second hand, on the existing optical methods used for monitoring microbiological risks.
Finally, this paper proposes new ways of research.
Keywords: optical methods; heavy rainfall; colloids; turbidity; pathogens
1. Introduction
In the context of international regulations, the contamination of water bodies by organic

micropollutants is the subject of constant interest and is always under investigation. Although at the
origin of numerous outbreaks of gastrointestinal diseases and public health concerns, microbial
contamination is rarely considered [1], even if some studies have shown consistent and significant
association between heavy rainfall events and waterborne disease outbreaks [2,3] and climate change
consequences [4]. A lack of studies concerning waterborne diseases correlated with extreme events
(such as droughts and floods) has recently been mentioned by a review of reviews [5]. Despite
a regulatory framework, microbial quality is often a source of impairment of the compliance of drinking
water supply [6], in particular for small scale water systems (SSWS) [7,8] and private water
supplies [9–11]. Additionally, water quality is known to be affected by increased microbial pollution
under extreme weather conditions (climate change) and requires more systematic studies [12,13].
Monitoring objectives consist in directly targeting the sources of contamination, by using simple and
rapid indicators but are mainly focused on parameters such as faecal bacteria (E. coli or Enterococci).
Enteric viruses, that play a major role in waterborne diseases, are rarely investigated due to the detection
limits of commonly applied methods [14,15]. Consequently, even when complying with microbiological
water quality standards, the endemic part of waterborne acute gastroenteritis (AGE) may vary from 0 to
40% [16,17] excluding the smallest supply water systems (<500 inhabitants) for which the syndromic
surveillance is ineffective.
For many years, analytical approaches have been developed for knowledge improvement concerning
microorganisms’ nature and origins. More recently, different studies have shown the importance of an
ecological (sediment and aquatic plants role) approach [18–23]. Although PCR is often used in
microbial source tracking, it has not found widespread application in microbial monitoring programs
due to its targeted approach for specific microbial genera or species [24,25]. MST tools proposed for the
identification of the origin of faecal pollution are well documented in the literature [21,26–30]. Except
for these methods and the conventional bacterial culture (standardized), some alternative ones are
available, such as methods based on the exploitation of the optical properties of water. The interest in
optical methods for microbial detection may be justified by the optical effects (mainly absorption and
light diffusion) of microorganisms often adsorbed on colloids or particles. The effect of interaction
between light and matter can be exploited to detect bacteria or viruses either as free particles or attached
to organo-mineral complexes capable of modifying the optical properties of surfaces.
This review focuses on: (1) sources, fate, behavior of microorganisms in water and factors
influencing pathogens’ presence, transport and mobilization, and (2), on optical methods for microbial
contamination monitoring.
2. Sources, Fate and Behavior of Microorganisms in Water
The microbial contamination of water is often of faecal nature related to humans (water sewage
treatment plants, combined sewage overflow (CSO), non-collective sewage systems), domesticated
animals (manure spreading, pit stock overflow), or wildlife. The main origins of microbial
contamination of natural aquatic resources are discharges of water treatment plants, decontamination
stations, hospitals, industries considered as point sources, etc. Correlation between pathogens
concentrations and urban activities is well documented (e.g., [31,32]). On the other hand, diffuse sources
(slurry, manure, sludge application…) may also be considered. The abundance and importance of
pathogens in water depend on factors such as the contamination level, pathogens’ persistence in water
bodies, biological reservoirs (including aquatic plants and sediments) and the ability of pathogens to be
transported [33]. The land use management practices and the size of the watershed also influence the
survival of microorganisms [34–36]. According to George et al. [37], streams flowing through areas
partly or fully covered with pastures are more contaminated than those running through forests and
cultivated areas.
2.1. Fate in Sediments and in Submerged Aquatic Vegetation
Sediments and submerged aquatic vegetation (SAV) are important reservoirs of

microorganisms [18,30,38]. For example, Badgley et al. [18] have shown that SAV harbored
significantly higher mean densities of Enterococci than sediments, which themselves harbored higher
densities than water. The high Enterococci densities observed in SAV are mainly due to increase
nutrient availability as well as protection from UV radiation. Sediments can contain 10-times more
viruses than water [33]. The microorganisms survival in different sediment types with different particle
size and organic carbon content have been studied by Garzio-Hadzick et al. [39]. Their results revealed
that sediments with small particle size and high organic carbon content could extended pathogens’
survival. Chandran et al. [40] concluded that sediments could act as a reservoir of pathogenic bacteria
and exhibit a potential health hazards from possible resuspension and subsequent ingestion during
recreational activities. The role of granulometric distribution of particles for microorganisms’ survival is
also important for their transport during high flow events.
The behavior of microorganisms is also influenced by hydrometeorological changes such as heavy
rainfalls which are likely to increase with climate change [12]. Thus, microorganisms associated with
sediments or SAV, are put back into suspension and could lead to water contamination [18,33].
Re-suspension of sediments during or shortly after rainfall events was studied by Cho et al. [41] in order
to investigate E. coli behavior regarding different bottom sediment textures of three streams for artificial
high-flow events. A bacteria transport model was proposed, but additional research was needed to
understand which and how sediment properties affect the parameters of streambed E. coli released into
the water column.
2.2. Transport and Fate of Microorganisms
The associations between microorganisms and sediments influence their survival but also their
transport characteristics [42]. Moreover the fate and transport of faecal bacteria are highly related to the
governing sediment transport processes [38]. The transport of viruses linked to particles is particularly
influenced by attachment to mineral surfaces and inactivation [43]. This adsorption can be reversible
under other environmental conditions such as pH variations [44]. The microbial partitioning between the
freely suspended and particulate attached phases during transport along overland flow pathways was
studied by Soupir and Mostaghimi [45]. Rainfall simulations conducted on large-scale field plots
showed that the majority of E. coli and Enterococci are transported from the fresh manure source in an
unattached state, with only 4.8% of E. coli and 13% of Enterococci associated with particles. The
average percent of attached E. coli or Enterococci is less than 3% with sparsely vegetated cover. In a
laboratory-scale model system developed to investigate the transport mechanisms of E. coli in overland
flow across saturated soils, E. coli appeared to be attached predominantly to small particles
(<2 μm) and hence remained unattenuated during transport [45]. Moreover, microorganisms could be
considered in hydric resources as biological particles transported by advection [33]. For example,
Bekhit et al. [46] were interested in the role of colloids on the mobility of microorganisms and
micropollutants in groundwater. They proposed a conceptual model to take into account the different
physiochemical and biological processes, reaction kinetics, and transport mechanisms of the combined
system (contaminant–colloids–bacteria). Different results have been obtained for natural aquatic media
such as lakes and reservoirs. In particular, a review of Brookes et al. [47] explained the factors controlling
pathogens transport and distribution, the respective role of dispersion, dilution and horizontal and
vertical transport in such media. Contrary to lab-scale models, settling of pathogens particles operates
with more complex hydrodynamic processes. Their results showed that all of the tested microbial
indicators (Escherichia coli, Enterococci, Clostridium perfringens, aerobic spores, somatic coliphages,
Cryptosporidium spp. and Giardia spp.) were associated with larger sized particle (>63.3 μm), except C.
perfringens spores which were associated with particles size in the range of 45.5−63.3 μm. This
granulometric distribution of the associated particles has important effects on the retention of
microorganisms by settling and transfer into sediments [48,49]. Abudalo et al. [50] and Searcy et al. [51]
precised that the environmental impact of protozoan parasites is closely related to their extended
survival in contrasting climatic conditions and disinfection processes, and to their ability to interact with
other organic or nonorganic particles [52]. The latter phenomenon governs their transport, retention (by
settling) and/or release, and survival in the transition from land to sea.
In addition to Auer and Niehaus [53], Ferguson et al. [54], Sinton et al. [55] and Noble et al. [56]
suggest that fecal indicator bacteria (FIB) inactivation rates (also referred to as a ―die-off‖ or ―decay‖
rate) vary under different environmental conditions, including solar radiation and water temperature.
The FIB inactivation rate variability in response to other factors, such as water column depth, is not as
well-understood, and has been recommended as an area for future research [57].
2.3. Influence of Hydrometeorological Conditions
Microorganisms’ transportation is particularly influenced by hydrometeorological conditions. Indeed

microbial contamination of surface water increases during heavy rainfalls with water combined sewage
overflows (CSOs) or agricultural land run-offs [33,58]. Rainfall increases the suspended matter content of
small streams as well as their fecal contamination, with fecal coliforms mainly adsorbed on particles [37].
Rainfall events can also increase human enteric virus load in natural waters impacted by CSOs and
stormwater leading to higher risk of gastroenteritis in recreational activities or shellfish consumption [59].
This risk has been evaluated using quantitative microbial risk assessment showing the dominance of
norovirus in swimming-associated gastroenteritis [60]. Heavy rainfalls may also led to catchment area
flooding with a strong increase of water turbidity, this last being considered as a non-questionable indicator
of potential contamination by aquatic pathogens [61,62]. Several prospective epidemiological studies in
France have also shown that groundwater influenced by surface water could be the cause of gastrointestinal
infections in an endemic level [61,63], especially during rainfall events in karstic environment.
Figure 1. Schematic synthesis of source and fate of allochtonous microorganisms in water.
Wet & dry

deposition**
(long distance
and local)
SEDIMENTATION
PHYSICO-CHEMICAL/
RESOURCE
Usage
BIOLOGICAL REACTIONS
BIOLOGICAL Water body
RESERVOIRS***
INFLUENCING FACTORS
(HYDROCHEMICAL, HYDROLOGICAL)
RESUSPENSION
* Including ponctual and diffuse sources: sewage overflows discharges, agricultural spreading…
** Including sand rains, dust, bioaerosols, sea sprays…
*** Including aquatic plants and sediments
Water temperature variation plays also a role in the hydrodynamic distribution of microorganisms. In
temperate areas, lakes are generally stratified during summer (with warmer water at the surface).
Destratification occurs with heavy rains or storms, modifying the convection movements of particles
containing microorganisms, moving them to the surface [64–66]. In a recent work, Wyer et al. [66]
studied the variation of fecal indicators in four streams during a moderate rainfall. They demonstrated,
thanks to microbial source tracking (MST), that the transfer time varies from minutes to hours,
according to the hydrological conditions and sites characteristics. Table 1 presents some relevant studies
dealing with pathogen interactions with particulate and colloidal phases in the aquatic compartment.
Finally, Figure 1 proposes a schematic synthesis of source and fate of allochtonous microorganisms in
water. From the resource to usages, the human pressure increases with production of drinking water and
public use of recreational waters. From sedimentation to resuspension, the biological reservoirs are
submitted to physicochemical and biological reactions on the one hand, and to hydrochemical and
hydrological factors in water bodies on the other hand. Moreover, the external runoffs from
anthropogenic and natural sources are considered as punctual and diffuse contribution, depending on
climate change.
Table 1. Pathogens interactions studies with particulate and colloidal phases in the aquatic compartment.
Type of pathogens Matrix Water Body References Comments
Bacteria Particles Karstic aquifer [67] Connections with the surface responsible for turbid and bacterial
contaminations
Bacteria groups Colloids Groundwater [46] Modelisation of transport mechanisms of the combined system
(contaminant–colloids–bacteria)
Lakes [47] Correlation between size particles and transport and distribution after
a storm
Recreational waters [66] Connection between microbial tracers and fecal indicator organisms
E. coli, Enterococci Particles Run-off [68] Rainfall simulations for erodible soil particles and sparsely vegetable soils
[45] run-off
E. coli Sediments Rivers [41] Modelisation of bacteria transport during rainfall events
Virus
Colloids Rivers [69] Direct spillage of wastewater in river during heavy rains
Norovirus
Mixture Colloids Distributed water [3] Correlation of heavy rains with gastroenteritis epidemics
Particles River, karstic water [17] Correlation of turbidity, flow rate and gastroenteritis epidemics
Others
Colloids Rivers [70] Correlation with rainy events
Giardia cyst
Particles Waterbeds soils [52] Interaction between parasites and particles (organic and inorganic)
Protozoan parasites groups
As a consequence, the demand for faster and reliable monitoring methods and approaches is also
increasing, with for example rapid molecular methods, although these methods do not allow
a real-time control, contrary to optical methods [24].
3. Optical Monitoring of Microbial Contamination: Current Methods, Trends and Needs
3.1. Current Trends: Turbidity, Particle Size Distribution (PSD) and Cytometry
The main simple and practical way to assess the microbiological risk of water contamination is
turbidimetry, largely used in water treatment plants. Despite its lack of sensitivity (there is no direct
relation between pathogens density in water and turbidity), the choice of this parameter is relevant
because: (i) the suspended particles are a safe and supportive environment for pathogens and a place of
protection against disinfectants; (ii) turbidity lowers the efficiency of disinfection processes by the
presence of oxidisable associated organic matter; (iii) turbidity in the water network creates particle
deposition and promotes the growth of biofilms with the potential presence of pathogens (Legionella,
Pseudomonas, Aeromonas, Mycobacterium, etc.) [17,71]. For 10 years, the interest in using turbidity for
the continuous monitoring of effluents quality has been widely demonstrated [72–74]. In a constant
environment, the respect for strict procedures gives precise and reliable turbidity data, in the laboratory
as well as in situ [75], with a high frequency acquisition (minute order). Moreover, if this parameter is
known as a good indicator of suspended matters in sewage networks [76–78], the turbidity measurement
of fresh waters is influenced by the particle size and shape of SS, the presence of plankton and the
presence of dissolved humic substances. Some authors have tried to establish correlations between
turbidity and parameters like water level, flow rate, or rain intensity [73,74] without success, excepted in
the case of a dilution phenomenon following a short dry period before raining [79]. Pronk et al. [80] have
studied the dynamics of organic carbon (OC), turbidity, faecal indicator bacteria and physicochemical
parameters in a karst system near Yverdon, Switzerland. OC appears to be a better indicator for bacterial
contamination than turbidity. Page et al. [81] have tried to correlate continuous physico-chemical
measurements (temperature, electrical conductivity, turbidity, spectral absorption coefficient (SAC),
and particle density) with the amount of faecal indicator bacteria, such as Escherichia coli and
Enterococcus sp., discontinuously sampled in the aquatic media. No individual proxy indicator for
bacterial contamination was found. However, the following of the river water infiltration could be
assessed by electrical conductivity, temperature and SAC measurements.
Turbidity measurements are also often used for drinking water quality monitoring and within water
treatment plants considering experimental approximate relations with the number of germs [82]. It is
regulated by the Directive 98/83/CE for the distributed waters but not for resources, contrary to
microbiological parameters considered in both cases. Even if turbidity measurements are often used as a
run-off pollution indicator of resource water, the precise interpretation of this parameter as a sanitary
threat indicator, remains difficult because of its sensibility to environmental factors as described before.
Beaudeau et al. [17] have recently shown a significant and biologically plausible risk was highlighted on
rivers. Turbidity of treated water may be considered as an exposure indicator which produces the more
reproducible risks [71].
The physico-chemical characterization methods of particles (shape, size, surface properties) were
recently considered in particular because of the sensitivity of turbidity to these parameters. Among them,
Goldscheider et al. [83] used a portable field particles counter in order to explain the turbidity
measurements by following the temporal evolution of particle size distribution (PSD) between 0.9 and
139 µm diameter. For karstic resources, turbidity measurements fail to accurately estimate
microbiological contamination because of the variability of the hydroclimatic conditions and the
presence of more or less large particles, remobilized by the first flush, followed by a very fine particle
mix (about 1 µm), sometimes with the presence of pathogens coming from contaminated waters,
confirming the results of Atteia and Kozel [84]. A very good correlation between turbidity and E. coli
(R2 = 0.93) has been demonstrated for colloidal particles (0.9–1.5 µm).
Cytometry techniques are also based on optical principles and are now routinely used in
microbiology, particularly for microbial ecology studies, unfortunately not on site. Flow cytometry can
detect microorganisms at the cellular level from their optical properties (fluorescence and light
scattering), to quantify and provide information on the individual characteristics of cells and information
on the heterogeneity of cellular states in a population. It applies to the study of autotrophic and
heterotrophic microorganisms the fluorescence properties of which are natural (pigments) or induced by
probes targeting taxonomic targets and/or physiological. Particles analyzed in the environmental field
are diverse and range from viruses to protozoa. The analysis time is less than 1 h and the number of cells
analyzed is very high (around 107 colony forming unit/mL). In some cases it is possible to physically
separate the cell functions of some of their optical properties in order to complete their analysis by other
techniques (biochemical and molecular biology techniques) [85,86].
3.2. Trends in Optical Methods
3.2.1. Fluorescence Measurements
Although fluorescence measurements are well known for organics analysis (e.g., hydrocarbons), this
method could also be considered for its interest in microbial sources tracking. The origin and
contamination type of dissolved organic matter (DOM) has been characterized by fluorescence [87–89].
For example, in river water samples, both protein-like (tyrosine- and tryptophan-like) and fulvic-like and
humic-like fluorescence were detected and the relative intensity of these signals was used as an indicator
of slurry occurrence in drainage waters in agricultural area [89]. Jaffrezic et al. [90] propose the
geochemical fluorescence index (GFI), to point out the difference between farm waste and human waste.
GFI is the ratio between two regions of the fluorescence spectrum divided into biochemical and
geochemical regions. Lefcourt et al. [91] have developed a field-portable fluorescence imagining system
for detecting fecal contaminants for agricultural products.
Exploitation of the bacteriophage (or phage) life cycle and its accompanying species specificity can
reduce much of the time and cost associated to classical methods such as gene sequencing,
electrophoresis or repetitive sequence-based polymerase chain reaction (PCR). The well-characterized
bacterial luciferase genes, encoded by the lux operon from the aquatic Gram-negative Vibrio fischeri are
perhaps the most frequently used reporters for incorporation into recombinant phages and have many
applications, namely for acute toxicity prediction for chemicals [92]. Other-component bacteriophage-based
bioluminescent reporter systems were developed for the detection of Escherichia coli in environmental
samples, for instance [93].
3.2.2. Biosensors
Biosensors combining the selectivity of biology with the processing power of microelectronics may
also be considered through bio-recognition systems, generally enzymes or binding proteins such as
immobilized antibodies. Biosensors have the potential for substantial improvements over standard
methods and have been reported employing the full spectrum of biorecognition molecules and
transduction methods for detection of waterborne pathogens, with oligonucleotide probes and antibodies
being the most common [94]. These biosensors have been used in various fields such as environment or
food processing for biorecognition of specific molecules, since they have a high sensitive detection
capacity. However, several problems for waterborne pathogens detection may appear with regard to low
concentration and many interfering enzymatic reactions in natural media [94]. In the literature, many
examples for the detection of E. coli can be found, and among these studies, the transduction based
methods, with fluorescence and chemiluminescence probes are the most used [94]. (Bio)sensing systems
have benefited from the achievements of scientific and technological research from nanomaterials and
nanotechnologies science [95]. Especially, nanoparticles, gold nanoparticles, graphene, quantum dots,
have led to major advances in this field [96]. Various mass-sensitive techniques have also been applied
for bacteria [94]. For these techniques, the limits of detection are around 100 cells/mL. Others
techniques using nanocrystal biosensors are used for drinking water market, since they seem to be
reliable and are in-situ techniques for monitoring multiple pathogens mixtures. Alternatively, many
studies concerning protozoan biosensing detection can be related since the detection schemes applied to
parasites and viruses too [94].
3.2.3. Spectrophotometric Methods
Spectrophotometric methods are potentially interesting for the detection of microbial contamination
because of their use in particle characterization. As shown in the review and synthesis of Bowers and
Binding [97], spectrophotometric methods are used for quantitative assessment of particle concentration
in sea water, in order to study the effect on sunlight penetration into the sea. Astoreca et al. [98] have
correlated the basic properties of the particles (concentration, composition, and size of suspended
particles in sea water) with their light absorption properties in the visible and near-infrared regions.
Stramski et al. [99] have proposed a database of the single-particle optical properties of marine microbial
particles to better understand the ocean optics. This database (based on the optical properties depending
on the particle size) includes representatives from five classes of particles: viruses, heterotrophic
bacteria, cyanobacteria, small nanoplanktonic diatoms, and nanoplanktonic chlorophytes. More
recently, Stadler et al. [100] have obtained good correlation between SAC254 (spectral absorption
coefficient at 254 nm) and E. coli in karst water resources and propose that SAC254 may also be used as
a real-time proxy to estimate the magnitude of faecal pollution during rainfall events. They precise that
only diffuse faecal pollution sources with adequate soil contact are supposed to be quantifiable by the
described approach.
Enlarging this approach, UV/Visible spectroscopy has lots of advantages to study the interactions
between natural matrices and pollutants in dissolved and colloidal compartments [101]. The first
advantage of this technique is to be rapid, direct and low cost. Several studies have proven the usefulness
of such a technique in water quality monitoring, especially thanks to the complementarities with basic
analysis such as dissolved biodegradable organic carbon [102], total organic carbon [103], by developing
spectral exploitation techniques such as deconvolution or multi-wavelength treatment [104,105]. Owing to
its good sensitivity, this technique could be useful to provide semi-quantitative information on
the interactions between phases (organic carbon exchange) by direct monitoring, a non-specific
pre-analysis preparation and at natural media concentrations.
Table 2. Characterization of particles by UV (water and wastewater).

Particle size Suspended matter
References Optical domains Measurement/study
(µm) concentration (mg/L)
Visible and Relationships between the
[98] near-infrared spectral concentration, composition and 2.72–460 0–90
regions size of suspended particles
UV spectrophotometry Characterization of
[106] 0.4–2 × 103 100–670
and laser granulometry heterogeneous suspensions
Coupling Heterogeneous suspensions,
[107] UV-spectrophotometry quantitative approach (size and 0.05–103 10–350
and laser granulometry concentration)
Study of the impact of
mechanical treatments on
[108] UV spectrophotometry 10−3–103 10–220
wastewater solids by UV
spectrophotometry
UV spectrophotometry Study of UV–vis responses of
[109] 1–100 10–250
and laser granulometry mineral suspensions in water
The technique could also give information about compounds’ size by coupling UV/Visible
spectroscopy to laser granulometry analysis [106–109]. Behro et al. [109] proposed a typology of UV
responses of particles according to their size and a model spectrum of mineral colloids in order to obtain
a better understanding of physical absorption phenomena. They showed that the UV responses of
non-absorbing particles can be modeled even if the physical phenomena are complex and the refraction
index and the shape of particles have not been taken into account.
Table 2 synthesizes some results obtained by spectrophotometry for the characterization of particles
in heterogeneous suspensions including water and wastewater. Table 3 presents the main literature
studies presented in Section 3.2.2 of the manuscript.
4. Conclusions
This review has shown that knowledge about pathogen transport mechanisms are partial, even if the
role of particles and colloids is relatively well documented. Moreover there is a lack of methods (index,
sensor) allowing onsite identification of microbial contaminants and dynamic exchanges of pathogens
into different phases such as soluble, colloidal, and particulate. Concerning the first point, the
improvement of the optical responses currently provided by turbidimetry could be an interesting

solution to identify hazardous situations of water microbial contamination. Indeed, the pertinence of
the turbidity measurement is limited, since nephelometric measurements are dependent on the presence
of colloids that may interfere with the results.
Table 3. Synthesis of classical and trends in optical methods for pathogens detection.
Particle size/
Kind of media/applications Influencing parameters for
Parameter/References Number of cell
fields/pathogens the studies/interferences
detected
Turbidity/[7]–[72–80] Natural and wastewaters Plankton, Humic substances 10–103 µm
PSD/[83,84] Karstic waters Hydroclimatical 0.9–1.5 µm
Others fluorescent species From virus to
Cytometry/[85,86] All fluorescent species (e.g., humic-like substances) bacteria/107 colony
+ light scattering forming unit/mL
River waters (tyrosine,
Fluorescence, Light-scattering, inner filters
tryptophan and fulvic-like From molecule to
Bacteriophage effects, bioluminescence
substances, E. Coli, bacteria
life cycle/[87–93] interferences
Vibrio fischeri
Environment, food process, Virus to protozoan
Biosensors/[94–96] Interfering enzyme reactions
military > 100 cells/mL
Virus, bacteria,
Spectrophotometry cyanobacteria,
Light scattering 10−3–2 × 103 µm
Methods/[97–109] nanoplanktonic and
chlorophytes diatoms
* to have a correlation between the size and the measurement value.
Among the ways of improvement encountered, the relevance of UV-Visible spectra exploitation to
obtain significant information about the chemical and granulometric composition of the particulate phase
is underlined. Thus, the development of a simple physico-chemical proxy (UV visible exploitation)
associated to other optical methods (particle size distribution, turbidimetry, fluorimetry…) should be
a research path to obtain an indicator for hazardous situations.
For the second point, the development of a characterization procedure (static and dynamic) allowing
the study of the exchanges between particles and the microbial agents, namely during heavy rainfall
events (aggregates, resuspension…) is needed. The exchanges being related to the size, the
concentration, and particles charges and nature, all these parameters should be considered. Moreover,
different operational conditions (e.g., pH, conductivity, organic carbon concentration) must be also
assessed to improve the knowledge of the matrix effect (environmental hazardous situations). As
shown by the review, there is a real lack of temporal studies integrating the seasonal and hydrological
parameters associated to heavy rainfall events with regard to health risk assessment.
All these research needs must be considered at fair value because of, in addition to the population
health impact, the importance of cost management of waterborne diseases such as acute gastroenteritis
is undeniable. The cost reduction, by a better anticipation of the sanitary degradation of natural water
resources quality, is therefore an important economical issue. For the scientific point of view, the main
question will be to complete the main exposure indicator (the turbidity measurement) with additional
more specific parameters, especially hydrological conditions. This needs a multidisciplinary approach
between hydrologists, chemists, microbiologists, water treatment managers and epidemiologists.
The outcomes of these studies could give arguments to water treatment managers to modify or
adapt the water treatment processes, or a different way to manage their plants by taking into account an
alert procedure proposed for hazardous situations, based on new indicators of microbial contamination.
Acknowledgments
The authors would like to thank IJERP Editors and the reviewers for their helpful comments.
Author Contributions
Aude-Valérie Jung had the original idea for the manuscript and, with all co-authors, carried out the
design. Aude-Valérie Jung and Marie-Florence Thomas drafted the manuscript, which was revised by all
authors. Pierre Le Cann and Benoit Roig verified especially the state of the art of the microbiological
knowledges. Olivier Thomas conceived the study idea, critically revised the manuscript and corrected
the English. All authors read and approved the final manuscript.
Conflicts of Interest
The authors declare no conflict of interest.
References
1. Hlavsa, M.C.; Roberts, V.A.; Anderson, A.R.; Hill, V.R.; Kahler, A.M.; Orr, M.; Garrison, L.E.;
Hicks, L.A.; Newton, A.; Hilborn, E.D.; et al. Surveillance for waterborne disease outbreaks and
other health events associated with drinking water—United States, 2007–2008. MMWR Surveill.
Summ. 2011, 60, 1–32.
2. Auld, H.; MacIver, D.; Klaassen, J. Heavy rainfall and waterborne disease outbreaks: The
Walkerton example. J. Toxicol. Environ. Health. A 2004, 67, 1879–1887.
3. Curriero, F.C.; Patz, J.A.; Rose, J.B.; Lele, S. The association between extreme precipitation and
waterborne disease outbreaks in the United States, 1948–1994. Am. J. Public Health 2001, 91,
1194–1199.
4. Hunter, P.R. Climate change and waterborne and vector-borne disease. J. Appl. Microbiol. 2003, 94,
37S–46S.
5. Bouzid, M.; Hooper, L.; Hunter, P.R. The effectiveness of public health interventions to reduce
the health impact of climate change: A systematic review of systematic reviews. PLoS One 2013,
8, doi:10.1371/journal.pone.0062041.
6. Blackburn, B.G.; Craun, G.F.; Yoder, J.S.; Hill, V.; Calderon, R.L.; Chen, N.; Lee, S.H.; Levy,
D.A.; Beach, M.J. Surveillance for waterborne-disease outbreaks associated with drinking
water—United States, 2001–2002. MMWR 2004, 53, 23–45.
7. Pitkänen, T.; Karinen, P.; Miettinen, I.T.; Lettojärvi, H.; Heikkilä, A.; Maunula, R.; Aula, V.;
Kuronen, H.; Vepsäläinen, A.; Nousiainen, L.-L.; et al. Microbial contamination of groundwater
at small community water supplies in Finland. Ambio 2010, 40, 377–390.
8. Risebro, H.L.; Breton, L.; Aird, H.; Hooper, A.; Hunter, P.R. Contaminated small drinking water
supplies and risk of infectious intestinal disease: A prospective cohort study. PLoS One 2012, 7,
doi:10.1371/journal.pone.0042762.
9. Rutter, M.; Nichols, G.L.; Swan, A.; De Louvois, J. A survey of the microbiological quality of
private water supplies in England. Epidemiol. Infect. 2000, 124, 417–425.
10. Said, B.; Wright, F.; Nichols, G.L.; Reacher, M.; Rutter, M. Outbreaks of infectious disease
associated with private drinking water supplies in England and Wales 1970–2000. Epidemiol.
Infect. 2003, 130, 469–479.
11. Kay, D.; Watkins, J.; Francis, C.A.; Wyn-Jones, A.P.; Stapleton, C.M.; Fewtrell, L.; Wyer, M.D.;
Drury, D. The microbiological quality of seven large commercial private water supplies in the
United Kingdom. J. Water Health 2007, 5, 523–538.
12. Bartram, J.; Corrales, L.; Davison, A.; Deere, D.; Drury, D.; Gordon, B.; Rinehold, A.; Stevens, M.
Water Safety Plan Manual: Step-by-Step Risk Management for Drinking-Water Suppliers; World
Health Organization: Geneva, Switzerland, 2009.
13. Bates, B.C.; Kundzewicz, Z.W.; Wu, S.; Palutikof, J.P. Climate Change and Water. Technical
Paper of the Intergovernmental Panel on Climate Change. IPCC Secretariat: Geneva, Switzerland,
2008; pp. 1–210.
14. Hundesa, A.; Maluquer de Motes, C.; Bofill-Mas, S.; Binana-Gimenez, N.; Girones, R.
Identification of human and animal adenoviruses and polyomaviruses for determination of sources
of fecal contamination in the environment. Appl. Environ. Microbiol. 2006, 72, 7886–7893.
15. Poma, H.R.; Gutiérrez Cacciabue, D.; Garcé, B.; Gonzo, E.E.; Rajal, V.B. Towards a rational
strategy for monitoring of microbiological quality of ambient waters. Sci. Total Environ. 2012,
433, 98–109.
16. Beaudeau, P.; Valdes, D.; Damien, M.; Stemfelet, M.; Seux, R. Natural and technical factors in faecal
contamination incidents of drinking water in small distribution networks, France, 2003–2004:
A geographical study. J. Water Health 2010, 8, 20–33.
17. Beaudeau, P.; Rambaud, L.; Galey, C.; le Tertre, A.; Zeghoun, A. Risque D’infections
Sporadiques Lié à L’ingestion D’eau Du Robinet: L’émergence D’une Approche
Epidémiologique. In Proceeding of Second National Congress Société Française Santé
Environnement (SFSE), Paris, France, 14–15 December 2011.
18. Badgley, B.D.; Nayak, B.S.; Harwood, V.J. The importance of sediment and submerged aquatic
vegetation as potential habitats for persistent strains of Enterococci in a subtropical watershed.
Water Res. 2010, 44, 5857–5866.
19. Meays, C.L.; Broersma, K.; Nordin, R.; Mazumder, A. Source tracking fecal bacteria in water:
A critical review of current methods. J. Environ. Manag. 2004, 73, 71–79.
20. Blanch, A.R.; Belanche-Munoz, L.; Bonjoch, X.; Ebdon, J.; Gantzer, C.; Lucena, F. Tracking the
origin of faecal pollution in surface water: An ongoing project within the European Union
Research Programme. J. Water Health 2004, 2, 249–260.
21. Field, K.G.; Samadpour, M. Fecal source tracking, the indicator paradigm, and managing water
quality. Water Res. 2007, 41, 3517–3538.
22. Gourmelon, M.; Caprais, M.P.; Mieszkin, S.; Marti, R.; Wéry, N.; Jardé, E.; Derrien, M.;
Jadas-Hécart, A.; Communal, P.Y.; Jaffrezic, A.; et al. Development of microbial and chemical
MST tools to identify the origin of the faecal pollution in bathing and shellfish harvesting waters
in France. Water Res. 2010, 44, 4812–4824.
23. Furtula, V.; Osachoff, H.; Derksen, G.; Juahir, H.; Colodey, A.; Chambers, P. Inorganic nitrogen,
sterols and bacterial source tracking as tools to characterize water quality and possible
contamination sources in surface water. Water Res. 2012, 46, 1079–1092.
24. Mesquita, S.; Noble, R.T. Recent Developments in Monitoring of Microbiological Indicators of
Water Quality across a Range of Water Types. In Water Resources Planning, Development and
Management; Wurbs, R., Ed.; Texas A&M University: College Station, TX, USA, 2013; Chapter 2.
25. Plummer, J.D.; Long, S.C. Monitoring source water for microbial contamination: Evaluation of
water quality measures. Water Res. 2007, 41, 3716–3728.
26. Edge, T.A.; Hill, S.; Seto, P.; Marasalek, J. Library-dependent and library-independent microbial
source tracking to identify spatial variation in faecal contamination sources along a lake Ontario
beach (Ontario, Canada). Water Sci. Technol. 2010, 62, 719–727.
27. Kortbaoui, R.; Locas, A.; Imbeau, M.; Payment, P.; Villemur, R. Universal mitochondrial PCR
combined with species-specific dot-blot assay as a source-tracking method of human, bovine,
chicken, ovine, and porcine in fecal-contaminated surface water. Water Res. 2009, 43, 2002–2010.
28. Lyautey, E.; Lu, Z.; Lapen, D.R.; Berkers, T.E.; Edge, T.A.; Topp, E. Optimization and
validation of rep-PCR genotypic libraries for microbial source tracking of environmental
Escherichia coli isolates. Can. J. Microbiol. 2010, 56, 651–659.
29. Lee, D.Y.; Weir, S.C.; Lee, H.; Trevors, J.T. Quantitative identification of fecal water pollution
sources by TaqMan real-time PCR assays using Bacteriodales 16S rRNA genetic markers. Appl.
Microbiol. Biotechnol. 2010, 88, 1373–1383.
30. Staley, C.; Reckhow, K.H; Lukasik, J.; Harwood, V.J. Assessment of sources of human
pathogens and fecal contamination in a Florida freshwater lake. Water Res. 2012, 46, 5799–5812.
31. Marsalek, J.; Rochfort, Q.J. Urban wet-weather flows: Sources of fecal contamination impacting on
recreational waters and threatening drinking-water sources. J. Toxicol. Environ. Health A. 2004, 6,
1765–1777.
32. Selvakumar, A.; Borst, M.J. Variation of microorganism concentrations in urban stormwater
runoff with land use and seasons. J. Water Health 2006, 4, 109–124.
33. Dechesne, M.; Soyeux, E.; Loret, J.F.; Westrell, T.; Stenström, T.A.; Gornik, V.; Koch, C.;
Exner, M.; Stanger, M.; Agutter, P.; et al. Pathogens in Source Water, Microbiological Risk
Assessment: A Scientific Basis for Managing Drinking Water Safety from Source to Tap;
Microrisk European Project: Nieuwegein, The Netherlands, 2006; pp. 1–42.
34. Ferguson, C.M.; Charles, K.; Deere, D.D. Quantification of microbial sources in drinking-water
catchments. Crit. Rev. Environ. Sci. Technol. 2008, 39, 1–40.
35. Harmel, R.D.; Karthikeyan Gentry, R.; Srinivasan, T.R. Effects of agricultural management, land
use and watershed scale on E. coli concentrations in runoff and stream flow. Trans. ASABE 2010,
53, 1833–1841.
36. James, E.; Joyce, M. Assessment and management of watershed microbial contaminants.
Crit. Rev. Environ. Sci. Technol. 2004, 34, 109–139.
37. George, I.; Anzil, A.; Servais, P. Quantification of fecal coliform inputs to aquatic systems
through soil leaching. Water Res. 2004, 38, 611–618.
38. Gao, G.; Falconer, R.A.; Lin, B. Numerical modelling of sediment bacteria interaction processes
in surface waters. Water Res. 2011, 45, 1951–1960.
39. Garzio-Hadzick, A.; Shelton, D.R.; Hill, R.L.; Pachepsky, Y.A.; Guber, A.K.; Rowland, R.
Survival of manure-borne E. coli in streambed sediment: Effects of temperature and sediment
properties. Water Res. 2010, 44, 2753–2762.
40. Chandran, A.; Varghese, S.; Kandeler, E.; Thomas, A.; Hatha, M.; Mazumder, A. An assessment
of potential public health risk associated with the extended survival of indicator and pathogenic
bacteria in freshwater lake sediments. Int. J. Hyg. Environ. Health 2011, 214, 258–264.
41. Cho, K.H.; Pachepsky, Y.A.; Kim, J.H.; Guber, A.K.; Shelton, D.R.; Rowland, R. Release of
Escherichia coli from the bottom sediment in a first-order creek: Experiment and reach-specific
modeling. J. Hydrol. 2010, 391, 322–332.
42. Jamieson, R.; Joy, D.M.; Lee, H.; Kostaschuk, R.; Gordon, R. Transport and deposition of
sediment-associated Escherichia coli in natural streams. Water Res. 2005, 39, 2665–2675.
43. Muirhead, R.W.; Collins, R.P.; Bremer, P.J. Interaction of Escherichia coli and soil particles in
runoff. Appl. Environ. Microbiol. 2006, 72, 3406–3411.
44. Loveland, J.P.; Ryan, J.N.; Amy, G.L.; Harvey, R.W. The reversibility of virus attachment to
mineral surfaces. Colloids Surf. A 1996, 107, 205–221.
45. Soupir, M.L.; Mostaghimi, S. Escherichia coli and Enterococci attachment to particles in runoff
from highly and sparsely vegetated grassland. Wat. Air Soil Poll. 2011, 216, 167–178.
46. Bekhit, H.M.; El-Hordy, M.A.; Hassan, A.E. Contaminant transport in groundwater in the
presence of colloids and bacteria: Model development and verification. J. Contam. Hydrol. 2009,
108, 152–167.
47. Brookes, J.D.; Antenucci, J.; Hipsey, M.; Burch, M.D.; Ashbolt, N.J.; Fergusson, C. Fate and
transport of pathogens in lakes and reservoirs. Environ. Int. 2004, 30, 741–759.
48. Garcia-Armisen, T.; Servais, P. Partitioning and fate of particle-associated E. coli in river waters.
Water Environ. Res. 2009, 81, 21–28.
49. Gutierrez, L.; Nguyen, T.H. Interactions between rotavirus and Suwannee River organic matter:
Aggregation, deposition, and adhesion force measurement. Environ. Sci. Technol. 2012, 21,
8705–8713.
50. Abudalo, R.A.; Ryan, J.N.; Harvey, R.W.; Metge, D.W.; Landkamer, L. Influence of organic
matter on the transport of Cryptosporidium parvum oocysts in a ferric oxyhydroxide-coated
quartz sand saturated porous medium. Water Res. 2010, 44, 1104–1113.
51. Searcy, K.E.; Packman, A.I.; Atwill, E.R.; Harter, T. Association of Cryptosporidium parvum
with suspended particles: Impact on oocyst sedimentation. Appl. Environ. Microbiol. 2005, 71,
1072–1078.
52. Dumètre, A.; Aubert, D.; Puech, P.H.; Hohweyer, J.; Azas, N.; Villena, I. Interaction forces drive
the environmental transmission of pathogenic protozoa. Appl. Environ. Microbiol. 2012, 78,
905–912.
53. Auer, M.T; Niehaus, S.L. Modeling fecal-coliform bacteria. 1. Field and laboratory
determination of loss kinetics. Water Res. 1993, 27, 693–701.
54. Ferguson, C.; Husman, A.M.D.; Altavilla, N.; Deere, D.; Ashbolt, N. Fate and transport of
surface water pathogens in watersheds. Crit. Rev. Environ. Sci. Technol. 2003, 33, 299–361.
55. Sinton, L.W.; Finlay, R.K.; Lynch, P.A. Sunlight inactivation of fecal bacteriophages and
bacteria in sewage-polluted seawater. Appl. Environ. Microbiol. 1999, 65, 3605–3613.
56. Noble, R.; Lee, I.; Schiff, K. Inactivation of indicator microorganisms from various sources of
faecal contamination in seawater and freshwater. J. Appl. Microbiol. 2004, 96, 464–472.
57. Gronewold, A.D.; Myers, L.; Swall, J.L.; Noble, R.T. Addressing uncertainty in fecal indicator
bacteria dark inactivation rates. Water Res. 2011, 45, 652–664.
58. Delpla, I.; Baures, E.; Jung, A.-V.; Thomas, O. Impacts of rainfall events on runoff water quality
in an agricultural environment in temperate areas. Sci. Total Environ. 2011, 409, 1683–1688.
59. Hata, A.; Katayama, H.; Kojima, K.; Sano, S.; Kasuga, I.; Kitajima, M.; Furumai, H. Effects of
rainfall events on the occurrence and detection efficiency of viruses in river water impacted by
combined sewer overflows. Sci. Total Environ. 2013, 468, 757–763.
60. McBride, G.B.; Stott, R.; Miller, W.; Bambic, D.; Wuertz, S. Discharge-based QMRA for
estimation of public health risks from exposure to stormwater-borne pathogens in recreational
waters in the United States. Water Res. 2013, 47, 5282–5297.
61. Beaudeau, P. Impact Sanitaire D’un Accident Sur le Réseau D’adduction en eau Potable du
Havre, D’une Panne De Désinfection à FECAMP et de 4 Episodes de Turbidité Dans Des
Secteurs Ruraux (Seine-Maritime, 1998). Rapport de la Direction Départementale des Affaires
Sanitaires et Sociales de Seine-Maritime et du Laboratoire D’études et D’analyses de la Ville du
Havre; Report for the Social and Sanitary Departmental Direction (Seine-Maritime) and the
studies and analyses Laboratory (Havre): Havre, MT, USA, 1999.
62. Beaudeau, P.; de Valk, H.; Vaillant, V.; Mannschott, C.; Tillier, C.; Mouly, D.; Ledrans, M.
Lessons learned from ten investigations of waterborne gastroenteritis outbreaks, France,
1998–2006. J. Water Health 2008, 6, 491–502.
63. Zmirou, D.; Ferley, J.P.; Collin, J.F.; Charrel, M.; Berlin, J. A follow-up study of
gastro-intestinal diseases related to bacteriologically substandard drinking water. Am. J. Public
Health 1987, 77, 582–584.
64. Brookes, J.D.; Hipsey, M.R.; Burch, M.D.; Regel, R.H.; Linden, L.G.; Ferguson, C.M.;
Antenucci, J.P. Relative value of surrogate indicators for detecting pathogens in lakes and
reservoirs. Environ. Sci. Technol. 2005, 39, 8614–8621.
65. Hipsey, M.R.; Antenucci, J.P.; Brookes, J.D.; Burch, M.D.; Regel, R.H.; Davies, C.M.; Ashbolt,
N.J.; Ferguson, C. Hydrodynamic of Pathogens in Lakes and Reservoirs; American Water Works
Research Foundation: Denver, CO, USA, 2005; Report 91073F.
66. Wyer, M.D.; Kay, D.; Watkins, J.; Davies, C.; Kay, C.; Thomas, R.; Porter, J.; Stapleton, C.M.;
Moore, H. Evaluating short-term changes in recreational water quality during a hydrograph event
using a combination of microbial tracers, environmental microbiology, microbial source tracking and
hydrological techniques: A case study in Southwest Wales, UK. Water Res. 2011, 44, 4783–4795.
67. Gargala, G. Evaluation Des Risques Humains et Environnementaux Associés àLa Présence de
Cryptosporidium Dans L’environnement Hydrique de Haute-Normandie. In Proceeding of
Second National Congress SociétéFrançaise SantéEnvironnement (SFSE), Paris, France, 14–15
December 2011.
68. Soupir, M.L.; Mostaghimi, S.; Dillaha, T. Attachment of Escherichia coli and Enterococci to
particles in runoff. J. Environ. Qual. 2010, 39, 1019–1027.
69. Tillaut, H.; Encrenaz, N.; Checlair, E.; Alexandre-Bird, A.; Gomes Do Esperito Santo, E.;
Beaudeau, P. Epidémie de gastro-entérite, Isère, novembre 2002. Bull. Environ. Hydrol. 2004, 12,
47–48.
70. Delbec, M.; Chesnot, T.; Mignard, C.; Duchemin, J. Risques Microbiologiques Emergents Pour
la Ressource en Eau, Cas de L’agglomération Parisienne. In Proceeding of Second National
Congress SociétéFrançaise SantéEnvironnement (SFSE), Paris, France, 14–15 December 2011.
71. Beaudeau, P.; Pascal, M.; Mouly, D.; Galey, C.; Thomas, O. Health risks associated with
drinking water in a context of climate change in France: A review of surveillance requirements.
J. Water Clim. Change 2011, 2, 230–246.
72. Aumond, M.; Joannis, C. Turbidity Monitoring in Sewage. In Proceeding of 10th International
Conference on Urban Drainage, Copenhague, Denmark, 21–26 August 2005.
73. Henckens, G.; Veldkamp, R.; Schuit, T. On Monitoring of Turbidity in Sewers. Global Solutions
for Urban Drainage. In Proceedings of the Ninth International Conference on Urban Drainage
(9ICUD), Portland, OR, USA, 8–13 September 2002.
74. Langeveld, J.G.; Veldkamp, R.G.; Clemens, F. Suspended solids transport: An analysis based on
turbidity measurements and event based fully calibrated hydrodynamic models. Water Sci. Technol.
2005, 52, 93–101.
75. Ruban, G.; Bertrand-Krajewski, J.L.; Chebbo, G.; Gromaire, M.C.; Joannis, C. Accuracy and
reproducibility of turbidity measurements in urban waste water. Houille Blanche. Revue
Internationale de l'Eau 2006, 4, 129–135.
76. Chebbo, G.; Bachoc, A.; Laplace, D.; Leguennec, B. The transfer of solids in combined sewer
networks. Water Sci. Technol. 1995, 31, 95–105.
77. Deletic, A.B.; Maksimovic, C.T. Evaluation of water quality factors in storm runoff from paved
areas. J. Environ. Eng. 1998, 124, 869–879.
78. Maréchal, A. Relations Entre Caractéristiques de la Pollution Particulaire et Paramètres Optiques
Dans Les Eaux Résiduaires Urbaines. PhD Thesis, Institut national polytechnique de Lorraine:
Nancy, France, 2000.
79. Grüning, H.; Orth, H. Investigations of the dynamic behaviour of the composition of combined
sewage using on-line analyzers. Water Sci. Technol. 2002, 45, 77–83.
80. Pronk, M.; Goldscheider, N.; Zopfi, J. Dynamics and interaction of organic carbon, turbidity and
bacteria in a Karst aquifer system. Hydrogeol. J. 2006, 14, 473–484.
81. Page, R.M.; Scheidler, S.; Polat, E.; Svoboda, P.; Huggenberger, P. Faecal indicator bacteria:
Groundwater dynamics and transport following precipitation and river water infiltration. Water
Air Soil Pollut. 2012, 223, 2771–2782.
82. Hannouche, A.; Chebbo, G.; Ruban, G.; Tassin, B.; Joannis, C. Relation entre la turbiditéet les
matières en suspension en réseau d’assainissement unitaire. Techniques Sciences et Méthodes
2011, 10, 42–50.
83. Goldscheider, N.; Pronk, M.; Zopfi, J. New insights into the transport of sediments and
microorganisms in Karst groundwater by continuous monitoring of particle-size distribution. Geol.
Croat. 2010, 63, 137–142.
84. Atteia, O.; Kozel, R. Particle size distributions in waters from a Karstic aquifer: From particles to
colloids. J. Hydrol. 1997, 201, 102–119.
85. Ferrari, B.C.; Stoner, K.; Bergquist, P.L. Applying fluorescence based technology to the recovery and
isolation of Cryptosporidium and Giardia from Industrial wastewater streams. Water Res. 2006, 40,
541–548.
86. King, D.N.; Brenner, K.P.; Rodgers, M.R. A critical evaluation of a flow cytometer used for
detecting Enterococci in recreational waters. J. Water Health 2007, 5, 295–306.
87. Parlanti, E.; Wörz, K.; Geoffroy, L.; Lamotte, M. Dissolved organic matter fluorescence
spectroscopy as a tool to estimate biological activity in a coastal zone submitted to anthropogenic
inputs. Org. Geochem. 2000, 31, 1765–1781.
88. Hudson, N.; Baker, A.; Reynolds, D. Fluorescence analysis of dissolved organic matter in
natural, waste and polluted waters—A review. River Res. Appl. 2007, 23, 631–649.
89. Naden, P.S.; Old, G.H.; Eliot-Laize, C.; Granger, S.J.; Hawkins, J.M.B.; Bol, R.; Haygarth, P.
Assessment of natural fluorescence as a tracer of diffuse agricultural pollution from slurry spreading
in intensively-farmed grasslands. Water Res. 2010, 44, 1701–1712.
90. Jaffrezic, A.; Jardé, E.; Pourcher, A.M.; Gourmelon, M.; Caprais, M.P.; Heddadj, D.; Cottinet, P.;
Bilal, M.; Derrien, M.; Marti, R.; et al. Microbial and chemical markers: Runoff transfer in
animal manure-amended soils. J. Environ. Qual. 2011, 40, 959–968.
91. Lefcourt, A.M.; Kim, M.S.; Chen, Y.-R. A transportable fluorescence imagining system for
detecting fecal contaminants. Comput. Electron. Agr. 2005, 48, 63–74.
92. Shahid, P.; Venkataraman, C.; Mukherji, S. A review on advantages of implementing
luminescence inhibition test (Vibrio fischeri) for acute toxicity prediction of chemicals. Environ.
Int. 2006, 32, 265–268.
93. Birmele, M.; Ripp, S.; Jegier, P.; Roberts, M.S.; Sayler, G; Garland, J.L. Characterization and
validation of a bioluminescent bioreporter for the direct detection of Escherichia coli. J.
Microbiol. Methods 2008, 75, 354–356.
94. Connelly, J.T.; Baeumner, A.J. Biosensors for the detection of waterborne pathogens. Anal.
Bioanal. Chem. 2012, 402,117–127.
95. Xu, X.; Ying, Y. Microbial biosensors for environmental monitoring and food analysis. Food
Rev. Int. 2011, 27, 300–329.
96. Pérez-Lopez, B.; Merkoçi, A. Portable Chemical Sensors: Weapons against Bioterrorism.
In Biosensors for Safety and Security Applications; Nikolelis, D.P., Ed.; Springer: Berlin,
Germany, 2012; pp. 43–61.
97. Bowers, D.G.; Binding, C.E. The optical properties of mineral suspended particles: A review and
synthesis. Estuarine Coastal Shelf Sci. 2006, 67, 1–2 .
98. Astoreca, R.; Doxaran, D.; Ruddick, K.; Rousseau, V.; Lancelot, C. Influence of suspended
particle concentration, composition and size on the variability of inherent optical properties of
the Southern North Sea. Continental Shelf Res. 2012, 35, 117–128.
99. Stramski, D.; Mobley, C.D. Effects of microbial particles on oceanic optics: A database of
single-particle optical properties. Limnol. Oceanography 1997, 42, 538–549.
100. Stadler, H.; Klock, E.; Skritek, P.; Mach, R.L.; Zerobin, W.; Farnleitner, A.H. The spectral
absorption coefficient at 254nm as a real-time early warning proxy for detecting faecal pollution
events at alpine karst water resources. Wat. Sci. Technol. 2010, 62, 1898–1906.
101. Thomas, O.; Burgess, C. From Spectra to Qualitative and Quantitative Results. In UV-Visible
Spectrophotometry of Water and Wastewater; Thomas, O., Burgess, C., Eds.; Elsevier: New
York, NY, USA, 2007; pp. 1–360.
102. Thomas, O.; Mazas, N.; Massiani, C. Determination of biodegradable dissolved organic carbon
in waters with the use of UV absorptiometry. Environ. Technol. 1993, 14, 487–493.
103. Thomas, O.; El Khorassani, H.; Touraud, E.; Bitar, H. TOC vs. UV spectrophotometry for
wastewater quality monitoring. Talanta 1999, 50, 743–749.
104. Thomas, O.; Gallot, S. UV multiwavelength absorptiometry (UVMA) for the examination of
natural waters and wastewaters. Fresenius J. Anal. Chem. 1990, 338, 234–237.
105. Thomas, O.; Theraulaz, F.; Domeizel, M.; Massiani, C. UV spectral deconvolution: A valuable
tool for wastewater quality determination. Environ. Technol.1993, 14, 1187–1192.
106. Azema, N.; Pouet, M.-F.; Berho, C.; Thomas, O. Wastewater suspended solids study by optical
methods. Colloids Surf. A 2002, 204, 131–140.
107. Bayle, S.; Azéma, N.; Berho, C.; Pouet, M.-F.; Lopez-Cuesta, J.-M.; Thomas, O. Study of
heterogeneous suspensions: A new quantitative approach coupling laser granulometry and
UV-visible spectrophotometry. Colloids Surf. A 2005, 262, 242–250.
108. Berho, C.; Pouet, M.-F.; Thomas, O. Study of the impact of mechanical treatments on waste
water solids by UV spectrophotometry. Environ. Technol. 2003, 24, 1545–1551.
109. Berho, C.; Pouet, M.-F.; Bayle, S.; Azema, N.; Thomas, O. Study of UV-visible responses of
mineral suspensions in water. Colloids Surf. A 2004, 248, 9–16.
© 2014 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article
distributed under the terms and conditions of the Creative Commons Attribution license
(http://creativecommons.org/licenses/by/3.0/).
Water and Society II 133
Oracle: objectivizing cyanobacteria-associated

risks in recreational waters
E. Baurès1,2, F. Pitois3, A. V. Jung4 & O. Thomas1,2
1
EHESP Rennes, Sorbonne Paris Cité, Rennes, France
2
INSERM, UMR IRSET 1085, LERES, Rennes, France
3
LIMNOLOGIE SARL, Rennes, France
4
School of Environmental Engineering (EME), Bruz, France
Abstract
The Oracle project intends to ameliorate the management methods of
cyanobacteria and associated toxins crisis in a changing climate context. Various
recent studies emphasize that in a changing climate environment, higher summer
water temperatures could lead to an increase of algal proliferations. This could in
turn necessitate a reinforcement of the monitoring of toxic cyanobacteria in
recreational waters. The aims of this project are a better understanding of
potentially toxic cyanobacteria crisis occurrences in recreational waters, and the
development of predictive monitoring routines.
Our study is based on public health survey data from 26 recreational lakes
monitored every year since 2004 in Brittany (north-west France) in order to
encompass spatial and interannual patterns of cyanobacteria development on a
regional scale.
This project includes 4 steps: (i) characterizing the annual rhythm of
cyanobacteria development, (ii) defining site-specific reference conditions
(iii) developing or improving remote survey techniques, (iv) developing new
cyanobacteria monitoring protocols based on site-specific crisis indicators.
Exploring the links between cyanobacteria cell densities, toxin occurrences
and interannual meteorological factors on one hand and characterizing the
climate change consequences on local meteorological conditions on another hand
can give us insights on the potential evolution of cyanobacterial crisis in the
future.
Keywords: cyanobacteria, microcystins, climate change, recreational water
quality.
WIT Transactions on Ecology and The Environment, Vol 178, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
doi:10.2495/WS130121
134 Water and Society II
1 Introduction
Since their introduction in the public health debate [1–3], cyanobacteria and their
toxins have gradually become a major concern for public health authorities.
Literature reviews focusing on toxin production have been regularly published in
the last 10 years, so biosynthetic pathways and toxicological issues are relatively
well known [4–7].
There is however no evidence of any direct relation between cyanobacterial
biomass and toxin concentration, nor any consensus on toxins origin and
ecological role in the cyanobacterial cell life cycle. This lack of knowledge of
toxin causality and of understanding of toxins purpose prevents any attempt in
the prediction of risk occurrence.
The relations between cyanobacterial biomass and trophic enrichment
(eutrophication processes) are widely admitted, whereas the influence of
meteorological parameters on in situ algal growth is rarely discussed. Some
abiotic parameters have nevertheless been studied in vitro in terms of selection
and growth rates of potentially toxic species [8–13] and through their influence
on toxin biosynthesis [14–18].
Reports of in vitro influence of light and temperature on cyanobacterial
growth and toxin production emphasize the possible consequences of climate
change on public health concerns. Exploring the relations between in situ abiotic
factors, climate change consequences, cyanobacteria distribution and public
health parameters on a regional scale is the main objective of the ORACLE
project. This project was initiated in 2012 and is based on an analysis of
monitoring data collected by local health authorities since 2004 in Brittany (NW
France). This analysis should enlighten some aspects of the interannual dynamics
of cyanobacterial development in 26 lakes used as recreative waters, and is
directed toward the development of indices predictive of cyanobacteria-
associated risk crisis.
2 Methodology
2.1 Data sources
Our study is based on public health survey data from 26 recreational lakes
monitored every year since 2004 in Brittany in the north-western quarter of
France.
Cyanobacteria data, i.e. cell densities, species composition and microcystin
concentrations, were collected from the regional public health authorities
(Agence Régionale de Santé de Bretagne).
All other parameters were gathered from dedicated institutional databanks for
hydrology (Minister for the Environment: BDHydro) and climate (MétéoFrance:
Climathèque).
Lake morphology, i.e. depth, lake and watershed area, volume… came from
Agence de l’Eau Loire-Bretagne or directly from lake owners.
0 50 100Km. 0 100 km
Figure 1: Brittany location and spatial distribution of the 26 investigated

sites.
2.2 Data limitations
Several categories of limitation were identified with the monitoring data:

- Survey duration can vary every year on a local basis. Finding a common annual
basis for data comparison in all sites reduced the time range to 16 weeks from
early June to mid-September. This time range couldn’t include all the initial
cyanobacteria appearance phases (from March to May), nor their decay phase
(from October to November): our data thus only cover 50% of the common time
range where cyanobacteria can be observed in Brittany.
- Survey continuity: interannual exploitation of recreative waters cannot be
assured, especially in case of low water quality. Only 26 sites out of a total of 40
could be considered as continuously monitored since 2004. Our data could then
only account for 58% of the monitoring data potentially available under a
continuous sampling program.
- Species composition lists are often incomplete, as sometimes all taxa
considered as non-potentially toxic (chlorophyts, diatoms…) were left
unidentified. Incomplete lists account for 23% of all the results available on the
26 selected sites.
- Hydrology: 30% of all sites are subjected to hydrometric surveys, leaving 70%
of all discharges and residence times to be extrapolated from pluviometric data.
- Water quality monitoring for nutrients are only present in 2 lakes: no trophic
status index could be integrated initially to the study.
- Toxin analysis: Microcystin (MCs) monitoring can rely on various methods
(ELISA, HPLC, LC/MS) according to laboratories. None of these methods can
give comparable results. The available MCs data were converted from
quantitative results to positive detection frequencies, where a result higher than
the highest quantification limit of all methods is considered a positive detection.
3 Results
3.1 Cyanobacteria and associated risk parameters
The first phase of the study focused on generic risk parameters for all 26 sites:
exposition intensity, i.e. cyanobacteria cell density; exposition duration, i.e. the
number of weeks every year with cell density higher than WHO alert level 3
(100 000 cell/ml, [1]); exposition recurrence, i.e. the number of years with cell
density higher than WHO level 3 whatever the duration; and MCs annual
detection frequencies.
This phase underlined:
- an interannual increase of the frequency of samples showing cell densities
higher than WHO alert levels 2 and 3 (fig. 2a)
- a positive, annual relation between exposition duration and maximal cell
densities (fig. 2b)
- a positive relation between interannual recurrence and annual exposition
duration (fig. 2c)
- an inverse relation between MCs detection frequency and cyanobacteria cell
density (fig. 2d)
2a 2b
90
10000000
WHO level frequency rank (% samples)
mean cyanobacteria cell density (cell/ml)
60 1000000
WHO level 2
WHO level 3
30 100000
0 2 4 6 8 10 12 14 16 18
2004 2006 2008 2010 2012
total time above WHO level 3 (weeks/year)
14
12 2c 5 000 000 2d
Cell density > WHO lev. 3 (weeks)
2 000 000
10
cell density (cell.ml )
-1
1 000 000
8
500 000
6 200 000
100 000
4
20 000
2
10 000
0
0 5 10 15 20 25 30 35
2 4 6 8 10
percent positive detection (>0.2 µg.l-1)
Yrs. with observed WHO lev.3 occurence
Figure 2: Cyanobacteria exposition parameters summarized, 2004–2011.
The abiotic factors identified as correlated with these health risk parameters
are summarized in table 1.
Table 1: Abiotic factors correlated with cyanobacterial risk parameters

where +/- are the Pearson correlation signs for p < 0.05.
Cell density Duration > Interannual MCs

WHO lev. 3 Recurrence detection
frequency
Morphology
Depth - - -
Surface -
Watershed drainage index -
Meteorology
Summer temperature +
Winter temperature
Summer sunshine - +
Winter sunshine + + -
Summer pluviometry -
Winter pluviometry -
Position
Longitude vs N. - -
Latitude vs W.
3.2 Cyanobacteria species composition
Cyanobacteria species composition was separated into 5 subgroups according to

taxonomic and cell dimensions criteria, and following the strategists’ definition
by Reynolds [19, 20] of c-strategists (competitor species), r-strategists (ruderal
species) and s-strategists (stress-resistant species):
- r-Nostocales (e.g. Anabaena sp., Aphanizomenon sp.)
- c-Chroococcales (e.g. Aphanothece sp., Aphanocapsa sp., Merismopedia sp,
Snowella sp.…)
- s-Chroococcales (Microcystis sp., Worochininia sp.)
- c-Oscillatoriales (e.g. Limnothrix sp., Pseudanabaena sp.,…)
- r-Oscillatoriales (Planktothrix sp.)
The species known for potential toxin production are then the r-Nostocales, s-
Chroococcales and r-Oscillatoriales. Our data show however that cell densities of
r-Nostocales and s-Chroococcales are positively correlated with MCs detection
frequency (r²=0.55 et 0.61, p<0.05), which is not the case for r-Oscillatoriales
(r²=-0.74, p<0.05).
We defined group dominance as the frequency of one strategist group
accounting for at least 50% of a sample total cell density. Under this assumption,
only meteorological factors could be identified as correlated with group
dominance episodes (table 2). Surprisingly, some winter conditions appeared
correlated with cyanobacterial development in the next summer.
Table 2: Factors correlated with dominance episodes of cyanobacterial

strategist groups (+/-: Pearson correlation signs for p<0.05).
r-Nostoc. c-Chrooc. s-Chrooc. c-Oscil. r-Oscil.

Meteorology
Summer temperature + - +
Winter temperature +
Summer sunshine + -
Winter sunshine - + -
Summer rain - +
Interestingly, the s-Chroococcales (Microcystis, Worochininia) showed no

correlation with any abiotic parameter, i.e. neither meteorology nor morphology.
This can indicate that the relevant factors (trophic status index, residence time)
are still to be integrated in the study.
3.3 Cyanobacteria at the regional scale
A dominance episode frequency mapping showed distinct NW-SE separations

between 3 strategist groups: c-Chroococcales are concentrated in the western part
of Brittany, whereas r-Nostocales and r-Oscillatoriales are concentrated in the
eastern part (fig. 3). Dominance frequencies for c-Oscillatoriales and
s-Chroococcales did not show notable geographic distribution.
c-Chroococcales r-Nostocales
r-Oscillatoriales
20 %
40 %
60 %
80 %
100 %
Figure 3: Spatial distribution of cyanobacterial group dominance frequency.
This geographic distribution of c-Chroococcales, r-Nostocales and

r-Oscillatoriales can be compared with their dependency to meteorological
factors (table 2), and with regional climatic distribution of temperature, sunshine
and rain during the growth period (i.e. May–October, fig. 4).
Contour Graph Contour Graph 2
Mean temperature (°C, May-October)

6.9e+6 6.9e+6 Mean cumulated rain (mm, May-October)
6.9e+6 14.0 6.9e+6 150

14.5 200
15.0
250
15.5
300
6.8e+6 16.0 6.8e+6
16.5 350
17.0 400
Y Y
D 17.5 D 450
a
ta 6.8e+6 a
ta 6.8e+6 500
6.7e+6 6.7e+6
1e+5 2e+5 2e+5 3e+5 3e+5 4e+5 4e+5 5e+5 5e+5 1e+5 2e+5 2e+5 3e+5 3e+5 4e+5 4e+5 5e+5 5e+5
X Data X Data
Contour Graph 2 Contour Graph 2
6.9e+6 Mean cumulated sunshine (hrs., May-October) 6.9e+6 Mean cumulated radiation (W/m², May-October)
540
6.9e+6 6.9e+6
560
2.7e+5
580
2.8e+5
600
2.9e+5
6.8e+6 620 6.8e+6
3.0e+5
640
Y Y 3.1e+5
660
D D 3.2e+5
a 680 a
ta 6.8e+6 ta 6.8e+6 3.3e+5
700
720
6.7e+6 6.7e+6
1e+5 2e+5 2e+5 3e+5 3e+5 4e+5 4e+5 5e+5 5e+5 1e+5 2e+5 2e+5 3e+5 3e+5 4e+5 4e+5 5e+5 5e+5
X Data X Data
Figure 4: Climate parameters in Brittany inferred from 2001–2011 data.
Oceanic climate in Brittany shows a distinct separation along a NW to SE

axis parallel to the sites distribution line, with increasing temperatures and
sunshine and decreasing pluviometry in the SE direction. This can be compared
with distribution data for r-Nostocales (temperature and sunshine: positive
correlations) and s-Chroococcales (temperature: negative correlation), with cell
densities and exposition duration (longitude: negative correlation), and with MCs
detection frequency (temperature and sunshine: positive correlations).
Relating meteorological factors to cyanobacteria in a changing climate
context is part of the ORACLE project. The temperature and pluviometry
evolution for the May–October period were extrapolated from 1st order
regression slopes and mapped (fig. 5). Temperatures show a strong tendency to
increase in all Brittany, with a maximal increase rate on the northern coast. On
the other hand, summer pluviometry tends to increase on the western and eastern
sectors while decreasing in the center of Brittany.
Sunshine data are currently being completed at the regional scale, but the
available data show a tendency to increase in the NW and SE quadrants, while
decreasing in the NE and SW quadrants. This is of particular concern, as higher
sunshine values and temperatures are positively correlated to r-Nostocales
dominance phases and MCs detection frequencies.
Mean May-October temperature evolution Cumulated May-October rain evolution

(1st order regression slope, °C/100 ans ) (1st order regression slope, mm/100 ans )
-300
-200
1 -100
2 0
3 100
4 200
5
6
7
Figure 5: Evolution of temperature and pluviometry in Brittany extrapolated

from monthly data 1981–2011.
4 Conclusion
At the end of its first year, the ORACLE project results are indicative of further
potential development:
- site-specific morphological and climatic parameters could be integrated in a
risk index indicative of each site ability to a recreative exploitation. This index
could be used as a support for Bathing Water Directive 2006/7/CE [21], as this
directive asks for the integration of cyanobacteria parameters without giving
guidelines or risk parameter definitions.
- at the annual scale, parameters such as cell density, MCs detection frequency
and exposure duration appear correlated with winter meteorology (e.g.
pluviometry and sunshine hours). This could be developed as a predictive
indicator of hazard situations before the beginning of recreative water use in the
next summer.
- at the survey time scale, parameters such as temperature, sunshine hours and
pluviometry appear to be indicators of crisis situation occurring in the next 20 to
30 days.
A closer use of meteorological surveys in the prediction of cyanobacteria and
toxin occurrences appears to be an important perspective, potentially leading to
alerting procedures directed specifically toward high cell densities and/or toxin
occurrences.
As a wider development perspective, it can be noted that climate change
consequences in Brittany are noticeable and show high geographic variability on
a 300 km scale.
Several studies already described long-term modifications of cyanobacterial
populations responses to climate change, in terms of proliferation frequencies
(see for example [22–24]) or in terms of species composition [25–27], although
in some cases no evolution can be highlighted [28].
Our results show a dependency of summer cyanobacteria populations not only
on summer conditions, but also on winter meteorology. Moreover, long-term
evolution of the summer and winter parameters can differ locally, or even show
opposite slopes at small geographic scale. Integrating this small scale variability
into a larger, regional scale model could enlighten the apparently opposite results
of more site-specific studies.
Acknowledgements
The ORACLE project is funded by the French National Agency of Sanitary
Security (ANSES) under the national program of Environment-Health-Work
research (PNRSET 2012). The authors sincerely thank the French Agence
Régionale de Santé (ARS) of Brittany and the French Agence de l’Eau Loire-
Bretagne (AELB) for the data provided and are grateful to Caroline Farenc,
student at the School of Environmental Engineering (EME), for the data
exploitation.
References
[1] Chorus I., Bartram J., (1999), Toxic Cyanobacteria in Water: a Guide to
Public Health Significance, Monitoring and Management. WHO, Chapman
& Hall, London, 416 pp
[2] Falconer I.R., (1989), Effects on human health of some toxic cyanobacteria
(blue-green algae) in reservoirs, lakes, and rivers, Environmental
Toxicology, 4, 175-184
[3] Falconer I.R., (1996), Potential impact on human health of toxic
cyanobacteria. Phycologia, 35, 6-11
[4] Leflaive J., Ten-Hage L., (2007), Algal and Cyanobacterial secondary
metabolites in freshwaters: a comparison of allelopathic compounds and
toxins, Freshwater Biology, 52, 199-214
[5] Smith J.L., Boyer G.L., Zimba P.V., (2008), A review of Cyanobacterial
odorous and bioactive metabolites: impacts and management alternatives in
aquaculture, Aquaculture, 280, 5-20
[6] Aràoz R., Molgo J., Tandeau de Marsac N., (2010), Neurotoxic
cyanobacterial toxins, Toxicon, 56, 813-828
[7] Pearson L., Mihali T., Moffitt M., Kellmann R., Neilan B., (2010), On the
chemistry, toxicology and genetics of the cyanobacterial toxins,
microcystin, nodularin, saxitoxin and cylindrospermopsin, Marine drugs, 8,
1650-1680
[8] De Nobel W.T., Matthijs H.C.P., Von Elert E., Mur L.R., (1998),
Comparison of the light-limited growth of the nitrogen-fixing cyanobacteria
Anabaena and Aphanizomenon, New Phytologist, 138, 579-587
[9] Baker P.D., Bellifemine D., (2000), Environmental influences on akinete
germination of Anabaena circinalis and implications for management of
cyanobacterial blooms, Hydrobiologia, 427, 65-73
[10] Hadas O., Pinkas R., Malinsky-Rushansky N., Shalev-Alon G., Delphine
E., Berner T., SukenikA., Kaplan A., (2002), Physiological variables
determined under laboratory conditions may explain the bloom of
Aphanizomenon ovalisporum in Lake Kinneret, European Journal of
Phycology, 37, 259-267
[11] McCausland M.A., Thompson P.A., Blackburn S.I., (2005),

Ecophysiological influence of light and mixing on Anabaena circinalis
(Nostocales, Cyanobacteria), European Journal of Phycology, 40, 9-20
[12] Yamamoto Y., Nakahara H., (2006), Importance of interspecific
competition in the abundance of Aphanizomenon flos-aquae
(Cyanophyceae), Limnology, 7, 163-170
[13] Imai H., Chang K-H., Kusaba M., Nakano S-I., (2009), Temperature-
dependent dominance of Microcystis (Cyanophyceae) species: M.
aeruginosa and M. weisenbergii, Journal of Plankton Research, 31, 171-178
[14] Rapala J., Sivonen K., (1998), Assessment of environmental conditions that
favor hepatotoxic and neurotoxic Anabaena spp. strains cultured under light
limitation at different temperatures, Microbial Ecology, 36, 181-192
[15] Wiedner C., Visser P.M., Fastner J., Metcalf J.S., Codd G.A., and Mur
L.R., (2003), Effects of light on the microcystin content of Microcystis
strains PCC 7806, Applied and Environmental Microbiology, 69, 1475-
1481
[16] Tonk L., Visser P.M., Christiansen G., Dittmann E., Snelder E.O.F.M.,
Wiedner C., Mur L.R., Huisman J., (2005), The microcystine composition
of the cyanobacterium Planktothrix agardhii changes toward a more toxic
variant with increasing light intensity, Applied and Environmental
Microbiology, 71(9), 5177-5181
[17] Briand E., Yéprémian C., Humbert J-F., Quiblier C., (2008), Competition
between microcystin- and non-microcystin-producing Planktothrix agardhii
(cyanobacteria) strains under different environmental conditions,
Environmental Microbiology, 10(12), 3337-3348
[18] Preußel K., Wessel G., Fastner J., Chorus I., (2009), Response of
cylindrospermopsin production and release in Aphanizomenon flos-aquae
(Cyanobacteria) to varying light and temperature conditions, Harmful
Algae, 8, 645-650
[19] Reynolds, 1998, Functional morphology and the adaptive strategies of
freshwater phytoplankton, in C. D. Sandgren (ed.), Growth and
reproductive strategies of freshwater phytoplankton. Cambridge University
Press, New York: 388–433
[20] Reynolds, C. 2006. Ecology of Phytoplankton. Cambridge University Press.
Cambridge, UK
[21] Official Journal of the European Union, 2006, Directive 2006/7/EC of 15
February 2006 concerning the management of bathing water quality and
repealing Directive 76/160/EEC
[22] Carvalho L., Kirika A., (2003), Changes in shallow lake functioning:
response to climate change and nutrient reduction, Hydrobiologia, 506-509,
789-796
[23] Malmaeus J.M., Blenckner T., Markensten H., Persson I., (2006), Lake
phosphorus dynamics and climate warming: A mechanistic model
approach, Ecological Modelling, 190, 1-14
[24] Mooij W.M., Janse J.H., De Senerpont Domis L.N., Hülsmann S., Ibelings
B.W., (2007), Predicting the effect of climate change on temperate shallow
lakes with the ecosystem model PCLake, Hydrobiologia, 584, 443-454
[25] Nõges T., Nõges P., Laugaste R., (2003), Water level as the mediator
between climate change and phytoplankton composition in a large shallow
temperate lake, Hydrobiologia, 506-509, 257-263
[26] Wiedner C., Rücker J., Brüggemann R., Nixdorf B., (2007), Climate change
affects timing and size of populations of an invasive cyanobacterium in
temperate regions, Oecologia, 152(3), 473-484
[27] Liu X., Lu X., Chen Y., (2011), The effects of temperature and nutrient
ratios on Microcystis blooms in Lake Taihu, China: An 11-year
investigation, Harmful Algae, 10, 337-343
[28] Vasconcelos V., Morais J., Vale M., (2011), Microcystins and
cyanobacteria trends in a 14 year monitoring of a temperate eutrophic
reservoir (Aguieira, Portugal), J. Environ. Monit., 13, 668-672
See discussions, stats, and author profiles for this publication at:
http://www.researchgate.net/publication/272414576
Variability of N Export in Water: A

Review
ARTICLE in CRITICAL REVIEWS IN ENVIRONMENTAL SCIENCE AND TECHNOLOGY ·

FEBRUARY 2015
Impact Factor: 3.47 · DOI: 10.1080/10643389.2015.1010432
5 AUTHORS, INCLUDING:
Jean Causse
3 PUBLICATIONS 1 CITATION
SEE PROFILE
Estelle Baurès
Ecole des hautes études en santé …
29 PUBLICATIONS 254 CITATIONS
SEE PROFILE
Aude-Valérie Jung
Ecole des Métiers de l'Environnem…
19 PUBLICATIONS 230 CITATIONS
SEE PROFILE
Available from: Jean Causse

Retrieved on: 03 September 2015
This article was downloaded by: [85.69.119.83]
On: 01 September 2015, At: 05:04
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: 5 Howick Place, London, SW1P 1WG
Critical Reviews in Environmental

Science and Technology
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/best20
Variability of N Export in Water: A

Review
abc ab c abd
Jean Causse , Estelle BaurèS , Yoann Mery , Aude-ValéRie Jung
ab
& Olivier Thomas
a
EHESP Rennes, Sorbonne Paris Cité, Rennes Cedex, France
b
INSERM U 1085-IRSET, LERES, Rennes Cedex, France
c
Coop de France Ouest, Maison de l'Agriculture, Rennes Cedex,
France
d
School of Environmental Engineering, Bruz, France
Accepted author version posted online: 03 Feb 2015.
To cite this article: Jean Causse, Estelle BaurèS, Yoann Mery, Aude-ValéRie Jung & Olivier Thomas
(2015) Variability of N Export in Water: A Review, Critical Reviews in Environmental Science and
Technology, 45:20, 2245-2281, DOI: 10.1080/10643389.2015.1010432
To link to this article: http://dx.doi.org/10.1080/10643389.2015.1010432
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. Taylor & Francis, our agents,
and our licensors make no representations or warranties whatsoever as to the accuracy,
completeness, or suitability for any purpose of the Content. Versions of published Taylor
& Francis and Routledge Open articles and Taylor & Francis and Routledge Open Select
articles posted to institutional or subject repositories or any other third-party website are
without warranty from Taylor & Francis of any kind, either expressed or implied, including,
but not limited to, warranties of merchantability, fitness for a particular purpose, or non-
infringement. Any opinions and views expressed in this article are the opinions and views
of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of
the Content should not be relied upon and should be independently verified with primary
sources of information. Taylor & Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or
howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Terms &
Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions
It is essential that you check the license status of any given Open and Open
Select article to confirm conditions of access and use.
Downloaded by [85.69.119.83] at 05:04 01 September 2015
Critical Reviews in Environmental Science and Technology, 45:2245–2281, 2015
Published with license by Taylor & Francis
ISSN: 1064-3389 print / 1547-6537 online
DOI: 10.1080/10643389.2015.1010432
Variability of N Export in Water: A Review
JEAN CAUSSE,1,2,3 ESTELLE BAURÈS,1,2 YOANN MERY,3

AUDE-VALÉRIE JUNG,1,2,4 and OLIVIER THOMAS1,2
1
EHESP Rennes, Sorbonne Paris Cité, Rennes Cedex, France
2
INSERM U 1085-IRSET, LERES, Rennes Cedex, France
3
Coop de France Ouest, Maison de l’Agriculture, Rennes Cedex, France
4
School of Environmental Engineering, Campus de Ker-Lann, Bruz, France
In spite of several studies on factors affecting nitrogen (N) export in

water, the spatiotemporal variability of N export is rarely addressed.
This review aims at discussing the different factors involved in N
export in water and the intra-annual variability of N occurrences
in rivers at a watershed scale. From the analysis of the existing
works some recommendations on future research work are pro-
posed regarding (i) other N forms than nitrate being systematically
considered for routine analysis or experimental studies (at least
Kjheldal nitrogen), (ii) hydrologists and biogeochemists working
together to better understanding of N dynamics according to water
pathways, especially during “situations at risk,” (iii) water quality
monitoring being reinforced both spatially and temporally, espe-
cially thanks to high frequency instrumentation, and (iv) accurate
data on watersheds needed to give better explanations of N vari-
ations. We argue that a better understanding of spatiotemporal
variability could greatly enhance the remediation of N export im-
pacts, which is crucial to anticipate the impacts of global changes.
© Jean Causse, Estelle Baurès, Yoann Mery, Aude-Valérie Jung, and Olivier Thomas
This is an Open Access article distributed under the terms of the Commons Attribution
License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distri-
bution, and reproduction in any medium, provided the original work is properly cited. The
moral rights of the named author(s) have been asserted.
Address correspondence to Estelle Baurès, EHESP Rennes, Sorbonne Paris Cité, Av-
enue du Professeur Léon Bernard - CS 74312 35043, Rennes Cedex, France. E-mail: Es-
telle.Baures@ehesp.fr
Color versions of one or more of the figures in the article can be found online at
www.tandfonline.com/best.
2245
2246 J. Causse et al.
KEY WORDS: hydrological processes, nitrogen sources and

export, rural watershed, situations at risk, spatiotemporal
variability
1. INTRODUCTION
Excess of nutrients in water is a recognized threat for ecosystems and hu-

man health. Increases of nitrate concentrations in rivers remain one of the
most serious environmental concerns, as it affects both developing and de-
veloped countries, with a high degree of advancement (Galloway et al.,
2008). The main factors are the demographic explosion with development
of anthropogenic activities resulting in increases of waste volumes, defor-
estation for agricultural lands, increases of fertilizer use, intensification of
animal production . . . . Scientists have pointed out the effect of such un-
controlled development on human health since 1945 regarding high con-
centration of nitrate in drinking water (Comly, 1945; Fewtrell, 2004; Greer
and Shannon, 2005) and others pathologies like cancers and reproductive
outcomes (Ward et al., 2005) or on the sustainability of ecosystems since
the beginning of the 1970s (Ryther and Dunstan, 1971). Concerning factors
affecting the export of nitrogen a lot of works have been published during
the last decade (Drewry et al., 2006; Pellerin et al., 2006; Schindler et al.,
2006; Alvarez-Cobelas et al., 2008; Howarth, 2008; Sutton et al., 2011; Pärn
et al., 2012). In spite of many efforts made to reduce nitrate discharge from
waste water treatment plants (De Wit and Bendoricchio, 2001; Thieu et al.,
2010) diffuse contaminations from agricultural activities and urban infras-
tructure malfunctions are still unresolved problems. Regarding agricultural
issues, many publications aimed at assessing the beneficial effects of chang-
ing agricultural practices [best management practices (BMPs)] (Behera and
Panda, 2006; Rao et al., 2009; Laurent and Ruelland, 2011), natural buffer
areas (Fisher and Acreman, 2004; Anbumozhi et al., 2005; Sahu and Gu,
2009; Borin et al., 2010; Ranalli and Macalady, 2010), and development of
constructed agro-ecological infrastructures (constructed wetland, denitrifying
barriers) (Blowes et al., 2000; Vymazal, 2007; Mousavi et al., 2012; Schmidt
and Clark, 2012; Tanner et al., 2012). Globally, substantial efforts have been
made by the agricultural community with respect to a tighter legislation (e.g.,
European Directive 91/676/CEE, 1991). However, the complexity of natural
systems and their interactions with anthropogenic activities are still hinder-
ing determination of remediation priorities. If current knowledge is sufficient
to carry out mitigation actions, the improvement of scientific knowledge is
still needed for a better understanding of nutrient dynamics at various spa-
tiotemporal scales at least to answer the following questions. How does
spatiotemporal variability impacts mitigation actions prioritization? What are
the effects of heavy rainfalls and/or variations in anthropogenic activities on
Variability of N Export in Water 2247
global nitrogen export at the watershed scale during one hydrological year?
To what extent are other nitrogen forms than nitrate involved in global N ex-
port during these periods “at risk”? In this context, the aim of this review is to
summarize the current knowledge on nitrogen export at short-term (event)
and mid-term (season). Long-term studies are also considered in order to
show the impacts of land use and climate change on N export. Finally, some
actions for future research work are proposed.
2. N SOURCES
2.1 N Forms
Dissolved nitrogen in water can take several forms, mainly nitrate (NO3 −)
but also nitrite (NO2 −), ammonium (NH4 +) and organic nitrogen (Norg ), this
last coming from organic matter decomposition and giving ammonium by

mineralization. Ammonium can be oxidized into nitrite and then nitrate by
nitrification. Nitrate is the main dissolved compound found in freshwater as
it is both the most soluble and mobile. It originates from nitrification and
fertilizers and can be denitrified to N gas (denitrification), uptaken by plants,
or more rarely transformed into ammonium under reducing conditions (am-
monification) (Robertson and Groffman, 2007).
Several studies show contradictory results on N form proportions be-
tween surface runoff and groundwater. On one hand, NO3 − is rather trans-
ported via groundwater because of its high mobility whereas the other forms
are exported via surface runoff with particles (Robertson and Groffman,
2007). On the other hand, ammonium and organic nitrogen can also be
transported in groundwater (Kroeger et al., 2006; Jiao et al., 2010) whereas
nitrate can be a substantial form of N loads in surface runoff/lateral flow
under high fertilization plots (Jaynes et al., 2001; Zheng et al., 2004; Chen
et al., 2012).
At the watershed scale, balance between all forms of nitrogen in river
can widely vary depending on land use, N sources, human activities, hy-
droclimatic variability, and watershed characteristics. For example, Udawatta
et al. (2006) observed in three adjacent agricultural watersheds that the mean
nitrate percentage in total nitrogen load in runoff was less than 43% during
soybean years against 63% for corn years. Another example is provided by
Taylor et al. (2005), who found similar proportions of dissolved inorganic and
organic nitrogen (50/50) in Melbourne urban catchment, while they found
about two-third proportion of organic nitrogen in international literature.
In Europe, nitrogen flows and budgets have been estimated for terrestrial
ecosystems (agriculture, forest, and other ecosystems) and for all combined
(systems including urban, transport, industrial, and aquatic flows) (Sutton
et al., 2011). The distribution of reactive forms (especially nitrate and ammo-
nium) in aquatic systems is dominated by nitrate losses, which are larger in
areas with high livestock density and precipitation excess, while more local-
ized peaks are associated with urban wastewaters. The budget highlights the
central role of crop production and livestock farming. The annual reactive
forms brought to agricultural soils represent 27.5 Tg N (among 11.2 Tg N as
synthetic fertilizers, 7.1 Tg N as manure, 2.4 Tg N as atmospheric deposition,
1.0 Tg N through biological nitrogen fixation, and 5.8 Tg N as crop residues),
which are over the requirements for crop production (17.6 Tg N).
The review of current studies dealing with nitrogen environmental
sources and fate, also highlights a tendency to address individual reactive
species (NO3 −, NH4 +) from specific sectorial sources (agriculture, traffic, in-
dustry), media (air, freshwater, marine), and for specific issues in general
(climate, urban air pollution, biodiversity, water quality, etc.).
2.2 Type of Sources

Two types of river contamination sources are usually distinguished, accord-
ing to production modes: point source (PS) and non-point source (NPS).
Beside this well-known typology, some authors recently proposed to distin-
guish between continuous and sporadic or episodic or discontinuous sources
(Withers and Jarvie, 2008; Badruzzaman et al., 2012).
2.2.1 POINT SOURCE
Continuous PS emissions are subject to few variations with time. This is the
case for aquaculture (Phan et al., 2009; Sindilariu et al., 2009; Amirkolaie,
2011) and wastewater treatment plant (WWTP) discharges under normal
conditions (dry weather), which consist in relatively constant volumes and
total nitrogen concentration (Carey and Migliaccio, 2009; Igbinosa and Okoh,
2009; Graham et al., 2010).
On the other side discontinuous PS emissions depend on various pa-
rameters. An example in rural area is the access of animals to stream with
possible in-stream defecation representing a significant source of N with herd
crossing the river (Collins et al., 2007; Davies-colley et al., 2004). Other dis-
continuous PS in towns are combined sewage overflows (CSOs) discharging
directly diluted wastewater when flow exceeds the system capacity during
heavy rainfall events (Kim et al., 2007; Bernhardt et al., 2008). CSOs can
happen from few times a year (Bremerton CSO Annual Report, 2014) to
more than 50 times a year in towns where no infrastructure improvement
have been made (Edmonton CSO Control Strategy, 2000; USEPA, 2009). If
many programs have been conducted to reduce this pollution for nearly
thirty years, it still remains a problem for most of urbanized areas with no
separate sewer system, as remediation represents large financial investments.
Furthermore, illicit connections (especially in developing countries) can can-
cel the benefits from a separate sewer system and create similar conditions
than combined systems (Li et al., 2013).
2.2.2 NON-POINT SOURCES

Continuous NPS emissions are diffused and take different pathways to reach
streams. This is the case for on-site sanitation facilities (cesspools, septic
tanks, and pit latrines) and leaky sewer, discharging contaminated wastew-
aters in soils and groundwater with poor removal of nutrients (Reay, 2004).
Amounts and frequency of release depends on daily human excreta and
water use, which are relatively constant. However, in a recent review,
Bernhardt et al. (2008) showed that N loads from these systems can be re-
tained or accumulated in soil profiles during very dry weather, whereas they
are directly routed to groundwater during wetter conditions, when connec-
tivity of hydrological flow path is high. Thus, this source can be considered
as “semi-continuous,” as it is not directly affected by rainfall events but rather
by water table depth and soil water content.
Discontinuous NPS emissions are the most complex sources because of
the need for both availability and transport to reach streams. This is the case
for fertilizer spreading, animal excreta (from livestock on grasslands and
wildlife), manure or biosolids. Mineral fertilizers, manure (animals waste),
and biosolids (treated sewage sludge) are spread at plot scale during spe-
cific periods of the year. Depending on agricultural system, crop rotation,
farmers practices or regulatory constraints, quantities can differ (Sheriff, 2005;
Thompson et al., 2007; Wei et al., 2009). N mineral fertilizers are directly us-
able by vegetation but can also be exported easily in large amounts if climatic
hazard occur, both by surface (Masetti et al., 2008; Liang et al., 2011) and
subsurface run-off. Home garden fertilization is controlled by similar factors,
but also by social and psychological ones. For example, maintaining a lawn
(and fertilizing it) is an avenue to engage with one’s neighbors, to fulfill
expectations of what it means to be a positive member of a community
(Carrico et al., 2012). N losses from organic fertilizers follow different dy-
namics, as organic matter mineralization is a limiting factor of N export.
Generally, nitrogen derived from organic matter accumulates in soil profile
during dry periods and is exported during wet periods (Rimski-Korsakov
et al., 2004; Mantovi et al., 2006). Finally, animal excreta directly produced
on grasslands are generally considered like organic fertilizers, with calcula-
tion of density and time spent by animals in the field (Di and Cameron, 2000).
However, the impact may be higher in this case than for organic waste appli-
cation because of a congregating phenomenon and nutrient supply at times
of low level uptake by plants (Hubbard et al., 2004; McGechan and Topp,
2004).
For non-impervious disposal sites, landfill leachates may vary in quality
and quantity depending on many factors such as site age, precipitation, evap-
otranspiration, temperature, waste type, and composition (Oman and Junest-
edt, 2008; Renou et al., 2008). Leachates may reach groundwater slowly and
consequently affect surface waters with a high delay in normal conditions.
FIGURE 1. Classification of N sources (WWTP: Waste Water Treatment Plant).
During wet weather, leachate can however affect water quality in a very
short time, when surface runoff occurs in large volumes (Marques, 2007;
Mangimbulude et al., 2009). Under humid climate, landfill leachate can also
be considered as a continuous NPS.
Besides nitrate export from agricultural spreading and landfill leakages,
nitrous oxides and ammonia can be emitted in air by natural ecosystems
and human activities (transportation, agriculture, industries, and waste) but
they can be deposited to soils and rivers by both wet and dry input (Asman
et al., 1998; Carrillo, 2002; Paerl et al., 2002). In some areas, inorganic nitro-
gen deposition seems to be the primary factor of eutrophication (Bergstrom
and Jansson, 2006). Dry input is however a continuous NPS, because there
is always deposition, even if quantities vary with time. Finally, the decom-
position of living organisms after their death could also be considered as
non-negligible releases of nitrogen (Hättenschwiler et al., 2005; Duffy et al.,
2007). Based on such classification of N sources, Figure 1 is proposed to
illustrate the differences between type and origin of N in watersheds.
As shown on Figure 1, nitrogen source types depend on spatiotempo-
ral factors: climatic hazards, dysfunction of infrastructures, animals/human
mobility . . . . Urban sources generate continuous wastes which are the most
controllable if appropriate investments are made. On the contrary, agricul-

tural sources are associated with professional practices and a way to re-
duce nitrogen sources should logically be the livestock pressure decrease
on the watersheds but the economic viability of farms must be taken into
consideration. Furthermore, agricultural sources are subject to multifactorial
explanation of N export: the wide context diversity compared to urban areas
implies many research needs to reach a good understanding of N dynamics
in landscape, and to prioritize remediation actions.
PS directly affect river water quality and their impact can be locally
important compared to NPS as they do not benefit of natural remediation
before reaching the stream. On the other hand, NPS are driven by several
external factors, which can have mixed effects before reaching streams.
In the following sections, two condition groups are considered within
watershed boundaries: (1) “normal conditions” defined by infrastructure rou-
tine functioning, dry weather, stable discharge and constant population; (2)
“exceptional conditions” defined by rainfall event and thus unstable dis-
charge, possibly strengthened by high variation of population, e.g., during
touristic period, when the population can highly increase in rural coastal
areas.
3. VARIABILITY OF N EXPORT
N export can be defined in different ways depending on the studied scale.

For each source, N produced under different forms is transported more or
less quickly through one of three water pathways: surface runoff (also called
overland flow), subsurface runoff (also called interflow or lateral flow), and
infiltration to groundwater. Surface runoff can affect very quickly the river
water quality (minute, hour), while export delays associated with subsur-
face and groundwater flows are, respectively, intermediate (hour, day) and
long/very long (likely years). Thus, N concentrations in rivers depend both
on N sources on the watershed, and on the ratio of export through the 3
flow paths (with their respective N load), but also on N transformations in the
landscape before reaching the stream, and then on in-stream transformations
by biological activity.
3.1 Drivers of N Export Depending on N Sources

The amount of N exported can widely vary depending on land use, N
availability in soil, soil column characteristics, hydroclimatic variability, and
human activities and practices (Table 1).
TABLE 1. N export and drivers (Hydro. Path.: hydrological pathway studied; SR: surface runoff; SubR: subsurface runoff; Inf: infiltration; DIN:
dissolved inorganic nitrogen; TN: total nitrogen; DN: dissolved nitrogen; DON: dissolved organic nitrogen; TON: total organic nitrogen; N∗ : all
2252
forms of N)
Flow paths of N export
Surface/subsurface runoff Leaching
Factor description
Type of Factor influencing N maximizing N N form/Hydro. Factor description N form/Hydro.
variability export export path. References maximizing N export path. References
Natural Soil column Soil Texture Coarse, high sand NO3 − N∗ NO3 − Beaudoin et al.,
drivers characteris- Hydraulic High laterally SubR SR McDaniel et al., 2008 content NO3 − DIN 2005; Kurunc
tics conductivity pH Small amount Cerdà, 2001 Shrinking–swelling et al., 2011
Rock fragments clays High vertically Zhou et al.,
Low 2006; Mantovi
et al., 2006
Fenton et al.,
2009 Mian
et al., 2009
Geology
Connectivity High Dense and SubR SubR SR Hopp and McDonnell,
(slope angle, soil slowly SubR 2009; Tromp-van
depth, bedrock permeable Meerveld et al., 2007
permeability) Gburek et al., 2006;
Presence of Needelman et al.,
fragipan 2004; McDaniel et al.,
2008
Hydroclimatic Rainfall event High Threshold to SR DN, TN TN Le Bissonnais et al., High Low NO3 − N∗ Liang et al., 2011
variability Volume Intensity occur High SubR N∗ 2005; Y. Liu et al., Tang et al.,
Variability in time Constant rain SubR N∗ 2012; Z. Liu et al., 2008
2012 Tromp-van
Meerveld and
McDonnell, 2006;
Uchida et al., 2005;
Tang et al., 2008;
Uchida et al., 2005
Franklin et al., 2007
Soil column state High Shallow Long SR/SubR SR SR Meyles et al., 2003; High NO3 − Burgos, 2006;
Soil temperature time ago High N∗ DON, Penna et al., 2011; Mantovi et al.,
Antecedent water volume Frequent NO3 − N∗ Truman et al, 2011; 2006; Patil
content Water DON N∗ Wei et al., 2007; Tian et al., 2010
table Antecedent et al., 2012 Cooper
rainfall/runoff et al., 2007; Ojeda
Cycles et al., 2006 Tian et al.,
Wetting/drying 2012 Cooper et al.,
2007; Tian et al., 2012
Anthropogenic Agricultural Fertilization Type Mineral High No TN DIN NO3 − Z. Liu et al., 2012 High (organic) High NO3 − NO3 − Daudén et al., 2004; Liang
drivers practices Application rate incorporation N∗ Kurz et al., 2005; (inorganic) et al., 2011 Decau
Application method Zheng et al., et al., 2004; Goulding
2004; Jaynes et al., 2000
et al., 2001 Chen
et al., 2012
Crop system Soil cover Bare soil SR TN Puigdefábregas, Bare soil NO3 − Gabriel et al., 2012;
2005; Ramos and Beaudoin et al., 2005
Martı́nez-
Casasnovas,
2006
Anthropogenic Agricultural Soil working Tillage Conventional (SR) SR/SubR N∗ SR Bosch et al., 2012; Strip Inf. Bosch et al., 2012
drivers practices Strip (SubR) SR/SubR Franklin et al.,
2007; Truman
et al., 2011 Bosch
et al., 2012
Irrigation Type Amount Surface, furrow Over Inf. NO3 − Hassanli et al., 2009
irrigation, rainfall NO3 − Daudén et al., 2004;
after irrigation NO3 − Feng et al., 2005;
Vázquez et al., 2006
Livestock Grazing High density TN N∗ DIN Bilotta et al., 2007;
Elliott and
Carlson, 2004;
Kurz et al., 2005
Urban/suburban Home gardens High rate, bad DIN NO3 − N∗ Morton et al., 1988;
Fertilization, watering timing, Petrovic, 1990; Wakida
overwatering and Lerner, 2005
Urbanization High percentage High N∗ TN N∗ TN Wollheim et al., 2005 Old age, poor DIN N∗ NO3 −, Hosono et al., 2011;
Imperviousness of density High rainfall NH4 + TN Baker et al., 2001 maintenance Old TN DIN N∗ Umezawa et al., 2009;
soils Leaky sewer volume Illicit N∗ Bernhardt et al., age, poor NO3 − N∗ Wakida and Lerner,
Domestic animal connections High 2008; Kim et al., maintenance N∗ TON, 2005 Kaushal et al.,
waste On-site sewage rainfall volume 2007; Previous land NH4 + 2006; Reay, 2004;
disposal SSO, CSO Soonthornnonda use, hot period Wakida and Lerner,
House building and Christensen, Inadequate 2005 Wakida and
Industry Landfill 2008 Li et al., handling, disposal Lerner, 2006 Wakida
2013 or use of N and Lerner, 2005 Oman
Mangimbulude Young age, and Junestedt, 2008;
et al., 2009; nitrogen content Renou et al., 2008
Marques, 2007
Atmospheric N Gaseous emissions High Acidic pH Low Liu et al., 2011; Zhang
deposition Soil properties Rainy et al., 2009 Dise and
Vegetation cover Wright, 1995 Hyvönen
Climatic conditions et al., 2007 Anisfeld
et al., 2007; Liu et al.,
2006
2253
3.1.1 NATURAL DRIVERS

3.1.1.1 Soil Column Characteristics. The soil column characteristics
determine the ability to N export. High nitrate leaching is generally observed
under coarse-texture soils, with high sand content. Nitrate concentration in
leachate is primarily affected by soil type. Beaudoin et al. (2005) observed
concentrations ranging from 31 mg/L in deep loamy soils up to 92 mg/L in
shallow sandy soils in Northern France. Zhou et al. (2006) found at labora-
tory scale using a soil column, higher values in sandy loam soils and lower
ones in clay loam soils. In a vast study over 161 wells in Turkey, Kurunc
et al. (2011) confirmed that nitrate leaching hot spots are generally associated
with high sand content in surface soil. Indeed, coarse texture is associated
with high saturated vertical permeability (Ks ) and less water holding capac-
ity, thus letting short time for denitrification to occur (Fenton et al., 2009).
On the contrary, losses in runoff are enhanced by fine texture. In a modeling
approach, Panagopoulos et al. (2007) have shown that rank of sensitivity of

TN losses by surface runoff was silty-clay-loam > silty-loam > clay > loamy
> sandy-loam > sandy whereas sensitivity to leaching was sandy > loamy
> sandy-loam > silty-loam > silty-clay-loam > clay. However, shrinking-
swelling of clays can also create fast drainage conditions (Mantovi et al.,
2006). Rock fragments in the field also play a role in increasing infiltration.
Their presence in a small amount can thus increase the runoff risk (Cerdà,
2001). In acid soils, less ammonium adsorption can also lead to leaching of
this form of nitrogen, whereas nitrate is generally considered as the only so-
lute which can leach (Mian et al., 2009). A high horizontal Ks associated with
fragipan presence in the soil column (i.e., subsurface layer slowly perme-
able) can enhance subsurface runoff and even lead to a flashy hydrological
events, when this flow pathway is of major importance (McDaniel et al.,
2008). Hopp and McDonnell (2009) and Tromp-van Meerveld et al. (2007)
identified connectivity (related to the subsurface layer characteristics, i.e.,
slope angle, bedrock permeability, and soil depth) as a key element of sub-
surface runoff. A presence of fragipan can also increase the possibility of
saturation-excess surface runoff to occur, by significantly reducing the depth
between saturated layers in soil surface (Needelman et al., 2004; Gburek
et al., 2006).
3.1.1.2 Hydroclimatic Variability. Hydroclimatic variability drives N
export at the plot scale depending on both the rainfall event characteristics
and the “watershed state,” i.e., antecedent conditions favorable to export.
In the Bourville catchment (France), Le Bissonnais et al. (2005) found poor
correlation between rainfall volume and runoff volume for a wide range of
soil antecedent conditions, but a good correlation for rainfall with similar
antecedent conditions. Rainfall amounts also showed positive linear cor-
relations with runoff N losses from agricultural plots (Y. Liu et al., 2012).
Furthermore, a constant rainfall in time seems to produce higher losses than
a variable rains (Franklin et al., 2007). Once rainfall reaches a threshold
value, subsurface flow occurs (Uchida et al., 2005; Tromp-van Meerveld and
McDonnell, 2006) and can transport high nutrient quantities. The subsurface
flow volume is then function of rain intensity and duration (Uchida et al.,
2005). Finally, according, respectively, to Gu and Riley (2010) and Liang et al.
(2011), low frequency, heavy rainfall are the most hazardous conditions for
loss by leaching. Nevertheless, Tang et al. (2008) showed that a high in-
tensity of rainfall can produce negligible compared to surface runoff, and
inversely for low intensity rainfall. Regarding watershed state, antecedent
water content seems to play a major role in surface runoff production (Wei
et al., 2007; Truman et al., 2011). Meyles et al. (2003) showed very small
increases in stream discharge during watershed dry state, but significantly
higher discharge rates during the wet state. In an alpine headwater catch-
ment, high runoff ratios (surface and subsurface flows) occurred during wet
antecedent conditions, when a soil moisture threshold value of 45% was ex-
ceeded in the first 30 cm (Penna et al., 2011). Moreover, irrigation practices

can greatly increase the “natural” wetness of the soil. Nitrogen availability
in soil surface can also be limited, depending if it has already been flushed
many times. Cooper et al. (2007) found that the release of nitrate is mostly
influenced by the time between the last flush of soil and the rewetting of the
H horizon. In a study on nitrogen losses in runoff waters from plots treated
with sewage sludge, the first five rainfall events represented the major part
of nitrate and ammonium export on 14 events (80.6% to 90.5% of cumulated
loads) (Ojeda et al., 2006). In forested drained area, subsurface flow seems to
be influenced by water table depth and antecedent precipitation for nitrate,
and frequency of wetting/drying cycles for dissolved organic nitrogen (Tian
et al., 2012). Leaching is influenced by drainage and availability of nitrogen
in soils. During dry periods, high temperature induces high mineralization
rate of organic nitrogen (optimum between 30 and 35 ◦ C), and thus nitrate
availability (Burgos, 2006). Concentrations can reach up to 300 mg/L in soil
water during warm period, in organically fertilized fields (Mantovi et al.,
2006). However, most of the drainage occurs during wet periods. A model-
ing of these mechanisms in Northern Europe in a context of climate change
showed a great complexity. The increase in heavy rainfalls frequency should
then increase nitrate leaching, whereas soil warming could reduce it through
an increase of evapotranspiration (Patil et al., 2010).
3.1.2 ANTHROPOGENIC DRIVERS

3.1.2.1 Agricultural Practices. Agricultural practices are widely in-
volved in nitrogen export and many works have been carried out during the
last decades. The application rate of fertilizer is one of the most important
factors of N export. High rates of organic or inorganic fertilizer applied on
fields induce large loss in surface, and subsurface runoff and by leaching
(Goulding et al., 2000; Jaynes et al., 2001; Daudén et al., 2004; Decau et al.,
2004; Zheng et al., 2004; Kurz et al., 2005; Liang et al., 2011). A poor knowl-
edge of N content in the soil can partly explain overfertilization practices
(Thompson et al., 2007). Because of higher mobility, conventional mineral
fertilizers are more susceptible to be exported by rainfall, and thus present
higher risk than organic or slow-release fertilizers (Z. Liu et al., 2012). Fer-
tilizer incorporation into soil instead of aerial spreading can reduce loads
in surface runoff. A bare soil increases the risk of nitrogen export in all
water pathways, because of easier drainage, less transpiration by plants,
and less N absorption (Beaudoin et al., 2005; Puigdefábregas, 2005; Ramos
and Martı́nez-Casasnovas, 2006; Gabriel et al., 2012). Thus, cover crop dur-
ing post-harvesting periods is needed to reduce the risk of N export. De-
pending on crop rotation, nitrate concentrations in soil can vary widely.
For example, Beaudoin et al. (2005) observed a lower concentration with a
sugarbeet–wheat rotation (38 mg/L) than for a pea–wheat rotation (66 mg/L).
However, considering the variety of possible rotations, this factor is not in-
dicated in Table 1. Monoculture, by reducing crop yield (Smith et al., 2008)
can also increase export of nitrogen compared to diversified crop produc-
tion. Bosch et al. (2012) found that strip-till practice increased infiltration
compared to conventional tillage, increasing nitrogen export via subsurface
losses. In surface runoff, even if strip-till systems lose more soluble fractions
than conventional ones, TN losses are lower (Franklin et al., 2007). Over-
irrigation leads to large drainage, which is the primary factor of N export
by leaching. It also moistens the soil profile, which can orient biogeochemi-
cal processes and increase N availability. Thus, many studies reported high
losses by leaching under irrigated fields (Daudén et al., 2004; Feng et al.,
2005; Vázquez et al., 2006). Irrigation modes may have an impact on water
uptake uniformity and efficiency by crops (Hassanli et al., 2009). Livestock
grazing can considerably increase loss by surface runoff, especially with
higher animal densities. Defoliation reduces soil protection against rainfall
and increases risks of crusting (Elliott and Carlson, 2004; Bilotta et al., 2007).
Treading damage increases the compaction of soil especially for fine-texture
soils and during wet periods (Bilotta et al., 2007). Finally, high livestock
densities produce a large amount of nutrients which is easily transported by
runoff (Kurz et al., 2005).
3.1.2.2 Urban/Suburban Sources. In urban, suburban and rural land-

scapes, several others sources are involved in N export depending on the
specific land use. Similarly to agricultural land, N export from home gardens
increases with fertilization and overwatering (Morton et al., 1988; Petrovic,
1990; Wakida and Lerner, 2005). A high percentage of impervious soils in-
creases surface water runoff potential and decreases N retention (Wollheim
et al., 2005). Even if N loads are deposited in small amount on these sur-
faces, a great volume of water may increase N export significantly (Brabec
et al., 2002). Kojima et al. (2011) and Baker et al. (2001), respectively, iden-
tified road dust and pet waste as majors deposited N source on impervious
surfaces. In developing countries, open defecation may be a significant N
source in surface runoff. However, its impact has not been directly investi-
gated yet. Other sources of N in surface runoff are discharges from sewage
overflows (CSO or SSO), responsible of high nutrients export via surface
runoff even if not frequent (USEPA, 2004). Sanitary sewage is the major
contributor of ammonium for CSO (Soonthornnonda and Christensen, 2008).
Kim et al. (2007) observed high TN load during sewage overflow events in
a small Korean urban catchment. Concerning groundwater contamination,
Wakida and Lerner (2005) reported that nitrate concentrations in some ur-
ban aquifers can be similar or even greater than those in agricultural areas.
They identified wastewater disposal as the major source. Septic tank and sim-
ilar on-site sanitation systems are a common practice in both developed and
developing countries. Population and system density, improper design, poor

maintenance, and shallow water tables are factors of groundwater pollution
(Wakida and Lerner, 2005). Reay (2004) observed N fluxes from such areas
similar to the lower values reported for agricultural lands in Chesapeake
Bay. In residential areas of the Blue River, Kaushal et al. (2006) found that
about 20% of the annual N export from developed tributaries was derived
from septic systems, and more than 50% during touristic season for one site.
GIS-based monitoring and isotopic analysis suggested that wastewater via se-
vere sewer leakages was the major source of nutrient contaminants in Metro
Manila and Jakarta urban areas (Umezawa et al., 2009). Similar results were
found in urban parts of Taipei Basin (Hosono et al., 2011). In sub-Saharan
Africa mega-cities, more than 60% of wastewater is disposed via onsite sani-
tation systems, sometimes without any treatment, and can represent half the
total precipitation entering these urban areas (Nyenje et al., 2010). Despite
such importance, only few studies estimate this N input (Wakida and Lerner,
2005). Landfills are great sources of N in peri-urban/rural areas. Young land-
fills have generally higher potential for N release in groundwater, but the
load is still important in intermediate and mature ones (Renou et al., 2008).
The composition of wastes and the hydro-climatic variability play obviously
a role in N losses amounts. N export is higher during wet periods because
of great drainage volume, while TN concentrations are higher during dry
periods (Mangimbulude et al., 2009). Leaching to groundwater constitutes
the main N loss pathway, while heavy rainfall also leads to high losses via
surface runoff (Marques, 2007; Mangimbulude et al., 2009). Indeed, land-
fills are generally open wastes dump in developing countries. In developed
countries, landfill roofing avoids surface runoff in adapted infrastructures.
House building is a source comparable to grassland ploughing in temperate
climates, especially if previous land use lead to high nitrogen content and
if building took place during the warm period (Wakida and Lerner, 2006).
Finally, Wakida and Lerner (2005) underlined that inadequate handling, dis-
posal or use of N compounds in industry (leaky tanks for example) can lead
to groundwater contamination.
3.1.2.3 Atmospheric Deposition. Atmospheric deposition is classified

here in anthropogenic sources because quantities of N per surface units
are now widely more important that under pristine conditions (Aber et al.,
2003; Galloway et al., 2008). Indeed, Dise and Wright (1995) have shown
that below a threshold of 10 kg/ha of N input, there is no leaching under
European forests. Deposition is an indirect anthropogenic source of Nitrogen
in water as the first step is the emission in the atmosphere of N gaseous forms
mainly by agricultural practices and urban traffic (Luo et al., 2003; Zhang
et al., 2009; Liu et al., 2011). This N is then deposited by wet or dry deposition
in (short-time transport) or out (long-time transport) of the hydrologic unit of
the emission (Luo et al., 2003; Holland et al., 2005). Dise and Wright (1995)
found that nitrate leaching in European forests is associated with strong
N inputs and low soil pH. Harriman et al. (1998) found similar results by
comparing N deposition and river water composition. Dise and Wright (1995)
also pointed out that the combination of N-input and soil pH explained 87%
of N output-. Furthermore, vegetation cover can be a negative factor for N
export by deposition as canopy intercepts nitrogen deposition thus lowering
the loss risks (Hyvönen et al., 2007). Globally, the proportions of wet and
dry deposition on watersheds seem to be equal (Lawrence et al., 2000;
Holland et al., 2005). However, some authors have shown that deposition
can be unsignificant during baseflow and very significant during stormflow
(Anisfeld et al., 2007). Similarly, Liu et al. (2006) found that more than 80%
of total N deposition in the Beijing area occurred during the rainy season,
following fertilizer application.
Through this review of factors enhancing N export depending on each
N source, it is shown that all human activities are associated with more or
less export of exceeding nitrogen. This export can be controlled by natural
drivers, which are primarily related to hydroclimatic conditions and local
watersheds characteristics. It can be relatively easy to assess what are best
practices at a small scale, for example by simulating rain on two agricul-
tural plots under different conditions of tillage and fertilization. However,
the site-characteristic specificities must be taken into consideration before
drawing general conclusions. Furthermore, hydroclimatic variability is very
important in “real” (non-simulated) situations. Particularly, water pathways
must be well known to assess qualitatively and quantitatively the nitrogen
forms exported at different time-scales. For example, practices reducing N
export via surface runoff (rapid transport, especially due to heavy rainfalls)
can also enhance N export via leaching (long transport). Based on this state-
ment, hydrologists and biogeochemists should join their efforts to obtain a
better understanding of both hydrological mechanisms and nitrogen export,

especially concerning subsurface runoff (lateral flow) for which uncertainties
are the highest. Moreover, factors maximizing N export (e.g., leaky sewers or
poor maintenance of on-site disposal) can also make the impact assessment
difficult to estimate at the watershed scale.
Regarding the different N forms involved and depending on the water
pathways (Table 1), nitrate is the main form concerned by leaching, even
if some authors underline the potential importance of others forms (DON,
NH4 +). Concerning N export via surface and subsurface runoff, TN can also
be considered. Thus, monitoring at least nitrate and Kjeldahl nitrogen could
greatly enhance the understanding of export phenomenon whatever the
various situations.
3.2 N in Drainage Network Depending on N Sources,

Transportation and Transformation
At the watershed scale, the difficulty to monitor and acquire data on water-
sheds limit the interpretation of all factors studied in the previous section.
N concentrations in river depend on land use and hydroclimatic variabil-
ity. Land use determines types of N sources, sensitivity to N export, and
transformations across the landscape, like buffering capacity (i.e., the po-
tential to retain/remove nitrogen). Hydroclimatic variability determines flood
regime, environmental conditions and rainfall events, which in turn affect
the response of landscape. Additionally, evolutions in land use and anthro-
pogenic activities have an effect on the variability of N concentrations in
rivers: population variation (e.g., tourism), agricultural practices (e.g., fer-
tilization) and land use evolution (e.g., deforestation). Figure 2 proposes a
representation of all influences on N export at different time scale, from
rainfall event to decades. In this review, variations of N concentrations in
rivers are studied at the annual scale. Because the concept of season is very
dependent on the geographic position, only wet and dry periods are con-
sidered, assuming that in many parts of the world both are relatively well
differentiated.
3.2.1 LAND USE
The primary factor of N export at the watershed scale is the land use.
“Forested,” “urban/suburban,” or “agricultural” are usually employed to char-
acterize a watershed type having a great proportion of one of these patterns.
A fourth type of watershed is “rural/lightly developed,” corresponding to
a mix of land use mostly dominated by agriculture. Undisturbed forested
catchments are generally used as a reference for pristine environment. Even
if anthropogenic atmospheric deposition can be involved in N export from
these watersheds, many studies have shown higher N concentrations in agri-
cultural and urban/suburban catchments (Chang, 2008; Mulholland et al.,
FIGURE 2. Time scale of environmental change and human practices (CC: climate change).
2008; Buda and Dewalle, 2009; Burns et al., 2009; Li et al., 2009; Kaushal
et al., 2011). TN concentrations are generally positively correlated with im-
pervious/urbanized soil (Chang, 2008; Shields et al., 2008) and agricultural
fields percentages (Donohue et al., 2005; Lassaletta et al., 2009). However,
the impact of land use depends also on the watershed characteristics. For
example, the percentage of forest can be related to very low N concen-
trations, when the contribution of base flow to the total flow is important.
Conversely, it can have no effect if the watershed is characterized by high
stream slopes, fine soil texture and high runoff potential (Norton and Fisher,
2000). Similarly, it seems not possible to associate greater N loads with a
given type of unforested watershed, because of the variety of contexts in
each type of watersheds. For example, lightly developed catchments can
deliver twice the N load of urban catchments because of a large number of
on-site disposal (Shields et al., 2008), and big towns can give similar N load
per area unit that agricultural areas (Wakida and Lerner, 2005).
3.2.2 HYDROCLIMATIC VARIABILITY

3.2.2.1 Stream Order. The second factor of N export at the watershed
scale is the hydroclimatic variability. Alexander et al. (2007) found that first-
order streams from headwaters contribute approximately to 65% of N fluxes
of the second-order streams and 40% of the flux of 4th- and higher-order
streams. Thus, agriculture and urban impacts must be particularly considered
in these areas (Freeman et al., 2007; Elmore and Kaushal, 2008). Headwa-
ters are also more reactive concerning N transformations, particularly for
ammonium forms (Peterson et al., 2001).
3.2.2.2 River Morphology. The linearization of stream decreases the
residence time of nitrate by increasing flow velocities and streams restoration
is therefore a possible way to reduce nitrogen export from watersheds (Chen

et al., 2011). Craig et al. (2008) reviewed stream restoration strategies for re-
ducing river nitrogen loads and proposed to promote actions that increase
in-stream carbon availability, contact between the water and benthos, and
connections between streams and adjacent terrestrial environments. How-
ever, there are many uncertainties on the efficiency of these actions. For
example, Wagenschein and Rode (2008) calculated a possible reduction of
N loads by only 5.4% on the Weisse Elster River (Germany).
3.2.2.3 Contribution of (Sub-)Surface Runoff and Groundwater Flow

to the River Flow. The hydroclimatic variability drives the contribution of
each hydrological component (surface, subsurface runoff and groundwater
flow) to the total flow. In average, TN concentrations are much lower during
dry period than during wet period for catchments with impervious bedrock
(Shrestha and Kazama, 2007; Burns et al., 2009; Li et al., 2009). This dif-
ference is generally attributed to a lower groundwater contribution to total
flow during this period (Schilling and Zhang, 2004; Donohue et al., 2005).
Indeed, recent studies have shown the primary significance of groundwater
in controlling stream water nitrate concentrations ([NO3 −]SW ) (Ruiz et al.,
2002a, 2002b; Blanco et al., 2010). Molenat et al. (2008) proposed a scheme
identifying the water table dynamic and spatial differences in nitrate con-
centrations in groundwater ([NO3 −]GW ) as the main drivers of [NO3 −]SW in
normal conditions. More generally, Legout et al. (2007) showed that solute
transfer in groundwater is not homogeneous, but rather complex and with
a two porosity compartment (a slow-mobility and rapid-mobility porosity).
Recently, Rouxel et al. (2011) showed that solute transfer via groundwater
should be affected by a strong spatiotemporal variability and argued that
it must be taken into account in hydrochemical models. However, these
mechanism-process studies were made on small headwaters catchments,
which may not integrate all types of N sources and the complexity of water
pathways in larger landscape.
Blanco et al. (2010) observed that nitrate in river increases with increas-
ing contribution in groundwater, and suggest that groundwater N concentra-
tion is the main driver to N export. At the opposite, some authors suggested
that surface and subsurface runoffs are a major vector of N export during
rainfall events. Intensively drained agricultural watersheds seem to present
a “flashy” response to rainfall events by exporting large quantities via sub-
surface tile-drain (Tomer et al., 2003; Royer et al., 2006). More than 80% of
agricultural land is drained by surfaces ditches or subsurface tiles in some
areas (e.g., Midwestern United states) (Blann et al., 2009), and N loads in
drainage can reach more than 80 kg N/ha in some cases (Kaspar et al., 2007).
Thus, the coincidence of temporal variation in TN concentrations with local
seasonal calendar of agricultural activities and rainfall distribution may reflect
the influence of rapid flow pathways (Royer et al., 2006; Bu et al., 2010).
3.2.2.4 Hydrological Regime. Discharges of wastewater treatment

plants (WWTP) can have an important effect on N concentration in streams
during some periods of the year. For example, in Blue River watershed, a
WWTP discharge represents only 15% of total annual N flux but up to 80%
of the N flux during dry period when WWTP discharge becomes significant
compared to the river flow (Graham et al., 2010). Similarly, Kaushal et al.
(2011) found that wastewater is a major source of nitrate during baseflow
conditions in Baltimore urbanized streams, and that its importance decreases
with increasing runoff. On the opposite, WWTP discharges can have a very
small effect on TN concentrations in river if other sources in the watershed
are much more important (Merseburger et al., 2005, 2011; Burns et al., 2009).
Furthermore, watersheds can show different global behaviors during the year
depending on the main land use. In the Chesapeake Bay watershed, most
of nitrate is exported at low to moderate flow (<1 mm/day) from forested,
lightly developed and agricultural catchments, but at high flow (>1 mm/day)
from urban catchments (Shields et al., 2008). Extreme events (associated here
to “exceptional conditions”) can also be of major importance for N export
from watersheds. In many watersheds, a large fraction of N export can occur
only in a limited period of the year, during rainfall periods and snowmelt
events (Borah et al., 2003; Royer et al., 2006; Sebestyen et al., 2008). For
example, Borah et al. (2003) observed in Illinois that added nitrate fluxes
in stormflow from March to June can reach nearly 40% of the average an-
nual flux during a dry year and 100% during a wet year. In a small forested
Mediterranean catchment, Bernal et al. (2002) found that 80% of the annual
nitrate flux occurred during a single storm event, because of lower N reten-
tion, and pulses of nutrient cycling after a long dry period. In river, particulate
nitrogen from diverse land use settled in riverbed, but can be resuspended
during high flow (Schulz et al., 2003; Donohue et al., 2005). In urban water-
shed, no significant differences in nitrate concentrations is generally found
between baseflow and stormflow, indicating that fluxes are higher during
storm events (Taylor et al., 2005; Wollheim et al., 2005; Kaushal et al., 2008).
Furthermore, urban watersheds are more submitted to extreme hydrologi-
cal conditions (droughts and storms) compared to agricultural and forested
catchements (Paul and Meyer, 2001). Borah et al. (2003) studied the nitrate
dynamics in river in three agricultural catchments (from 98 to 932 km2) dur-
ing storm events, and showed an inverse relationship with water discharge
during intense storms and a weak relationship during less intense storms.
They suggested that this was due to a mobilization of nitrate-rich porewater
during light rainfall events and dilution during heavy rainfall events.
Finally, a recent study on three French rivers over 27 years, Baurès et al.
(2013) showed that nitrate fluxes increase with discharge and become con-
stant at very high flow, indicating a mass-limitation until a discharge threshold
value is reached. N sources also have different effects during storm events,
depending on land use and period. For example, atmospheric deposition
contribute up to 30% of N load during stormflow in forested catchments

(Sebestyen et al., 2008; Buda and Dewalle, 2009), while it may be higher in
urban catchments (>50%) (Anisfeld et al., 2007; Buda and Dewalle, 2009;
Burns et al., 2009).
3.2.3 N TRANSFORMATION
3.2.3.1 Denitrification. The vicinity of N sources to river seems to be
also an important factor of N delivery to large rivers (Alexander et al., 2000). If
some parts of the watershed are particularly sensitive to N export, some other
parts can attenuate N loads. In fact, more than 50% of inputs can “disappear”
before reaching the outlet (Seitzinger et al., 2006; Alexander et al., 2007).
Some N compounds are either retain in temporary sinks (e.g., plant growth
uptake) or turned to gaseous forms such as N2 O, NO, or N2 that will be
transformed into aerosol phase compounds (ammonium and nitrate) and can
fall back partly with precipitation (Sutton et al., 2011). Riparian zones are
recognized as very important areas involved in these processes (Ranalli and
Macalady, 2010) and “hot spots” of N removal/retention are found especially
at the interface between coarse permeable material and fine organic rich
material in the subsurface soil layers (McClain et al., 2003; Vidon et al.,
2010). In this way, Wohlfart et al. (2012) found for example that the fraction
of coarse texture and organic soils in an intensively managed agricultural
watershed has a major influence on N export. Moreover, a fraction of N input
is lost in stream by denitrification via the exchange between water column
and hyporreic sediments (Seitzinger et al., 2006; Mulholland et al., 2008).
Thus, the retention is higher in tributaries than in mains stream (Chen et al.,
2011), and decline from upstream to downstream with increasing stream size
(Alexander et al., 2000). Retention and denitrification are also more efficient
in rural than in urban/suburban area (Groffman and Crawford, 2003).
The efficiency of buffer zones is also very dependent of hydrocli-
matic variability. In riparian areas, important amounts of nitrogen can be
retained/remove during specific period of the year, decreasing nitrogen flux
in water. But these zones can be bypassed during high flow (Paul and Meyer,
2001) or be inefficient in very dry context, by lack of water to denitrifying mi-
croorganisms (Bernal et al., 2012). In the same way, the retention efficiency of
sediments is generally inversely proportional to the stream discharge because
exchange between water column and sediments is less important during high
flow (Royer et al., 2004). However, flooding can increase denitrification by
improving contact with microbiologically reactive sediments in some rivers
with extensive floodplains or riparian areas (Alexander et al., 2007). In sum-
mer and early autumn, macrophyte abundance in river channels can increase
N retention and water residence time (Schulz et al., 2003). In general, pro-
cesses that have minor consequences at high flow can become significant at
low flow, especially during dry periods (Alexander et al., 2007).
3.2.3.2 Other in Stream Transformations. Improving knowledge

about processes controlling in stream N transformations is critical to assess
watershed vulnerability to water quality changes. The relationships between
nitrogen concentration and runoff flowrate for watersheds of varying land use
have provide a better understanding on how watershed hydrology and land
use interactively regulate nitrogen inputs to stream water and downstream
aquatic media. For instance, Goodridge and Melack (2012) have determined
a hyperbolic relation for nitrate concentration observed in stream water dur-
ing periods of low runoff (i.e., base flow) and high runoff (i.e., stormflow).
They have also observed the contribution of land use nature: in undeveloped
watershed, an enrichment of nitrate-runoff relationship was demonstrated,
whereas invariance and dilution were found, respectively, for urban water-
sheds and agricultural watersheds. The seasonal variation in hydrological
conditions was found to be the predominant controlling factor of in stream
N-transformations (Ohte, 2012). A seasonal evolution between spring (high

nitrate/low DOC) and autumn (high DOC/low nitrate) were also shown by
Thomas et al. (2014), who proposed a simple linear model for nitrate–DOC
relationship in freshwaters in intensive agricultural environments where al-
gae blooms may occur in waterbodies (rivers, lakes and coastal waters).
Nowadays, different approaches are proposed a better understanding of nu-
trient cycling, and are based on ecological theory and stoichiometry (Allen
and Gillooly, 2009). According to Ohte (2012), the ecosystem nutrient de-
mand as well as the solute transport capability and in-stream nutrient use
and supply are the main factors impacting the seasonal patterns of stream
nitrate concentrations.
Concerning DON it was found that headwater streams can quickly con-
vert inorganic N into organic forms, so that the production of autochthonous
DON can represent a substantial transformation of stream N, although the
ultimate fate of DON remains unclear (Johnson et al., 2013). Improving our
knowledge on the riparian zone role in nitrate attenuation in undisturbed
watersheds is a real challenge since these watersheds are very sensitive to
land use and climate changes, wildfire, and increases in atmospheric nitrogen
deposition (Ranalli and Macalady, 2010).
3.3 Long Term Studies

Anthropogenic activities have created complex ecosystem structures, with
different land uses and landscapes which determine the flows and the fate
of nitrogen reactive forms from local scale (plot) to more extended scale
(watershed). Alvarez-Cobelas et al. (2008), analyzed N export rates from 946
rivers in the world, and suggested that spatial and temporal scale are very
important to predict N export, and that regional approaches are more useful
than global-scale analyses. Furthermore, many authors underlined the neces-

sity to implement high resolution monitoring programs to evaluate both spa-
tial and temporal variability at the watershed scale (Bengraıne and Marhaba,
2003; Donohue et al., 2005; Chang, 2008; Bu et al., 2010). Many works have
been published concerning long-term studies of N export, taking into ac-
count the global effects of anthropogenic evolution impacts such as climate
change. For instance, some studies correlate several nutrients fluxes (C, N, P,
Si) with livestock densities increase or with sewage treatment plants (Hum-
borg, 2007). However, the main published results concern modeling works,
sometimes integrating both anthropogenic evolutions, land use and climate
change. Indeed, even current terrestrial ecosystem models fail to explain
some long-term nutrient evolutions (Bernhardt et al., 2005). Other factors
such as changes in the chemical composition of atmosphere, soil frost, or
interannual climate fluctuations should be taken into account. The high tem-
poral variability in nitrogen concentrations both among catchments and into

each site highlights the critical need for long-term networked water quality
monitoring to gain a better knowledge of nutrient fluxes evolution (Argerich
et al., 2013). The application of models for future land use, integrating the
management practices such as organic farming methods, and climate sce-
narii suggest that riverine N export is likely to significantly increase both in
response to heavier fertilizer use and to annual discharge fluctuations (Han
et al., 2009). Some statistical nutrient loss models were developed to simulate
the combined effects of climate–change on hydrology, nutrient losses, and
nitrogen retention processes at the river basin scale (Andersen et al., 2006).
This last study concludes that mean annual TN export from the river basin
could increase from the control to the scenario period by 7.7%. Even though
a 4.2% increase in nitrogen retention in the river system was simulated in the
scenario, an increase in-stream TN export resulted because of the simulated
increase in the diffuse TN transfers from land to surface waters.
Climate change also increase the fractional export of anthropogenic N
inputs to river. The lag effect, soil N release, and climate change, delay-
ing riverine TN export reductions with respect to decreases in N inputs and
should be considered in developing and evaluating N management measures
(Chen et al., 2014). Wu et al. (2012) show that global climate change impact
on pollution load are relatively greater when compared to future livestock
and poultry breeding increases or agricultural population reductions. The
effects observed from runoff increment lead to approximately 28.6% and
22.5% increases of TN and TP pollution load respectively. Generally, the
impacts of land-use change have shown insignificant effect due to soil con-
servation measures, whereas the impacts of rural residential area account for
high proportion changes, among which around 5% related to the increase
in livestock and poultry breeding. The largest contribution rate comes from
the output of different land use types.
In the previous paragraphs, we demonstrated that local factors

including catchment characteristics and natural disturbance events in-
fluence the evolution of N export. Moreover, synthesis of long-
term stream chemistry data from different catchments is valu-
able to understand these trends. Indeed, a watershed ecosystem is
a temporally evolving structure. Measuring how nutrient flux and
cycling in both stream and terrestrial ecosystems change through time is
crucial for analyzing and interpreting nutrient flux and cycling at the over-
all watershed–landscape scale. These results also emphasize the importance
of site-specific strategies that are relevant for the choice of catchments and
sampling schemes. Such information is crucial when considering trends, re-
finement of existing programs, and establishment of new monitoring sites.
In this section, we showed that nitrogen in rivers depends on a com-
bination of factors described in Section 3.1 (via land use and hydro-climatic
variability on the watersheds) but also on N transformations across the land-

scape and within the river system. However, all these factors are very difficult
to assess together as a watershed is a sum of various land uses and condi-
tions. Therefore, most of studies focus only on some of them to estimate their
effect on N variability in streams. Evaluating accurately the main contributors
to total nitrogen export and the effects of “situations at risk” on this export
is a hard task depending on local conditions. Generally, modeling is used
to combine all possible factors and define strategies for N export reduction.
If it is sufficient to plan mitigation actions at a global scale, the responses
given by models are limited by our knowledge of combined factors and by
a lack of high frequency data, especially during “situations at risk.” Scientists
should also join their efforts for a better understanding of variations in N
export at different spatiotemporal scales. Accurate data on watersheds are
also needed to associate/dissociate the different anthropogenic influences
from other and natural influences.
Nitrate is the main N compound studied at the watershed scale even
if organic compound as much or more relevant in numerous case studies:
nitrate and Kjeldahl nitrogen should at least be measured in all routine
watershed monitoring.
Furthermore, global change could lead to radical increases of N export
despite the mitigation efforts undertaken by all actors. Therefore, there is
an urgent need for a better understanding of spatiotemporal variability of N
export to reduce actual loads and anticipate future changes.
4. CONCLUSION & PERSPECTIVES
This review of N sources and N export conditions shows that such en-
vironmental and health issues cannot be reduced to simple remediation
actions. On one hand, the export of N is affected by numerous natural
and anthropogenic factors, which are partly uncertain (e.g., climate) and
partly not well known (e.g., hydrological pathways). On the other hand,
in-streams N dynamics at the watershed scale, from rainfall event to intra-
annual variability is associated with many other factors such as the dis-
tribution of land use in the landscape, watershed characteristics, calendar
of agricultural activities, etc. Scientific studies on factors promoting N ex-
port (e.g., fertilization rate or type) applied to a wide range of environ-
ment allow to establish a large catalog of “best management practices” for
different types of N sources. The application of such practices is a key
component for the efficiency of remediation actions and should be widely
promoted.
At the watershed scale, some observations on N concentrations variabil-
ity and load in stream water are largely assumed by authors (e.g., impact of
land use) but some others rather seem to be very linked to the watershed
characteristics and to hydroclimatic context. If these observations have ma-

jor implications in understanding global processes at a large scale, it seems
very difficult, for example, to accurately assess the risk under rainy events,
for different hydrological conditions in a watershed and at the intra-annual
scale. Given the difficulty to accurately monitor water quality at watershed
scale, various methods have been proposed to evaluate the effects of an-
thropogenic activities and the related variability: experimental headwater
catchments, N source tracking, multivariate analysis, interannual compar-
ison, hydrologic separation . . . . These methods allow many advances in
understanding of N export but the diversity of watersheds and the lack in
mechanisms knowledge still remain a difficulty in the analysis of experi-
mental data. Analysis of long-term stream studies from multiple catchments
is also a real challenge for trend understanding. These studies reveal that
spatial and temporal scales are important to predict N export, so that high
resolution monitoring programs are a real need for a more precise under-
standing of global variability. Eventually, the increasing use of rapid on site
methods for very high frequency monitoring (minute step) of some water
quality parameters opens new opportunities. Furthermore, the development
of different analytical methods to explain N export variations should be very
useful to take full advantage of the acquired data. Advices in prioritization
of remediation techniques could be greatly enhanced by this integrative ap-
proach.
ACKNOWLEDGMENTS
The authors sincerely thank Dr Frédéric Pitois for the proofreading.

FUNDING
This work is supported by the French Agence de l’Eau Loire-Bretagne (AELB),

the French “Conseil Régional de Bretagne,” the French Association Na-
tionale de la Recherche et de la Technologie (ANRT), and Coop de France
Ouest.
REFERENCES
Aber, J. D., Goodale, C. L., Ollinger, S. V., Smith, M.-L., Magill, A. H., Martin, M.
E., Hallett, R. A., and Stoddard, J.L. (2003). Is nitrogen deposition altering the
nitrogen status of northeastern forests? Bioscience 53, 375–389.
Alexander, R. B., Boyer, E. W., Smith, R. A., Schwarz, G. E., Moore, R. B., and
Richard, B. (2007). The role of headwater streams in downstream water quality.
J. Am. Water Resour. Assoc. 43, 41–59.
Alexander, R. B., Smith, R., and Schwarz, G. (2000). Effect of stream channel size on
the delivery of nitrogen to the Gulf of Mexico. Nature 403, 758–761.
Allen, A. P., and Gillooly, J. F. (2009). Towards an integration of ecological stoi-
chiometry and the metabolic theory of ecology to better understand nutrient
cycling. Ecol. Lett. 12, 369–384.
Alvarez-Cobelas, M., Angeler, D. G., and Sánchez-Carrillo, S. (2008). Export of nitro-
gen from catchments: A worldwide analysis. Environ. Pollut. 156, 261–269.
Amirkolaie, A. K. (2011). Reduction in the environmental impact of waste discharged
by fish farms through feed and feeding. Rev. Aquacult. 3, 19–26.
Anbumozhi, V., Radhakrishnan, J., and Yamaji, E. (2005). Impact of riparian buffer
zones on water quality and associated management considerations. Ecol. Eng.
24, 517–523.
Andersen, H. E., Kronvang, B., Larsen, S. E., Hoffmann, C. C., Jensen, T. S., and
Rasmussen, E. K. (2006). Climate-change impacts on hydrology and nutrients
in a Danish lowland river basin. Sci. Total Environ. 365, 223–237.
Anisfeld, S. C., Barnes, R. T., Altabet, M. A., and Wu, T. (2007). Isotopic apportion-
ment of atmospheric and sewage nitrogen sources in two Connecticut rivers.
Environ. Sci. Technol. 41, 6363–6369.
Argerich, A., Johnson, S. L., Sebestyen, S. D., Rhoades, C. C., Greathouse, E., Knoepp,
J. D., Adams, M. B., Likens, G. E., Campbell, J. L., McDowell, W. H., Scatena, F.
N., and Ice, G. G. (2013). Trends in stream nitrogen concentrations for forested
reference catchments across the USA. Environ. Res. Lett. 8.
Asman, W. E.H., Sutton, M. A., and Schjorring, J. K. (1998). Ammonia : Emission,
atmospheric transport and deposition. New Phytol. 139, 27–48.
Badruzzaman, M., Pinzon, J., Oppenheimer, J., and Jacangelo, J. G. (2012). Sources
of nutrients impacting surface waters in Florida: A review. J. Environ. Manage.
109, 80–92.
Baker, L. A., Hope, D., Xu, Y., Edmonds, J., and Lauver, L. (2001). Nitrogen balance
for the Central Arizona-Phoenix (CAP) ecosystem. Ecosystems 4, 582–602.
Baurès, E., Delpla, I., Merel, S., Thomas, M.-F., Jung, A.-V., and Thomas, O. (2013).
Variation of organic carbon and nitrate with river flow within an oceanic regime
in a rural area and potential impacts for drinking water production. J. Hydrol.
477, 86–93.
Beaudoin, N., Saad, J. K., Van Laethem, C., Machet, J. M., Maucorps, J., and Mary,
B. (2005). Nitrate leaching in intensive agriculture in Northern France: Effect
of farming practices, soils and crop rotations. Agric. Ecosyst. Environ. 111,
292–310.
Behera, S., and Panda, R. K. (2006). Evaluation of management alternatives for
an agricultural watershed in a sub-humid subtropical region using a physical
process based model. Agric. Ecosyst. Environ. 113, 62–72.
Bengraıne, K., and Marhaba, T. F. (2003). Using principal component analysis to
monitor spatial and temporal changes in water quality. J. Hazard. Mater. 100,
179–195.
Bergstrom, A.-K., and Jansson, M. (2006). Atmospheric nitrogen deposition has
caused nitrogen enrichment and eutrophication of lakes in the northern hemi-
sphere. Global Change Biol. 12, 635–643.
Bernal, S., Butturini, A., and Sabater, F. (2002). Variability of DOC and ni-
trate responses to storms in a small Mediterranean forested catchment
Site description of the Fuirosos catchment. Hydrol. Earth Syst. Sci. 6,
1031–1041.
Bernal, S., Schiller, D., Sabater, F., and Martı́, E. (2012). Hydrological extremes
modulate nutrient dynamics in mediterranean climate streams across different
spatial scales. Hydrobiologia 719, 31–42.
Bernhardt, E. S., Band, L. E., Walsh, C. J., and Berke, P. E. (2008). Understanding,
managing, and minimizing urban impacts on surface water nitrogen loading.
Ann. N.Y. Acad. Sci. 1134, 61–96.
Bernhardt, E. S., Likens, G. E., Hall, R. O., Buso, D. C., Fisher, S. G., Burton, T. M.,
Meyer, J. L., McDowell, W. H., Mayer, M. S., Bowden, W. B., Findlay, S. E.G.,
Macneale, K. H., Stelzer, R. S., and Lowe, W. H. (2005). Can’t see the forest for
the stream? In-stream processing and terrestrial nitrogen exports. BioScience 55,
219–230.
Bilotta, G. S., Brazier, R. E., and Haygarth, P. M. (2007). The impacts of grazing
animals on the quality of soils, vegetation, and surface waters in intensively
managed grasslands. Adv. Agron. 94, 237–280.
Blanco, A. C., Nadaoka, K., Yamamoto, T., and Kinjo, K. (2010). Dynamic evolution
of nutrient discharge under stormflow and baseflow conditions in a coastal
agricultural watershed in Ishigaki Island, Okinawa, Japan. Hydrol. Processes 24,
2601–2616.
Blann, K. L., Anderson, J. L., Sands, G. R., and Vondracek, B. (2009). Effects of
agricultural drainage on aquatic ecosystems: A review. Crit. Rev. Environ. Sci.
Technol. 39, 909–1001.
Blowes, D. W., Ptacek, C. J., Benner, S. G., McRae, C. W., Bennett, T. A., and Puls,
R. W. (2000). Treatment of inorganic contaminants using permeable reactive
barriers. J. Contam. Hydrol. 45, 123–137.
Borah, D. K., Bera, M., and Shaw, S. (2003). Water, sediment, nutrient, and pesticide
measurements in an agricultural watershed in Illinois during storm events. Am.
Soc. Agric. Eng. 46, 657–674.
Borin, M., Passoni, M., Thiene, M., and Tempesta, T. (2010). Multiple functions of
buffer strips in farming areas. Eur. J. Agron. 32, 103–111.
Bosch, D. D., Truman, C. C., Potter, T. L., West, L. T., Strickland, T. C., and Hub-
bard, R. K. (2012). Tillage and slope position impact on field-scale hydro-
logic processes in the South Atlantic Coastal Plain. Agric. Water Manage. 111,
40–52.
Brabec, E., Schulte, S., and Richards, P. L. (2002). Impervious surfaces and water
quality: A review of current literature and its implications for watershed plan-
ning. J. Plann. Lit. 16, 409–514.
Bremerton CSO Annual Report. (2014). City of Bremerton: Combined sewer overflow
annual report 2013, City of Bremerton Press, Bremerton.
Bu, H., Tan, X., Li, S., and Zhang, Q. (2010). Temporal and spatial variations of
water quality in the Jinshui River of the South Qinling Mts., China. Ecotoxicol.
Environ. Saf. 73, 907–913.
Buda, A. R., and Dewalle, D. R. (2009). Dynamics of stream nitrate sources and
flow pathways during stormflows on urban, forest and agricultural watersheds

in central Pennsylvania, USA. Hydrol. Processes 23, 3292–3305.
Burgos, P. (2006). Nitrogen mineralization and nitrate leaching of a sandy
soil amended with different organic wastes. Waste Manage. Res. 24,
175–182.
Burns, D. A., Boyer, E. W., Elliott, E. M., and Kendall, C. (2009). Sources and
transformations of nitrate from streams draining varying land uses: Evidence
from dual isotope analysis. J. Environ. Qual. 38, 1149–1159.
Carey, R. O., and Migliaccio, K. W. (2009). Contribution of wastewater treatment
plant effluents to nutrient dynamics in aquatic systems: A review. Environ.
Manage. 44, 205–217.
Carrico, A. R., Fraser, J., and Bazuin, J. T. (2012). Green with envy: Psychological and
social predictors of lawn fertilizer application. Environ. Behav. 20(10), 1–28.
Carrillo, J. H. (2002). Atmospheric deposition of inorganic and organic nitrogen and
base cations in Hawaii. Global Biogeochem. Cycles 16, 1076.
Cerdà, A. (2001). Effects of rock fragment cover on soil infilltration, interrill runoff
and erosion. Eur. J. Soil Sci. 52, 59–68.
Chang, H. (2008). Spatial analysis of water quality trends in the Han River basin,
South Korea. Water Res. 42, 3285–3304.
Chen, D., Huang, H., Hu, M., and Dahlgren, R. A. (2014). Influence of lag effect, soil
release, and climate change on watershed anthropogenic nitrogen inputs and
riverine export dynamics. Environ. Sci. Technol. 48, 5683–5690.
Chen, D., Lu, J., Shen, Y., Gong, D., and Deng, O. (2011). Spatio-temporal variations
of nitrogen in an agricultural watershed in eastern China: Catchment export,
stream attenuation and discharge. Environ. Pollut. 159, 2989–2995.
Chen, Y.-H., Wang, M.-K., Wang, G., Chen, M.-H., Luo, D., and Li, R. (2012). Nitrogen
runoff under simulated rainfall from a sewage-amended lateritic red soil in
Fujian, China. Soil Tillage Res. 123, 35–42.
Collins, R., Mcleod, M., Hedley, M., Donnison, A., Close, M., Hanly, J., Horne,
D., and Ross, C. (2007). Best management practices to mitigate faecal con-
tamination by livestock of New Zealand waters. N. Z. J. Agric. Res. 50,
267–278.
Comly, H.H. (1945). Cyanosis in infants caused by nitrates in well water. J. Am. Med.
Assoc., 129, 112.
Cooper, R., Thoss, V., and Watson, H. (2007). Factors influencing the re-
lease of dissolved organic carbon and dissolved forms of nitrogen from a
small upland headwater during autumn runoff events. Hydrol. Processes 633,
622–633.
Craig, L. S., Palmer, M. A., Richardson, D. C., Filoso, S., Bernhardt, E. S., Bledsoe,
B. P., Doyle, M. W., Groffman, P. M., Hassett, B. A., Kaushal, S. S., Mayer, P.
M., Smith, S. M., and Wilcock, P. R. (2008). Stream restoration strategies for
reducing river nitrogen loads. Front. Ecol. Environ. 6, 529–538.
Daudén, A., Quı́lez, D., and Vera, M. V. (2004). Pig slurry application and irrigation
effects on nitrate leaching in Mediterranean soil lysimeters. J. Environ. Qual. 33,
2290–2295.
Davies-colley, R. O.B.J., Nagels, J. W., Smith, R. O.B.A., Zealand, N., Young, R. G.,
and Phillips, C. J. (2004). Water quality impact of a dairy cow herd crossing a
stream. N. Z. J. Mar. Freshwater Res. 38, 569–576.

De Wit, M., and Bendoricchio, G. (2001). Nutrient fluxes in the Po basin. Sci. Total
Environ. 273, 147–161.
Decau, M. L., Simon, J. C., and Jacquet, A. (2004). Nitrate leaching under grassland
as affected by mineral nitrogen fertilization and cattle urine. J. Environ. Qual.
33, 637–644.
Di, H. J., and Cameron, K. C. (2000). Calculating nitrogen leaching losses and critical
nitrogen application rates in dairy pasture systems using a semi-empirical model.
N. Z. J. Agric. Res. 43, 139–147.
Dise, N. B., and Wright, R. F. (1995). Nitrogen leaching from European forests in
relation to nitrogen deposition. For. Ecol. Manage. 71, 153–161.
Donohue, I., Styles, D., Coxon, C., and Irvine, K. (2005). Importance of spatial and
temporal patterns for assessment of risk of diffuse nutrient emissions to surface
waters. J. Hydrol. 304, 183–192.
Drewry, J. J. A. , Newham, L. T. H. A., Greene, R. S. B. B., and Jakeman, A. J.A.
(2006). A review of nitrogen and phosphorus export to waterways: Context for
catchment modelling. Mar. Freshwater Res. 57, 757–774.
Duffy, J. E., Cardinale, B. J., France, K. E., McIntyre, P. B., Thébault, E., and Loreau, M.
(2007). The functional role of biodiversity in ecosystems: Incorporating trophic
complexity. Ecol. Lett. 10, 522–538.
Edmonton CSO Control Strategy. (2000). The city of Edmonton: Combined sewer
overflow control strategy (Implementation Plan), City of Edmonton Press, Ed-
monton.
Elliott, A. H., and Carlson, W. T. (2004). Effects of sheep grazing episodes on sedi-
ment and nutrient loss in overland flow. Aust. J. Soil Res. 42, 213.
Elmore, A. J., and Kaushal, S. S. (2008). Disappearing headwaters: Patterns of stream
burial due to urbanization. Front. Ecol. Environ. 6, 308–312.
European Directive 91/676/CEE. (1991). Retrieved April 17, 2015, from http://ec.
europa.eu/environment/water/water-nitrates/directiv.html.
Feng, Z.-Z., Wang, X.-K., and Feng, Z.-W. (2005). Soil N and salinity leaching after
the autumn irrigation and its impact on groundwater in Hetao Irrigation District,
China. Agric. Water Manage. 71, 131–143.
Fenton, O., Richards, K. G., Kirwan, L., Khalil, M. I., and Healy, M. G. (2009). Factors
affecting nitrate distribution in shallow groundwater under a beef farm in south
eastern Ireland. J. Environ. Manage. 90, 3135–3146.
Fewtrell, L. (2004). Drinking-water nitrate, methemoglobinemia, and global burden
of disease: A discussion. Environ. Health Perspect. 112, 1371–1374.
Fisher, J., and Acreman, M. C. (2004). Wetland nutrient removal: A review of the
evidence. Hydrol. Earth Syst. Sci. 8, 673–685.
Franklin, D., Truman, C., Potter, T., Bosch, D., Strickland, T., and Bednarz, C. (2007).
Nitrogen and phosphorus runoff losses from variable and constant intensity
rainfall simulations on loamy sand under conventional and strip tillage systems.
J. Environ. Qual. 36, 846–854.
Freeman, M. C., Pringle, C. M., and Jackson, C. R. (2007). Hydrologic connectivity
and the contribution of stream headwaters to ecological integrity at regional
scales. J. Am. Water Resour. Assoc. 43, 5–14.
Gabriel, J. L., Muñoz-Carpena, R., and Quemada, M. (2012). The role of cover crops
in irrigated systems: Water balance, nitrate leaching and soil mineral nitrogen
accumulation. Agric. Ecosyst. Environ. 155, 50–61.
Galloway, J. N., Townsend, A. R., Erisman, J. W., Bekunda, M., Cai, Z., Freney, J.
R., Martinelli, L. A., Seitzinger, S. P., and Sutton, M. A. (2008). Transformation
of the nitrogen cycle: Recent trends, questions, and potential solutions. Science
320, 889–892.
Gburek, W. J., Needelman, B. A., and Srinivasan, M. S. (2006). Fragipan controls on
runoff generation: Hydropedological implications at landscape and watershed
scales. Geoderma 131, 330–344.
Goodridge, B. M., and Melack, J. M. (2012). Land use control of stream nitrate
concentrations in mountainous coastal California watersheds. J. Geophys. Res.
Biogeosci. 117(G2), G02005.
Goulding, K. W.T., Poulton, P. R., Webster, C. P., and Howe, M. T. (2000). Nitrate
leaching from the Broadbalk Wheat Experiment, Rothamsted, UK, as influ-
enced by fertilizer and manure inputs and the weather. Soil Use Manage. 16,
244–250.
Graham, J. L., Stone, M. L., Rasmussen, T. J., and Poulton, B. (2010). Effects of
wastewater effluent discharge and treatment facility upgrades on environmental
and biological conditions of the Upper Blue River, Johnson County, Kansas and
Jackson County, Missouri, January 2003 through March 2009: U.S. Geological
Survey Scientific Investigations Report, U.S. Geological Survey Press, Reston.
Greer, F. R., and Shannon, M. (2005). Infant methemoglobinemia: The role of dietary
nitrate in food and water. Pediatrics 116, 784–786.
Groffman, P. M., and Crawford, M. K. (2003). Denitrification potential in urban
riparian zones. J. Environ. Qual. 32, 1144–1149.
Gu, C., and Riley, W. J. (2010). Combined effects of short term rainfall patterns and
soil texture on soil nitrogen cycling—A modeling analysis. J. Contam. Hydrol.
112, 141–154.
Han, H., Allan, J. D., and Scavia, D. (2009). Influence of climate and human activities
on the relationship between watershed nitrogen input and river export. Environ.
Sci. Technol. 43, 1916–1922.
Harriman, R., Curtis, C., and Edwards, A. C. (1998). An empirical approach for
assessing the relationship between nitrogen deposition and nitrate leaching
from upland catchments in the United Kingdom using runoff chemistry. In R.
K. Wieder, M. Novák, and J. Černý (Eds.), The third international symposium
on ecosystem behavior June 21–25, 1997 (pp. 193–203). Dordrecht: Springer.
Hassanli, A. M., Ebrahimizadeh, M. A., and Beecham, S. (2009). The effects
of irrigation methods with effluent and irrigation scheduling on water use
efficiency and corn yields in an arid region. Agric. Water Manage. 96,
93–99.
Hättenschwiler, S., Tiunov, A. V., and Scheu, S. (2005). Biodiversity and lit-
ter decomposition in terrestrial ecosystems. Annu. Rev. Ecol. Evol. Syst. 36,
191–218.
Holland, E. L.A., Braswell, B. H., and Sulzman, J. (2005). Nitrogen deposition onto
the United States and Western Europe: Synthesis of observations and models.
Ecol. Appl. 15, 38–57.
Hopp, L., and McDonnell, J. J. (2009). Connectivity at the hillslope scale: Identifying
interactions between storm size, bedrock permeability, slope angle and soil
depth. J. Hydrol. 376, 378–391.
Hosono, T., Wang, C.-H., Umezawa, Y., Nakano, T., Onodera, S., Nagata, T.,
Yoshimizu, C., Tayasu, I., and Taniguchi, M. (2011). Multiple isotope (H, O,
N, S and Sr) approach elucidates complex pollution causes in the shallow
groundwaters of the Taipei urban area. J. Hydrol. 397, 23–36.
Howarth, R. W. (2008). Coastal nitrogen pollution: A review of sources and trends
globally and regionally. Harmful Algae 8, 14–20.
Hubbard, R. K., Newton, G. L., and Hill, G. M. (2004). Water quality and the grazing
animal. J. Anim. Sci. 82, E255–E263.
Humborg, C., Mörth, C.M., Sundbom, M. and Wulff, F. (2007). Riverine transport
of biogenic elements to the Baltic Sea – past and possible future perspectives.
Hydrol. Earth Syst. Sci., 11, 1593–1607.
Hyvönen, R., Agren, G. I., Linder, S., Persson, T., Cotrufo, M. F., Ekblad, A., Freeman,
M., Grelle, A., Janssens, I. A., Jarvis, P. G., Kellomäki, S., Lindroth, A., Loustau,
D., Lundmark, T., Norby, R. J., Oren, R., Pilegaard, K., Ryan, M. G., Sigurdsson,
B. D., Strömgren, M., van Oijen, M., and Wallin, G. (2007). The likely impact
of elevated [CO2 ], nitrogen deposition, increased temperature and management
on carbon sequestration in temperate and boreal forest ecosystems: A literature
review. New Phytol.; 173, 463–480.
Igbinosa, E. O., and Okoh, A. I. (2009). Impact of discharge wastewater effluents
on the physico-chemical qualities of a receiving watershed in a typical rural
community. Int. J. Environ. Sci. Technol. 6, 175–182.
Jaynes, D. B., Colvin, T. S., Karlen, D. L., Cambardella, C. A., and Meek, D. W.
(2001). Nitrate loss in subsurface drainage as affected by nitrogen fertilizer rate.
J. Environ. Qual. 30, 1305–1314.
Jiao, J. J., Wang, Y., Cherry, J. A., Wang, X., Zhi, B., Du, H., and Wen, D. (2010). Ab-
normally high ammonium of natural origin in a coastal aquifer-aquitard system
in the Pearl River Delta, China. Environ. Sci. Technol. 44, 7470–7475.
Johnson, L. T., Tank, J. L., Hall Jr., R.O., Mulholland, P. J., Hamilton, S. K., Valett, H.
M., Webster, J. R., Bernot, M. J., McDowell, W. H., Peterson, B. J., and Thomas,
S. M. (2013). Quantifying the production of dissolved organic nitrogen in head-

water streams using 15N tracer additions. Limnol. Oceanogr. 4, 1271–1285.
Kaspar, T. C., Jaynes, D. B., Parkin, T. B., and Moorman, T. B. (2007). Rye cover crop
and gamagrass strip effects on NO3 concentration and load in tile drainage. J.
Environ. Qual. 36, 1503–1511.
Kaushal, S. S., Lewis, W. M., and McCutchan, J. H. (2006). Land use change and
nitrogen enrichment of a Rocky Mountain watershed. Ecol. Appl. 16, 299–312.
Kaushal, S. S., Groffman, P. M., Band, L. E., Elliott, E. M., Shields, C. A., and Kendall,
C. (2011). Tracking nonpoint source nitrogen pollution in human-impacted
watersheds. Environ. Sci. Technol. 45, 8225–8232.
Kaushal, S. S., Groffman, P. M., Band, L. E., Shields, C. A., Morgan, R. P., Palmer, M.
A., Belt, K. T., Swan, C. M., Findlay, S. E.G., and Fisher, G. T. (2008). Interaction
between urbanization and climate variability amplifies watershed nitrate export
in Maryland. Environ. Sci. Technol. 42, 5872–5878.
Kim, G., Yur, J., and Kim, J. (2007). Diffuse pollution loading from urban stormwater
runoff in Daejeon city, Korea. J. Environ. Manage. 85, 9–16.

Kojima, K., Murakami, M., Yoshimizu, C., Tayasu, I., Nagata, T., and Furumai, H.
(2011). Evaluation of surface runoff and road dust as sources of nitrogen using
nitrate isotopic composition. Chemosphere 84, 1716–1722.
Kroeger, K. D., Cole, M. L., and Valiela, I. (2006). Groundwater-transported dis-
solved organic nitrogen exports from coastal watersheds. Limnol. Oceanogr.
51, 2248–2261.
Kurunc, A., Ersahin, S., Uz, B. Y., Sonmez, N. K., Uz, I., Kaman, H., Bacalan, G.
E., and Emekli, Y. (2011). Identification of nitrate leaching hot spots in a large
area with contrasting soil texture and management. Agric. Water Manage. 98,
1013–1019.
Kurz, I., Coxon, C., Tunney, H., and Ryan, D. (2005). Effects of grassland man-
agement practices and environmental conditions on nutrient concentrations in
overland flow. J. Hydrol. 304, 35–50.
Lassaletta, L., Garcı́a-Gómez, H., Gimeno, B. S., and Rovira, J. V. (2009). Agriculture-
induced increase in nitrate concentrations in stream waters of a large Mediter-
ranean catchment over 25 years (1981–2005). Sci. Total Environ. 407, 6034–6043.
Laurent, F., and Ruelland, D. (2011). Assessing impacts of alternative land use and
agricultural practices on nitrate pollution at the catchment scale. J. Hydrol. 409,
440–450.
Lawrence, G. B., Goolsby, D. A., Battaglin, W. A., and Stensland, G. J. (2000).
Atmospheric nitrogen in the Mississippi River Basin–emissions, deposition and
transport. Sci. Total Environ. 248, 87–99.
Le Bissonnais, Y., Cerdan, O., Lecomte, V., Benkhadra, H., Souchère, V., and Martin,
P. (2005). Variability of soil surface characteristics influencing runoff and interrill
erosion. Catena 62, 111–124.
Legout, C., Molenat, J., Aquilina, L., Gascuel-Odoux, C., Faucheux, M., Fauvel, Y.,
and Bariac, T. (2007). Solute transfer in the unsaturated zone-groundwater con-
tinuum of a headwater catchment. J. Hydrol. 332, 427–441.
Li, S., Liu, W., Gu, S., Cheng, X., Xu, Z., and Zhang, Q. (2009). Spatio-temporal
dynamics of nutrients in the upper Han River basin, China. J. Hazard. Mater.
162, 1340–1346.
Li, T., Zhang, W., Feng, C., and Shen, J. (2013). Performance assessment of separate
and combined sewer systems in metropolitan areas in southern China. Water
Sci. Technol. 69(2), 422–429.
Liang, X.-Q., Xu, L., Li, H., He, M.-M., Qian, Y.-C., Liu, J., Nie, Z.-Y., Ye, Y.-S., and
Chen, Y. (2011). Influence of N fertilization rates, rainfall, and temperature on
nitrate leaching from a rainfed winter wheat field in Taihu watershed. Phys.
Chem. Earth 36, 395–400.
Liu, X., Duan, L., Mo, J., Du, E., Shen, J., Lu, X., Zhang, Y., Zhou, X., He, C., and
Zhang, F. (2011). Nitrogen deposition and its ecological impact in China: An
overview. Environ. Pollut. 159, 2251–2264.
Liu, X., Ju, X., Zhang, Y., He, C., Kopsch, J., and Fusuo, Z. (2006). Nitrogen deposition
in agroecosystems in the Beijing area. Agric. Ecosyst. Environ. 113, 370–377.
Liu, Y., Tao, Y., Wan, K. Y., Zhang, G. S., Liu, D. B., Xiong, G. Y., and Chen, F.
(2012). Runoff and nutrient losses in citrus orchards on sloping land subjected
to different surface mulching practices in the Danjiangkou Reservoir area of
China. Agric. Water Manage. 110, 34–40.

Liu, Z., Yang, J., Yang, Z., and Zou, J. (2012). Effects of rainfall and fer-
tilizer types on nitrogen and phosphorus concentrations in surface runoff
from subtropical tea fields in Zhejiang, China. Nutr. Cycling Agroecosyst. 93,
297–307.
Luo, Y., Yang, X., Carley, R. J., and Perkins, C. (2003). Effects of geographical location
and land use on atmospheric deposition of nitrogen in the State of Connecticut.
Environ. Pollut. 124, 437–448.
Mangimbulude, J. C., van Breukelen, B. M., Krave, A. S., van Straalen, N. M., and
Röling, W. F.M. (2009). Seasonal dynamics in leachate hydrochemistry and nat-
ural attenuation in surface run-off water from a tropical landfill. Waste Manage.
29, 829–838.
Mantovi, P., Fumagalli, L., Beretta, G. P., and Guermandi, M. (2006). Nitrate leaching
through the unsaturated zone following pig slurry applications. J. Hydrol. 316,
195–212.
Marques, M. (2007). Stormwater runoff pollution at waste management sites. Urban
Water J. 4, 173–181.
Masetti, M., Poli, S., Sterlacchini, S., Beretta, G. P., and Facchi, A. (2008). Spatial and
statistical assessment of factors influencing nitrate contamination in groundwa-
ter. J. Environ. Manage. 86, 272–281.
McClain, M. E., Boyer, E. W., Dent, C. L., Gergel, S. E., Grimm, N. B., Groff-
man, P. M., Hart, S. C., Harvey, J. W., Johnston, C. A., Mayorga, E., Mc-
Dowell, W. H., and Pinay, G. (2003). Biogeochemical hot spots and hot
moments at the interface of terrestrial and aquatic ecosystems. Ecosystems 6,
301–312.
McDaniel, P. A., Regan, M. P., Brooks, E., Boll, J., Barndt, S., Falen, A., Young,
S. K., and Hammel, J. E. (2008). Linking fragipans, perched water tables, and
catchment-scale hydrological processes. Catena 73, 166–173.
McGechan, M., and Topp, C. F. (2004). Modelling environmental impacts of depo-
sition of excreted nitrogen by grazing dairy cows. Agric. Ecosyst. Environ. 103,
149–164.
Merseburger, G., Martı́, E., and Sabater, F. (2005). Net changes in nutrient concen-
trations below a point source input in two streams draining catchments with
contrasting land uses. Sci. Total Environ. 347, 217–229.
Merseburger, G., Martı́, E., Sabater, F., and Ortiz, J. D. (2011). Point-source effects
on N and P uptake in a forested and an agricultural Mediterranean streams. Sci.
Total Environ. 409, 957–967.
Meyles, E., Williams, A., Ternan, L., and Dowd, J. (2003). Runoff generation in
relation to soil moisture patterns in a small Dartmoor catchment, Southwest
England. Hydrol. Processes 17, 251–264.
Mian, I. A., Riaz, M., and Cresser, M. S. (2009). The importance of ammonium
mobility in nitrogen-impacted unfertilized grasslands: A critical reassessment.
Environ. Pollut. 157, 1287–1293.
Molenat, J., Gascuel-Odoux, C., Ruiz, L., and Gruau, G. (2008). Role of water table
dynamics on stream nitrate export and concentration in agricultural headwater
catchment (France). J. Hydrol. 348, 363–378.
Morton, T. G., Gold, A. J., and Sullivan, W. M. (1988). Influence of overwatering and
fertilization on nitrogen losses from home lawns. J. Environ. Qual. 17, 124.
Mousavi, S., Ibrahim, S., Aroua, M. K., and Ghafari, S. (2012). Development of
nitrate elimination by autohydrogenotrophic bacteria in bio-electrochemical
reactors—A review. Biochem. Eng. J. 67, 251–264.
Mulholland, P. J., Helton, A. M., Poole, G. C., Hall, R. O., Hamilton, S. K., Peterson,
B. J., Tank, J. L., Ashkenas, L. R., Cooper, L. W., Dahm, C. N., Dodds, W. K.,
Findlay, S. E.G., Gregory, S. V., Grimm, N. B., Johnson, S. L., McDowell, W. H.,
Meyer, J. L., Valett, H. M., Webster, J. R., Arango, C. P., Beaulieu, J. J., Bernot,
M. J., Burgin, A. J., Crenshaw, C. L., Johnson, L. T., Niederlehner, B. R., O’Brien,
J. M., Potter, J. D., Sheibley, R. W., Sobota, D. J., and Thomas, S. M. (2008).
Stream denitrification across biomes and its response to anthropogenic nitrate
loading. Nature 452, 202–205.
Needelman, B. A., Gburek, W. J., Petersen, G. W., Sharpley, A. N., and Kleinman,
P. J.A. (2004). Surface runoff along two agricultural hillslopes with contrasting
soils. Soil Sci. Soc. Am. J. 68, 914–923.
Norton, M., and Fisher, T. (2000). The effects of forest on stream water quality in
two coastal plain watersheds of the Chesapeake Bay. Ecol. Eng. 14, 337–362.
Nyenje, P. M., Foppen, J. W., Uhlenbrook, S., Kulabako, R., and Muwanga, A. (2010).
Eutrophication and nutrient release in urban areas of sub-Saharan Africa: A
review. Sci. Total Environ. 408, 447–455.
Ohte, N. (2012). Implications of seasonal variation in nitrate export from forested
ecosystems: A review from the hydrological perspective of ecosystem dynamics.
Ecol. Res. 27, 657–665.
Ojeda, G., Tarrasón, D., Ortiz, O., and Alcañiz, J. M. (2006). Nitrogen losses in runoff
waters from a loamy soil treated with sewage sludge. Agric. Ecosyst. Environ.
117, 49–56.
Oman, C. B., and Junestedt, C. (2008). Chemical characterization of landfill
leachates–400 parameters and compounds. Waste Manage. 28, 1876–1891.
Paerl, H. W., Dennis, R. L., and Whitall, D. R. (2002). Atmospheric deposition of
nitrogen: Implications for nutrient over-enrichment of coastal waters. Estuaries
25, 677–693.
Pärn, J., Pinay, G., and Mander, Ü. (2012). Indicators of nutrients transport from
agricultural catchments under temperate climate: A review. Ecol. Indic. 22, 4–15.
Patil, R. H., Laegdsmand, M., Olesen, J. E., and Porter, J. R. (2010). Effect of soil
warming and rainfall patterns on soil N cycling in Northern Europe. Agric.
Ecosyst. Environ. 139, 195–205.
Paul, M. J., and Meyer, J. L. (2001). Streams in the urban landscape. Urban Ecol. 32,
333–365.
Pellerin, B. A., Kaushal, S. S., and McDowell, W. H. (2006). Does anthropogenic
nitrogen enrichment increase organic nitrogen concentrations in runoff from
forested and human-dominated watersheds? Ecosystems 9, 852–864.
Penna, D., Meerveld, H. J., Gobbi, A., Borga, M., and Dalla Fontana, G. (2011). The
influence of soil moisture on threshold runoff generation processes in an alpine
headwater catchment. Hydrol. Earth Syst. Sci. 15, 689–702.
Peterson, B. J., Wollheim, W. M., Mulholland, P. J., Webster, J. R., Meyer, J. L., Tank,
J. L., Marti, E., Bowden, W. B., Valett, H. M., Hershey, A. E., McDowell, W. H.,
Dodds, W. K., Hamilton, S. K., Gregory, S., and Morrall, D. D. (2001). Control
of nitrogen export from watersheds by headwater streams. Science 292, 86–90.
Petrovic, A. M. (1990). The fate of nitrogenous fertilizers applied to turfgrass. J.
Environ. Qual. 19(1), 1–14.
Phan, L. T., Bui, T. M., Nguyen, T. T.T., Gooley, G. J., Ingram, B. A., Nguyen, H. V.,
Nguyen, P. T., and De Silva, S. S. (2009). Current status of farming practices of
striped catfish, Pangasianodon hypophthalmus in the Mekong Delta, Vietnam.
Aquaculture 296, 227–236.
Puigdefábregas, J. (2005). The role of vegetation patterns in structuring runoff and
sediment fluxes in drylands. Earth Surf. Processes Landforms 30, 133–147.
Ramos, M. C., and Martı́nez-Casasnovas, J. A. (2006). Nutrient losses by runoff in
vineyards of the Mediterranean Alt Penedès region (NE Spain). Agric. Ecosyst.
Environ. 113, 356–363.
Ranalli, A. J., and Macalady, D. L. (2010). The importance of the riparian zone
and in-stream processes in nitrate attenuation in undisturbed and agricultural
watersheds—A review of the scientific literature. J. Hydrol. 389, 406–415.
Rao, N. S., Easton, Z. M., Schneiderman, E. M., Zion, M. S., Lee, D. R., and Steen-
huis, T. S. (2009). Modeling watershed-scale effectiveness of agricultural best
management practices to reduce phosphorus loading. J. Environ. Manage. 90,
1385–1395.
Reay, W. G. (2004). Septic tank impacts on ground water quality and nearshore
sediment nutrient flux. Ground Water 42, 1079–1089.
Renou, S., Givaudan, J. G., Poulain, S., Dirassouyan, F., and Moulin, P. (2008).
Landfill leachate treatment: Review and opportunity. J. Hazard. Mater. 150,
468–493.
Rimski-Korsakov, H., Rubio, G., and Lavado, R. S. (2004). Potential nitrate losses
under different agricultural practices in the pampas region, Argentina. Agric.
Water Manage. 65, 83–94.
Robertson, G. P., and Groffman, P. (2007). Nitrogen transformations. Soil Microbiol.
Biochem. Ecol. 341–364.
Rouxel, M., Molénat, J., Ruiz, L., Legout, C., Faucheux, M., and Gascuel-Odoux, C.
(2011). Seasonal and spatial variation in groundwater quality along the hillslope
of an agricultural research catchment (Western France). Hydrol. Processes 25,

831–841.
Royer, T. V., David, M. B., and Gentry, L. E. (2006). Timing of riverine export of
nitrate and phosphorus from agricultural watersheds in Illinois: Implications for
reducing nutrient loading to the Mississippi River. Environ. Sci. Technol. 40,
4126–4131.
Royer, T. V., Tank, J. L., and David, M. B. (2004). Landscape and watershed processes
transport and fate of nitrate in headwater agricultural streams in Illinois. J.
Environ. Qual. 33, 1296–1304.
Ruiz, L., Abiven, S., Durand, P., Martin, C., Vertès, F., and Beaujouan, V. (2002a).
Effect on nitrate concentration in stream water of agricultural practices in small
catchments in Brittany: I. Annual nitrogen budgets. Hydrol. Earth Syst. Sci. 6,
497–506.
Ruiz, L., Abiven, S., Martin, C., Durand, P., Beaujouan, V., and Molénat, J. (2002b).
Effect on nitrate concentration in stream water of agricultural practices in small
catchments in Brittany: II. Temporal variations and mixing processes. Hydrol.

Earth Syst. Sci. 6, 507–514.
Ryther, J. H., and Dunstan, W. M. (1971). Nitrogen, phosphorus, and eutrophication
in the coastal marine environment. Science 171, 1008–1013.
Sahu, M., and Gu, R. R. (2009). Modeling the effects of riparian buffer zone and
contour strips on stream water quality. Ecol. Eng. 35, 1167–1177.
Schilling, K., and Zhang, Y.-K. (2004). Baseflow contribution to nitrate-nitrogen
export from a large, agricultural watershed, USA. J. Hydrol. 295, 305–316.
Schindler, D. W., Dillon, P. J., and Schreier, H. (2006). A review of anthropogenic
sources of nitrogen and their effects on Canadian aquatic ecosystems. Biogeo-
chemistry 79, 25–44.
Schmidt, C. A., and Clark, M. W. (2012). Efficacy of a denitrification wall to treat
continuously high nitrate loads. Ecol. Eng. 42, 203–211.
Schulz, M., Kozerski, H.-P., Pluntke, T., and Rinke, K. (2003). The influence of
macrophytes on sedimentation and nutrient retention in the lower River Spree
(Germany). Water Res. 37, 569–578.
Sebestyen, S. D., Boyer, E. W., Shanley, J. B., Kendall, C., Doctor, D. H., Aiken, G.
R., and Ohte, N. (2008). Sources, transformations, and hydrological processes
that control stream nitrate and dissolved organic matter concentrations during
snowmelt in an upland forest. Water Resour. Res. 44, W12410.
Seitzinger, S., Harrison, J. A., Böhlke, J. K., Bouwman, A. F., Lowrance, R., Peterson,
B., Tobias, C., and Van Drecht, G. (2006). Denitrification across landscapes and
waterscapes: A synthesis. Ecol. Appl. 16, 2064–2090.
Sheriff, G. (2005). Efficient waste? Why farmers over-apply nutrients and the impli-
cations for policy design. Rev. Agric. Econ. 27, 542–557.
Shields, C. A., Band, L. E., Law, N., Groffman, P. M., Kaushal, S. S., Savvas, K.,
Fisher, G. T., and Belt, K. T. (2008). Streamflow distribution of non-point source
nitrogen export from urban-rural catchments in the Chesapeake Bay watershed.
Water Resour. Res. 44, 1–13.
Shrestha, S., and Kazama, F. (2007). Assessment of surface water quality using multi-
variate statistical techniques: A case study of the Fuji river basin, Japan. Environ.
Modell. Software 22, 464–475.
Sindilariu, P.-D., Reiter, R., and Wedekind, H. (2009). Impact of trout aquaculture
on water quality and farm effluent treatment options. Aquat. Living Resour. 22,
93–103.
Smith, R. G., Gross, K. L., and Robertson, G. P. (2008). Effects of crop diversity on
agroecosystem function: Crop yield response. Ecosystems 11, 355–366.
Soonthornnonda, P., and Christensen, E. R. (2008). Source apportionment of pollu-
tants and flows of combined sewer wastewater. Water Res. 42, 1989–1998.
Sutton, M. A., Howard, C. M., Erisman, J. W., Billen, G., Bleeker, A., Grennfelt, P.,
van Grinsven, H., and Grizzetti, B. (2011). The European nitrogen assessment:
Sources, effects and policy perspectives. Cambridge University Press: Cambridge.
Tang, J.-L., Zhang, B., Gao, C., and Zepp, H. (2008). Hydrological pathway and
source area of nutrient losses identified by a multi-scale monitoring in an agri-
cultural catchment. Catena 72, 374–385.
Tanner, C. C., Sukias, J. P.S., Headley, T. R., Yates, C. R., and Stott, R. (2012).
Constructed wetlands and denitrifying bioreactors for on-site and decentralised
wastewater treatment: Comparison of five alternative configurations. Ecol. Eng.

42, 112–123.
Taylor, G. D., Fletcher, T. D., Wong, T. H.F., Breen, P. F., and Duncan, H. P. (2005).
Nitrogen composition in urban runoff—Implications for stormwater manage-
ment Water Res., 39, 1982–1989.
Thieu, V., Garnier, J., and Billen, G. (2010). Assessing the effect of nutrient mitigation
measures in the watersheds of the Southern Bight of the North Sea. Sci. Total
Environ. 408, 1245–1255.
Thomas, O., Jung, A. V., Causse, J., Louyer, M. V., Piel, S., Baurès, E., and
Thomas, M. F. (2014). Revealing organic carbon–nitrate linear relationship
from UV spectra of freshwaters in agricultural environment. Chemosphere 107,
115–120.
Thompson, R. B., Martı́nez-Gaitan, C., Gallardo, M., Giménez, C., and Fernández,
M. D. (2007). Identification of irrigation and N management practices that
contribute to nitrate leaching loss from an intensive vegetable produc-
tion system by use of a comprehensive survey. Agric. Water Manage. 89,
261–274.
Tian, S., Youssef, M. A., Skaggs, R. W., Amatya, D. M., and Chescheir, G. M. (2012).
Temporal variations and controlling factors of nitrogen export from an artificially
drained coastal forest. Environ. Sci. Technol. 46, 9956–9963.
Tomer, M. D., Meek, D. W., Jaynes, D. B., and Hatfield, J. L. (2003). Evaluation of
nitrate nitrogen fluxes from a tile-drained watershed in Central Iowa. J. Environ.
Qual. 32, 642–653.
Tromp-van Meerveld, H. J., and McDonnell, J. J. (2006). Threshold relations in
subsurface stormflow: 1. A 147-storm analysis of the Panola hillslope. Water
Resour. Res. 42, W02410.
Tromp-van Meerveld, H. J., Peters, N. E., and Mcdonnell, J. J. (2007). Effect of
bedrock permeability on subsurface stormflow and the water balance of a
trenched hillslope at the Panola Mountain Research Watershed, Georgia, USA.
Hydrol. Processes 769, 750–769.
Truman, C. C., Potter, T. L., Nuti, R. C., Franklin, D. H., and Bosch, D. D. (2011).
Antecedent water content effects on runoff and sediment yields from two Coastal
Plain Ultisols. Agric. Water Manage. 98, 1189–1196.
Uchida, T., Meerveld, H. J., and McDonnell, J. J. (2005). The role of lateral pipe
flow in hillslope runoff response: An intercomparison of non-linear hillslope
response. J. Hydrol. 311, 117–133.
Udawatta, R.P., Motavalli, P.P., Garrett, H.E. and Krstansky, J.J., (2006). Nitrogen
losses in runoff from three adjacent agricultural watersheds with claypan soils,
Agric. Ecosyst. Environ., 117, 39–48.
Umezawa, Y., Hosono, T., Onodera, S., Siringan, F., Buapeng, S., Delinom,
R., Yoshimizu, C., Tayasu, I., Nagata, T., and Taniguchi, M. (2009). Er-
ratum to “Sources of nitrate and ammonium contamination in ground-
water under developing Asian megacities”. Sci. Total Environ. 407(9),
3219–3231.
USEPA. (2004). Report to Congress on the Impacts and Control of CSOs and
SSOs. Chapter 5 - Environmental Impacts of CSOs and SSOs. US. EPA Press,
Washington.
USEPA. (2009). Combined sewer overflow controls in Southeast Michigan. Retrieved
April 17, 2015, from http://www.epa.gov/med/grosseile site/indicators/cso.
html.
Vázquez, N., Pardo, A., Suso, M. L., and Quemada, M. (2006). Drainage and ni-
trate leaching under processing tomato growth with drip irrigation and plastic
mulching. Agric. Ecosyst. Environ. 112, 313–323.
Vidon, P., Allan, C., Burns, D., Duval, T. P., Gurwick, N., Inamdar, S., Lowrance,
R., Okay, J., Scott, D., Sebestyen, S., and Low, R. (2010). Hot spots and hot
moments in riparian zones: potential for improved water quality management.
J. Am. Water Resour. Assoc. 46(2), 278–298.
Vymazal, J. (2007). Removal of nutrients in various types of constructed wetlands.
Sci. Total Environ. 380, 48–65.
Wagenschein, D., and Rode, M. (2008). Modelling the impact of river morphology
on nitrogen retention—A case study of the Weisse Elster River (Germany). Ecol.
Modell. 211, 224–232.
Wakida, F. T., and Lerner, D. N. (2005). Non-agricultural sources of groundwater
nitrate: A review and case study. Water Res. 39, 3–16.
Wakida, F. T., and Lerner, D. N. (2006). Potential nitrate leaching to groundwater
from house building. Hydrol. Processes 20, 2077–2081.
Ward, M. H., deKok, T. M., Levallois, P., Brender, J., Gulis, G., Nolan, B. T., and Van-
Derslice, J. (2005). Workgroup report: Drinking-water nitrate and health—recent
findings and research needs. Environ. Health Perspect. 113, 1607–1614.
Wei, L., Zhang, B., and Wang, M. (2007). Effects of antecedent soil moisture on
runoff and soil erosion in alley cropping systems. Agric. Water Manage. 94,
54–62.
Wei, Y., Davidson, B., Chen, D., and White, R. (2009). Balancing the economic, so-
cial and environmental dimensions of agro-ecosystems: An integrated modeling
approach. Agric. Ecosyst. Environ. 131, 263–273.
Withers, P. J.A., and Jarvie, H. P. (2008). Delivery and cycling of phosphorus in
rivers: A review. Sci. Total Environ. 400, 379–395.
Wohlfart, T., Exbrayat, J.-F., Schelde, K., Christen, B., Dalgaard, T., Frede, H.-G., and
Breuer, L. (2012). Spatial distribution of soils determines export of nitrogen and
dissolved organic carbon from an intensively managed agricultural landscape.
Biogeosciences 9, 4513–4525.
Wollheim, W. M., Pellerin, B. A., Vörösmarty, C. J., and Hopkinson, C. S. (2005). N
retention in urbanizing headwater catchments. Ecosystems 8, 871–884.
Wu, L., Long, T., Liu, X., and Guo, J. (2012). Impacts of climate and land-use changes
on the migration of non-point source nitrogen and phosphorus during rainfall-
runoff in the Jialing River Watershed, China. J. Hydrol. 475, 26–41.
Zhang, Q., Streets, D. G., Carmichael, G. R., He, K. B., Huo, H., Kannari, A., Klimont,
Z., Park, I. S., and Reddy, S. (2009). Asian emissions in 2006 for the NASA
INTEX-B mission. Atmos. Chem. Phys. 9, 5131–5153.
Zheng, F.-L., Huang, C.-H., and Norton, L. D. (2004). Effects of near-surface hydraulic
gradients on nitrate and phosphorus losses in surface runoff. J. Environ. Qual.
33, 2174–2182.
Zhou, J.-B., Xi, J.-G., Chen, Z.-J., and Li, S.-X. (2006). Leaching and transformation
of nitrogen fertilizers in soil after application of N with irrigation: A soil column
method. Pedosphere 16, 245–252.
Colloids and Surfaces A: Physicochem. Eng. Aspects 481 (2015) 567–576
Colloids and Surfaces A: Physicochemical and

Engineering Aspects
journal homepage: www.elsevier.com/locate/colsurfa
Fluorescence spectroscopy as a specific tool for the interaction study

of two surfactants with natural and synthetic organic compounds
Jung A.-V. a,∗ , Frochot C. b , Bersillon J.-L. c
a
Ecole des Métiers de l’Environnement (EME), Campus de Ker Lann, Avenue Schuman, 35170 Bruz, France
b
Laboratoire Réactions et Génie des procédés (LRGP), UMR 7274, CNRS, Université de Lorraine 1, Rue Grandville BP 20451, 54001 Nancy, France
c
Laboratoire Interdisciplinaire des Environnements Continentaux (LIEC), CNRS UMR 7360 UFR SciFA, Université de Lorraine, Site Charmois, 15 avenue du
Charmois, 54500 Vandoeuvre les Nancy, France
g r a p h i c a l a b s t r a c t
Article history: Four different techniques were used to study the binding of cationic cetyltrimethylammonium bro-
Received 10 February 2015 mide (CTAB) and non-ionic nonylphenylethoxyl (NPE) surfactants to three synthetic organic components
Received in revised form 18 June 2015 that mimic humic-like aggregates and to two natural aggregated humic substances (HS) extracted from
Accepted 19 June 2015
aquatic suspended matter. The composition of synthetic organic components were chosen to be simi-
Available online 22 June 2015
lar to high molecular weight highly processed terrigenous HS and low and high molecular weight less
processed terrigenous (or aquatic terrigenous) HS. The natural HS were extracted under two different
Keywords:
meteorological conditions (rainy and dry periods). No significant interaction between the non-ionic sur-
Cetyltrimethylammonium bromide (CTAB)
surfactant
factant and any of the studied compounds was found. Concerning CTAB; pH, conductivity and turbidity
Nonylphenylethoxyl (NPE) surfactant measurements, along with fluorescence spectroscopy were combined to provide a better understanding
Humic-like aggregates of interactions between organic aggregates and the surfactant. The spectroscopic data show that a “highly
Natural and synthetic organic compounds processed terrigenous HS” fluorophore interacts in a different way with the cationic surfactant than an
Abbreviations: A, area under the emission fluorescence curve for mixture of surfactants and organic materials; A0 , area under the emission fluorescence curve of
organic material alone without surfactant; CG, catechol glycine synthetic organic compound; CMC, critical micellar concentration; CTAB, cetyltrimethylammonium bromide
surfactant; CT1, catechol tri-glycine (1–3.5 kDa) synthetic organic compound; CT2, catechol tri-glycine (>8 kDa) synthetic organic compound; DIS, derivative isotherms
summation; FI, fluorescence intensity; HA, humic acid; HS, humic-like aggregate; n, refractive index of the media organic matter – surfactant solutions; n0 , refractive index
of organic matter solutions; N1, natural humic compound collected during rainy period (autumn), extracted from suspended matters according to Thurman and Malcolm
(1981); N2, natural humic compound collected during drought period (summer), extracted from suspended matters according to Thurman and Malcolm (1981); NOM, natural
organic matter; NPE, non-ionic nonylphenylethoxyl surfactant; NTU, nephelometric turbidity unit; OD, optical density for mixture of surfactants and organic materials at the
excitation wavelength selected for fluorescence emission measurements; OD0 , optical density for organic materials without surfactant at the excitation wavelength selected
for fluorescence emission measurements; ϕf /ϕ0 , fluorescent quantum yield calculated for each surfactant and organic matter solution.
∗ Corresponding author. Fax: +33 2 99 05 88 09.
E-mail address: av2jung@live.fr (A.-V. Jung).
http://dx.doi.org/10.1016/j.colsurfa.2015.06.037
0927-7757/© 2015 Elsevier B.V. All rights reserved.
568 A.-V. Jung et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 481 (2015) 567–576
“aquatic terrigenous (or less processed terrigenous) HS” fluorophore does. Under similar conditions, some
spectral changes in the fluorescence signal are correlated to changes in non-specific physical-chemical
parameters (pH, turbidity, conductivity) for the organic compounds tested. The complexation mechanism
is essentially governed by charge neutralization, which can be monitored specifically by the fluorescence
of the organic moieties.
© 2015 Elsevier B.V. All rights reserved.
1. Introduction chromatography and octanol-water partitioning. This approach

yielded evidence of various interaction modes between humic col-
There has been an exponential increase in the production loids and cationic/anionic surfactants.
and use of surfactants over the past century [1]. Indeed, their Although many papers have reported investigations into the
amphiphilic nature makes them invaluable as detergents, for sequestration of organic pollutants by either natural organic matter
surface hydrophobization, and for stabilization/destabilization or surfactants, the interactions at the particle level between surfac-
processes in both industrial and domestic applications [2]. Despite tants and humic aggregates are not yet well-understood. Moreover,
the high biodegradability required by law for these molecules [3], all the previously cited characterization methods are quite time-
surfactants are introduced into the environment by waste-water consuming (due to the analytical techniques themselves, but also
discharge, point-charge pollution, and, ironically, during the reme- due to the time needed for pre-analytical preparation of the
diation of polluted soils or groundwaters [4–5]. sample) and could be destructive for HS samples. Therefore, the
Humic compounds are complex, heterogeneous materials experiments described in the present work were designed to exam-
formed in soils, sediments, and natural waters by biochemical and ine the nature of surfactant-humic aggregate interactions using
chemical reactions that occur during the decay and transformation fast, non-destructive measurements. Fluorescence spectroscopy
of plant and microbial remains. Humic aggregates are involved in has commonly served to characterize organic matter (OM)-metal
many processes in soils and natural water, such as soil weather- or OM-polycyclic aromatic hydrocarbon interactions (e.g. [21]),
ing, plant nutrition, pH buffering, trace metal mobility and toxicity, but the technique is also well-suited to explore the complexity of
bioavailability, degradation and transport of hydrophobic organic surfactant-OM interactions [22]. Indeed, fluorophores are sensitive
chemicals, formation of disinfection by-products during water to their surroundings, i.e. they are indicators of local changes, rather
treatment and heterotrophic production in blackwater ecosys- than macroscopic changes. The aim of this report was to com-
tems [6–7]. The mechanisms of binding of contaminants to humic pare the interaction of synthetic organic components (chosen to
aggregates have been discussed in the literature [8–10], as they mimic various molecular weight HSs of different physical–chemical
affect the distribution, transport and mobility of contaminants composition) with natural HS (extracted from suspended aquatic
in waters and soils. Moreover, it is well-established that natu- matter under two different meteorological conditions (rainy and
ral organic matter itself (NOM) has natural surfactant properties dry periods)) with two different charged surfactants (non-ionic
[11–13]. nonyphenylethoxyl (NPE) and cationic CTAB). The measurements
Up until now, several techniques have been used to investigate were performed at a given pH range from 4.5 to 8 depending on
the interactions between surfactants and colloidal humic aggre- the organic compounds (initial pH 8 for all compounds) and tem-
gates. Shang and Rice [14] used X-ray scattering to reveal that perature using fluorescence emission spectroscopy combined with
the structure of hexadecyltrimethylammonium bromide micelles conductivity, turbidity and pH measurements. The different tech-
was disturbed by the addition of various humic acids (HA). Binding niques are complementary and their combined results allow a more
isotherms [8], charge measurements with a particle charge detector detailed interpretation of the surfactant binding results.
[9], X-ray and interfacial tension measurements [15], HS/aqueous-
phase partition coefficients measured by fluorescence quenching
[16], light scattering and electrophoretic mobility studies [17] 2. Materials and methods
have been performed on various surfactant-humate complexes.
The possible structures of the complexes formed were discussed 2.1. Organic materials
and sometimes modeled [10]. The main conclusions of all these
studies were that HAs are removed from the aqueous phase by 2.1.1. Preparation of synthetic organic components
forming hydrophobic complexes with cationic surfactants which Three types of synthetic organic components (CG, CT1, CT2)
are neutral at the iso-electric point of the complexes [8–10]. Otto were synthesized according to the Andreux procedure [23–24]. This
et al. [18] measured diffusion coefficients with pulse-field gradi- synthesis method consists of an oxidative polymerization reaction
ent nuclear magnetic resonance spectroscopy for various humic using catechol and two different amino-acids: a glycine (the organic
aggregates in the presence of sodium dodecyl sulfate (SDS) and compound will be called CG) or a tri-glycine (the corresponding
cetyltrimethylammonium bromide (CTAB). It was thus shown that organic compounds will be called CT1 and CT2 depending on their
humic aggregates enhance the aggregation of SDS prior to micel- molecular weight). The synthetic compounds present different N/C
lization, a more pronounced effect being observed for the more elemental atomic ratios and molecular weight fractions.
hydrophobic humic materials. The interaction of CTAB with humic NOM can be divided into two major subsets: (i) autochtonous
aggregates was also studied for the removal of HA by flotation [19]. organic compounds (active or detrital) (ii) allochtonous organic
Another study showed that ion-pair formation between dissoci- materials (from vegetation, soil and rocks of the watershed). Each
ated functional groups of humic substances and CTAB could be type of organic substance has a different age range. Autochtonous
analyzed quantitatively by a potentiometric method using an ionic substances are freshly synthesized so are easily assimilated, while
surfactant selective electrode. Furthermore, the results correlated the bulk of the allochtonous organic materials are older, reflecting
with the acidity of HS [20]. Other authors [5] examined the com- a more refractory nature to such damage [25]. The main sources of
bined effects of HA and surfactants (cationic and anionic) on the autochtonous substances are (i) phytoplankton and bacteria in sus-
water solubility of pyrene, using high-performance size exclusion pension (free or bound) (ii) aquatic plants (macrophytes), mosses,
A.-V. Jung et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 481 (2015) 567–576 569
Table 1
Organic materials under investigation.
Organic fraction O/Ca N/Ca H/Ca pKA1 b (acidity meq/g) pKA2 b (acidity meq/g) pKA3 b (acidity meq/g) CMCc (mg/L) HId
CG 0.97 0.89 0.56 4.6 (2.3) 5.1 (0.8) 7.2 (0.6) * 84.6
CT2 1.02 3.33 0.78 3.9 (4.1) 7.7 (0.2) 9.8 (1.3) 320 82.8
*
CT1 1.86 4.92 1.38 3.8 (3.8) 7.6 (0.1) 9.6 (1.5) 76.6
N1 0.53 2.74 1.87 3.7 (2.1) 8.9 (0.4) 10.1 (10.6) 500 51.1
N2 0.75 1.77 1.61 3.2 (2.7) 8.5 (0.3) 9.5 (15) 450 52.2
a
Elemental analysis (all in wt%) (O = 100%-C, H, N).
b
Three pKA were determined by the derivative isotherms summation (DIS) method ([33,34]).
c
Critical micellar concentration (CMC) determined by surface tension measurements using Wilhemy technique [51] with a maximum standard deviations values for all
fractions of 50 mg/L.
d
Hydrophoby index (HI) determined by reversed-phase HPLC. HI corresponds to the Hydrophilic/hydrophobic ratio characterisation of the humic acid fractions. It was
performed with a Shimadzu HPLC (DGU-14A, Kyoto, Japan), equipped with a C18 reversed-phase column (5 ␮m, 150 × 4 mm) from Lichrosorb (Agilent, Palo Alto, CA, USA),
a fluorescent RF-10AXL detector, an LC-10AT UP pump and an FCV-10ALVP autosampler, all from Agilent. The reversed-phase HPLC conditions were chosen according to
[39] Components with a retention time below 11 min were considered as “hydrophilic” moieties. Peaks with elution time over 20 min were integrated for the hydrophobic
contribution quantification.
*
No stabilization value obtained.
invertebrate excretions and microflora associated with trash and International Humic Substances Society (IHSS) standard procedure
epilithon (bioderm benthic). Autochtonous aquatic humic-like sub- for the isolation of the humic fraction of bulk solid earth mate-
stances possess a higher N/C ratio than terrigenous processed rials, with slight modifications [30]. During the dissolution step
humic-like substances ([26,27]). The N/C ratios of organic com- of the HA fraction, NaOH 0.1 M was used instead of KOH. This
pounds were chosen to mimic the N/C ratio of aquatic, terrigenous base was chosen to obtain the sodium form of the final HA, so
and/or autochtonous and allochtonous HS. HA have a higher molec- as to better compare these extracts with the synthetic products.
ular weight than fulvic acids (FA). The same can be said of soil (or For the same reason, crystallized NaCl was added instead of KClto
ground) humic substances compared to aquatic humic substances. flocculate the mineral fraction. Because of the difference in the floc-
Indeed, according to Thurman and Malcolm [28], FA corresponds to culation potential of these two salts, the NaCl dose was recalculated
substances from 500 to 2000 Daltons, whereas HA would encom- according to results from Bonneau and Souchier [31]. Dialysis was
pass substances from 24 000 to 1360 000 Daltons Hayes et al. [29]. performed against distilled water with an 8000 molecular weight
Molecular weights of organic compounds were also chosen to be cut-off (Spectra/Por® 7, Roth). It should be noted that no organic
as close as possible to natural HAs. materials below this molecular weight cut-off were obtained.
Catechol (6.6 g) was dissolved in 500 mL of a solution of
NaH2 PO4 /Na2 HPO4 sodium phosphate buffer at pH 8 to obtain a
0.03 M catechol solution. This was then mixed with an equal vol- 2.1.3. Characterization of synthetic and natural substances
ume of an0.03 M solution of buffered glycine or tri-glycine. The Elemental analysis, pKa, critical micellar concentration (CMC)
resulting mixture was allowed to react in a 2 L flask in the dark and hydrophobicity indexes (HI) of the synthetic organic com-
at 25 ◦ C under a pure oxygen constant pressure (0.9 bar) for 5 pounds and natural substances are reported in Table 1. Other
days. At the end of the reaction, the solutions were transferred to extensive characterization results are presented elsewhere [32].
Spectra/Por® 7 dialysis bags (Roth Sochiel, Lauterbourg, France). For The composition is given as an elemental ratio to clarify the com-
the polymer prepared with glycine (CG), the fraction above a 8 kDa parison with previously published data. According to elemental
molecular cut-off was collected. For the polymers prepared with analysis, all data lies within ranges usually reported in the litera-
tri-glycine (CT), successive dialysis steps with 3 different molec- ture for humic-like compounds [30] except for CT1, which has a low
ular weight cut-offs were used to isolate 2 different tri-glycine carbon content. According to our previous studies [32], CT1 com-
molecular weight fractions (1 3.5 kDa, CT1; above 8 kDa, CT2). The position could be close to that of the final step of the evolution of
fraction below 1 kDa, which encompassed non-reactive or remain- humic substances in natural media, where microbial-derived mate-
ing free monomeric and oligomeric compounds, was not kept. rial constitutes the major component. The synthetic CG compound
The final forms of all the polymeric HAs were sodium-saturated. shows an elemental composition related to that of a soil humic
The different fractions of polymer solutions were freeze-dried by material prior to any microbial degradation. The elemental ratios
freeze-drying and weighed. The powders were carefully stored in also indicate that the natural substances are more aliphatic than
glass bottles at 4 ◦ C in the dark. To obtain the different HS solutions, the synthetic materials.
the powders were then re-dispersed (see Section 2.3). Acidity dissociation constants have been evaluated thanks
to derivative isotherms summation (DIS) method [33–34]. The
applied method is based on that developed by Prélot [33]. Potentio-
2.1.2. Natural substance extraction metric titration data (pH in function of the volume of the solution
Samples of natural suspended matter were collected with a added by titrating) are considered as an adsorption isotherm of
17,000 rpm flow-through (1 m3 /h) centrifuge at the water intake protons (or hydroxyl). The curve is reconstructed by theoreti-
of the Urban Community of Nancy, France. This sampling point cal derived isotherms type Bragg–William–Temkin [35–36]. The
is located between the pumping station in the Moselle River and decomposition of the experimental isotherm in local derivative
the retention basins used as an alternative water resource. The isotherms provides different settings for each peak: position, mid-
water was pumped from 20 cm below surface level. Suspended height width, number of protons or hydroxyl adsorbed. The lowest
solids were collected on Teflon plates covering the inner surface value (pKA1 ) can be attributed to carboxylic acidity, the second
of the centrifuge. Each sample was placed in a glass bottle and (pKA2 ) to amino-phenolic acidity and the third (pKA3 ) to phenolic
stored at 4 ◦ C during transport and storage prior to treatment. acidity [23]. As expected since the same functional acidic groups are
The extraction of humic substances began within 24 h after sam- present in the final products, both CT fractions have similar pKA
pling the river water. Two samples were collected, N1 under dry values. Interestingly, their acidic contents are similar. The differ-
and “hot” summer conditions, and N2 under rainy and “cold” fall ences still have to be investigated, especially by studying relaxation
weather conditions. Extraction was carried out according to the during the titration process as suggested in Jung et al. [32]. Car-
Fig. 1. Typical absorbance spectra (a) and fluorescence spectra (c) of organic matter in presence of NPE. (b) NPE absorbance spectra (10−3 M).
boxylic groups due to glycine incorporation are more abundant 2.2. Surfactants
in the synthetic polymers, whereas phenolic groups are ten times
more numerous in natural substances than in synthetic products CTAB and NPE were purchased from Sigma–Aldrich and used
[32]. These results clearly show that CT exhibits the carboxylic and without further purification. To gain information about surfactant
phenolic acidities that are found in natural humic compounds from purity, CMC values were determined thanks to surface tension
the Moselle River. pKA1 and pKA3 values are consistent with data measurements and it was checked that the addition of NaHCO3
found in the literature [37–38]. The pKA2 value is consistent with did not affect the CMC values (data not shown). The experimental
the data of Andreux et al. [23], attributing this second acidity to values CMCCTAB = 9.10−4 mol/L and CMCNPE = 6.3.10−5 mol/L found
glycine/carboxylic-like structures. This pKA2 accounts for less than were consistent with data in the literature [40–41].
5% of the total acidity. Potentiometric titration shows that synthetic
organic components feature the two main carboxylic and pheno- 2.3. Mixtures
lic acidities, even though phenolic moieties contribute a higher
proportion of the total acidity for natural materials. This suggests The same mixing protocol was used for both surfactants:
that the aromatic parts of the synthetic organic compounds have a 10 mg/L solutions of CG, CT1, CT2 or N2 were prepared at an ini-
higher polymerization degree than the natural compounds isolated tial pH of 8, and the surfactant was added to reach a concentration
in this work. The carboxylic acidity (given in meq/g for pKA1 and from 10−3 to 1.5 of their respective CMCs. A 30 mg/L concentration
pKA2 ) is greater for synthetic organic compounds than for natural was used for N1, because the fluorescence signal was too weak at
ones. It appears that the synthetic organic compounds look like near 10 mg/L. The solutions of humic aggregates and surfactants were
to final products of the oxidation evolution of natural substances mixed in the dark for 4–6 h to reach chemical equilibrium before
[32]. measurements were taken [42]. For all experiments, the temper-
The hydrophilic to hydrophobic balance (HI ratio in Table 1) ature was maintained at 25 ◦ C and the initial pH was stabilized
determined by reversed phase high performance liquid chromatog- using NaHCO3 (336 ppm) that corresponds to 5.39.10−4 mol/L for
raphy [39] is lower for synthetic compounds (more hydrophobic) the ionic strength in the media, similar to the Moselle river aver-
than for natural ones. This is consistent with the previous conclu- age ionic strength [43]. In such conditions, CTAB contributes from
sion about the higher aromatic contribution in synthetic organic 0.15 (10−3 CMC) to 100% (CMC) of the ionic strength for the chosen
compounds. The hydrophobic character of synthetic organic com- concentration range. The same preparation process was used for
ponents is also confirmed. all measurements. No attempts were made to take into account the
According to all these results, CG could be a kind of model sorption of surfactants on reactor walls or on to the quartz cuvettes
organic compound for some high molecular weight highly used for spectroscopic measurements.
processed terrigenous HS, CT1 for the low molecular weight
(fulvic-like) aquatic autochtonous (microbially processed) or less 2.4. General physical–chemical measurements
processed terrigenous HS and CT2 the high molecular weight
(humic-like) aquatic autochtonous (microbially processed) or less pH, conductivity and turbidity were measured with a Hach sen-
processed terrigenous HS. sion3 pH meter, an electrical conductivity EC215 cell from VWR
Table 2
Fluorescence parameters considered.
Fractions Excitation Emission domain recorded

wavelengths (nm) to calculate the
(nm) quantum yielda
CG 288 350–500 (␣ )

337.5 350–600 (␣)
CT2 272.5 350–550 (␣ )
335 350–600 (␣)
CT1 272 300–550 (␣ )
330 350–600 (␣)
N1 270 275–550 (␦, ␣ )
330 350–600 (␣)
N2 270 275–550 (␦, ␣ )
330 350–600 (␣)
a
The name of the fluorescent emission components (␦, ␣ or ␣) detected in the
domain according to [32] and the literature is indicated under brackets.
and a Hach 2100AN turbidimeter, respectively. Before measure-

ment, the samples were conditioned by stirring with a magnetic
bar for about 2 min. Turbidity measurements were performed after Fig. 2. pH Evolution of the various OM solutions investigated in presence of CTAB.
2 min at rest, 5 cm under the sample surface. Each measurement

was duplicated and the average is reported. A reproducibility study
(series of measurements of standard solutions) was performed for
all solutions: the maximum standard deviations for pH, turbidity,
and conductivity were 0.05, 1NTU, and 10 ␮S/cm, respectively.
2.5. Spectroscopic measurements
2.5.1. UV/Visible measurements

The UV/visible absorption spectra were recorded between 200
and 700 nm with a 2501 PC computer-driven spectrophotometer
from Shimadzu (Kyoto, Japan). Solutions were placed in quartz glass
cells (Whatman® ) with a 10 mm light path length.
2.5.2. Fluorescence measurements

Steady-state fluorescence measurements were performed on a
Spex Fluorolog-3 spectrofluorometer (Jobin-Yvon, Stanmore, UK)
using a slit width of 0.5 mm for the excitation and emission
monochromators. The instrument was equipped with a 450 W
Xenon lamp (Jobin-Yvon). The sample temperature was maintained
at 25.0 ± 0.1 ◦ C by a LTD6G temperature control unit (Grant, Sarose,
UK). To compare the fluorescence intensities of all spectra and to
avoid time drift, fluorescence was normalized with a 0.16 mg/L flu-
oresceine solution buffered at pH 10. The absorption of organic
matter was corrected for the inner filter effect (absorption inter-
fering with fluorescence measurements) according to Auger [44]
using UV/Visible data. The excitation wavelength was selected in
agreement with the position of the fluorescence maximum of each
fraction of organic matter. Further details are provided in Table 2.
To obtain quantitative fluorescence measurements, the fluores-
cent quantum yield (ϕf /ϕ0 ) was determined for each surfactant and
organic matter solution. ϕf /ϕ0 Characterizes the ability of the flu-
orophore to emit the energy absorbed, in the form of light. It is
defined as the ratio of the number of photons emitted to the num-
ber of absorbed photons for the same time period. It is calculated
from the following equation [45]:
ϕf A × OD0 × n2
=
ϕ0 A0 × OD × n20
where A is the area under the emission fluorescence curve for the
mixture of surfactant and organic matter, A0 is the area under the
emission fluorescence curve of organic matter without surfactant,
OD is the optical density for the mixture of surfactant and organic Fig. 3. Conductivity (a) and Turbidity (b) as a function of CTAB concentration for
matter at the excitation wavelength selected for fluorescence mea- organic matter-surfactant mixtures.
surements, OD0 is the optical density for organic matter without
surfactant at the same excitation wavelength, n and n0 are the
Fig. 4. Fluorescence emission spectra (␣ fluorophore) of humic compounds in presence of various CTAB concentrations.
refractive indexes of the media, their ratio is considered close to 3.2. CTAB
1 since the experiments took place in aqueous solution of NaHCO3
(refractive index = 1.5 [46]). The refractive index of OM and CTAB In contrast, the results obtained with CTAB indicate drastic
solutions are considered close to 1.5 on average [47–50]. changes upon the addition of a cationic surfactant.
The NPE absorption capacity is negligible compared to all the
compound absorptions for the concentrations tested (Fig. 1b). CTAB
3.2.1. pH, Conductivity and turbidity measurements
does not absorb light at the specific wavelengths considered here.
In the presence of CTAB, the pH decreased significantly for all
The two surfactants do not give any fluorescence signal at the exci-
solutions of humic compounds and reached a plateau at around
tation wavelengths used in this study.
[CTAB]/CMC = 0.1 (Fig. 2). At the pH range of the study (4.5–8
depending on the organic compounds see Fig. 2), only the carboxylic
groups of organic matter were dissociated (corresponding to pKA1
3. Results and discussion and sometimes pKA2 in Table 1, depending on the organic matter
considered). No direct correlation was found between the differ-
In the following sections, the indicated surfactant concentra- ent pH profiles and the corresponding acidities. Specific functional
tion is the total surfactant concentration and not the equilibrium groups and hydrophobic interactions as well as generic forces with
monomer concentration. respect to the electrostatic interactions can favor, or on the con-
trary block, proton release in solution. After an initial drop, the pH
remained constant at higher CTAB concentrations.
3.1. NPE In Fig. 3a, conductivity is plotted against CTAB concentration.
It can be seen that the conductivity of the solution was related to
As illustrated in Fig. 1, the addition of increasing amounts of the nature of humic compounds, since synthetic organic compo-
NPE to the various solutions of HS (CG, CT1, CT2, N1 and N2) did nents gave conductivity values some 100 times lower than those of
not induce any changes. The UV/Visible and fluorescence spectra natural aggregates. This could be explained by matrix effects. For
and the classical physical-chemical parameter measurements (pH, natural HS, the presence of different kinds of counter ions (other
conductivity, turbidity) remained unchanged, i.e. less than repeata- than Na+ ) is highly probable. This can occur even if the purification
bility error upon NPE addition. Such results are consistent with procedure is performed the same way as for synthetic organic com-
previous published studies, which showed that a non-ionic surfac- pounds. This is attributed to quality variations in natural aquatic
tant does not exert any influence on the octanol–water partition media. For CT1, CT2 and CG, the conductivity remained more or
of humic acids [5]. It seems that NPE does not interact with the less constant up to 0.3CMC than increased. For N1, the same gen-
different compounds. eral trend was followed. However the increase started at a higher
Fig. 5. Fluorescence emission spectra (␦ and ␣ fluorophores) of humic compounds in presence of various CTAB concentrations.
CTAB concentration (0.6CMC). It is possible that N2 conductivity ima can be distinguished; namely at the lowest concentration
only increased very weakly. A charge neutralization mechanism tested and around 0.2CMC. The aggregation state of a natural sub-
occurred between HS and CTAB, as suggested by the stability of the stance is also higher and more compact than those of the organic
conductivity. This is attributed to a reaction between carboxylate synthetic compounds and the breaking-up of aggregates is more
moieties of HS and the cationic head of surfactant molecules. When fluctuant.
all accessible negatively charged sites were saturated, the addition Brum and Oliveira [17] observed precipitates for a 30–60 ppm
of CTAB accounted for the conductivity increase. Indeed, for CTAB addition of CTAB to a 20 ppm commercial ALDRICH HA solution at
concentrations higher than 0.3CMC, the surfactant contributed to at pH 8. Our concentration range for the precipitation of all fractions
least 50% of the ionic strength of the initial NaHCO3 solution. This corresponds to around 0.05CCTAB /CMC to around 0.5CCTAB /CMC (i.e.
would suggest that proton release or liberation of organic mate- 20–200 ppm). Aldrich HA precipitation CTAB concentration is thus
rial counterions appears to be a minor parameter encountering of the order of magnitude of all our results. It is difficult to propose
conductivity increase in the second step of the mechanism. an explanation for the differences observed in the different stud-
The turbidity curves are shown in Fig. 3b. These can give ies, since many factors can affect the results: the HS nature, the
indirect information on complexing between CTAB and OM. Tur- efficient collision frequencies (these in turn depend on surfactant
bidity may indicate aggregation, or the breaking-up of aggregates. and HS concentration as well as temperature), the reaction kinetic
The turbidity measurements confirm the visual observations of rate constants. The most likely assumption is that surfactant ions
the solutions: a solid precipitate is formed in suspension around move fast compared to HS particles and that complexing between
0.1CMC for all HS fractions, corresponding to the turbidity peak the surfactant and HS precedes aggregation, provided surfactant
of the curves (Fig. 3b). For N1 and N2, two main turbidity max- concentration is sufficiently high.
Fig. 6. Evolution of the fluorescence quantum yields ϕf /ϕ0 for the various fluorophores ( for ␦, • for ␣ and 䊐 for ␣ ) of organic materials in presence of CTAB.
3.2.2. Fluorescence spectroscopy results orescence was stabilized beyond 0.5–0.6 [CTAB]/CMC, as defined
Various fluorophores are commonly identified in natural organic by stabilization of the complexes, the last stage of particle reor-
matter and they present different fluorescent emission bands ganization in solution. Generally, a decrease in fluorescence signal
[51–52]. The so-called alpha components absorb at 250–260 nm reveals some quenching by organic aggregates, whereas an increase
and emit at 420–480 nm, whereas alpha components absorb at might correspond to either a condensation process of organic mate-
330–350 nm and emit at 450–490 nm. We detected both alpha and rial or the formation of new fluorescing material [55]. In all cases
alpha bands in all compounds investigated, with different intensi- except for N2, the alpha fluorescence is higher with CTAB than with-
ties related to the CTAB concentration (Figs. 4 and 5). For N1 and N2, out. This suggests that CTAB induces an alternative conformation
another fluorescence band called the delta peak can be evidenced or aggregation state with less quenching, which could be due to
(Fig. 5). Table 2 summarizes the different fluorescence peak ranges. hydrophobic, electrostatic and any other specific interactions.
According to several studies ([51,53,54]), the alpha , alpha and delta
peaks, respectively, correspond to fluorophores associated with
3.2.3. General discussion concerning CTAB-organic matter
highly processed terrigenous, aquatic/autochtonous microbial or
interaction
less terrigenous humic-like substances and tryptophan moieties.
Brum and Oliveira [17] showed that the cationic polar head of
The fluorescence emission spectra of the different aggregates
the surfactant binds electrostatically to the dissociated carboxylic
in the presence of various CTAB concentrations are shown in
groups present in the humic aggregates, to form precipitates. As the
Figs. 4 and 5. The evolution of fluorescence intensity with CTAB
surfactant concentration is increased the nonpolar portions of the
concentration is shown in Fig. 6. Three stages could be distin-
surfactant molecules interact through van der Waals forces forming
guished for all the natural and synthetic organic components: (i)
hemi-micelles. Shang and Rice [14], investigating the complexation
up to 0.1 [CTAB]/CMC, fluctuations in fluorescence were observed,
of HA with CTAB using small-angle X-ray scattering, showed that
corresponding to an unstable transitional phase, (ii) at around 0.3
after a neutralization step the humic network collapsed, increasing
[CTAB]/CMC, an increase of ␣ fluorescence (and a concomitant
the hydrophobicity of the system. In exchange, the humic substance
decrease for ␣ except for CT1) was seen that could be attributed to
induced the deformation of two-dimensional liquid crystals formed
a 3D reorganization of the compounds, (iii) for all samples, the flu-
from CTAB through an electrostatic interaction between the CTAB
head groups and the functional groups of HAs. All these findings the fluorescence signal are correlated to changes in non-specific
are complementary and allow the interpretation of our results. physical-chemical parameters (pH, turbidity, conductivity) for the
Thanks to four different techniques we were able to observe the organic compounds tested.
influence of NPE and CTAB addition onto different natural humic The role of cationic surfactants in aquatic media is certainly
or synthetic humic-like aggregates. For all the organic materials, underestimated. Both natural organic matter (NOM) and surfac-
we detected an interaction with CTAB but not NPE. For the first tants are known to enhance the apparent aqueous solubility of
time, a fluorescence emission tool was used to visualize surfactant hydrophobic organic contaminants [57]. As suggested by several
interactions with HS and similar organic materials. The interac- studies [58–59], thanks to their natural surfactant properties, HS
tions are different according to the amount of CTAB added, as are found to reduce sorption of organic contaminants on spiked
proved by the 4 different techniques. For [CTAB] < 0.1CMC, pH and soils. For further natural attenuation in washed soils for instance,
fluorescence decrease, while turbidity increases. We can suggest we can also benefit from the buffering capacity of organic com-
that the complexes formed result mainly from the neutraliza- pounds in the soils with respect to pollution, which should not be
tion of humic anionic charges by the cationic surfactants. First, viewed as a way of avoiding responsibility concerning environmen-
a release of protons into the media might occur (decreasing pH) tal challenges.
to allow the neutralization of the negatively charged carboxylic
moieties by CTAB polar ammonium heads. This release can be
Acknowledgments
more or less favored due to hydrophobic forces of the compounds
that can induce a quenching of alpha fluorescence emission and
This work was achieved within the framework of “Zone Ate-
an activation of alpha fluorescence emission (except for CT1). As
lier du bassin de la Moselle”, with financial support from Region
the complexes are formed, an increase of turbidity appears in
Lorraine and CNRS (Contrat de Plan Etat – Région and Programme
the medium, since a visual precipitation occurs. The precipitate
“Environnement Vie et Société”). Moreover, we would like to sin-
remains in suspension. In the next step, the HS are saturated
cerely thank Pr. B. Lartiges and an anonymous reviewer for his/her
and the conductivity increases again since there are more acces-
very helpful and constructive comments on our work.
sible “carboxylate” groups or free ones to form a complex by
charge interactions with CTAB. This is attributed to a CTAB increase
and a minor contribution to OM-counterion liberation. Turbidity References
decreases in favor of a less aggregated state, supported by the sta-
bilization of fluorescence intensities. [1] European Chemical Agency (ECHA) (2014) Available on http://echa.europa.
eu/fr/
Other interactions than pure charge interactions can occur: the
[2] P. Mukherjee, S.K. Oadhan, S. Dash, S. Patel, B.K. Mishra, Clouding behavior in
natural surfactant property of HS can play a role. Hydrophobic bind- surfactant systems, Adv. Colloid Interface Sci. 162 (2011) 59–79.
ing forces can also explain some results. Natural materials (N1 and [3] REACH Regulation (EC) No 1907/2006 of the European Parliament and of the
N2) need more CTAB to be saturated (regarding conductivity). How- Council of 18 December 2006 concerning the Registration, Evaluation,
Authorisation and Restriction of Chemicals (REACH), establishing a European
ever, the number of carboxylate moieties (Table 1 examining the 2 Chemicals Agency.
first acidities) for these substances is on average smaller than for [4] J.H. Harwell, D.A. Sabatini, R.C. Knox, Surfactants for ground water
synthetic organic compounds. In contrast, based on the HI index remediation, Colloids Surf. A 151 (1999) 255–269.
[5] H. Lippold, U. Gottschalch, H.H. Kupsch, Joint influence of surfactants and
(Table 1), natural HS are less hydrophobic than synthetic organic humic matter on PAH solubility. Are mixed micelles formed, Chemosphere 70
compounds are. This correlates with the CMC determined in Jung (2008) 1979–1986.
et al. [56] for CT2, N1 and N2 (Table 1). Synthetic organic com- [6] J. Buffle, Complexation Reactions in Aquatic Systems, Ellis Horwood Ltd.,
Chichester, 1988.
pounds tend to cluster more easily than natural HS, increasing the [7] International humic aggregates society (IHSS) (2007) Available on http://
hydrophobic binding forces due to the bound surfactant. It can also www.humicaggregates.org/index.html
be expected that the binding sites for natural HS are also more [8] L.K. Koopal, T.P. Goloub, T.A. Davis, Binding of ionic surfactants to purified
humic acid, J. Colloid Interface Sci. 275 (2004) 360–367.
accessible to the surfactant CTAB than are those for the synthetic
[9] M. Ishiguro, W. Tan, L.K. Koopal, Binding of cationic surfactants to humic
organic compounds. aggregates, Colloids Surf. 306 (2007) 29–39.
[10] M. Ishiguro, L.K. Koopal, Predictive model of cationic surfactant binding to
humic aggregates, Colloids Surf. A 379 (2011) 70–78.
[11] R.L. Wershaw, A new model for humic materials and their interactions with
4. Conclusion hydrophobic organic chemicals in soil–water or sediment-water systems, J.
Contam. Hydrol. 1 (1986) 29–45.
Comparing NPE and CTAB binding to various HSs, it was found [12] H.-H. Cho, J. Choi, M.N. Goltz, J.-W. Park, Combined effect of natural organic
matter and surfactants on the apparent solubility of polycyclic aromatic
that electrostatic attraction plays an important role in the inter- hydrocarbons, J. Environ. Qual. 31 (2002) 275–280.
action. The role of the aliphatic tail cannot be excluded too with [13] B. Berkowitz, Contaminant Geochemistry: Interactions and Transport in the
the present experiments. Future experiments could be carried out Subsurface Environment, Springer Science & Business, 2014, 2015, 582 pp.
[14] C. Shang, J.A. Rice, Investigation of humate-cetyltrimethylammonium
with simple quaternary ammonium ion with only methyl groups complexes by small-angle X-ray scattering, J. Colloid Interface Sci. 305 (2007)
for instance or with different cationic surfactants with various tail 57–61.
lengths. [15] E. Tombacz, K. Varga, F. Szanto, A X-ray diffraction study of
alkylammoniumhumate complexes, Colloid Polym. Sci. 266 (1988) 734–738.
The main conclusion of this study is that the addition of a [16] S.J. Traina, D.C. McAvoy, D.J. Verstee, Association of
cationic surfactant such as CTAB can generate drastic modifica- linearalkylbenzenesulfonates with dissolved humic aggregates and its effect
tions of HS fluorescence characteristics, even at concentrations on bioavailability, Environ. Sci. Technol. 30 (1996) 1300–1309.
[17] M.C. Brum, J.F. Brum, Removal of humic acid drom water by precipitate
much lower than CMC. The three main fluorophores (alpha, alpha
flotation using cationic surfactants, Miner. Eng. 20 (2007) 945–949.
and delta) usually observed for humic-like substances can also [18] W.H. Otto, D.J. Britten, C.K. Larive, NMR diffusion analysis of surfactant-humic
be viewed as specific probes to monitor the interaction. Aquatic aggregate interactions, J. Coll. Interface Sci. 261 (2003) 508–513.
[19] A.I. Zouboulis, W. Jun, I.A. Katsoyiannis, Removal of humic acids by flotation,
and terrigenous humic-like fluorophores interact differently with
Colloids Surf. A. 231 (2003) 181–193.
CTAB. The structure of HS in the presence of CTAB favors alpha [20] E. Illés, E. Tombácz, New method for pH-dependent charge determination of
fluorescence emission, corresponding to the aquatic humic-like humic acids and its potential use in DOM analysis (2006) 13th Symposium on
fluorophore (rather autochtonous). Fluorescence is also a pow- Analytical and Environmental Problems, Szeged.
[21] A.-V. Jung, C. Frochot, F. Villieras, B.S. Lartiges, S. Parant, M.-L. Viriot, J.-L.
erful specific tool to monitor humic-like substance interactions Bersillon, Interaction of pyrene fluoroprobe with natural and synthetic
with CTAB. Under similar conditions, some spectral changes in organic components humic aggregates: examining the local particle
organization from photophysical and interfacial processes, Chemosphere 80 [40] C.A. Bunton, G. Savelli, Organic reactivity in aqueous micelles and similar
(2010) 228–234. assemblies, Adv. Phys. Org. Chem. 22 (1986) 213–309.
[22] R. von Wandruska, R.R. Engebretson, L.M. Yates, Humic acid pseudomicelles [41] Sigma–Aldrich (2004) Available on http://www.sigmaaldrich.com/catalog/
in dilute aqueous solution: fluorescence and surface tension measurements, [42] A.W. Adamson, Physical chemistry of surfaces, 5th ed., Wileys and Son, New
in: E.A. Ghabbour, G. Davies (Eds.), Understanding Humic Aggregates, York, 1990.
Advanced methods, properties and applications, 1999, 2015, pp. 79–85. [43] A.-V. Jung, Interaction De La Matière Organique Avec Les Micropolluants
[23] F. Andreux, D. Golebiowska, F. Chone, M. Metche, Characterization and Anthropiques Et Devenir Lors De La Coagulation. Phd Thesis, Institut National
transformation in mud middle of a humic model issued from autoxidation of Polytechnique de Lorraine (INPL), 2004.
catechol/glycine system and selectively marked with 14C, in: M.A. Anderson [44] R.L. Auger, A.M. Jacobson, M.M. Domach, Aqueous phase fluorescence
(Ed.), Soil Organic Matter Studies, vol 2, IAEA, Vienna, 1977, pp. 43–57. quenching technique for measuring naphtalene partition coefficients in
[24] F. Andreux, D. Golebiowska, M. Metche, Oxydative Polymerization of nonionic surfactant micelles, Environ. Sci. Technol. 29 (1995) 1273–1278.
O-diphenols in Presence or Absence of Amino-acids, Topics on [45] J.R. Lakowicz, Principles of Fluorescence Spectroscopy, third ed, Springer, New
(Catechol-glycocolle) and (Catechol-diglycylglycine), General Assembly of York, 2007.
Polyphenols Group Report, Logrono (Spain), 1980178–188, Bulletin 9. [46] http://www.chemicalbook.com
[25] M. Meybeck, C, N, P and S in rivers: from sources to global inputs in [47] O. Schmid, D. Chand, E. Karg, P. Guyon, G. Frank, E. Wietlicki, M.O. Andrea,
Biogeochemical Cycles and Global Change, in: R. Wollast, F.T. Mackenzie, L. Derivation of the density and refractive index of organic matter and
Chou (Eds.), Springer-Verlag, 1993, 2015, pp. 163–193. elemental carbon from closure between physical and chemical aerosol
[26] L.R.G. Wetzel, Gradient-dominated ecosystems: sources and regulatory properties, Environ. Sci. Technol. 43 (2009) 1166–1172.
functions of dissolved organic matter in freshwater ecosystems, [48] J.A. Koark, F.L. Rosario-Ortiz, R.S. Summer, Fluorescence Characterization of
Hydrobiologia 229 (1992) 181–198. Humic Substance Coagulation: Application of New Tools to an Old Process in
[27] E. Veyssy, Transferts De Matières Organiques Des Bassins Versants Aux Advances in the Physicochemical Characterization of Dissolved Organic
Estuaires De La Gironde Et De L’adour (sud-ouest De La France), PhD Univ., Matter: Impact on Natural and Engineered Systems, in: Rosario-Ortiz (Ed.),
Bordeaux I, 1998, 262 pp. ACS Symposium Series, American Chemical Society, Washington 2014, 2015,
[28] E.M. Thurman, R.L. Malcolm, Molecular size of aquatic humic substances, Org. pp. 282–300.
Geochem. 4 (1982) 27–35. [49] M. Avena, L. Koopal, Desorption of humic acids from an iron oxide surface,
[29] M.H.B. Hayes, P. MacCarthy, R.L. Malcolm, Search for Structure: Setting the Environ. Sci. Technol. 32 (1998) 2572–2577.
Scene Structures of Humic Substances: the Emergence of « Forms » Humic [50] P. Ekwall, L. Mandell, P. Solyom, The aqueous cetyltrimethylammonium
Substances II, J. Wileys and sons, London, 1989 (Chapters 1 and 24). bromide solutions, J. Colloid Int. Sci. 35 (1971) 519–528.
[30] E.M. Thurman, R.L. Malcolm, Preparative isolation of aquatic humic [51] B.J.H. Matthews, A.C. Jones, N.K. Theodorou, A.W. Tudhope,
aggregates, Am. Chem. Sci. J. 4 (1981) 463–466. Excitation–emission-matrix fluorescence spectroscopy applied to humic acid
[31] M. Bonneau, B. Souchier, Constituants et Propriétés du Sol, 2nd ed., Masson, bands in coral reefs, Mar. Chem. 55 (1996) 317–332.
Paris, 1977, pp. 1–230. [52] E. Parlanti, K. Worz, L. Geoffroy, M. Lamotte, Dissolved organic matter
[32] A.-V. Jung, C. Frochot, S. Parant, B.S. Lartiges, C. Selve, M.-L. Viriot, J.-L. fluorescence spectroscopy as a tool to estimate biological activity in a coastal
Bersillon, Synthesis of amino-phenolic humic-like aggregates and comparison zone submitted to anthropogenic inputs, Org. Geochem. 31 (2000)
with natural aquatic humic acids: a multi-analytical techniques approach, 1765–1781.
Org. Geochem. 36 (2005) 1252–1271. [53] P.G. Coble, Characterization of marine and terrestrial DOM in seawater using
[33] B. Prélot, Mesure Et Modélisation De L’hétérogénéité Énergétique À excitation-emission matrix spectroscopy, Mar. Chem. 51 (1996) 325–346.
L’interface Oxyde/electrolyte/métaux, Phd, Thesis, Institut National [54] A. Stubbins, J.-F. Lapierre, M. Berggren, Y.T. Prairie, T. Dittmar, P.A. del Giorgio,
Polytechnique de Lorraine, 2001, 2015, 246 pp. What is in an EEM? molecular signatures associated with dissolved organic
[34] F. Villieras, L.J. Michot, F. Bardot, J.M. Cases, M. François, W. Rudzinski, An fluorescence in Boreal Canada, Environ. Sci. Technol. 48 (2014) 10598–10606.
improved derivative method to study sulfate heterogeneity of clays minerals, [55] I. Saadi, M. Borisover, R. Armon, Y. Laor, Monitoring of effluent DOM
Langmuir 13 (1997) 1104–1117. biodegradation using fluorescence, UV and DOC measurements,
[35] W.L. Bragg, E.G. Williams, The effect of thermal agitation on atomic Chemosphere 63 (2006) 530–539.
arrangement in alloys (I), Proc. R. Soc. 145 (1934) 699. [56] A.-V. Jung, C. Frochot, F. Villieras, B.S. Lartiges, S. Parant, M.-L. Viriot, J.-L.
[36] W.L. Bragg, E.G. Williams, The effect of thermal agitation on atomic Bersillon, Interaction of pyrene fluoroprobe with natural and synthetic humic
arrangement in alloys (II), Proc. R. Soc. 151 (1935) 699. substances: examining the local molecular organization from photophysical
[37] S.K. Banerjee, S.K., Acidity, quotient values, and metal retention power of and interfacial processes, Chemosphere 80 (2010) 228–234.
humic acids of varying molecular weight, J. Indian Soc. Soil Sci. 27 (1979) [57] H.-H. Cho, J. Choi, M.N. Goltz, J.-W. Park, Combined effect of natural organic
197–205. matter and surfactants on the apparent solubility of polycyclic aromatic
[38] G.R. Aiken, D.M. McKnight, R.L. Wershaw, Humic substances in soil, sediment, hydrocarbons, J. Environ. Qual. 31 (2002) 275–280.
and water, in: D.M. Aiken, R.L. McKnight (Eds.), Geochemistry, Isolation and [58] S. Krishnamurthy, Humic aggregates and contaminant transport: a review,
Characterisation, Wiley-Interscience, New York, 1985, pp. 1–692. Northeastern Geol. 14 (1992) 59–64.
[39] F.C. Wu, R.D. Evans, P.J. Dillon, High-performance liquid chromatographic [59] P. Conte, A. Agretto, R. Spaccini, A. Piccolo, Soil remediation: humic acids as
fractionation and characterisation of fulvic acid, Anal. Chim. Acta. 464 (2002) natural surfactants in the washings of highly contaminated soils, Environ.
47–55. Pollut. 135 (2005) 515–522.
See discussions, stats, and author profiles for this publication at: http://www.researchgate.net/publication/277975253
Heavy Rainfall Impacts on Trihalomethane

Formation in Contrasting Northwestern
European Potable Waters
ARTICLE in JOURNAL OF ENVIRONMENTAL QUALITY · MAY 2015

Impact Factor: 2.65 · DOI: 10.2134/jeq2014.10.0442
READS
16
8 AUTHORS, INCLUDING:
Timothy G. Jones Estelle Baurès

Bangor University Ecole des hautes études en santé publique
21 PUBLICATIONS 466 CITATIONS 29 PUBLICATIONS 312 CITATIONS
SEE PROFILE SEE PROFILE
Aude-Valérie Jung Christopher Freeman

Ecole des Métiers de l'Environnement (EME) Bangor University
19 PUBLICATIONS 255 CITATIONS 188 PUBLICATIONS 5,141 CITATIONS
SEE PROFILE SEE PROFILE
All in-text references underlined in blue are linked to publications on ResearchGate, Available from: Ianis Delpla
letting you access and read them immediately. Retrieved on: 10 November 2015
Published July 10, 2015
Journal of Environmental Quality TECHNICAL REPORTS

Surface Water Quality
Heavy Rainfall Impacts on Trihalomethane Formation in

Contrasting Northwestern European Potable Waters
Ianis Delpla, Timothy G. Jones,* Don T. Monteith, David D. Hughes, Estelle Baurès, Aude-Valérie Jung,
Olivier Thomas, and Chris Freeman
Abstract
There is emerging concern over the impact of extreme events
such as heavy rainfall on the quality of water entering the
drinking water supply from aboveground sources, as such events
D issolved organic matter (DOM) is an
important and dynamic component of global biogeo-
chemical cycles. The movement of DOM from soils to
freshwaters represents an important transfer of organic carbon
(as dissolved organic carbon [DOC]) and nutrients from ter-
are expected to increase in magnitude and frequency in response
to climate change. We compared the impact of rainfall events on restrial to aquatic environments, and it plays a number of roles
streamwater quality in four contrasting upland (peatland and in the functioning of rivers and lake ecosystems. High intensity
mineral soil) and lowland agricultural catchments used to supply precipitation and storm events can lead to the rapid transfer of
drinking water in France (Brittany) and the United Kingdom (North large quantities of DOM from catchments to water bodies, sig-
Wales) by analyzing water samples collected before, during, and
nificantly altering the concentration and composition of DOM
after specific events. At all four streams, heavy rainfall led to a
considerable rise in organic matter concentration ranging from 48 in rivers and lakes (Yang et al., 2013; Nguyen et al., 2013). Storm
to 158%. Dissolved organic carbon (DOC) quality, as determined events may be responsible for a significant proportion of the total
using specific ultraviolet absorbance, changed consistently DOM export budget within a catchment; Hinton et al. (1997),
at all sites during rainfall events, with a greater proportion of for example, reported that periods of storm flow contributed 36
aromatic and higher molecular weight compounds following
to 64% of annual DOC export for two watersheds in Canada,
the onset of rainfall. However, the change in DOC quality and
quantity did not significantly alter the trihalomethane formation despite such flow conditions only occurring 11 to 21% of the
potential. We observed small increases in trihalomethane (THM) time. Similarly, Raymond and Saiers (2010) reported that the
generation only at the Welsh peatland and agricultural sites and a highest 4.8% of streamflow for 30 small US watersheds was
small decrease at the Brittany agricultural site. The proportion of responsible for 57% of annual DOC export.
brominated THMs in chlorinated waters was positively correlated
Given that short-term events can be responsible for large
with bromide/DOC ratio in raw waters for all sites and hydrological
conditions. These results provide a first indication of the potential flushes of DOC to freshwaters, it has been acknowledged that
implications for surface-based drinking water resources resulting monthly or weekly sampling regimes do not capture the complete
from expected future increases in rainfall event intensity and annual DOC export profile of a catchment and can therefore
extension of dry periods with climate changes. lead to underestimation of long-term DOC fluxes (Clark et al.,
2007; Neal et al., 2012).
The presence of DOM in freshwaters is a problem for the
production of clean, safe drinking water (Fenner and Freeman,
2011). Concentrations of DOM must be reduced using water
treatment processes such as coagulation to improve water
taste and odor and to minimize the formation of disinfection
by-products (DBPs), such as trihalomethanes (THMs),
I. Delpla, E. Baurès, A.-V. Jung, and O. Thomas, Advanced School of Public Health
(EHESP), Sorbonne Paris Cité, Avenue du Professeur Léon Bernard- CS 74312,
Rennes Cedex, France, Inserm, U 1085-Institute of Research in Environmental and
Copyright © 2015 American Society of Agronomy, Crop Science Society of Ameri- Occupational Health (IRSET), LERES, Rennes 1, Rennes, France, and European Univ.
ca, and Soil Science Society of America. 5585 Guilford Rd., Madison, WI 53711 USA. of Brittany (UEB); T.G. Jones, D.D. Hughes, and C. Freeman, Wolfson Carbon Capture
All rights reserved. No part of this periodical may be reproduced or transmitted Lab., School of Biological Sciences, Brambell Bldg., Bangor Univ., Deiniol Rd.,
in any form or by any means, electronic or mechanical, including photocopying, Bangor, Gwynedd, LL57 2UW, UK; D.T. Monteith, Centre for Ecology and Hydrology,
recording, or any information storage and retrieval system, without permission in Lancaster Univ., Lancaster, LA1 4AP, UK; A.-V. Jung, School of Environmental
writing from the publisher. Engineering (EME), Campus de Ker Lann, Avenue Robert Schumann, Bruz, France.
Assigned to Associate Editor Ying Ouyang.
J. Environ. Qual. 44:1241–1251 (2015)
doi:10.2134/jeq2014.10.0442 Abbreviations: BR, Brittany; DBP, disinfection by-product; DBPFP, disinfection
Supplemental material is available online for this article. by-product formation potential; DOC, dissolved organic carbon; DOM, dissolved
Received 26 Oct. 2014. organic matter; HAA, haloacetic acid; LOQ, limit of quantification; NW, North Wales;
Accepted 14 Apr. 2015. POC, particulate organic carbon; SUVA, specific ultraviolet absorbance; THM,
*Corresponding author (t.jones@bangor.ac.uk). trihalomethane; THMFP, trihalomethane formation potential.
1241
haloacetic acids (HAAs), and haloacetonitriles, which may be THMFP (expressed in micrograms of THM by milligrams of C)
harmful to human health (Krasner et al., 2006). Storm events was not measured, inferences could not be drawn as to whether
can have adverse impacts on drinking water quality as they can the rainfall event changed the proportion of THM precursors.
produce rapid increases in sediment, organic matter, turbidity, Using fluorescence techniques, Nguyen et al. (2013) identified
and pathogens and can decrease clarifier performance (Curriero significant inputs of terrestrially derived DOM with a high DBP
et al., 2001; Hurst et al., 2004). Furthermore, increases in formation potential during two storm events in a large watershed
organic matter in raw waters have a negative effect on coagulant in South Korea, with the more intense storm contributing greater
and chlorine demand, resulting in elevated treatment costs quantities of precursor compounds. This study and others, such
(Matilainen and Sillanpää, 2010). as Nguyen et al. (2010), which reported increases in aromatic
Concentrations of DOC have increased in many northern and fluorescent DOM compounds during a flushing event in a
hemisphere freshwater lakes and rivers over the last two decades forested watershed, and Hong et al. (2012), which demonstrated
(Freeman et al., 2001a; Monteith et al., 2007). Reductions in that protein-like DOM decreases during storm events, suggest
sulfur deposition, and the resulting recovery from acidification, that heavy rainfall events flush proportionally more DBP
has been identified as a potential key driver of the rising DOC precursor compounds than low-flow conditions and therefore
trend (Evans et al., 2012), with this trend likely to persist provided have significant impacts on water quality.
deposition continues to fall. Increasing atmospheric CO2 and It would seem likely that DOM and DBP responses to heavy
consequent changes in climate such as rising temperature may rainfall will vary depending on catchment characteristics, but to
lead to further increases in DOC export to freshwaters by date we are not aware of any studies that have compared effects
stimulating soil decomposition and net primary production in on drinking water resources located in contrasting environments
catchments (Freeman et al., 2001a, 2004). Climate change is also using consistent methodologies. Moreover, the impact of rainfall
predicted to increase the frequency of heavy rainfall events in events on specific DBPs is still unclear. In this study, we aimed
northern Europe (Bates et al., 2008; Freeman et al., 1998; IPCC, to assess the effect of spring rainfall events on the quality of
2013), and winter precipitation intensity and the number of wet water feeding four drinking water reservoirs in catchments with
days in the UK and northern France have been reported to have differing soil types and land uses in the UK and France.
increased during the second half of the 20th century (Klein-Tank
et al., 2002). Therefore, the magnitude of DOC export during Materials and Methods
storm events may be greater in the future, warranting greater
attention with respect to impacts on drinking water quality.
Sites Description
In the UK, many sources of drinking water are located in Our study comprised four catchments in oceanic temperate
upland regions. In these high rainfall environments, peat soils areas of northwestern Europe with contrasting altitudes, land
dominate many reservoir catchments. This soil type is almost use, and land management. Figure 1 provides an overview map
entirely organic and leaches high concentrations of DOC to of the sampling location, and Table 1 synthesizes the main
freshwaters; DOC export has been shown to be linearly related characteristics of the catchments.
to the percentage cover of peatlands within the catchment Raw water used by water treatment plants to produce
(Urban et al., 1989; Dillon and Molot, 1997; Monteith et al., potable water is abstracted from natural or manmade lakes or
2015). Because long-term increases in DOC have been greatest reservoirs, directly from rivers, or from groundwater. In the
in regions where many organic soils such as peats dominate, there UK, approximately two-thirds of drinking water originates
are concerns within the UK water industry that the treatment of from surface reservoirs. In France, there is a greater reliance on
water will become progressively more costly (Ritson et al., 2014). groundwater, yet 80% of drinking water is provided by surface
Drinking water in Brittany, one of the most intensive water resources in Brittany (Newell-Price and Daumer, 2006),
agricultural regions of France (Molénat et al., 2002), is also one of the study areas. Our choice of sites in this study aimed
provided largely by surface water resources. Over the past 60 to represent typical drinking water sources in Brittany (France)
yr, landscapes in this region have been profoundly modified: and North Wales (UK). The French site is a small river system
grassland areas have been plowed, hedgerows removed, fields used as a raw water abstraction point, while the UK sites are
enlarged, wetlands drained, and animal density increased streams that either flow into reservoirs that are used as sources of
progressively (Gascuel-Odoux et al., 2010). As a result of drinking water or are least typical of such upland waters. We did
intensive farming practices, most drinking water resources in not sample from the reservoirs themselves because any impact
the area face strong agricultural pressures (Molénat et al., 2002). of heavy rainfall on DOC concentrations and characteristics
Flushing of organic matter to raw waters following rainfall within the reservoir would be dampened substantially by the
events is common and of major concern for the drinking water effect of dilution. Sampling the inflowing streams enabled a
industry. However, the impact of rainfall events on drinking more accurate determination of the flush of DOM from soils
water quality parameters such as trihalomethane (THM) into freshwaters during a storm event.
precursor concentration and formation potential have not been Brittany
widely reported to date, and the few published studies are partly In Brittany (BR), land use is dominated by intensive
contradictory. farming (Gascuel-Odoux et al., 2010). Samples were collected
Alkhatib and Peters (2008) reported that rainfall events from the Meu River (BR1 [48°08¢ N and 1°57¢ O]), which
more than doubled total THM formation potential (THMFP) is a small tributary of the river Vilaine, the largest river in the
in a US river and that elevated THM precursor concentrations region. The catchment is dominated by an oceanic regime with
were measured up to 3 d after peak rainfall. However, as specific
1242 Journal of Environmental Quality
Fig. 1. Sampling locations of the four streams/rivers in the study, in France and the UK. BR = Brittany; NW = North Wales.
a mean annual precipitation of 711 mm and a mean annual air upstream from the sampling point and were later obtained via
temperature of 11.8°C over the last 30 yr (Météo France, http:// the Banque Hydro database managed by the French Ministry of
france.meteofrance.com/). The source of the tributary occurs Environment.
at an altitude of 225 m, and the catchment is characterized by North Wales
gentle slopes and schist bedrock lying on loamy soils (67%).
These relatively impermeable soil characteristics induce large In North Wales (NW), as in many other regions of the
hydrological variations and render the soils highly sensitive to UK, drinking water is often, but not exclusively, derived from
erosion. upland catchments where anthropogenic pressures from direct
Agricultural land forms 86% of the catchment area of 350 km2. and diffuse pollution are generally light relative to those in the
Catchment land use is characteristic of the intensive agricultural lowlands. Despite this, however, water treatment costs tend to
practices of the region, with a high density of livestock (cattle, be high due to inputs of organic matter from the organic soils
pigs, and poultry). The river is used for water abstraction by a that tend to dominate the catchments of upland environments
small-scale water service supplying water for 5600 people. (Milne and Brown, 1997; Hughes et al., 2013). Three sampling
A weather station, installed 200 m from the sampling site, sites were chosen in North Wales, representing a range of land
records the amount of precipitation in real time using a tipping use types typical of catchments feeding many reservoirs in the
bucket rain gauge (ARG 100). Data are continuously gathered UK. Two sites are representative of upland catchments (NW1
by a data logger (Campbell Scientific CR800) and transmitted and NW2), but to provide direct comparisons with the French
via a cellular modem. Over the period of our study, hydrological agricultural site, we also included one lowland site (NW3).
data were recorded hourly by a gauging station located 1 km The NW1 and NW2 catchments are relatively smaller in size
Table 1. Main characteristics of the river basins studied.

Number of
Site Altitude Catchment Mean
Site name Country Land use Pedology Geology inhabitants
code range area flowrate
supplied
m km2 m3 s-1
BR1 Montfort s/Meu France 35–225 350 High intensity 3.0 Silt: 67%, Brioverian schist 5,600
agriculture clay: 17%,
(poultry, dairy sand: 16%
and pigs)
NW1 Nant-y-Brwyn UK 380–460 1.2 Peat 0.05 Deep peat: 100% Siltstone: 100% 2,300
NW2 Afon y Foel UK 380–520 1.1 Mineral NA† Deep peat: 76%, Siltstone: 60.7%, NA
shallow peat: 24% rhyolite: 39.3%
NW3 Afon Cefni UK 15–76 21 Low intensive 5.4 Fine loam: 82%, Granite: 41.2%, 19,000
agriculture fine silty clay: 18% sandstone/mudstone:
(improved 27.1%, schist: 24.5%,
grassland, other: 7.3%
rough grazing)
† Not applicable.
Journal of Environmental Quality 1243
(1.1–1.2 km2) and located at a higher altitude (380–520 m) >0.45 mm, after measuring total organic carbon using the same
than NW3 (area: 21 km2; altitude: 15–76 m). method as for DOC in Brittany. Dissolved organic carbon for
The two closely located upland sites in Wales (NW1 and the NW samples was determined using a Multi N/C 2100
NW2) drain the Migneint-Arenig-Dduallt Special Area of Analytik Jena (BR) and an Analytical Sciences Thermalox TOC
Conservation, which consists mostly of blanket bog. Both sites analyzer (NW).
are in close proximity to each other, so the meteorological data Ultraviolet/visible absorbance spectra were also acquired
collected by a weather station at NW1 was also used for NW2 on filtered samples with PerkinElmer Lambda 35 (BR) and
and demonstrates the relatively wet and cool local environment Molecular Devices Spectramax M2e (NW) UV-visible
(mean annual precipitation: 2400 mm; mean temperature: spectrometers, using a 10 mm optical path length quartz cell for
5.6°C). NW1 (52°99¢ N, 3°80¢ W) drains predominantly all sites (acquisition step: 1 nm, slit width: 2 nm, and scan speed:
shallow slopes characterized by peat soils of approximately 1 to 1920 nm min-1).
2 m depth (Evans et al., 2012), while NW2 (52°99¢ N, 3°84¢ W) Bromide was analyzed by ion chromatography (Dionex
drains a much steeper catchment dominated by mineral soils DX 120; IonPac AS19 hydroxide-selective anion-exchange
such as podzols. NW3 (53°27¢ N, 4°36¢ W) is the largest stream column) (NF EN ISO, 2009) (BR samples) or a Metrohm 850
to drain into the lowland Llyn Cefni, on the island of Anglesey ion chromatograph and Dionex AS14A anion column (NW
( Jones, 2006). The catchment is dominated by low intensity samples).
agriculture and improved grassland; it receives a mean annual Trihalomethane formation potential of all samples was tested
precipitation of 850 mm and has a mean annual temperature of using a version of standard method 5710 in APHA, AWWA,
10.2°C (data provided by UK Meteorological Office for weather WEF (1992) and adapted by Goslan et al. (2002) to determine the
station 12 km away from stream). propensity of the DOC to form THMs during water treatment.
Continuous hydrological data were available for streams Samples were diluted to 1 mg L-1 DOC and buffered at pH 6.8;
supplying NW1 and NW3, the former collected by the UK sodium hypochlorite was added to provide 5 mg L-1 Cl2 (excess of
Environment Agency and the latter by the Centre for Ecology chlorine). Samples were incubated in capped amber 100-mL bottles
and Hydrology, and water sampling of both was undertaken in the dark at 25°C for 7 d. At the end of the 7-d reaction period,
within a few meters of where these are collected. No flow data sodium sulfite was added to the samples to quench residual chlorine.
was available for the stream supplying NW2. The four THM compounds (CHCl3, CHCl2Br, CHClBr2, and
CHBr3) were determined by gas chromatography using an Agilent
Sampling and Analysis Protocols 6890A Series GC (Agilent Technologies) coupled to a Agilent
At all study sites, a 24-bottle 500-mL autosampler (BR, ISCO 5973 Network mass spectrometer (Agilent Technologies) (BR) or
3700; NW, Hach Lange Xian 1000) was installed on the river a Varian GC-450 utilizing solid phase micro extraction and a 63Ni
or stream bank and programmed to sample at set time intervals. Electron Capture Detector (NW). Identification of THMs was
Sampling was conducted during the spring from midwater depth conducted by retention time matching. The quantification was
(10 cm for the NW sites and 60 cm for the BR site). Rainfall achieved by three point internal calibration. Calibration solutions
events were anticipated on the basis of local weather forecasts, containing the 4 THMs at three concentrations in the range 0.5
and the autosamplers were set to sample for a few hours to 70 mg L-1 were prepared by dilution of a stock solution of 100
before the rainfall event and during the rainfall/flow peak and mg L-1. The limit of quantification (LOQ) was 0.5 mg L-1 for the
the falling limb of the hydrograph. The frequency of sample four species.
collection varied slightly between sites to account for differences
in hydrological response times, i.e., 3 hourly at the NW sites and Data Pretreatment
2 hourly at the BR site. To standardize data analysis between sites, three hydrological
Autosamplers were retrieved within a few hours of the end sampling periods were defined: prestorm (before the onset of
of each sampling period. For the samples collected in Brittany, rain to the beginning of the rising limb), storm (rising limb to
on-site measurements of pH and conductivity were conducted peak flow/stage and rainfall end), poststorm (after peak flow and
with Hach Lange probes, before transportation back to the rainfall end).
laboratory. All measurements on all NW samples were conducted Rainfall was quantified in the form of accumulated rainfall,
in the laboratory: pH was measured on unfiltered samples using a accumulated rainfall between each water sample, and peak. Peak
Mettler Toledo S20 SevenEasy pH meter and conductivity with intensity was defined as the maximum amount of rainfall in any
a Primo 5 conductivity meter. Samples from both sites were then 10-min period between two samplings, expressed in millimeters
filtered using either 0.45-mm HAWP nitrocellulose filter papers per hour. Heavy rainfall was defined as an amount higher to
(Millipore) (BR site) or pre-rinsed Whatman glass-fiber GF/F 10 mm in 24 h on the watersheds studied (Aguilar et al., 2005;
(0.7 mM pore size) filters (NW sites) for the determination of Déqué, 2007). Chemical concentrations below the LOQ were
particulate organic carbon (POC) and then again through pre- replaced by the LOQ divided by the square root of 2 (Bessonneau
rinsed 0.45-mM cellulose nitrate filters to remove all particulate et al., 2011).
matter. All samples were stored at 4°C in the dark until further
analysis. Statistical Analysis
Dissolved organic C was determined for sites in both regions Statistical analyses were performed using GraphPad Prism
by thermal oxidation coupled with infrared detection after the v5.01 for Windows (GraphPad, 2011). The D’Agostino and
removal of inorganic carbon by acidification with hydrochloric Pearson normality test was used to test if variables follow a
acid (ISO, 1999). Particulate organic C was defined as all carbon
Gaussian distribution. Statistical analyses were conducted to The storm monitored at the BR site occurred between 19:00,
find possible correlations between THM formation and the 24 Apr. to 17:00, 26 Apr. 2012 after an extended period of rain
indicator parameters, and to test for significant differences in (83.6 mm from 9 to 23 Apr. 2012). The period of sampling
the distributions of DOC and THM levels between catchments. covered the highest flow experienced at any time during the
Correlations between variables were calculated for the full times antecedent wet period (winter and spring). At its peak, flow
series and within sampling periods by using the nonparametric rate reached 16.8 m3 s-1 (03:00, 26 Apr. 2012). The amount
Spearman’s rank correlation test since the variables were mostly of rain to fall during the monitoring period (25.6 mm) was
not normally distributed. Differences between the rainfall comparable to amounts recorded at the NW sites. Rainfall
patterns across sites and significant changes in water quality intensity was relatively light throughout the monitoring period,
variable concentrations between the three sampling periods and reaching a maximum of 3.6 mm h-1 at 09:00 on 25 Apr. 2012.
between the four sites were assessed by using a one-way ANOVA The relative higher lag observed for the BR site reflects the higher
test followed by a Bonferroni correction. Since no statistical hydrological response time of the larger river basin.
differences in accumulated rainfall between sites were identified, Differences in patterns of flow were even more marked even
rainfall patterns were also assumed to be similar and comparable. though all three streams for which data were available receiving
similar amounts of rainfall. NW3 was the most responsive,
Results showing a rapid increase and return to near-baseflow levels
Rainfalls Characteristics within the period of the study. Flow rates in BR agricultural site
(BR1) showed a more a gradual increase and decrease, possibly
The storm monitored in NW (early May 2011) occurred
because there had been significant rainfall in the preceding
after a relatively long period of very dry weather. In the previous
weeks, while the flow rate at NW1 (upland peatland) was
29 d, rainfall was limited to just 18.6 mm at the upland site and
initially slow to respond but then increased rapidly during the
7.4 mm at the lowland site. During the 69 h of sampling (from
storm and declined slowly once the rain ceased.
21:00, 6 May to 18:00, 9 May 2011) 20.8 mm of rain fell at the
upland site, with a maximum intensity of 5.6 mm h-1 occurring Overall Differences in Water Quality between Sites
at approximately 03:00 on 8 May 2011. The same storm system
A summary of water quality parameter concentrations is
brought rain slightly later to the lowland site, amounting to 24.6
provided in Table 2.
mm, and a maximum intensity of 18.0 mm h-1 at approximately
The highest baseline and median DOC values were measured
04:30 on 8 May 2011. At the upland peat site, stream discharge
at BR1 and NW1 (median respectively 12.2 and 9.7 mg L-1)
peaked at 0.01 m3 s-1 3 h after peak rainfall intensity. For the
while POC values were highest at BR1. The two lowland sites
lowland agricultural site, peak stream flow (3.76 m3 s-1) was
had the highest conductivities and were the most alkaline. NW2
recorded 15 h after peak rainfall.
Table 2. Descriptive summary of water quality parameters concentrations during the monitored rainfall events (N = 24 for each site).
Site
Site† pH Conductivity DOC‡ POC‡ SUVA‡ Bromide THMFP‡
code
mS cm-1 mg L-1 mg L-1 L mg-1 m-1 mg L-1 mg L-1
BR1 Agricultural
Min. 6.9 251.0 9.0 0.2 4.0 0.069 60.6
(BR) 25th percentile 7.2 260.0 10.6 3.5 4.3 0.078 69.7
Median 7.2 265.0 12.2 5.2 4.5 0.083 76.7
75th percentile 7.3 269.5 13.9 12.7 4.6 0.092 80.0
Max. 7.7 275.0 15.6 19.3 5.2 0.097 93.5
NW1 Peatland (NW) Min. 6.7 41.0 7.6 0.7 3.1 0.008 19.3
25th percentile 6.7 42.0 8.7 1.2 3.4 0.010 29.4
Median 6.8 44.5 9.7 1.6 3.7 0.015 46.4
75th percentile 7.0 46.0 13.3 2.7 4.1 0.017 65.0
Max. 7.0 48.0 15.0 3.7 4.2 0.020 87.2
NW2 Mineral (NW) Min. 4.8 34.0 2.1 0.3 2.5 0.010 22.7
25th percentile 5.0 36.0 2.2 0.5 3.2 0.011 28.1
Median 5.3 36.0 3.8 0.7 4.4 0.015 36.8
75th percentile 5.9 37.0 7.0 0.7 4.8 0.019 63.2
Max. 5.9 39.0 10.0 4.1 7.3 0.020 115.4
NW3 Agricultural Min. 7.6 273.0 3.1 NA§ 2.5 0.078 10.6
(NW) 25th percentile 7.7 301.3 4.1 NA 2.7 0.082 30.8
Median 7.8 304.5 6.3 NA 2.9 0.084 49.4
75th percentile 7.8 308.8 8.3 NA 3.2 0.085 58.0
Max. 7.9 317.0 9.2 NA 7.3 0.091 63.1
† Br, Brittany; NW, North Wales.
‡ DOC = dissolved organic carbon; POC = particulate organic carbon; SUVA = specific ultraviolet absorbance; THMFP = trihalomethane formation
potential.
§ NA = not analyzed.
was the most acidic. Bromide concentrations showed a similar Dissolved organic C concentration rose after the onset of rain
range at the two agricultural sites and were at least an order in all four streams. Dissolved organic C concentrations increased
of magnitude higher than the upland Welsh sites. The highest between pre- and poststorm periods by 47.7% for BR1, 51.6% for
median and lowest range of THMFP concentrations were NW1, 101.7% for NW3, and 158.0% for NW2, with all of these
measured at BR1 (median: 76.7 mg L-1 for BR site versus 36.8– increases significant (p < 0.05). The smallest overall proportional
49.4 mg L-1 for NW sites). NW2 showed the largest range and increase in DOC concentration was observed at BR1 and
highest maximum concentrations in THMFP. NW1, which showed the highest prestorm DOC values. The
Analysis of variance revealed that the distribution of both apparent reduction in poststorm DOC concentration, which
DOC and bromodichloromethane concentrations at most sites was observed only at NW2, was not statistically significant. For
were statistically different from each other (p < 0.05) with the this stream, the DOC concentration rose much more sharply
exception of the BR1 and NW3. In contrast, chloroform and from the prestorm to storm (i.e., peak flow) period than the
THMFP distributions did not differ significantly between the other sites, increasing from 2.2 to 7.1 mg L-1, equating to a
three NW sites (NW1, NW2, and NW3) (p > 0.05) but were 217.3% increase, as compared to 20.0, 24.5, and 38.1% in NW3,
significantly lower than BR1 (p < 0.05). BR1, and NW1, respectively. The SUVA-inferred quality of the
The DOC concentration of samples showed strong positive organic matter following rainfall also changed for the upland
correlations with accumulated rainfall (r = 0.70–0.92; p < NW sites, with a greater proportion of aromatic and higher
0.01) and stage height (r = 0.79–0.88; p < 0.01) at all sites. In molecular weight compounds following rainfall, but did not for
contrast, POC showed a significant positive correlation with the two agricultural sites.
stage height and flow rate at BR1 only (r = 0.81, p < 0.01; r = In contrast to the temporal behavior of DOC, concentrations
0.79, p < 0.01, respectively). Particulate organic C was also more of POC were highest during, rather than after, the storm period
weakly correlated with maximum rainfall intensity at NW1 (r for all three streams for which data was collected. Concentrations
= 0.41, p < 0.05), but there was no comparable link between of bromide were higher at the two agricultural sites (BR1 and
maximum rainfall intensity and DOC concentration. Bromide NW3) than at the two upland sites (NW1 and NW2). Dissolved
concentration was significantly negatively correlated with organic carbon concentrations were the only measured chemical
accumulated rainfall (r = −0.75 to -0.90; p < 0.01) and stage parameter to peak during the poststorm period.
height (r = −0.53 to -0.89; p < 0.01) for all sites except NW3. Figure 3 displays variation of total THMs and percentage
brominated species during the three sampling periods. Despite
Patterns of Change in Distributions between the substantial increases in DOC concentrations in response to
Different Sampling Periods rainfall at all sites, a changes in SUVA-inferred DOC quality at
Figure 2 summarizes average water quality parameters (DOC, the upland sites, there were no significant differences in THMFP
POC, specific ultraviolet absorbance [SUVA], and bromide) between the three sampling periods (Fig. 3a) at any site.
and their variation during the three sampling periods covering Overall, the highest THMFP occurred at BR1 (69.7–80.9
prestorm, storm and poststorm periods. mg L-1) during the three sampling periods. At all four sites,
chloroform dominated the THM load, representing 70 to
Fig. 2. Dissolved organic carbon (DOC), particulate organic carbon (POC), specific ultraviolet absorbance (SUVA), and bromide variation (with
standard error) for the three rainfall event time periods at the four studied sites. Differences between rainfall periods for a same site are significant
when letters (a, b or c) are different. BR = Brittany; NW = North Wales.
DOC were found to be higher during the descending limb than
during the rising limb at a given discharge, displaying hysteresis
defined as anticlockwise A2 by Evans and Davies (1998)
when plotted as concentration–discharge graphs (Fig. 4). This
hysteresis behavior has also been reported for a Welsh peatland
watershed (Austnes et al., 2010) and two Korean agricultural-
forested watersheds (Nguyen et al., 2010; Yang et al., 2013).
The greater flow of BR1 resulted in a considerably higher
DOC flux than at the other sites. The DOC flux increased from
280.2 kg h-1 at baseflow to 902.3 kg h-1 at peak flow. For the
other sites, the DOC flux increased from an average of 0.1 kg h-1
at baseflow to 0.4 kg h-1 at peak flow for NW1 stream and from
7.5 kg h-1 to 111.0 kg h-1 for NW3.
To determine the total amount of DOC mobilized as a
consequence of the storm events, we made the assumption
that DOC decreased exponentially during the falling limb
of the hydrograph and extrapolated the trend line back to a
point representing the average prestorm concentration. We
acknowledge that DOC does not always return to prestorm
concentrations so rapidly, as shown in Fig. 4 of this study
and demonstrated in Austnes et al. (2010). This calculation
was performed mainly to demonstrate site differences and to
demonstrate that a single storm event can wash significant
quantities of DOC from soils to freshwaters. We calculated the
area under the line to derive the amount of extra carbon flushed
into the stream. The greatest amount of mobilized carbon was
recorded for BR1, with an extra 20.7 Mg, representing a mean
extra flux of 313.5 kg h-1 over the entire storm period. Values
were lower for the two Welsh sites, at 10.17 kg (0.14 kg h-1)
for NW1 and 2.6 Mg (33.1 kg h-1) for NW3. These differences
largely reflect the contrasting discharge volumes of the streams.
Discussion
This study aimed to compare the effects of heavy spring rainfall
on the water quality of four contrasting northwestern European
streams typical of potable surface water supplies. There clearly are
insufficient numbers of sites to make broad generalizations about
differences between soil types or land uses, but the results reveal a
Fig. 3. (a) Trihalomethane formation potential (THMFP), (b) surprising level of shared behavior from which broad conclusions
percentage of brominated trihalomethane (THM), and (c) bromide/
dissolved organic carbon (DOC) ratio variations (with standard error)
can be drawn.
during the three sampling periods at the four sites. Differences Differences in mean concentrations of DOC and POC
between rainfalls periods for a same site are significant when letters between sites will be due to factors such as soil type, altitude,
(a, b or c) are different. BR = Brittany; NW = North Wales.
land use, and climate. The high DOC concentration of the river
96% of all THM species in all samples. Brominated species in Brittany likely partly reflects the intensive agricultural land
were present in relatively low concentrations (4–30%), with use (crops and pastures, high poultry and dairy animal density)
dichlorobromomethane being the principal brominated THM in the catchment (Agreste DRAAF Bretagne, 2008), which in
species formed at all four sites. NW3, which also had the highest this region of France is known to leach organics and nutrients
bromide (Br–) concentrations, showed a considerably higher into freshwaters ( Jardé et al., 2007). High POC values for this
proportion of brominated species (25–40%) than all other sites. site probably reflect higher erosion in the catchment due to soil
The formation of brominated THMs correlated significantly tillage during the spring sampling period (Veum et al., 2009).
with the Br–/DOC ratio when comparing all sites (Fig. 3b and In contrast, the high DOC concentration of the upland peat
3c) (p < 0.001). draining stream is typical of upland wetland catchments more
widely and likely reflects barriers to mineralization of organic
Dissolved Organic Carbon Temporal Trends and Fluxes matter as a consequence of cool anaerobic conditions and routing
Comparisons through mineral soil horizons where significant adsorption can
occur.
Figure 4 presents the DOC concentration–discharge
The influence of anthropogenic activities within the
relationships for the three sampling sites for which flow data is
agricultural sites (BR1 and NW3) on surface water quality is
available. At all three sites monitored for flow, concentrations of

Fig. 4. Relationships between dissolved organic carbon (DOC) concentration and discharge for the different sites.
clearly shown by the higher bromide concentrations measured behavior during rainfall events has been widely reported in
that could be linked with leaching from agricultural fields and other studies (e.g., Hood et al., 2006; Yang et al., 2013). The
roads or industrial and urban wastewater discharges (Luong et rise and fall of the water table in these types of soils during wet
al., 1983). It may also be explained as an influence of both sites’ and dry periods will lead to highly variable concentrations of
close proximity to the sea, which will elevate the concentration DOC in the streams draining these soils (Tang et al., 2013),
of bromide in the precipitation in these regions. with the drying and rewetting cycles within the organic layer
creating a situation whereby a large pool of potentially mobile
Impacts on Dissolved Organic Carbon C can be generated through the opening of the enzymic latch
Our study reinforces observations made elsewhere that (Freeman et al., 2001b; Fenner and Freeman, 2011). Reservoirs
rainfall events can exert an important impact on water quality whose catchments contain this common soil type may therefore
degradation with respect to concentrations and fluxes of DOC experience short-term pulses in DOC due to flushing of these
and POC, and it demonstrates the greater importance of flow upper organic layers.
energy in determining POC as opposed to DOC concentrations. BR1 had the highest DOC concentration for each of
Dissolved organic C concentrations in our study sites responded the three rainfall period categories. Given that this site’s
differently to the rainfall events, although for the three sites catchment is very poor in naturally organic soils, which often
for which discharge data was available similar concentration– provide the principle source of terrestrial DOC to freshwaters
discharge hysteresis loops were generated. The anticlockwise (e.g., NW1), and that hysteresis plots revealed that the soils
rotation indicates that soil water, rather than surface run-off and were the most important source of DOC, our data suggest that
groundwater, was the most important source of DOC to these the intensive agricultural practices taking place at this site are
three streams (Evans and Davies, 1998). contributing to a significant deterioration in water quality.
Our results demonstrate a relatively prolonged effect of This is also confirmed by the high POC concentrations, which
rainfall on DOC. Higher DOC fluxes during the descending were significantly higher during the storm and poststorm
limb than during the rising limb could be explained by the periods, and BR1 was the only site to exhibit this behavior.
flushing of a large pool size of terrestrial DOM and a relatively The results obtained by Jardé et al. (2007) in small Brittany
small dilution effect due to the storm. The storms generated catchments also suggest a connection between certain types of
substantial levels of precipitation (20.8–25.6 mm) over a short manure spread (dairy and poultry) and the increasing trends
period of time, but these appear to have been insufficient for in organic matter in rivers. Moreover, it has been shown that
dilution to dominate over flow-path related effects. Moreover, soil management practices involving frequent soil mechanical
low soil permeability in the area could lead to the flushing of cultivation, as in the BR1 catchment, lead to higher soil
soil organic horizons during storm events (Birgand et al., 2004). and soil organic C losses, especially during high intensity
Higher DOC concentrations occurring during the poststorm rainfall storms ( Jacinthe et al., 2004). Finally, it should not
period could reveal a strong contribution of riparian wetland be overlooked that the topography and geology of the site,
subsurface soil water during the recession limb of the hydrograph which is dominated by gentle slopes and schist bedrock,
and have also been observed in an another agricultural catchment could favor organic C leaching in the organic horizon during
in Brittany (Morel et al., 2009). rainfall and, consequently, increase DOC concentrations in
The greatest temporal change in DOC concentration was rivers (Birgand et al., 2004).
observed at NW2, the stream draining mineral soils. Dissolved
organic C also fell most rapidly at this site during the falling Impacts on Trihalomethane Formation Potential
limb of the hydrograph. The catchment of NW2 is dominated The main purpose of the study was to investigate how
by steep topography and freely draining acid organic soils. streamwater DOC character, and the propensity of the DOC
Although we are unable to plot concentration–discharge to form DBPs, specifically THMs, responded to heavy rainfall.
hysteresis graphs to infer sources of DOC, the magnitude of The results suggest that despite the substantial increase in
the increase and rapidity of the fall in DOC suggest a change DOC load during the storm, there was no significant change in
in the hydrological flowpath from the deeper, mineral horizons DOC reactivity. Overall, the highest THMFP occurred in BR1
to the upper organic horizon during the peak of rainfall. Such

(69.7–80.9 mg L-1) during the three sampling periods and could (UKWIR, 2011). With respect to water treatment, although
reflect the more intensive agricultural system. we did not observe a significant increase in THMFP at any of
For the three NW sites, the sampled rainfall event occurred the four sites (i.e., per unit DOC reactivity), the impact of an
at the end of a relatively long dry period, thereby maximizing the increase in organic matter flux in a receiving source water body
likelihood of observing changes in DOC character. As shown could lead to an increase in coagulant demand, DBP formation,
in Fig. 3, we did not observe significant differences in THMFP and treatment costs. Moreover, although this study focuses on
between the three event periods for any of the four sites, chlorinated DBPs (i.e., THMs), the presence of nitrogenous
indicating that DOM at baseline and high flow did not differ in DBPs such as haloacetonitriles, halonitromethanes, and
its propensity to form THMs. These results are in contrast to the haloacetamides in drinking water should not be overlooked due
two known studies on rainfall effects on disinfection by-product to their potentially higher cytotoxicity and genotoxicity than
formation potential (DBPFP); Reckhow et al. (2004) published THMs and HAAs (Bond et al., 2011).
results showing reduced HAA formation potential in storm This study has focused specifically on the impact of significant
waters, and Nguyen et al. (2013) observed greater THMFP spring rainfall events on northwestern European raw and
of storm and poststorm compared with baseline samples, chlorinated water quality. However, the quality and availability of
particularly for the most intense of the sampled storms. The soluble organic matter is known to vary seasonally, generally being
latter study also reported correlations between DBPFP and highest in autumn, and further work is necessary to determine
measurements of DOM characteristics, such as fluorescence and the extent to which our observations may apply at other times
UV absorbance, and concluded that the export of terrestrially of year. For example, high THM levels could also be measured
derived DOM was responsible for increased DBP formation. during extended summer dry periods, especially during algal
In this study, we observed for the three NW sites significantly blooms. Algae have been recognized as an important precursor
higher SUVA for either storm or poststorm samples compared for DBP production and can correlate with THM production
with baseflow, indicating a shift in character to more terrestrial- (Chen et al., 2008). Moreover, summer low rainfall periods
like DOM, but this was not accompanied by a change in could shift organic matter composition from hydrophobic to
THMFP. We must therefore reject our original hypothesis that hydrophilic predominant fraction (Sharp et al., 2006), which are
rainfall flushed DOM compounds with higher DBP reactivity. more difficult to remove by common water treatment processes
Specific UV absorbance has traditionally been associated with such as coagulation (Matilainen et al., 2010). In this study, we did
THMFP, as compounds that absorb UV light, such as aromatics, not include the effects of temperature due to the methodological
often have high reactivity with chlorine. However, the literature constraints inherent to the THMFP test (constant temperature
is mixed regarding correlations between SUVA and DBPFP, of incubation). Further studies should assess the influence of
with Reckhow et al. (1990) and Singer (1999) reporting strong temperature as it is shown to have significant influence on THM
positive relationships but Weishaar et al. (2003) and White et al. formation (Rodriguez et al., 2003). As an example, simulated
(2003) concluding that SUVA is a weak predictor of THMFP distribution systems tests could be used to assess its effect on
due to the reactivity of nonaromatic compounds. waters distributed to the consumers.
The absence of any increase in THMFP at BR1 during the
storm and poststorm periods may stem at least partly from an Conclusions
abundance of colloidal material in the water column before the This study, in which a consistent sampling protocol was
rainfall event (Alkhatib and Peters, 2008) as a consequence of the applied to rivers draining contrasting soil types, confirms the
wet weather before the sampling event and slow colloidal settling importance of rainfall events on DOC concentrations and
rates. Nevertheless, it seems that the extra organic matter leaching quality across a wide range of environments (agricultural,
into the river during the rainfall event did not have propensity to mineral, peatland). A long-lasting effect of rainfall on DOC
form more THMs. The lack of significant effects for the NW was observed to persist several hours after the rainfall ended,
sites, which experienced dry weather in the preceding weeks, is whereas POC behavior appeared to be highly controlled
more difficult to explain. Despite the evidence for a change in by rainfall flow events. Conversely, the study indicates that
DOC character from pre- to poststorm, there is no indication of such events may not alter DOC reactivity, specifically THM
a change in DOC reactivity with chlorine. formation and speciation, as has been reported in a small
Although the THMFP results obtained in this study shows number of other studies, at least during the spring period of
that there should be no change in the efficiency of treatment, it investigation. Additionally, the Br–/DOC ratio was found to
should be noted that we observed an increase in THM load in all be useful in predicting the percentage of brominated THM
sites (due to increased DOC). This rainfall effect on THM load in chlorinated waters. Nevertheless, the overall increase in
is highly relevant for safe drinking water production. organic matter during such events will inevitably lead to overall
Complementary discussion on the factors influencing increases in treatment costs, micropollutants, and bacterial
THM levels monitored in the study can be found in the online regrowth within the distribution system. Research should
supplemental material. be conducted to assess the impact on micropollutants and
Potential future increases in the intensity and frequency of pathogen contamination of the resources at the scale event.
heavy rainfall events are likely to exert more wide-reaching effects Studies should also be conducted to identify the wide range
than have been discussed above. These could include a possible of chlorination and other disinfection by-products produced
rise in heavy metals following remobilization of sediments and during rainfall events, together with the potential for microbial
an increase in particle bound pesticides during autumn events survival after treatment processes.

Acknowledgments Fenner, N., and C. Freeman. 2011. Drought-induced carbon loss in peatlands.
Nat. Geosci. 4(12):895–900. doi:10.1038/ngeo1323
This research was funded by the French National Agency of Sanitary Freeman, C., C.D. Evans, D.T. Monteith, B. Reynolds, and N. Fenner.
Security (ANSES) and the Natural Environment Research Council 2001a. Export of organic carbon from peat soils. Nature 412:785.
(NERC) under the first ERA-EnvHealth call (FP7-ENV-2007- doi:10.1038/35090628
CSA-1.2.3-01). The authors are grateful to Olivier Nauleau (Banque Freeman, C., N. Fenner, N.J. Ostle, H. Kang, D.J. Dowrick, B. Reynolds, M.A.
Hydro) and the French meteorological service Météo France for Lock, D. Sleep, S. Hughes, and J. Hudson. 2004. Export of dissolved
providing hydrological and climate data, respectively, for the Brittany organic carbon from peatlands under elevated carbon dioxide levels.
site, to the UK Meteorological Office for providing climate data for Nature 430:195–198. doi:10.1038/nature02707
the UK sites, and to the Centre for Ecology and Hydrology (CEH) Freeman, C., G.B. Nevison, S. Hughes, B. Reynolds, and J.A. Hudson. 1998.
and Environment Agency for providing stream discharge data. We are Enzymic involvement in the biogeochemical responses of a Welsh peatland
to a rainfall enhancement manipulation. Biol. Fertil. Soils 27:173–178.
also grateful to David Norris at CEH Bangor for collating GIS data doi:10.1007/s003740050417
for the UK sites. Freeman, C., N.J. Ostle, and H. Kang. 2001b. An enzymic “latch” on a global
carbon store. Nature 409:149. doi:10.1038/35051650
References Gascuel-Odoux, C., P. Aurousseau, P. Durand, L. Ruiz, and J. Molenat. 2010.
Agreste DRAAF Bretagne. 2008. Agricultural census. Directions régionales de The role of climate on inter-annual variation in stream nitrate fluxes and
l’Alimentation, de l’Agriculture et de la Forêt, France concentrations. Sci. Total Environ. 408(23):5657–5666. doi:10.1016/j.
scitotenv.2009.05.003
Aguilar, E., T.C. Peterson, P. Ramírez Obando, R. Frutos, J.A. Retana, M. Solera,
J. Soley, I. González García, R.M. Araujo, A. Rosa Santos, V.E. Valle, M. Goslan, E.H., D.A. Fearing, J. Banks, D. Wilson, P. Hills, A.T. Campbell, and S.A.
Brunet, L. Aguilar, L. Álvarez, M. Bautista, C. Castañón, L. Herrera, E. Parsons. 2002. Seasonal variations in the disinfection by-product precursor
Ruano, J.J. Sinay, E. Sánchez, G.I. Hernández Oviedo, F. Obed, J.E. Salgado, profile of a reservoir water. J. Water Supply Res. Technol. AQUA 51:475–482.
J.L. Vázquez, M. Baca, M. Gutiérrez, C. Centella, J. Espinosa, D. Martínez, GraphPad. 2011. GraphPad Prism v5.01 for Windows. GraphPad Software Inc.,
B. Olmedo, C.E. Ojeda Espinoza, R. Núñez, M. Haylock, H. Benavides, San Diego, CA.
and R. Mayorga. 2005. Changes in precipitation and temperature extremes Hinton, M.J., S.L. Schiff, and M.C. English. 1997. The significance of storms
in Central America and northern South America, 1961–2003. J. Geophys. for the concentration and export of dissolved organic carbon from
Res. 110:D23107. doi:10.1029/2005JD006119 two Precambrian Shield catchments. Biogeochemistry 36:67–88.
Alkhatib, E., and R. Peters. 2008. Wet weather impact on trihalomethane doi:10.1023/A:1005779711821
formation potential in tributaries to drinking water reservoirs. Environ. Hong, H., L. Yang, W. Guo, F. Wang, and X. Yu. 2012. Characterization of
Monit. Assess. 139(1–3):173–181. doi:10.1007/s10661-007-9823-y dissolved organic matter under contrasting hydrologic regimes in a
APHA, AWWA, WEF. 1992. Standard methods for the examination of subtropical watershed using PARAFAC model. Biogeochemistry 109(1–
water and wastewater. 18th ed. American Public Health Association, 3):163–174. doi:10.1007/s10533-011-9617-8
American Water Works Association, and Water Environment Federation, Hood, E., M.N. Gooseff, and S.L. Johnson. 2006. Changes in the
Washington, DC. character of stream water dissolved organic carbon during flushing
Austnes, K., C.D. Evans, C. Eliot-Laize, P.S. Naden, and G.H. Old. 2010. Effects in three small watersheds, Oregon. J. Geophys. Res. 111(G1):1–8.
of storm events on mobilisation and in-stream processing of dissolved doi:10.1029/2005JG000082
organic matter (DOM) in a Welsh peatland catchment. Biogeochemistry Hughes, D.D., P.J. Holliman, T. Jones, and C. Freeman. 2013. Temporal variations
99(1–3):157–173. doi:10.1007/s10533-009-9399-4 in dissolved organic carbon concentrations in upland and lowland lakes in
Bates, B., Z.W. Kundzewicz, S. Wu, and J. Palutikof, editors. 2008. Climate North Wales. Water Environ. J. 27:275–283. doi:10.1111/wej.12025
change and water. Technical Paper of the Intergovernmental Panel on Hurst, A.M., M.J. Edwards, M. Chipps, B. Jefferson, and S.A. Parsons. 2004. The
Climate Change. IPCC Secretariat, Geneva. impact of rainstorm events on coagulation and clarifier performance in
Bessonneau, V., M. Derbez, M. Clément, and O. Thomas. 2011. Determinants of potable water treatment. Sci. Total Environ. 321:219–230. doi:10.1016/j.
chlorination by-products in indoor swimming pools. Int. J. Hyg. Environ. scitotenv.2003.08.016
Health 215:76–85. doi:10.1016/j.ijheh.2011.07.009 IPCC. 2013. Climate change 2013: The physical science basis. Contribution of
Birgand, F., E. Novince, T. Bioteau, and G. Gruau. 2004. Facteurs expliquant Working Group I to the Fifth Assessment Report of the Intergovernmental
la présence de matière organique dans les eaux superficielles en Bretagne: Panel on Climate Change. Cambridge University Press, New York.
Analyse des données existantes. Report DRASS et Région Bretagne. ISO. 1999. ISO 8245:1999: Water quality: Guidelines for the determination
Bond, T., J. Huang, M.R. Templeton, and N. Graham. 2011. Occurrence and of total organic carbon (TOC) and dissolved organic carbon (DOC).
control of nitrogenous disinfection by-products in drinking water: A International Organization for Standardization, Geneva.
review. Water Res. 45(15):4341–4354. doi:10.1016/j.watres.2011.05.034 Jacinthe, P.-A., R. Lal, L.B. Owens, and D.L. Hothem. 2004. Transport of labile
Chen, C., X.-J. Zhang, L.-X. Zhu, J. Liu, W.-J. He, and H. Han. 2008. carbon in runoff as affected by land use and rainfall characteristics. Soil
Disinfection by-products and their precursors in a water treatment plant in Tillage Res. 77:111–123. doi:10.1016/j.still.2003.11.004
North China: Seasonal changes and fraction analysis. Sci. Total Environ. Jardé, E., G. Gruau, and L. Mansuy-Huault. 2007. Detection of manure-derived organic
397:140–147. doi:10.1016/j.scitotenv.2008.02.032 compounds in rivers draining agricultural areas of intensive manure spreading.
Clark, J.M., S.N. Lane, P.J. Chapman, and J.K. Adamson. 2007. Export Appl. Geochem. 22:1814–1824. doi:10.1016/j.apgeochem.2007.03.037
of dissolved organic carbon from an upland peatland during storm Jones, T.G. 2006. Climate change and dissolved organic carbon: Impacts on
events: Implications for flux estimates. J. Hydrol. 347(3-4):438–447. drinking water supplies. Ph.D. thesis, University of Wales, Bangor, UK.
doi:10.1016/j.jhydrol.2007.09.030 Klein Tank, A.M.G., J.B. Wijngaard, G.P. Können, R. Böhm, G. Demarée,
Curriero, F.C., J.A. Patz, J.B. Rose, and S. Lele. 2001. The association between A. Gocheva, M. Mileta, S. Pashiardis, L. Hejkrlik, C. Kern-Hansen, R.
extreme precipitation and waterborne disease outbreaks in the United Heino, P. Bessemoulin, G. Müller-Westermeier, M. Tzanakou, S. Szalai, T.
States, 1948–1994. Am. J. Public Health 91(8):1194–1199. doi:10.2105/ Palsdottir, D. Fitzgerald, S. Rubin, M. Capaldo, M. Maugeri, A. Leitass,
AJPH.91.8.1194 A. Bukantis, R. Aberfeld, A.F.V. van Engelen, E. Forland, M. Mietus, F.
Déqué, M. 2007. Frequency of precipitation and temperature extremes over Coelho, C. Mares, V. Razuvaev, E. Nieplova, T. Cegnar, J.A. Lopez, B.
France in an anthropogenic scenario: Model results and statistical Dahlström, A. Moberg, W. Kirchhofer, A. Ceylan, O. Pachaliuk, L.V.
correction according to observed values. Global Planet. Change 57(1– Alexander, and P. Petrovic. 2002. Daily dataset of 20th-century surface air
2):16–26. doi:10.1016/j.gloplacha.2006.11.030 temperature and precipitation series for the European Climate Assessment.
Dillon, P.J., and L.A. Molot. 1997. Effect of landscape form on export of dissolved Int. J. Climatol. 22:1441–1453. doi:10.1002/joc.773
organic carbon, iron, and phosphorus from forested stream catchments. Krasner, S.W., H.S. Weinberg, S.D. Richardson, S.J. Pastor, R. Chinn, M.J.
Water Resour. Res. 33:2591–2600. doi:10.1029/97WR01921 Sclimenti, G.D. Onstad, and A.D. Thruston. 2006. Occurrence of a new
Evans, C., and T.D. Davies. 1998. Causes of concentration/discharge hysteresis generation of disinfection byproducts. Environ. Sci. Technol. 40:7175–
and its potential as a tool for analysis of episode hydrochemistry. Water 7185. doi:10.1021/es060353j
Resour. Res. 34(1):129–137. doi:10.1029/97WR01881 Luong, T.V., C.J. Peters, and R. Perry. 1983. Occurrence of bromide in source and
Evans, C.D., T.G. Jones, A. Burden, N. Ostle, P. Zieliński, M.D.A. Cooper, M. treated waters. Effluent Water Treat. J. 23:192–198.
Peacock, J.M. Clark, F. Oulehle, D. Cooper, and C. Freeman. 2012. Acidity Matilainen, A., and M. Sillanpää. 2010. Removal of natural organic matter from
controls on dissolved organic carbon mobility in organic soils. Glob. drinking water by advanced oxidation processes. Chemosphere 80:351–
Change Biol. 18(11):3317–3331. doi:10.1111/j.1365-2486.2012.02794.x 365. doi:10.1016/j.chemosphere.2010.04.067

Matilainen, A., M. Vepsäläinen, and M. Sillanpää. 2010. Natural organic matter Reckhow, D.A., P.C. Singer, and R.L. Malcolm. 1990. Chlorination of humic
removal by coagulation during drinking water treatment: A review. Adv. materials: Byproduct formation and chemical interpretations. Environ.
Colloid Interface Sci. 159:189–197. doi:10.1016/j.cis.2010.06.007 Sci. Technol. 24:1655–1664. doi:10.1021/es00081a005
Milne, R., and T.A. Brown. 1997. Carbon in the vegetation and soils of Great Ritson, J.P., N.J.D. Graham, M.R. Templeton, J.M. Clark, R. Gough, and C.
Britain. J. Environ. Manage. 49:413–433. doi:10.1006/jema.1995.0118 Freeman. 2014. The impact of climate change on the treatability of dissolved
Molénat, J., P. Durand, C. Gascuel-Odoux, P. Davy, and G. Gruau. 2002. organic matter (DOM) in upland water supplies: A UK perspective. Sci.
Mechanisms of nitrate transfer from soil to stream in an agricultural Total Environ. 473-474:714–730. doi:10.1016/j.scitotenv.2013.12.095
watershed of French Brittany. Water Air Soil Pollut. 133:161–183. Rodriguez, M.J., Y. Vinette, J.-B. Sérodes, and C. Bouchard. 2003.
doi:10.1023/A:1012903626192 Trihalomethanes in drinking water of greater Quebec region (Canada):
Monteith, D.T., P.A. Henrys, C.D. Evans, I. Malcolm, E.M. Shilland, and M.G. Occurrence, variations and modelling. Environ. Monit. Assess. 89:69–93.
Pereira. 2015. Spatial controls on dissolved organic carbon in upland doi:10.1023/A:1025811921502
waters inferred from a simple statistical model. Biogeochemistry 123:363– Sharp, E.L., S.A. Parsons, and B. Jefferson. 2006. The impact of seasonal variations
377. doi:10.1007/s10533-015-0071-x in DOC arising from a moorland peat catchment on coagulation with
Monteith, D.T., J.L. Stoddard, C.D. Evans, H.A. de Wit, M. Forsius, T. Høgåsen, iron and aluminium salts. Environ. Pollut. 140:436–443. doi:10.1016/j.
A. Wilander, B.L. Skjelkvåle, D.S. Jeffries, J. Vuorenmaa, B. Keller, J. envpol.2005.08.001
Kopácek, and J. Vesely. 2007. Dissolved organic carbon trends resulting Singer, P.C. 1999. Humic substances as precursors for potentially harmful
from changes in atmospheric deposition chemistry. Nature 450:537–540. disinfection by-products. Water Sci. Technol. 40:25–30. doi:10.1016/
doi:10.1038/nature06316 S0273-1223(99)00636-8
Morel, B., P. Durand, A. Jaffrezic, G. Gruau, and J. Molénat. 2009. Sources of Tang, R., J.M. Clark, T. Bond, N. Graham, D. Hughes, and C. Freeman.
dissolved organic carbon during stormflow in a headwater agricultural 2013. Assessment of potential climate change impacts on peatland
catchment. Hydrol. Proc. 23(20):2888–2901. doi:10.1002/hyp.7379 dissolved organic carbon release and drinking water treatment from
Neal, C., B. Reynolds, P. Rowland, D. Norris, J.W. Kirchner, M. Neal, D. Sleep, laboratory experiments. Environ. Pollut. 173:270–277. doi:10.1016/j.
A. Lawlor, C. Woods, S. Thacker, H. Guyatt, C. Vincent, K. Hockenhull, envpol.2012.09.022
H. Wickham, S. Harman, and L. Armstrong. 2012. High-frequency water UKWIR. 2011. Climate change implications for water treatment. Vol. 1.
quality time series in precipitation and streamflow: From fragmentary Overview report (11/CL/08/2). UK Water Industry Research, London.
signals to scientific challenge. Sci. Total Environ. 434:3–12. doi:10.1016/j. Urban, N.R., S.E. Bayley, and S.J. Eisenreich. 1989. Export of dissolved organic
scitotenv.2011.10.072 carbon and acidity from peatlands. Water Resour. Res. 25:1619–1628.
Newell-Price, P., and M.-L. Daumer. 2006. A case study: Adoption of best doi:10.1029/WR025i007p01619
management practice in Brittany (France) using economic instruments and Veum, K.S., K.W. Goyne, P.P. Motavalli, and R.P. Udawatta. 2009. Runoff and
regulation. In: L. Gairns, K. Crighton, and B. Jeffrey, editors, Proceedings dissolved organic carbon loss from a paired-watershed study of three
of the SAC and SEPA Biennial Conference, 5–6 Apr. 2006. Edinburgh. adjacent agricultural watersheds. Agric. Ecosyst. Environ. 130:115–122.
NF EN ISO. 2009. NF EN ISO 10304-1: Qualité de l’eau- Dosage des anions doi:10.1016/j.agee.2008.12.006
dissous par chromatographie des ions en phase liquide—Partie 1: Dosage Weishaar, J.L., G.R. Aiken, B.A. Bergamaschi, M.S. Fram, R. Fujii, and K.
du bromure, chlorure, fluorure, nitrate, nitrite, phosphate et sulfate. Mopper. 2003. Evaluation of specific ultraviolet absorbance as an indicator
Nguyen, H.V.-M., J. Hur, and H.-S. Shin. 2010. Changes in spectroscopic and of the chemical composition and reactivity of dissolved organic carbon.
molecular weight characteristics of dissolved organic matter in a river Environ. Sci. Technol. 37:4702–4708. doi:10.1021/es030360x
during a storm event. Water Air Soil Pollut. 212:395–406. doi:10.1007/ White, D.M., D.S. Garland, J. Narr, and C.R. Woolard. 2003. Natural organic
s11270-010-0353-9 matter and DBP formation potential in Alaskan water supplies. Water Res.
Nguyen, H.V.-M., M.-H. Lee, J. Hur, and M.A. Schlautman. 2013. Variations 37:939–947. doi:10.1016/S0043-1354(02)00425-6
in spectroscopic characteristics and disinfection byproduct formation Yang, L., W. Guo, N. Chen, H. Hong, J. Huang, J. Xu, and S. Huang. 2013.
potentials of dissolved organic matter for two contrasting storm events. J. Influence of a summer storm event on the flux and composition of dissolved
Hydrol. 481:132–142. doi:10.1016/j.jhydrol.2012.12.044 organic matter in a subtropical river, China. Appl. Geochem. 28:164–171.
Raymond, P.A., and J.E. Saiers. 2010. Event controlled DOC export from doi:10.1016/j.apgeochem.2012.10.004
forested watersheds. Biogeochemistry 100:197–209. doi:10.1007/
s10533-010-9416-7
Reckhow, D.A., P.L.S. Rees, and D. Bryan. 2004. Watershed sources of
disinfection byproduct precursors. Water Sci. Technol. 4:61–69.

HDR Version Finale COMPLET

Transféré par

Droits d'auteur :

Formats disponibles

HDR Version Finale COMPLET

Transféré par

Informations du document

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

HDR Version Finale COMPLET

Transféré par

Droits d'auteur :

Formats disponibles

Habilitation à diriger des recherches

CARACTERISATION ET FACTEURS DE DEGRADATION DE LA QUALITE DE

Chercheure associée LERES (EHESP‐IRSET U1085 INSERM, Rennes)

Soutenue le 12 décembre 2016 devant la commission d’examen :

Dr. Hélène Budzinski, Université Bordeaux 1, EPOC ‐ UMR 5805 CNRS

ECOLE DES METIERS DE L’ENVIRONNEMENT

Campus de Ker Lann – Tél : +33(0)2 99 05 88 00 Association loi 1901

ANTOINE DE SAINT‐ EXUPERY

Pr LINUS PAULING, Prix Nobel de chimie et Prix Nobel de la paix

J’adresse mes remerciements à Jean‐Luc Bersillon, ancien Professeur au Laboratoire

Merci à Olivier Thomas, Professeur émérite au Laboratoire d’Etudes et de Recherche en

Je remercie vivement Geneviève Feuillade, Professeure à l’Ecole Nationale Supérieure

En 2009, je retourne à la recherche plus académique et je deviens chercheure associée

2015 Comprendre la santé publique et le système de santé, MOOC EHESP

2014 IED: quelles obligations pour les ICPE ? CCI, Paris

Qualifications aux fonctions de Maître de Conférences

Activités d’enseignement et ingénierie pédagogique

Activités d’enseignements antérieures données dans d’autres établissements:

> Encadrement de projets d’ingéniérie (5ème année cycle ingénieur EME) :

> Encadrement de projets bibliographiques, étudiants 3ème année cycle ingénieur

 NACRE (NAvire Conduit pour le Respect de l’Environnement), Pôle Mer Bretagne,

Formation de personnel hautement qualifié

Co‐encadrante thèse Jean Causse (2012‐2015) Impact de facteurs émergents sur

JUNG A.‐V., 2014, témoignage « la femme ingénieure au travers de la formation à l’EME »,

JUNG A.‐V., 2008, Réglementation et méthodologie d’évaluation des risques chimiques,

JUNG A.‐V., 2002, Interactions de la matière organique avec les micropolluants

Rapports de contrats de recherche, contrats avec le monde socio‐économique et

Co‐organisatrice et participante à la Journée Technique du Réseau Coopératif de

Depuis 2007 : participation à des réunions du Collège des Hautes Études du

‐ Membre du conseil de laboratoire du LERES

‐ Animation scientifique journée de la Vilaine (Thèmatique 2014 : chlorure de

Réponses à appels à projet

‐ ANR Eco‐TS, 2013 : Projet ISIS Identification des Situations à rISque de

Nom du projet/ année Enjeux Travaux étudiants

Le défi de la qualité de l’eau et des milieux aquatiques est une préoccupation

Mes travaux de recherche s’inscrivent dans une problématique de

- Etude de l’interaction « matière organique – micropolluants » : les travaux

- Caractérisation et facteurs de dégradation de la qualité de l’eau/recherche

- Evaluation environnementale et prévention des risques sanitaires :

Figure 1 : synoptique et synergie entre les 3 axes de recherche développés

Cette thématique a été développée lors de la thèse de doctorat (2001‐2004),

Parmi l’ensemble des molécules naturelles du milieu aquatique, il existe une

Pour répondre à ces questions, une approche utilisant différentes techniques

1.1. Synthèse et caractérisation de composés humiques modèles –

Des acides humiques de synthèse de différentes tailles moléculaires, obtenus par

Figure 2: Représentation des matrices de fluorescence (excitation‐émission‐intensité de

Cette étude, par un couplage multi‐techniques (analyses élémentaires, pyrolyse

La voie de synthèse utilisée des substances « modèles » est donc intéressante et

1.2. Etude de l’organisation des substances humiques en solution aqueuse à

L’utilisation de sondes de fluorescence a essayé de répondre à la question du

Cette étude nous a permis d’appréhender les difficultés de l’utilisation de

La question de la pertinence de l’utilisation éventuelle de sondes avec les

1.3. Etude des interactions entre la matière organique et les polluants

La mise en évidence d’interactions avec plusieurs familles de polluants types

1.3.1 Interaction matière organique ‐ pyrène

Tout d’abord, l’interaction entre la matière organique et le pyrène a été regardée

Avec I0 et I les intensités de fluorescence du pyrène (excitation 332nm, émission

Les mesures simultanées de tension de surface ont permis de déterminer les

Cette étude a montré l’existence de 2 principaux domaines de concentration en matière