Continental Shelf Research

Continental Shelf Research ∎ (∎∎∎∎) ∎∎∎–∎∎∎
1 Contents lists available at ScienceDirect

2
3
4 Continental Shelf Research
5
6
journal homepage: www.elsevier.com/locate/csr
7
8
9
10
11
12
Methods of eutrophication assessment in the context of the water framework
13 directive: Examples from the Eastern Mediterranean coastal areas
14
15 Alexandra Pavlidou a,n, Nomiki Simboura a, Eleni Rousselaki a, Manolis Tsapakis b, Kalliopi Pagou a,
16
Q1
17
Paraskevi Drakopoulou a, Georgia Assimakopoulou a, Harilaos Kontoyiannis a, Panayotis Panayotidis a
18 a
Institute of Oceanography, Hellenic Centre for Marine Research, 46.7 km Athens-Sounio Avenue, Mavro Lithari, PO Box 712, 19013 Anavissos, Attika, Greece
b
19 Institute of Oceanography, Hellenic Centre for Marine Research, PO Box 2214, 71003 Heraklion, Greece
20
21
22 art ic l e i nf o a b s t r a c t
23
24 Article history: A set of methodological tools were tested in order to assess the eutrophication quality of selected coastal
Received 15 September 2014 areas in Eastern Mediterranean Sea, Greece, in the context of the Water Framework Directive under
25
Received in revised form various anthropogenic pressures. Three, five-step tools, namely, TRIX, chlorophyll-a (chl-a) biomass
26 22 May 2015
classification scheme, and eutrophication index (E.I.) were applied in oligotrophic waters for (a) the
27 Accepted 30 May 2015
whole water column and (b) the euphotic zone. The relationship among the eutrophication assessment
28
indices and the ecological quality status (EcoQ) assessment indices for benthic macroinvertebrates
29 Keywords: (BENTIX index) and macroalgae (ecological evaluation index-EEIc) was also explored. Agricultural ac-
30 Eutrophication index tivities and mariculture are the pressures mostly related to the eutrophication assessment of the selected
31 TRIX
Greek coastal water bodies. Chl-a proved to be the criterion with the best overall correlation with the
Chlorophyll-a
32 EcoQ indices, while TRIX with the lowest. Moreover, among the eutrophication indices, E.I. showed better
Biological indices
33 Class agreement overall agreement with BENTIX showing that probably it reflects the indirect relation of macro-
34 invertebrates with water eutrophication in a better way. Among the eutrophication indices used, TRIX
35 rather overestimated the eutrophication status of the selected coastal areas. The first stage of eu-
36 trophication was reflected more efficiently using E.I. than TRIX, but E.I. seems to be a rather sensitive
37 index. A future modification of the high to good boundary of E.I. may be needed in order to demonstrate
the high status of the relatively undisturbed Greek coastal sites.
38
& 2015 Elsevier Ltd. All rights reserved.
39
40
67
41
68
42
1. Introduction (2010) presents a system of applying the experience gained from 69
43
the WFD to implement the MSFD. This work outlines the points of 70
44
Coastal marine environments are usually influenced by human- overlap and conflict between the two directives and is regarding 71
45
induced and natural pressures, which may alter their functioning, the WFD as a ‘deconstructing structural approach’ whereas the 72
46
and finally contribute to ecosystem degradation and pollution MSFD is a ‘holistic functional approach’, i.e. the WFD has split the 73
47
ecosystem into several biological quality elements (BQEs) and 74
48 problems (Jickells, 2005; Aubry and Elliott, 2006; Borja et al.,
evaluates them individually before combining them and at- 75
49 2010). The legislation developed and applied worldwide includes
tempting to determine the overall condition. The other elements 76
50 restoration of degraded aquatic habitats as one of the primary
(hydromorphological and physicochemical) are only used to sup- 77
51 goals and require suitable methods to assess their quality in re-
port the BQEs. 78
52 lation to anthropogenic impacts on marine ecosystems using
In contrast the MSFD focuses on the set of 11 descriptors with 79
53 various elements of the ecosystem (Borja et al., 2008, 2011, 2012;
several indicators covering the ecological, physical, chemical and 80
54 Ferreira et al., 2011). In Europe, the umbrella regulations for ad-
anthropogenic components of the ecosystem that need to be in- 81
55 dressing the ecological quality of the water systems are the Water
tegrated at the indicator and descriptor levels (Van Hoey et al., 82
56 Framework Directive (WFD; 2000/60/EC), for lakes, rivers, transi-
2010). These 11 descriptors together summarize the way in which 83
57 tional (estuaries and lagoons) and coastal waters, and the Marine
the whole system functions. Moreover, MSFD has established a 84
58 Strategy Framework Directive (MSFD; 2008/56/EC) for all marine
framework for the development of strategies designed to achieve 85
59
waters (Van Hoey et al., 2010). An informative work by Borja et al., good environmental status (GEnS), which takes into account the 86
60
structure, function, and processes of the marine ecosystems to- 87
61
n gether with natural physiographic, geographic, and climatic fac- 88
62 Corresponding author. Fax: þ 30 2291076347.
E-mail address: aleka@hcmr.gr (A. Pavlidou). tors, as well as physicochemical conditions including those 89
63
90
64 http://dx.doi.org/10.1016/j.csr.2015.05.013 91
65 0278-4343/& 2015 Elsevier Ltd. All rights reserved.
92
66
Please cite this article as: Pavlidou, A., et al., Methods of eutrophication assessment in the context of the water framework directive:
Examples from the Eastern Mediterranean coastal areas. Continental Shelf Research (2015), http://dx.doi.org/10.1016/j.csr.2015.05.013i
2 A. Pavlidou et al. / Continental Shelf Research ∎ (∎∎∎∎) ∎∎∎–∎∎∎
1 resulting from human activities in the area concerned (Borja and forces such as urbanization, industrialization, changes in land use, 67
2 Collins, 2009). tourism development, aquaculture development, climate change, 68
3 The ecological quality status (EcoQ) within the WFD and the and others (Pavlidou and Souvermezoglou, 2006; UNEP, 2007; 69
4 environmental status (ES) within the MSFD should be harmonized Halpern et al., 2007; Karydis and Kitsiou, 2013; Aguilar et al., 2014; 70
5 and the two directives should be fully and seamlessly integrated Levin et al., 2014; Newton et al., 2014). In Greece, more than 80% of 71
6 (Borja et al., 2010). To this respect, intercalibrated indices which the industrial activities and 90% of tourism activities are located 72
7 are used under the WFD can also assess the ES within the MSFD in along the coastline (Anagnostou et al., 2005). Athens and Thes- 73
8 the respective interlinked criteria or indicators. Such an approach saloniki, the two biggest cities of Greece, exceeding 4 and 1 million 74
9 addressing interlinking quality elements’ indicators and de- people, respectively, are also located on the coastal zone and in- 75
10 scriptors of both directives has been applied in several cases i.e. 76
fluence the functioning of Saronikos and Thermaikos ecosystems,
11 the Basque country (Borja et al., 2011) and Greece (Simboura et al., 77
in the central and northwestern Aegean Sea, respectively (Ana-
12 2015). The intercalibrated and interlinked indicators used in the 78
gnostou et al., 2005; Karageorgis et al., 2005; Konstantinou et al.,
13 present work are the chlorophyll-a (chl-a) biomass (pertaining to 79
2012; Pavlidou et al., 2014).
14 eutrophication descriptor 5) and the benthic and macroalgae in- 80
This study was conducted in selected coastal areas of Greece,
15 dices (pertaining to biodiversity and sea floor integrity descriptors, 81
16 namely, D1 and D6, respectively). influenced by the human activities and which are subjected to 82
17 Regarding the eutrophication, the WFD intends to improve the different types of pressure. The eutrophication status of these 83
18 ecological status, including eutrophication status, of all European areas was studied using different indicators and different metho- 84
19 surface waters of which many are considered to be eutrophic dological approaches in the context of WFD. The WFD tries to 85
20 (European Environment Agency, 2001, 2003). However, according combine both aspects of pressures and biological elements (water 86
21 to Andersen et al. (2006), the WFD does not explicitly consider and benthic) into a sole ecological status. In these terms, results of 87
22 eutrophication because the treatment of eutrophication is indirect the eutrophication status of the coastal water bodies were juxta- 88
23 with the boundary between good and moderate ecological status posed and compared to the benthic indices results. 89
24 being defined as an environmental management objective. Con- The objectives of this work are (i) to identify, evaluate, and map 90
25 sequently, the need for a common understanding and definition of the different types of pressures affecting each area; (ii) to assess 91
26 eutrophication, as well as, the need for stronger coordination be- the eutrophication status of the studied coastal areas based on 92
27 tween directives dealing directly or indirectly with eutrophication three different assessment principles and methods usually applied 93
28 has been emerged (Andersen et al., 2006; Ferreira et al., 2011). It is in Greek ecosystems; (iii) to compare the resulting classifications 94
29 important to point out that the WFD is a dynamic directive and and evaluate them; (iv) to compare the eutrophication status with 95
30 permits further incorporation of new methodologies, or im- the benthic status of the coastal areas and investigate whether 96
31 provements of those already applied (Revilla et al., 2009). On the there is a good link between them or not. 97
32 other hand, MSFD takes a functional approach to eutrophication 98
33 establishing it as one of the 11 holistic quality descriptors, namely, 99
34 descriptor 5 (MSFD; 2008/56/EC; Ferreira et al., 2011). This is im- 100
35 portant because eutrophication problems have been reported from 2. Materials and methods 101
36 a wide variety of coastal ecosystems (Justic et al., 2005; O’ Higgins 102
37 and Gilbert, 2014). 2.1. Monitoring program 103
38 The guidance for the descriptor 5 (D5) defines that most eu- 104
39 trophication assessment methods recognize that the immediate WFD requires that EU member states must regularly monitor 105
40 biological response is increased primary production reflected as and report on the condition of the coastal water bodies within 106
41 chl-a and/or macroalgal abundance (Ferreira et al., 2010). These their jurisdiction (Ferreira et al., 2007). However, reviewing the 107
42 are “direct effects” or “primary symptoms” and indicate the first objectives and requirements of marine water quality monitoring, 108
43 stages of eutrophication. “Indirect effects” or “secondary symp- Karydis and Kitsiou (2013) highlighted the scarcity of the marine 109
44 toms” such as low dissolved oxygen (DO), losses of submerged monitoring programs. 110
45 aquatic vegetation (SAV), changes in macrozoobenthic species A national monitoring program for coastal waters is under- 111
46 composition, and occurrences of nuisance and toxic blooms in- taken and run by the Hellenic Centre for Marine Research (HCMR) 112
47 dicate more advanced problems. in Greece (Simboura et al., 2015). The monitoring network has 113
48 Various methods have been developed in the EU to assess eu- been designed for the implementation of WFD in coastal waters 114
49 trophication in order to fulfill requirements of legislation designed and is delegated by the Greek water management authorities. The 115
50 to monitor and protect coastal water bodies from degradation. Greek authorities report annually on the water quality status to 116
51 Some methods use only chl-a concentrations, while in others chl-a the European Environment Agency providing data sets of physical 117
52 concentrations are combined with other parameters to give a 118
characteristics and concentrations of inorganic and organic nu-
53 more integral picture of eutrophication (Borja et al., 2012). How- 119
trient, organic matter, chl-a, macroalgae and macroinvertebrates
54 ever, in many cases the various methods give different assessment 120
and hazardous substances together with the characterization of
55 results in terms of classes when they are applied to the same water 121
the main pressures and impacts from human and other activities
56 body. In such cases, we have to decide which method is more ef- 122
57 ficient and representative of the condition in determining the at each monitoring station, according to Annex V of WFD 2000/60/ 123
58 eutrophication status (Borja et al., 2012). EC (Anonymous, 2012). 124
59 The Eastern Mediterranean Sea has always been considered as For this study, we have used data from 27 coastal monitoring 125
60 one of the most oligotrophic areas in the world with extremely stations located in 15 water bodies of Greece which are subjected 126
61 low nutrient concentrations, 12 times lower than the Atlantic to different types of anthropogenic pressures. Among them, the 127
62 Ocean (Pavlidou and Souvermezoglou, 2006; Krom et al., 2010). station in Limnos Island in the Aegean Sea receives very minor 128
63 Despite the oligotrophic character of the Mediterranean Sea, ele- anthropogenic pressures (Fig. 1; Table 1; see Section 2.2). In this 129
64 vated nutrient concentration indicates coastal eutrophication work, we used data from five sampling periods during 2012–2014 130
65 problems because several coastal areas undergo intense and con- (March–April 2012; November 2012; March–April 2013; No- 131
66 tinuous environmental pressure derived from a number of driving vember 2013; March–April 2014). 132
A. Pavlidou et al. / Continental Shelf Research ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 3
1 67
2 68
3 69
4 70
5 71
6 72
7 73
8 74
9 75
10 76
11 77
12 78
13 79
14 80
15 81
16 82
17 83
18 84
19 85
20 86
21 87
22 88
23 89
24 90
25 91
26 92
27 93
28 94
29 95
30 96
31 97
32 98
33 99
34 100
35 101
36 102
37 103
38 104
39 105
40 106
41 107
42 108
43 Fig. 1. Monitoring stations and water bodies. 109
44 110
45 2.2. Sites description and pressures pressure index has been estimated (Aubry and Elliott, 2006; Borja 111
46 et al., 2010, 2011) for the different coastal areas to give the mag- 112
47 The 15 water bodies of Greece where this study is focused, are nitude of the anthropogenic pressures imposed. The pressure in- 113
48 Elefsis bay, Inner Saronikos gulf, Western Saronikos gulf, Kor- tensity scale of Borja et al., (2011) has been modified to include 114
49 inthiakos gulf, Patraikos gulf, Amvrakikos gulf, Argolikos gulf, N. five levels of evaluation and assigning scores from 3 to 0 for each 115
50 Evoikos gulf, S. Evoikos Gulf, Pagasitikos gulf, Thermaikos gulf, pressure type within the corresponding area. The scores were 116
51 estimated using expert judgment based on our knowledge of the 117
Thessaloniki bay, Argostoli bay, Coasts of Ionian Sea, and Aegean
52 study areas. The pressure index has been calculated as the average 118
Sea (Fig. 1; Table 1). The coastal areas are impacted by several
53 pressure scores and the selected areas have been grouped in five 119
human activities such as treated and untreated sewage and in-
54 categories from no or minor pressure intensity to heavy pressure 120
dustrial discharges, agricultural/livestock farm discharges, aqua-
55 intensity (Table S1). The pressure types include sewage discharge, 121
culture (in general, finfish but in some cases e.g. Thermaikos gulf
56 industrial discharge, agricultural discharge, spoil waste, mar- 122
57 and Maliakos gulf, shelfish production also takes place), urbani- 123
iculture, fishing, marinas, and ports based on the Water Informa-
58 zation and tourism, and so on, whereas the Aegean Sea (Limnos 124
tion System for Europe (WISE-SoE), a reporting European system
59 Island) has been considered as less impacted by any of the afore- 125
for coastal and marine waters (http://cdr.eionet.europa.eu/gr/eea).
60 mentioned anthropogenic activities area (Anagnostou et al., 2005; 126
Lastly, a map has been produced showing the pressure status and
61 UNEP, 2007; Pavlidou, 2012). Many of these pressures have been 127
composition on a station basis.
62 recognized to be related to eutrophication status (Ferreira et al., 128
63 2011). Table 2 summarizes the characteristics of the study areas 2.3. Eutrophication assessment 129
64 and anthropogenic pressures that affect them. 130
65 In this work, we have tried to assess the distribution of an- In this study, we have assessed the eutrophication quality of 131
66 thropogenic pressures on the coastal water bodies of Greece. A the selected coastal Greek areas using three, five-step, different 132
44
66
64
49
48
65
63
62
60
59
58
56
55
54
53
52
50
46
45
43
42
40
39
38
36
34
32
30
29
28
26
25
23
22
20
35
33
24
27
61
57
37
47
31
21
12
41
51
19
16
15
14
13
10
18
17
11
9
8
7
6
5
4
3
2
1
4
Table 1
Coastal monitoring stations used for analysis.
Station ID Monitoring Stations Coastal Water Body Water District Stations’ Characteristics
Name Name Coordinates (wgs84) Name Code Name Max Depth (m) Average Secchi Disk Depth
(m)
lon lat
A. Pavlidou et al. / Continental Shelf Research ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 S1 Elefsina Bay 23.55750 38.01810 Elefsis Bay GR000600010007H Attica 22 7
2 Faneromeni Faneromeni Bay 23.43110 37.97126 Elefsis Bay GR000600010008N Attica 20 7
3 S7 Inner Central Saronikos - Psitalleia 23.59580 37.92830 Inner Saronikos Gulf GR000600010006H Attica 70 12
4 S8 Inner (Central) Saronikos 23.53330 37.88330 Inner Saronikos Gulf GR000600010005N Attica 90 16
5 S11 Inner (Central) Saronikos 23.64170 37.87670 Inner Saronikos Gulf GR000600010005N Attica 75 15
6 S25 Western Saronikos Gulf 23.25560 37.64720 Western Saronikos Gulf GR000600010009N Attica 415 20
7 Argolikos Argolikos Gulf 22.78500 37.55970 Argolikos Gulf GR000300010002N Eastern Peloponnese 16 5
8 Vourlias Argolikos Gulf 22.90930 37.47700 Argolikos Gulf GR000300010002N Eastern Peloponnese 180 17
9 TP10 Thessaloniki Bay 22.95140 40.53580 Thessaloniki Bay GR001000010010H Central Macedonia 20 4
10 TP16 Inner Thermaikos Gulf - N. 22.71750 40.46920 Inner Thermaikos Gulf GR001000010009N Central Macedonia 28 4
Mhxaniona
11 Limnos Limnos coasts 25.38329 39.88268 Open Aegean Sea GR001400010002N Aegean Islands 80 19
12 Itea Itea Bay 22.42245 38.42384 Korinthos Gulf GR000700010014N Eastern Greece Hellas 20 9
13 Antikyra Antikyra Bay 22.66047 38.35119 Korinthos Gulf GR000700010013H Eastern Sterea Hellas 70 18
14 Korinthos Korinthos Bay 22.87206 37.97522 Korinthos Gulf GR000200010006N Northern Peloponnese 98 13
15 Theologos N. Evoikos Gulf 23.33060 38.67310 N. Evoikos Gulf GR000700010007N Eastern Sterea Hellas 65 11
16 Skouries N. Evoikos Gulf 23.37500 38.57780 N. Evoikos Gulf GR000700010007N Eastern Sterea Hellas 75 12
17 Larymna Larymna Bay 23.30080 38.57190 N. Evoikos Gulf GR000700010008H Eastern Sterea Hellas 18 7
18 Asopos Avlida Bay 23.74470 38.33920 S. Evoikos Gulf GR000700010006N Eastern Sterea Hellas 10 5
19 Volos Volos Bay 22.96670 39.33330 Pagasitikos Gulf GR000800010005H Thessaly 40 10
20 Patra Patraikos Gulf 21.67890 38.23620 Patraikos Gulf GR000200010004N Northern Peloponnese 100 11
21 W. Patraikos Patraikos Gulf 21.51000 38.25000 Patraikos Gulf GR000200010004N Northern Peloponnese 90 9
22 S. Patraikos Patraikos Gulf 21.57110 38.18140 Patraikos Gulf GR000200010004N Northern Peloponnese 60 11
23 Argostoli Argostoli Bay 20.45170 38.24170 Argostoli Bay GR000200010014N Northern Peloponnese 18 7
24 Kalamas Eastern coasts of Corfu Sea 20.14390 39.57750 Coats of Ionian Sea GR000500010007N Epirus 12 6
25 S.Amvrakikos S. Amvrakikos Gulf 21.09450 38.92150 Amvrakikos Gulf GR000500010002N Epirus 40 2
26 Louros Estuary Northern Amvrakikos Gulf 20.80463 39.03025 Amvrakikos Gulf GR000500010001N Epirus 16 3
27 Arachthos Estuary Northern Amvrakikos Gulf 21.09420 39.03920 Amvrakikos Gulf GR000500010001N Epirus 20 2
100
129
128
126
125
123
122
120
124
106
104
132
130
127
109
108
105
103
102
107
121
101
131
112
119
116
115
114
113
110
118
117
111
68
99
98
96
95
94
93
92
90
89
88
87
86
85
84
83
82
81
80
79
78
75
69
77
70
67
97
73
72
91
71
76
74
1 Table 2 67
2 Characteristics of the sampling sites and anthropogenic pressures affecting them. 68
3 69
Station ID Station name Water body name Pressures Other characteristics
4 70
5 1 S1 Elefsis Bay Industrial effluents (Refineries, shipyards, chemical plants, Hypoxic events, Algal Blooms 71
6 2 Faneromeni food, metal, cement industries etc) 72
7 3 S7 Inner Saronikos Gulf Treated Sewage from Athens 73
4 S8
8 5 S11
74
9 6 S25 Western Saronikos Gulf Maricultures and agricultural discharges. Hypoxic and anoxic events during the 75
10 last 22 years 76
11 7 Argolikos Argolikos Gulf Agricultural 77
8 Vourlias
12 78
9 TP10 Thessaloniki Bay Thessaloniki harbor, industrial, treated or partly treated Oxygen decrease
13 sewage 79
14 10 TP16 Inner Thermaikos Gulf Agricultural discharges from the heavily polluted Axios 80
15 River, mariculture 81
16 11 Limnos Open Aegean Sea Very minor pressures 82
12 Itea Korinthos Gulf Domestic and industrial effluents
17 13 Antikyra
83
18 14 Korinthos Harbor activities 84
19 15 Theologos N. Evoikos Gulf Agricultural, maricultures 85
20 16 Skouries Industrial (smelting plant discharge) 86
17 Larymna
21 87
18 Asopos S. Evoikos Gulf Industrial and agricultural
22 19 Volos Pagasitikos Gulf Sewage, industrial and harbor activities 88
23 20 Patra Patraikos Gulf Harbor and industrial activities 89
24 21 W. Patraikos Industrial and agricultural 90
25 22 S. Patraikos 91
23 Argostoli Argostoli Bay Aquacultures and tourism activities
26 24 Kalamas Coats of Ionian Sea Agriculture and other activities
92
27 25 S.Amvrakikos Amvrakikos Gulf Agriculture Anoxic conditions 93
28 26 Louros Estuary Hypoxic conditions 94
29 27 Arachthos Estuary 95
30 96
31 97
32 tools applied in the oligotrophic waters of the Eastern Medi- Table 3 98
33 terranean coastal areas: (i) the trophic index TRIX (Vollenweider Methodological tools, indicators, and ranges used for Greek coastal areas for the 99
34 eutrophication assessment. 100
et al., 1998; Primpas and Karydis, 2011); (ii) chl-a biomass classi-
35 fication scheme (Simboura et al., 2005; Pagou et al., 2002); and Methods Indicators Eutrophication Eutrophication
101
36 (iii) eutrophication index (E.I.) (Primpas et al., 2010). status Range 102
37 TRIX was calculated according to the following equation based 103
38 Q2 on Vollenweideretal (1998), whereas eutrophication ranges have TRIXa,b D%O2, DIN ( ¼ High o 1.6 104
NO3 þ Good 1.6–2.8
39 been modified and applied for the oligotrophic areas of Eastern 105
NO2 þ NH4 þ ), Moderate 2.8–4.0
40 Mediterranean according to Primpas and Karydis (2011): PO43 , Chl-a Poor 4.0–5.3 106
41 Bad 45.3 107
42 TRIX = ⎡⎣log10 (CPO4 *CDIN *CChla *D%O2 ) + 1.5⎤⎦/1.2 Chl-a biomass Chl-a High o 0.1 (mg m 3) 108
43 classification Good 0.1–0.4 (mg m 3) 109
The E.I. was calculated according to the following schemec,d Moderate 0.4–0.6 (mg m 3)
44 0.6–2.21 (mg m 3)
110
equation (Primpas et al., 2010): Poor
45 Bad 42.21 (mg m 3) 111
46 E. I . = 0.279*CPO4 + 0.261*CNO3 + 0.296*CNO2 + 0.275*CNH 4 + 0.261 E.I.e NO3 , NO2 , High o 0.04 112
47 NH4 þ , PO43 , Good 0.04–0.38 113
*CChla Chl-a Moderate 0.38–0.85
48 114
Poor 0.85–1.51
49 where Bad 41.51 115
50 CDIN is the concentration of dissolved inorganic nitrogen 116
51 (¼ CNO3 þCNO2 þ CNH4); CPO4 is the concentration of phosphate; a
Vollenweider et al. (1998). 117
b
52 CNO3 is the concentration of nitrate; CNO2 is the concentration of Primpas and Karydis (2011). 118
c
Simpoura et al. (2005)
53 nitrite; CNH4 is the concentration of ammonium (nutrient con- d
Pagou et al. (2002).
119
54 centrations for TRIX in mg*m-3;for E.I calculation in mmol m 3); e
Primpas et al. (2010). 120
55 CChla is the concentration of phytoplankton chl-a (in mg m 3). 121
56 D%O2 is the % deviation of the oxygen concentration from sa- 122
57 turation conditions. methodological tool. Nutrient, DO and chl-a data from the whole 123
58 Table 3 shows the different methods used for the eutrophica- water column, as well as from the layer of the euphotic zone 124
59 tion assessment, the indicators used for each methodological tool, (determined as three times the Secchi disk disappearance depth) 125
60 the classes of eutrophication status, and the eutrophication range. 126
were used. Thematic maps presenting the eutrophication assess-
61 Nutrient, DO and chl-a data were measured using standard 127
ment according to the three different methodological tools were
62 methods and quality assurance protocol according to the ISO 128
produced (a) for the whole water column and (b) for the euphotic
63 17025 certification procedures (Mullin and Riley, 1955; Murphy 129
64 and Riley, 1962; Holm-Hansen et al., 1965; Carpenter, 1965; Kor- zone. The thematic maps were produced in ArcGIS/ArcINFO en- 130
65 oleff, 1970; Strickland and Parsons, 1977; Welschmeyer, 1994). vironment on station basis. 131
66 The coastal areas have been classified according to each 132
1 2.4. Benthic indices station allowed us to assess the relative importance of each area 67
2 compared to the others (Table S1; Fig. 2). Industrial discharges, 68
3 Benthic ecological status indices are influenced mostly by the port activities, sewage discharges, aquaculture activities, and 69
4 pressures affecting the sea bottom, while water column indices by fishing are the most important pressures affecting the coastal 70
5 pressures in the water column. However, the benthic macro- areas of Greece. In fact, maricultures seem to affect more the se- 71
6 invertebrates are considered as good indicators of general en- lected coastal areas among the anthropogenic pressures defined 72
7 vironmental status due to their limited ability to move and avoid (Table S1), followed by fishing, other activities and industrial dis- 73
8 pressures and also due to their low temporal variability and long charges. This is interesting, since mariculture activities in Greece 74
9 life cycles (Borja et al., 2009; Prins et al., 2013). have expanded rapidly during the last years. Moreover, the impact 75
10 Therefore, two intercalibrated (GIG, 2013) EcoQ benthic indices of mariculture activities on the water column of some Greek 76
11 were used in order to juxtapose and compare the classification coastal areas has been reported, indicating relatively elevated 77
12 according to eutrophication indices with the one resulting from nutrient concentrations (Pavlidou and Rousselaki, 2014). 78
13 these ecological quality indices. The BENTIX index (Simboura and According to Pearson analysis, chl-a, TRIX, and E.I. correlated 79
14 Zenetos, 2002) was used for the classification of the EcoQ of the positively with the pressure index with correlation coefficients of 80
15 benthic macroinvertebrate communities, while for the ecological 0.72, 0.68, and 0.56, respectively. More specifically, in both data 81
16 evaluation of the biological element of macroalgae, the EEIc (Or- treatments, chl-a, E.I., and TRIX correlated significantly (at the 82
17 fanidis et al., 2001, 2003, 2011) was used. level 0.01) with agricultural pressures and mariculture, whereas 83
18 Within the intercalibration exercise, it has been found that they did not correlate with sewage and industrial discharges and 84
19 diversity measures of the benthic community did not show port activities (Tables S2 and S3). 85
20 monotonic patterns of response to the gradient of organic content, A factor analysis (FA) for the pressures and eutrophication in- 86
21 while strong correlations were found between indicator taxa indices was performed for the entire water column data treatment 87
22 dices and the pressure gradient. Diversity indices have also been (Figure is not presented) as well as for the euphotic zone (Fig. 3). 88
23 criticized as ecological status indicators due to their dependency The analysis for the entire water column resulted to three main 89
24 on habitat type, natural variations, and taxonomic effort. In order factors or components (75% of total system variability was ex- 90
25 to cover the MSFD requirements for structural components of the plained; 39% by the first factor, 27% by the second, and 9% by the 91
26 benthic community, a multimetric benthic formula has been de- third). Fishing activities, mariculture, and discharges from agri- 92
27 veloped (Simboura et al., 2015) including biotic and structural cultural activities together with chl-a and other activities acquired 93
28 components in a formula controlling diversity components con- high loadings in the first component, sewage and industrial dis- 94
29 tribution. However, this formula results in a similar EcoQ classifi- charges, spoil wastes, and ports in the second component, while 95
30 cation with the biotic index at most cases, and it was not used the eutrophication indices E.I. and TRIX cluster together in the 96
31 here. third component and acquire high loadings together with marinas 97
32 (Table S4). The eutrophication indices, that is, E.I. and TRIX, seem 98
33 2.5. Statistical analysis to reflect the integral effects of the pressures exerted on the 99
34 coastal water bodies of Greece. 100
35 To analyze the agreement among eutrophication and ecological FA for the euphotic zone resulted to two main factors (71% of 101
36 quality status indices, as well as among eutrophication indices, a total system variability was explained, 40% by the first factor, and 102
37 weighted Kappa analysis was undertaken (Cohen, 1960; Landis 31% by the second). Chl-a, E.I., and TRIX together with fishing, 103
38 and Kosch, 1977) applying the methodology presented in Borja agricultural activities, mariculture, and other activities acquired 104
39 et al., (2007). This analysis takes into account that the importance high loadings in the first component (Table S5). In the second 105
40 of misclassification is not the same among the close categories component, sewage and industrial discharges, ports, and spoil 106
41 (e.g. high or good, moderate or poor) as among other categories wastes are the main pressures. It seems that, agricultural and 107
42 (e.g. between high or good and moderate or poor). The Kappa maricultural activities, fishing, and other activities (riverine dis- 108
43 values reveal the next levels of agreement: (i) null o0.05; (ii) very charges, dredging, etc.) couple with all eutrophication status in- 109
44 low: 0.05–0.2; (iii) low: 0.2–0.4; (iv) moderate: 0.4–0.55; (v) good: dices when we refer to the euphotic zone only. 110
45 0.55–0.7; (vi) very good: 0.7–0.85; (vii) almost perfect: 0.85–0.99; From this analysis, it seems that two main types of pressures 111
46 and (viii) perfect: 1 (Monserud and Leemans, 1992). affect the selected coastal areas: the pressures related to agri- 112
47 In addition, the SPSS software program was used to explore cultural and maricultural activities and those related to sewages, 113
48 correlations across eutrophication results and selected indices. For industries, and ports. Thus, we can recognize two different drivers 114
49 Pearson correlations, the index values were standardized by using for these pressures: nutrients related to agricultural activities, and 115
50 log10 transformation of values. Factor analysis was rotated using inorganic and organic pollutants (metals and organic compounds) 116
51 the Varimax rotation method and was used in order to investigate related to industries, sewages, and ports. The eutrophication in- 117
52 the relative importance of the human pressures on eutrophication dices which are discussed in this work seem that are mainly 118
53 and nutrient ratios. connected or affected by the agricultural activities, mariculture, 119
54 fishing, and other activities. Thus, the eutrophication indices in the 120
55 euphotic zone spontaneously express the direct effects of eu- 121
56 3. Results and discussion trophication driven by the nutrient enrichment. 122
57 123
58 3.1. Human-induced pressures 3.2. Nutrient and nutrient ratios 124
59 125
60 The need to assess eutrophication in Greek coastal waters ap- Spatial variation of nutrient concentrations among the coastal 126
61 propriately and in relation to different pressures derives from the stations was observed. Nitrate concentrations ranged between 127
62 need to clarify the responsiveness of the ecological indices/metrics 0.01 and 23.5 mg m 3. The highest nitrate value was recorded at 128
63 elements to the different pressures within the implementation of station 10 in the inner Thermaikos gulf. Phosphate concentrations 129
64 the European Directive in Greece. ranged between 0.01 and 2.25 mg m 3 and the highest value was 130
65 The resulting ranking of the selected areas based on the priority recorded near the bottom of station 25 in the south Amvrakikos 131
66 score (pressure index) given to each type of pressure for each gulf, where anoxic conditions have been developed. Nitrite 132
1 67
2 68
3 69
4 70
5 71
6 72
7 73
8 74
9 75
10 76
11 77
12 78
13 79
14 80
15 81
16 82
17 83
18 84
19 85
20 86
21 87
22 88
23 89
24 90
25 91
26 92
27 93
28 94
29 95
30 96
31 97
32 98
33 99
34 100
35 101
36 102
37 103
38 104
39 105
40 106
41 107
42 108
43 109
44 110
45 111
46 112
47 113
48 114
49 Fig. 2. Assessment of the pressure level in terms of relevant pressures and water body. 115
50 116
51 117
52 concentrations ranged between 0.01 and 1.54 mg m 3 (maximum water column were much higher than the Redfield ratio (16:1) in 118
53 concentration near the bottom of station 27 in Amvrakikos gulf, most of the cases (82% of the stations) with average mean in- 119
54 close to the mouth of Arachthos). Ammonium concentrations in tegrated value of all stations 37 7 28, except in the inner Saronikos 120
55 the coastal areas of Greece ranged between 0.05 and 5.06 mg m 3 gulf (stations 3 and 5), Amvrakikos gulf (stations 25 and 26), and 121
56 (maximum value recorded near the bottom of station 21). Chl-a Thessaloniki bay (station 9), which were defined as nitrogen lim- 122
57 concentration in the Greek coastal areas ranged between 0.01 and ited areas. DIN:Si ratio was calculated lower than 1 (average value: 123
58 6.18 mg m 3. High phytoplankton biomass was recorded near the 0.477 0.20) in the majority of the study stations indicating silicate 124
59 bottom (17 m) of station 26 close to Louros mouth in Amvrakikos excess in the monitoring stations. The analysis of our data showed 125
60 gulf. spatial variation of the DIN:P ratio. 126
61 Concerning the different forms of dissolved inorganic nitrogen The atomic Si:DIN:P ratio of marine diatoms is about 16:16:1 in 127
62 (DIN), nitrite (NO2 ) was the predominant form only in 3% of the a nutrient-replete ecosystem (Redfield, 1958; Xu et al., 2008). 128
63 data. On the contrary, nitrate (NO3 ) was the dominant form in Deviation from the Redfield ratio indicates the potential for N, P, or 129
64 the 58% of data, whereas ammonium (NH4 þ ) was the pre- Si limitation of phytoplankton growth. In our assessment of stoi- 130
65 dominant form in the 39% of the data. chiometric limitations, we have calculated Redfield ratios follow- 131
66 The mean integrated DIN:P ratios calculated for the entire ing Pavlidou et al. (2004) and Xu et al. (2008) to predict 132
1 sewage discharges. According to those correlations, heavy in- 67

2 dustrial pressure and sewage discharges would result to the in- 68
3 crease of DIN:Si ratio and decrease of Si:P ratio, thus potentially to 69
4 the shift of the ecosystem towards Si-Limitation, whereas heavy 70
5 pressure from agriculture, mariculture and fishing activities would 71
6 potentially lead to the shift of the ecosystem towards N-Limitation. 72
7 It seems that at the selected coastal Greek areas, deviation of Si: 73
8 DIN:P ratios from the theoretical value are probably related mainly 74
9 to the mariculture, agriculture, fishing, industrial and sewage 75
10 discharges. 76
11 A factor analysis (FA) among the pressures, the eutrophication 77
12 indices (E.I. and TRIX) and the nutrient ratios was also performed 78
13 for the entire water column data treatment (Figure is not pre- 79
14 sented). The analysis resulted to three main factors or components 80
15 (71% of total system variability was explained; 36% by the first 81
16 factor, 23% by the second, and 12% by the third). According to this 82
17 analysis, mariculture, agriculture, fishing and other activities to- 83
18 Fig. 3. Distribution of the pressures and eutrophication assessment results, within gether with the eutophication indices acquired the highest load- 84
19 the new multidimensional space defined by the factor analysis (rotated, using the ings in the first component, sewage and industrial discharges to- 85
20 Varimax rotation method) in the euphotic zone. gether with DIN:Si ratio acquired the highest loadings in the 86
21 second component, while in the third component the DIN:P and 87
22 1. N limitation occurs when DIN:P o16 and DIN:Si o1; Si:P ratios acquire high loadings. 88
23 2. P limitation occurs when DIN:P 4 16 and Si:P 4 16; Among the N-limited coastal areas, the lowest mean integrated 89
24 3. Si limitation occurs when DIN:Si 4 1 and Si:P o16. DIN:P and DIN:Si values (4.8 and 0.05, respectively) together with 90
25 the highest Si:P (177) value were recorded in South Amvrakikos 91
26 The data indicated significant probable P-limitation (69%), gulf, where anoxic condition occurred. Thus, this area with high 92
27 whereas N-limitation was found at the 31% of the cases and Si- eutrophication problems was characterized by excess of silicate 93
28 limitation only at 7%. Among the study coastal areas, Saronikos, and nitrogen depletion probably due to its consumption by phy- 94
29 Thermaikos, and Amvrakikos gulfs indicated N-limitation. toplankton or other biochemical processes (e.g. denitrification). 95
30 Changes in DIN:P:Si ratios may produce “undesirable dis- The mean integrated value of phytoplankton biomass in South 96
31 turbances” (e.g. the potential effects of increased production and Amvrakikos gulf was 3.00 mg m 3. On the other hand the highest 97
32 the direct and indirect changes in the balance of organisms) on the DIN:P and DIN:Si and Si:P values were recorded in Patraikos gulf, 98
33 ecosystem structure and function, as well as on the ecosystem an area with light anthropogenic pressure. Limnos with very 99
34 goods and services used by humans. However, such effects do not minor anthropogenic impact is characterized by relatively high 100
35 always result from nutrient enrichment and may be triggered by mean integrated DIN:P, DIN:Si and Si:P values (54, 0.28 and 176, 101
36 other causes (Ferreira et al., 2011), while there is still a lot of dis- respectively), indicating surplus of inorganic nitrogen and silicate 102
37 cussion on how and which useful nutrient ratios are assessed in in this phosphorus limited area. The mean integrated values of 103
38 eutrophication studies (Newton et al., 2003). In our case, the re- phytoplankton biomass in Limnos island and Patraikos gulf were 104
39 latively lower DIN:P ratios were recorded in coastal areas where 0.05 and 0.19 mg m 3, respectively. 105
40 some of the direct and indirect effects of eutrophication have been Skogen et al. (2004) have used DIN:P ratio as a possible proxy 106
41 recognized. More specifically, in Amvrakikos gulf hypoxic and for eutrophication and have proposed five different classes of DIN: 107
42 anoxic conditions occurred, while in Thessaloniki bay (Thermaikos P. According to this classification scheme, west and south Pa- 108
43 gulf) increased abundance of certain harmful dinoflagellate species traikos, Limnos island, Patra, Korinthos and Elefsis bay (S1) are 109
44 was recorded. Indeed, in Thermaikos gulf, dinoflagellates were the grouped together with very high and high mean integrated DIN:P 110
45 dominant blooming species from 1996 onwards including the ratio values (50–100 and 4100) and low phytoplankton biomass 111
46 toxic species Dinophysis acuminata (Reizopoulou et al., 2004; concentration (0.18 70.08 mg m 3). These are areas characterized 112
47 Kouhypoxikaras and Nikolaidis, 2004; Pagou, 2005; Vakirtzi et al., by low impact of maricultures, agriculrure and fishing. On the 113
48 2006, 2010). other hand, areas with high and heavy anthropogenic pressures, 114
49 The nutrient ratio DIN:P correlated negatively and significantly mostly mariculture and agricultural, are grouped together with 115
50 (at level 0.05) with the pressure index, indicating that the increase low and/or very low (16–30 and o16, respectively) mean in- 116
51 of the anthropogenic pressures results in more N-limited coastal tegrated DIN:P ratio values and higher concentrations of phyto- 117
52 areas. In addition, the mean integrated DIN:P ratio calculated for plankton biomass (1.50 70.99 mg m 3). Unfortunately, the lack of 118
53 the whole water column showed significant negative correlation data on composition and abundance of the phytoplankton com- 119
54 (at the level 0.05) with mariculture, agriculture pressures (corre- munities do not allow us to investigate and discuss any differences 120
55 lation coefficients: 0.391 and 0.455, respectively) and with on the ecosystems structure and function. 121
56 other pressures (at level 0.01; correlation coefficient: 0.519) 122
57 whereas it did not correlate with sewage and industrial discharges 3.3. Eutrophication assessment with different methodological tools 123
58 and port activities. 124
59 The mean integrated DIN:Si ratio correlated positively and Few methods have been developed in the EU in order to assess 125
60 significant (at level 0.01) with industrial pressures (correlation eutrophication and evaluate its trends. The state of the art in 126
61 coefficient:þ0.636). In addition, DIN:Si ratio correlated negatively Greece, regarding the eutrophication assessment of the coastal 127
62 and significant (at level 0.01) with fishing and other activities waters, is the following: A Greek eutrophication scale was devel- 128
63 (correlation coefficients: 0.638 and 0.632, respectively), oped by Ignatiades et al. (1992), Karydis (1999) and Pagou et al. 129
64 whereas, at level 0.05 DIN:Si ratio correlated negatively and sig- (2002), and has been used extensively ever since. It involved four 130
65 nificant with mariculture (correlation coefficient: 0.403). Si:P levels of trophic status, thus, in order to fit the five step ecological 131
66 ratio correlated negatively and significant (at level 0.05) with status scale of WFD, chl-a values from the Greek eutrophication 132
1 67
2 68
3 69
4 70
5 71
6 72
7 73
8 74
9 75
10 76
11 77
12 78
13 79
14 80
15 Fig. 4. Eutrophication assessment of the Greek coastal water bodies according to chlorophyll-a, TRIX, and E.I. applied in the entire water column. 81
16 82
17 scale were modified by Simboura et al. (2005). The boundaries of characterized as in good eutrophication status, 55% in moderate 83
18 this scale were decided during the EU WFD 2nd intercalibration status, and 15% in poor status, while 0% of the stations were 84
19 phase and published in the “Commission Decision of 20 Septem- classified into high or bad status. On the contrary, according to the 85
20 ber 2013 establishing, pursuant to Directive 2000/60/EC of the TRIX index tool, 56% of the monitoring stations were assessed into 86
21 European Parliament and of the Council, the values of the Member the high eutrophication status, 28% into good status, 12% into 87
22 State monitoring system classifications as a result of the inter- moderate status, 4% into poor status, and 0% into bad status. The 88
23 eutrophication assessment according to phytoplankton biomass 89
calibration exercise and repealing Decision 2008/915/EC.
24 90
In addition, two multiparametric methods, E.I. and TRIX, were expressed as chl-a concentrations, showed that only 7% of the
25 91
used for the assessment of the trophic conditions of the Greek monitoring stations were classified into the high eutrophication
26 92
coastal waters, according to the WFD requirements. The bound- status, 48% into good status, 11% into moderate status, 30% into
27 93
aries of TRIX have been modified by Primpas et al. (2010) in poor status, and 4% into the bad eutrophication status. It is obvious
28 94
order to be used for the oligotrophic waters of the Eastern that there is a certain degree of divergence between the resulting
29 95
Mediterranean. classifications from the different methodological tools used for the
30 96
Phytoplankton is usually employed as an indicator of change in assessment of eutrophication in Greece. Thus, low class agreement
31 97
nutrient loads and as a key element for assessing eutrophication in between E.I and TRIX has been observed (Table 4).
32 98
marine systems. Indeed, its assessment has been required by dif- When we used data from the euphotic zone only, the eu-
33 99
ferent legislations (Borja et al., 2012; Garmendia et al., 2013), and trophication assessment showed that according to E.I. 54% of the
34 100
has also been applied in this study together with E.I. and TRIX monitoring stations were classified into good eutrophication sta-
35 101
indices. Figs. 4 and 5 present the eutrophication assessment ac- tus, 35% into moderate status, 11% into poor status, and 0% into
36 102
cording to the three different methodological tools applied in high and bad status. According to TRIX, the eutrophication status
37 103
Greece, (a) for the whole water column of the study water bodies was rather upgraded with 29% of the stations classified into high
38 104
and (b) for the euphotic zone. Euphotic zone reaches to 1.5 m in status, 54% into good, 17% into moderate, and 0% into poor or bad
39 105
the inner Thermaikos gulf to 87 m in Western Saronikos gulf. The status. According to chl-a criterion, the classification was more
40 106
estimation of the euphotic zone (Welch, 1948) is three times the severe, with only 4% of the monitoring stations classified into high
41 107
Secchi disk disappearance depth. Also, Simboura et al. (2015) status, 50% into good status, 11.5% into moderate status, 23% into
42 108
analyzing the variance of the Secchi disk disappearance depth in poor status, and 11.5% into bad status. In the euphotic zone E.I. and
43 109
relation to ecological status over Greek coastal waters, found that TRIX showed low agreement also (Table 4).
44 110
the moderate to good threshold limit is located at 11 m depth. This According to MSFD, the indirect effects of eutrophication are
45 111
means that on average and in relatively unpolluted waters, the abundance of perennial seaweeds and sea grasses impacted by
46 112
47 euphotic zone may extend to 30–33 m depth. decrease in water transparency and DO changes, for example, 113
48 The eutrophication assessment, when using data from the hypoxic or/and anoxic events as well as changes in macro- 114
49 whole water column of the monitoring stations, showed that ac- zoobenthic species composition (Ferreira et al., 2010; Borja et al., 115
50 cording to the E.I. 30% of the studied coastal stations were 2013). In our cases, hypoxic or/anoxic events have been recorded 116
51 117
52 118
53 119
54 120
55 121
56 122
57 123
58 124
59 125
60 126
61 127
62 128
63 129
64 130
65 131
66 Fig. 5. Eutrophication assessment of the Greek coastal water bodies according to chlorophyll-a, TRIX, and E.I. applied in the euphotic zone. 132
1 Table 4 67
2 Kappa values and agreement, matching and mismatching percentage, Pearson correlation between the different methods (a) entire water column,(b) euphotic zone. 68
3 69
Chl-a E.I. TRIX
4 70
5 Kappa value Pearson Kappa value Pearson Kappa value Pearson 71
6 72
7 a) Entire water column 73
BENTIX 0.43 0.7155nn 0.44 0.4986nn 0.31 0.4832n
8 Moderate Moderate Low
74
9 EEIc 0.6 0.7189nn 0.38 0.3074 0.23 0.3367 75
10 Good Low 0.1874 Low 0.1718 76
11 E.I. 0.4 0.5528nn 0.31 0.8937nn 77
Moderate Low
12 78
TRIX 0.28 Low 0.5373nn
13 79
14 b) Euphotic zone 80
BENTIX 0.39 0.7222nn 0.47 0.6289nn 0.23 0.6686nn
15 Low Moderate Low
81
16 EEIc 0.68 0.7993nn 0.47 0.5452n 0.31 0.6800nn 82
17 Good Moderate Low 83
18 E.I. 0.62 0.8693nn 0.38 0.7171nn 84
Good Low
19 85
TRIX 0.32 0.7143nn
20 low 86
21 87
n
22 Significant correlation at 0.05 level. 88
nn
23 Significant correlation at 0.01 level. 89
24 90
25 91
26 in S. Amvrakikos gulf and the deep layer of the western Saronikos According to Devlin et al. (2011), TRIX method, which uses a 92
27 gulf, whereas relatively lower oxygen concentrations have also combination of N, P, phytoplankton, chl-a, and DO saturation may 93
28 been recorded in Elefsis bay and Thessaloniki bay. In addition, produce biased results because it assumes that eutrophication 94
29 decrease in water transparency has been observed in Amvrakikos processes are mainly reflected as changes in phytoplankton bio- 95
30 gulf and Thessaloniki bay, but not in the western Saronikos gulf mass, but this does not hold true for some ecosystems, for ex- 96
31 (Table 1), whereas in Thessaloniki bay, harmful algal blooms have ample, shallow systems where other primary producers (e.g. 97
32 been developed. According to this information, these areas reflect macroalgae, seagrasses, etc.) may contribute a significant amount 98
33 some of the direct and indirect effects of eutrophication in the to total primary production. The E.I. method uses the same as- 99
34 water column. South Amvrakikos and western Saronikos gulf have sumption as the TRIX method as using a combination of N, P, and 100
35 been classified into poor status using both E.I. and TRIX. chl-a without though including oxygen saturation. Thus, it may 101
36 Oxygen deficiency can result from the sinking and decom- also introduce a bias in the results because oxygen and probably 102
37 position of the excess organic matter produced as a result of eu- other primary and/or secondary parameters are not included 103
38 trophication, but it can also be the result of other causes, including (Garmendia et al., 2012). The juxtaposition and comparison with 104
39 decreases in the ventilation of deep water caused, for example, by other biological indicators (e.g. macroalgal abundance and benthic 105
40 climate change. This is the case of the western Saronikos gulf with macroinvertebrate communities) may provide a more complete 106
41 high residence time of the deep water layer. On the other hand, in overview within the context of WFD. For this reason, we have 107
42 Amvrakikos gulf nutrients are introduced into the surface layer of examined the class agreement relations among eutrophication and 108
43 the semi-enclosed water body of Amvrakikos gulf from riverine benthic indices presented below. 109
44 discharges resulting to water column stratification and oxygen 110
45 deficiency in the near bottom water layer. These examples indicate 3.4. Class agreement between indices 111
46 the significant role of hydrodynamics, bathymetry, sedimentary 112
47 processes, and so on in the eutrophication assessment (Ferreira The E.I. and chl-a parameter are in good agreement in the eu- 113
48 et al., 2010; Newton et al., 2014). Amvrakikos and western Sar- photic zone and in moderate agreement in the entire water col- 114
49 onikos gulfs, which have been classified into poor status with both umn data treatment, whereas the agreement between TRIX and 115
50 E.I. and TRIX in the whole water column, reflect some of the direct chl-a is low in both cases. In addition, the class agreement of E.I. to 116
51 and indirect effects of eutrophication in the water column, which the macroalgae index EEIc is low in the euphotic zone and mod- 117
52 affect also the perennial seaweeds and sea grasses (Ferreira et al., erate in the entire water column, whereas the class agreement of 118
53 2010; Newton et al., 2014). However, according to Garmendia et al. TRIX with EEIc is low at both cases of data treatment. According to 119
54 (2012), the inclusion of secondary indicators (e.g. lower DO levels, this, we could assume that the E.I. probably reflects in a better way 120
55 appearance of HABs, or changes on the benthic community) in than TRIX the “direct effect” of nutrient enrichment, which is the 121
56 assessment methods is very important and provides a more robust increased primary production indicated as increased chl-a and/or 122
57 picture with a better perspective of the scale of the eutrophication macroalgal abundance. Indeed, the increase of primary production 123
58 problem than methods that consider only water-column chem- has been recognized as the biological response of nutrient en- 124
59 istry. Thus, the assessment methods that consider only primary richment in the water column (Ferreira et al., 2007; Borja et al., 125
60 indicators of the water column may downgrade the eutrophication 2008; Ferreira et al., 2011). In that way, among the two multi- 126
61 status of an area, as they do not include secondary indicators. parametric indexes (E.I. and TRIX), E.I. seems to have a better 127
62 Including additional parameters for the eutrophication assessment performance than TRIX and reflects in a more efficient way the 128
63 (e.g. D% O2 in TRIX) is in general useful, but it can also bias the first stages of eutrophication. However, this must be verified with 129
64 eutrophication status because it can be related to other pressures other biological indices according to the class agreement analysis 130
65 or factors (such as climate change) different from nutrients pres- presented below. Different quality elements address different 131
66 sure (Garmendia et al., 2012). pressures, but WFD requires the integrated impact from all 132
1 pressures (Caroni et al., 2013). The comparison among the eu- pressures, and/or intercalibrated with other methods. Indeed, the 67
2 trophication indices and the benthic indices aims to demonstrate intercalibrated BENTIX was designed for the Mediterranean 68
3 the differential sensitivity of the various indices to pressures with benthic ecosystem and has been tested using various anthro- 69
4 the overall goal to indicate the most responsive tool to the sum of pogenic pressures, such as eutrophication and organic pollution, 70
5 pressures. mining residues, and aquaculture in Greece, Cyprus, and the 71
6 Class agreement of eutrophication indices with EcoQ benthic Western Mediterranean through the intercalibration exercise (GIG, 72
7 indices was checked with data from the whole water column and 2013; Simboura et al., 2012 and references within). 73
8 with data from the euphotic zone (considered as three times the The observed mismatching has been attributed to the high 74
9 Secchi Disk depth; Welch, 1948), in order to examine the levels of spatial and temporal variability of some of the biological elements, 75
10 linkage or relevance of the eutrophication assessment tools with such as the macroalgae and phytoplankton (Borja et al., 2004). A 76
11 the EcoQ (Table 4) following, in both cases, the weighted Kappa similar analysis of WFD data originating from the implementation 77
12 analysis. The Pearson atation coefficient and p values among the of WFD in Greece also verifies this divergence between macroalgae 78
13 eutrophication indices (chl-a, E.I., and TRIX) and the benthic EcoQ and the benthic element, and the general EcoQ and attributes this 79
14 indices (BENTIX and EEIc) are also presented in Table 4. According to the specific sensitivity of macroalgae to the point source pres- 80
15 to the results for the whole water column, from the surface layer sures exerted on the coasts of a water body (Simboura et al., 2015). 81
16 to the near-bottom layer, the eutrophication indices E.I. and TRIX The later effect of the localized direct effect of eutrophication on 82
17 show a better correlation with the BENTIX than with EEIc (Table 4). the coast is getting more evident in the present work. 83
18 More specifically, chl-a parameter correlated positively and sig- Caroni et al. (2013) showed that the sensitivity of the different 84
19 nificantly with all eutrophication indices and significantly nega- biological quality elements (BQEs) to various pressures influences 85
20 tively with the EcoQ indices. The E.I. correlated positively with chl- the confidence level and comparability of the various methods for 86
21 a parameter and TRIX and negatively with the BENTIX, but it did combining the assessment results. The BQEs used for the WFD 87
22 not correlate with EEIc. The TRIX showed a significant and nega- assessment may be sensitive to the same pressure, be com- 88
23 tive correlation with the BENTIX and a positive one with chl-a and plementary in displaying the effects of different pressures, on 89
24 E.I. The BENTIX and the EEIc indices correlated with each other different spatial and or temporal scales, on different aspects of the 90
25 significantly. According to the above, among the eutrophication ecosystem functioning, or be responsive to multiple pressures. 91
26 indices and criteria, chl-a showed linear correlation with both Integrating both aspects of eutrophication including primary 92
27 EcoQand in good agreement with EEIc. Moreover, E.I. and TRIX and secondary effects such as macroalgae, benthic macro- 93
28 eutrophication indices correlated significantly only with the invertebrate communities, and water eutrophication indices 94
29 BENTIX, noting that E.I. had stronger correlation and better through the application of an integrative method such as decision 95
30 agreement with BENTIX than TRIX had. tree (Simboura et al., 2015) will ensure the capturing of all eu- 96
31 However, for the euphotic zone chl-a showed good agreement trophication effects on ecological and environmental status 97
32 with EEIc and low with BENTIX. Regarding E.I., the class agreement assessments. 98
33 was the same (moderate) both with BENTIX and EEIc. Pearson 99
34 correlations in the euphotic zone treatment of data are all sig- 100
35 nificant (Table 4). 4. Conclusions 101
36 In general, for both cases of data treatment, we observed that 102
37 class agreements and correlations between TRIX and benthic in- In this work, the eutrophication status of selected Greek coastal 103
38 dices were at all cases low. The class agreement between chl-a and areas was studied under various pressures using different in- 104
39 TRIX was always also low (Table 4). EEIc linked better with chl-a in dicators and different methodological approaches. The methods 105
40 both cases of treatment (good agreement) because macroalgae and applied for Greek coastal areas (TRIX, E.I., and chl-a biomass cri- 106
41 chl-a are both primary producers representing the same level of terion) did not give absolutely matching results of eutrophication 107
42 the trophic web. Macroalgal growth and chl-a increase are “direct” assessment for all the studied water bodies. 108
43 effects or “primary symptoms” indicating the first stages of eu- Agricultural activities, mariculture, fishing, and other activities 109
44 trophication. It is noteworthy that among the eutrophication in- (e.g. riverine discharges) are the pressures that mostly related to 110
45 dices, E.I. showed overall better (moderate) agreement with the eutrophication status assessment methodologies used for the 111
46 BENTIX. Macroinvertebrates are secondary producers and the selected Greek coastal water bodies. 112
47 “indirect” effects of eutrophication. This means that we expect no Chl-a linked better with EEIc in both cases of treatment (good 113
48 direct linkage between water eutrophication (nutrients) and agreement). It is noteworthy that among the eutrophication in- 114
49 benthic macroinvertebrates because it flows via plant growth and dices, E.I. showed overall better (moderate) agreement with 115
50 organic compounds. However, there is an indirect relation of BENTIX. 116
51 macroinvertebrates with water eutrophication, which is better Chl-a is the most relevant indicator to reflect eutrophication 117
52 reflected to E.I. (moderate agreement). impact on macroalgae. Moreover, E.I reflects better the indirect 118
53 The coastal macroalgae through the index EEIc are more sen- relation of eutrophication. 119
54 sitive in capturing the effect of the surface or euphotic zone eu- The evaluation of the eutrophication status applied in dynamic 120
55 trophication, expressed through chlorophyll biomass levels. coastal areas depends on the selected indicator as each one fo- 121
56 From the above, it seems that among BENTIX and EEIc there is a cuses on different aspects of the ecosystem (concentration of nu- 122
57 divergence concerning the relationships with eutrophication in- trients, DO, chl-a, benthic, and planktonic organisms). In Medi- 123
58 dices. Indeed, at many cases of coastal and transitional water terranean, and especially in the oligotrophic Eastern Mediterra- 124
59 bodies, macroalgae and chl-a biomass have produced a high dis- nean, it is still not well established which physicochemical and/or 125
60 agreement with the macroinvertebrates assessment, and conse- biological indices should be used. Among the methods used in this 126
61 quently, with the integrative assessment that assigns a special study, chl-a reflected more efficiently the first stage of eu- 127
62 weightage to benthic communities as being good indicators of trophication (macroalgae), while E.I. reflected better the integral 128
63 environmental quality (Borja and Rodríguez, 2010). The increased eutrophication status of a water body as a whole. In the integrative 129
64 reliability of a given method was based on the assumption that the status implementation of Greek coastal waters, E.I. has been used 130
65 method or index is used broadly by authors other than the pro- as a surrogate for the general physicochemical status used to- 131
66 posers of the method, was tested for several different human gether with the biological elements following the decision tree of 132
1 Borja et al. (2009) that gives a special weightage to the macro- Borja, A., Bald, J., Franco, J., Larreta, J., Muxika, I., Revilla, M., Rodrνguez, J.G., Solaun, 67
2 invertebrates (Simboura et al. 2015). Therefore, it could be con- O., Uriarte, A., Valencia, V., 2009. Using multiple ecosystem components, in 68
assessing ecological status in Spanish (Basque Country) Atlantic marine waters.
3 sidered as a reliable tool with regard to the assessment of eu- Mar. Pollut. Bull. 59, 54–64. 69
4 trophication status, as well as, to the implementation of nutrient Borja, A., Barbone, E., Basset, E., Borgersen, G., Brkljacic, M., Elliott, M., Garmendia, J. 70
5 management strategies under the EU WFD and the EU MSFD. M., Marques, J.C., Mazik, K., Muxika, I., Neto, J.M., Norling, K., Rodrνguez, J.G., 71
Rosati, I., Rygg, B., Teixeira, H., Trayanova, A., 2011. Response of single benthic
6 Nevertheless, it should be stressed that the station in Limnos Is- metrics and multi-metric methods to anthropogenic pressure gradients, in five
72
7 land (very minor pressures) has been classified into high status distinct European coastal and transitional ecosystems. Mar. Pollut. Bull. 62, 73
8 according to the TRIX, chl-a parameter and benthic indices, but 499–513. 74
Borja, A., Basset, A., Bricker, S., Dauvin, J., Elliot, M., Harrison, T., Marques, J.,
9 into good status according to E.I. According to this, we could 75
Weisberg, S., West, R., 2012. Classifying ecological quality and integrity of es-
10 preliminarily assume that probably the E.I. is too sensitive or tuaries. In: Wolanski, E., McLusky, D. (Eds.), Treatise on Estuarine and Coastal 76
11 stringent, resulting in the downgrade of the quality status of Greek Science. Academic Press, Waltham, pp. 125–162. 77
12 Borja, A., Bricker, S.B., Dauer, D.M., Demetriades, N.T., Ferreira, J.G., Forbes, A.T., 78
sites corresponding to minor pressure index values and a future
Hutchings, P., Jia, X., Kenchington, R., Marques, J.C., Zhu, C., 2008. Overview of
13 modification of the high to good boundary of E.I. may be needed in Integrative tools and methods in assessing ecological integrity in estuarine and 79
14 order to demonstrate the high status of some relatively un- coastal systems worldwide. Mar. Pollut. Bull. 56, 1519–1537. 80
15 Borja, A., Dauer, D.M., Elliott, M., Simenstad, C.A., 2010. Medium- and long-term 81
disturbed Greek coastal sites.
recovery of estuarine and coastal ecosystems: patterns, rates and restoration
16 effectiveness. Estuar. Coasts 33, 1249–1260. http://dx.doi.org/10.1007/s12237- 82
17 010-9347-5. 83
18 Borja, A., Elliott, M., Andersen, J.H., Cardoso, A.C., Carstensen, J., Ferreira, J.G., 84
Uncited references
Heiskanen, A.S., Marques, J.C., Neto, J.M., Teixeira, H., Uusitalo, L., Uyarra, M.C.,
19 Zampoukas, N., 2013. Good environmental status of marine ecosystems: what
85
20 European Commission (2000), European Commission (2008), is it and how do we know when we have attained it? Mar. Pollut. Bull. 76, 86
21 Borja and Dauer (2008). 16–27. 87
Borja, Á., Franco, F., Valencia, V., Bald, J., Muxica, I., Belzunce, M.J., Solaun, O., 2004.
22 88
Implementation of the European water framework directive from the Basque
23 country (northern Spain): a methodological approach. Mar. Pollut. Bull. 48 (3– 89
24 Acknowledgments 4), 209–218. 90
25 Caroni, R., Van de Bund, W., Clark, R.T., Johnson, R.K., 2013. Combination of multiple 91
biological quality elements into waterbody assessment of surface waters. Hy-
26 We would like to acknowledge Dr. Angel Borja (AZTI-Tecnalia) drobiologia 704, 437–451. 92
27 for providing access to the excel software sheet for computing the Carpenter, J.H., 1965. The accuracy of the Winkler method for the dissolved oxygen 93
28 analysis. Limnol. Oceanogr. 10, 135–140. 94
Kappa analysis results. Cohen, J., 1960. A coefficient of agreement for nominal scales. Educ. Psychol.
29 We would like to thank the anonymous reviewers for their time Measur. 20, 37–46. 95
30 and their valuable and helpful comments and suggestions for the Devlin, M., Bricker, S., Painting, S., 2011. Comparison of five methods for assessing 96
31 impacts of nutrient enrichment using estuarine case studies. Biogeochemistry 97
manuscript. 106, 177–205.
32 This work was performed under the Water Framework Direc- EEA, 2001. Eutrophication in Europe's Coastal Waters. Topic Report, 7/2001. Eur-
98
33 tive (WFD) implementation within the framework of the National opean Environment Agency, p. 166. 99
34 Q3 Monitoring Project of Greece. The project is financed by European
EEA, 2003. Europe's Water: An Indicator-based Assessment. Topic Report, 1/2003. 100
European Environment Agency, p. 97.
35 Union Funds through the National Strategic Reference Framework 101
European Commission, 2000. Directive 2000/60/EC of the European Parliament and
36 (NSRF). of the Council of 23 October 2000 establishing a framework for Community 102
37 actions in the field of water policy. Off. J. Eur. Commun. L327 (1) 22.12.2000. 103
38 European Commission, 2008. Directive 2008/56/EC of the European Parliament and 104
of the Council of 17 June 2008 establishing a framework for Community actions
39 in the field of marine environmental policy (Marine Strategy Framework Di- 105
Appendix A. Supplementary material
40 rective). Off. J. Eur. Commun. L164/19 25.06.2008. 106
41 Ferreira, J.G., Vale, C., Soares, C.V., Salas, F., Stacey, P.E., Bricker, S.B., Silva, M.C., 107
Supplementary data associated with this article can be found in Marques, J.C., 2007. Monitoring of coastal and transitional waters under the E.U.
42 Water Framework Directive. Environ. Monit. Assess. 135, 195–216. http://dx. 108
the online version at http://dx.doi.org/10.1016/j.csr.2015.05.013.
43 doi.org/10.1007/s10661-007-9643-0. 109
44 Ferreira, J.G., Andersen, J.H., Borja, A., Bricker, S.B., Camp, J., Cardoso da Silva, M., 110
Garcés, E., Heiskanen, A.S., Humborg, C., Ignatiades, L., Lancelot, C., Menesguen,
45 A., Tett, P., Hoepffner, N., Claussen, U., 2010. Marine Strategy Framework Di-
111
References
46 rective—Task Group 5 Report Eutrophication, EUR 24338 EN—Joint Research 112
47 Centre, Joint Report. EUR—Scientific and Technical Research Series—ISSN 1018- 113
Anagnostou, C.H., Drakopoulou, V., Kapsimalis, V., Papathanasiou, E., 2005. An- 5593.
48 thropogenic Activities along the Hellenic Coasts. In: Zenetos, A., Papathanasiou,
114
Ferreira, J.G., Andersen, J.H., Borja, A., Bricker, S.B., Camp, J., Cardoso da Silva, M.,
49 E. (Eds.), State of the Hellenic Marine Environment. HCMR, Athens, Greece, Garcés, E., Heiskanen, A.S., Humborg, C., Ignatiades, L., Lancelot, C., Menesguen, 115
50 pp. 301–352. A., Tett, P., Hoepffnerm, N., Claussen, U., 2011. Overview of eutrophication in- 116
Andersen, J.H., Schlüter, L., Aertebjerg, G., 2006. Coastal eutrophication: recent dicators to assess environmental status within the European Marine Strategy
51 117
developments in definitions and implications for monitoring strategies. J. Framework Directive. Estuar. Coast. Shelf Sci. 93, 117–131.
52 Plankton Res. 28 (7), 621–628. Garmendia, M., Bricker, S., Revilla, M., Borja, A., Franco, J., Bald, J., Valencia, V., 2012. 118
53 Anonymous. Data Dictionary, Definition of WISE-SoEReporting: Transitional, Eutrophication assessment in Basque estuaries: comparing a North American 119
54 Coastal and Marine Waters European Environment Agency, July 2012, 135 pp. and a European method. Estuar. Coasts 35, 991–1006. 120
Aubry, A., Elliott, M., 2006. The use of environmental integrative indicators to assess Garmendia, M., Borja, A., Franco, J., Revilla, M., J., 2013. Phytoplankton composition
55 seabed disturbance in estuaries and coasts: application to the Humber Estuary, indicators for the assessment of eutrophication in marine waters: Present state 121
56 UK. Mar. Pollut. Bull. 53, 175–185. and challenges within the European directives. Mar. Pollut. Bull. 66 7–6. 122
57 Borja, Á., Rodríguez, J.G., 2010. Problems associated with the “one-out, all-out” GIG, 2013. Technical Report Coastal Waters, Mediterranean GIG Benthic In- 123
principle, when using multiple ecosystem components in assessing the ecolo- vertebrate Fauna. JRC and the Benthic Macroinvertebrates Subgroup https://
58 gical status of marine waters. Mar. Pollut. Bull. 60, 1143–1146. circabc.europa.eu/faces/.
124
59 Borja, A., Collins, M., 2009. Regional Seas integrative studies, as a basis for an Halpern, B.S., Selkoe, K.A., Micheli, F., Kappel, C.V., 2007. Evaluating and Ranking the 125
60 ecosystem-based approach to management: the case of the Bay of Biscay. Cont. Vulnerability of Global Marine Ecosystems to Anthropogenic Threats, In: 126
Shelf Res. 29, 951–956. Threats to Marine Ecosystems, Conservation Biology, 21 (5): 1301–1315, Society
61 127
Borja, A., Dauer, D.M., 2008. Assessing the environmental quality status in estuarine for Conservation Biology 10.1111/j.1523-1739.2007.00752.x.
62 and coastal systems: comparing methodologies and indices. Ecol. Indic. 8, Holm-Hansen, O., Lorenzen, C.J., Hormes, R.N., Strickland, J.D.H., 1965. Fluorometric 128
63 331–337. determination of chlorophyll. J. Cons. Perm. Int. Explor. Mer. 30, 3–5. 129
Borja, A., Josefson, A.B., Miles, A., Muxika, I., Olsgard, F., Phillips, G., Germán Ignatiades, L., Vounatsou, P., Karydis, M., 1992. A possible method for evaluating
64 130
Rodríguez, J., Rygg, B., 2007. An approach to the intercalibration of benthic oligotrophy and eutrophication based on nutrient concentration scales. Mar.
65 ecological status assessment in the North Atlantic ecoregion, according to the Pollut. Bull. 24, 238–243. 131
66 European Water Framework Directive. Mar. Pollut. Bull. 55, 42–52. Jickells, T., 2005. External inputs as a contributor to eutrophication problems. J. Sea 132
1 Res. 54, 58–69. Aegean). Fresen. Environ. Bull. 13, 1509–1514. 67

2 Justic, D., Rabalais, N.N., Turner, R.E., 2005. Coupling between climate variability Pavlidou, A., Kontoyiannis, H., Zarokanelos, N., Hatzianestis, I., Assimakopoulou, G., 68
and coastal eutrophication: evidence and outlook for the northern Gulf of Psyllidou-Giouranovits, R., 2014. Seasonal and spatial nutrient dynamics in
3 Mexico. J. Sea Res. 54, 25–35. Saronikos Gulf: the impact of sewage effluents from Athens sewage. In: 69
4 Karageorgis, A.P., Skourtos, M.S., Kapsimalis, V., Kontogianni, A.D., Skoulikidis, N.Τh, Treatment Plant, I.N., Ansari, A.A., Gill, S.S. (Eds.), Eutrophication: Causes, 70
5 Pagou, K., Nikolaidis, N.P., Drakopoulou, P., Zanou, B., Karamanos, H., Levkov, Z., Consequences and Control. Springer Science þBusiness Media Dordrecht, 71
Anagnostou, Ch, 2005. An integrated approach to watershed management p. 2014. http://dx.doi.org/10.1007/978-94-007-7814-6_10.
6 within the DPSIR Framework: Axios River catchment and Thermaikos Gulf. Reg. Primpas, I., Karydis, M., 2011. Scaling the trophic index (TRIX) in oligotrophic
72
7 Environ. Change 5, 138–160. marine environments. Environ. Monit. Assess. 178, 257–269. http://dx.doi.org/ 73
8 Karydis, M., 1999. Evaluation of the trophic levels in Greek coastal ecosystems. 10.1007/s10661-010-1687-x. 74
Lesvos Island. University of Aegean, Scientific Report, Greece. Primpas, I., Tsirtsis, G., Karydis, M., Kokkoris, G.D., D., 2010. Principal component
9 Karydis, M., Kitsiou, D., 2013. Marine water quality monitoring: a review. Mar. analysis: development of a multivariate index for assessing eutrophication
75
10 Pollut. Bull. 77, 23–36. according to the European water framework directive. Ecol. Indic. 10, 178–183. 76
11 Konstantinou, Z.I., Krestenitis, Y.N., Latinopoulos, D., Pagou, K., Galinou-MitsoudiS. Prins, T., Van der Meulen, M., Boon, A., Simboura, N., Tsangaris, C., Borja., A., 77
and Savvidis, Y., 2012. Aspects of mussel-farming activity in Chalastra, Ther- Menchaca, I., 2013. Coherent geographic scales and aggregation rules in as-
12 78
maikos Gulf, Greece: an effort to untie a management Gordian Knot, Ecology sessment and monitoring of Good Environmental Status, Analysis and con-
13 and Society, 17(1): 1. 10.5751/ES-04455-170101. ceptual phase, Deltares—The Netherlands November 2013. Analytical Report 79
14 Koroleff, F., 1970. Revised version of “direct determination of ammonia in natural under Framework Contract No. NV.D2/FRA/2012/0019 http://ec.europa.eu/en 80
15 waters as indophenol blue I.C.E.S., C.M. 1970/C: 9. ICES information on Tech- vironment/marine/publications/. 81
niques and Methods for the sea water analysis. Interlab. Rep., 19–22. Revilla, M., Franco, J., Bald, J., Borja, A., Laza, A., Seoane, S., Valencia, V., 2009. As-
16 Koukaras, K., Nikolaidis, G., 2004. Dinophysis blooms in Greek coastal waters sessment of the phytoplankton ecological status in the Basque coast (northern 82
17 (Thermaikos Gulf, NW Aegean Sea). J. Plankton Res. 26, 445–457. Spain) according to the European Water Framework Directive. J. Sea Res. 61, 83
18 Krom, M.D., Emeis, K.-C., Van Cappellen, P., 2010. Why is the Eastern Mediterranean 60–67. 84
phosphorus limited? Prog. Oceanogr. 85, 236–244. Redfield, A.C., 1958. The biological control of chemical factors in the environment.
19 Landis, J.R., Kosch, G.G., 1977. The measurement of observer agreement for cate- Am. Sci. 46, 205–221. 85
20 gorical data. Biometrics 33, 159–174. Reizopoulou, S., Strogyloudi, E., Giannakourou, A., Hatzianestis, J., Varkitzi, I., 86
21 Levin, N., Coll, M., Fraschetti, S., Gal, G., Giakoumi, S., Göke, C., Heymans, J.J., Kat- Graneli, E., 2004. Accumulation of Okadaic acid in shellfish from Thermaikos 87
sanevakis, S., Mazor, T., Öztürk, B., Rilov, G., Gajewski, J., Steenbeek, J., Kark, S., Gulf (Eastern Mediterranean) in: Proceedings of the 11th International Con-
22 2014. Biodiversity data requirements for systematic conservation planning in ference on Harmful Algae.
88
23 the Mediterranean Sea. Mar. Ecol. Prog. Ser. 508, 261–281. http://dx.doi.org/ Simboura, N., Panayotidis, P., Papathanassiou, E., 2005. A synthesis of the biological 89
24 10.3354/meps10857. quality elements for the implementation of the European Water Framework 90
Monserud, R., Leemans, R., 1992. Comparing global vegetationmaps with the Kappa Directive in the Mediterranean ecoregion: the case of Saronikos Gulf. Ecol. In-
25 statistic. Ecol. Model. 62, 275–293. dic. 5, 253–266.
91
26 Mullin, J.B., Riley, J.P., 1955. The colorimetric determination of silicate with special Simboura, N., Zenetos, A., Pancucci-Papadopoulou, M.A., Reizopoulou, S., Streftaris, 92
27 reference to sea and natural waters. Anal. Chim. Acta 12, 162–176. N., 2012. Hellenic Seas indicators for sea-floor integrity under the European 93
Murphy, J., Riley., J.P., 1962. A modified solution method for determination of Marine Strategy Framework Directive: setting the thresholds and standards for
28 94
phosphate in natural waters. Anal. Chim. Acta 27, 31–36. Good Environmental Status. Mediterr. Mar. Sci. 13 (1), 140–152.
29 Newton, A., Icely, J., Falcão, M., Nobre, A., Nunes, J.P., Ferreira, J.G., Vale, C., 2003. Simboura, N., Tsapakis, M., Pavlidou, A., Assimakopoulou, G., Pagou, K., Kon- 95
30 Evaluation of eutrophication in the Ria Formosa coastal lagoon, Portugal. Cont. toyiannis, H., Zeri, Ch, Krasakopoulou, E., Rousselaki, E., Katsiaras, N., Diliberto, 96
31 Shelf Res. 23, 1945–1961. S., Naletaki, M., Tsiamis, K., Gerakaris, V., Drakopoulou, P., Panayotidis, P., 2015. 97
Newton, A., Icely, J., Cristina, S., Brito, A., Cristina Cardoso, A.C., Colijn, F., Dalla Riva, Assessment of the environmental status in Hellenic coastal waters (Eastern
32 S., Gertz, F., Hansen, J.W., Holmer, M., Ivanova, K., Leppäkoski, E., Donata Melaku Mediterranean): from the Water Framework Directive to the Marine Strategy 98
33 Canu, D.M., Mocennim, C., Mudge, S., Murray, N., Morten Pejrup, M., Razinko- Water Framework Directive. Mediterr. Mar. Sci. 16 (1), 46–64. 99
34 vas, A., Sofia Reizopoulou, S., Pérez-Ruzafa, A., Schernewski, G., Schubert, H., Simboura, N., Zenetos, A., 2002. Benthic indicators to use in ecological quality 100
Carr, L., Cosimo Solidoro, C., PierluigiViaroli, P., Zaldívar, J.M., 2014. An overview classification of Mediterranean soft bottom marine ecosystems, including a
35 of ecological status, vulnerability and future perspectives of European large new biotic index. Mediterr. Mar. Sci. 3 (2), 77–111. 101
36 shallow, semi-enclosed coastal systems, lagoons and transitional waters. Estuar. Skogen, M.D., Søiland, H., Svendsen, E., 2004. Effects of changing nutrient loads to 102
37 Coast. Shelf Sci. 140, 95–122. the North Sea. J. Mar. Syst. 46, 23–38. 103
O’Higgins, T.G., Gilbert, A.J., 2014. Embedding ecosystem services into the Marine Strickland, J.D., Parsons, T.R., 1977. A Practical Handbook of Sea Water Analysis. 167.
38 Strategy Framework Directive: illustrated by eutrophication in the North Sea. Fisheries Research Board of Canada, p. 310.
104
39 Estuar. Coast. Shelf Sci. 140, 146–152. UNEP, 2007. Approaches to the Assessment of Eutrophication in Mediterranean 105
40 Orfanidis, S., Panayotidis, P., Stamatis, N., 2003. An insight to the ecological eva- Coastal Waters, 98. 106
luation index (EEI). Ecol. Indic. 3, 27–33. Van Hoey, G., Borja, A., Birchenough, S., Buhl-Mortensen, L., Degraer, S., Fleischer,
41 107
Orfanidis, S., Panayotidis, P., Stamatis, N., 2001. Ecological evaluation of transitional D., Kerckhof, F., Magni, P., Muxika, I., Reiss, H., Schrӧder, A., Zettler, M.L., 2010.
42 and coastal waters: a marine benthic macrophytes model. Mediterr. Mar. Sci. 2 The use of benthic indicators in Europe: From the Water Framework Directive 108
43 (2), 46–65. to the Marine Strategy Framework Directive. Mar. Pollut. Bull. 60, 2187–2196. 109
Orfanidis, S., Panayotidis, P., Ugland, K.I., 2011. Ecological Evaluation Index con- Varkitzi, I., Pagou, K., Hatzianestis, I., Pyrgaki, C., Pavlidou, A., Asimakopoulou, G.,
44 110
tinuous formula (EEI-c) application: a step forward for functional groups, the Montesanto, B., Εkonomou-AmilliA., Graneli, E., 2006. Effects of nutrients
45 formula and reference condition values. Mediterr. Mar. Sci. 12 (1), 199–231. availability on the growth and toxicity of the dinoflagellate Prorocentrum lim. 111
46 Pagou, K., 2005. Eutrophication in Hellenic coastal areas. In: Papathnasiou, E., Ze- In: Proceedings of 8th Symposium of Oceanography and Fisheries, Thessaloniki, 112
47 netos, A. (Eds.), State of the Hellenic Marine Environment. HCMR, pp. 127–136. Greece. 113
Pagou, K., Siokou-Frangou, I., Papathanassiou, E., 2002. Nutrients and their ratios in Varkitzi, I., Pagou, K., Graneli, E., Hatzianestis, I., Pyrgaki, Ch, Pavlidou, A., Mon-
48 relation to eutrophication and HAB occurrence. The Case of Eastern Medi- tesanto, B., Economou-Amilli, A., 2010. Unbalanced N:P ratios and nutrient 114
49 terranean Coastal Waters Second Workshop on Thresholds of Environmental stress controlling growth and toxin production of the harmful dinoflagellate 115
50 Sustainability: The Case of Nutrients,18–19 June 2002, Brussels, Belgium. Prorocentrum lima (Ehrenberg) Dodge. Harmful Algae 9 (3), 304–311. 116
Pascual-Aguilar, J., Andreu, V., Gimeno-García, E., Picó, Y., 2014. Current anthro- Vollenweider, R.A., Giovanardi, F., Montanari, G., Rinaldi, A., 1998. Characterization
51
Q4 pogenic pressures on agro-ecological protected coastal wetlands. Sci. Total of the trophic conditions of marine coastal waters with special reference to the 117
52 Environ. 503, 190–199. http://dx.doi.org/10.1016/j.scitotenv.2014.07.007. NW Adriatic Sea. Proposal for a trophic scale, turbidity and a generalized water 118
53 Pavlidou, A., Rousselaki, E., 2014. Impacts of aquacultures on the waer column of quality index. Environmetrics 9, 329–357. 119
Hellenic coastal ecosystems: dissolved oxygen and nutrients. Fresen. Environ. Welch, P.S., 1948. Limnological Methods, Blankiston, Philadelphia, 381.
54 Bull. 23 (12b), 3325–3330. Welschmeyer, N., 1994. A method for the determination of chlorophyll a in the
120
55 Pavlidou, A., 2012. Nutrient distribution in selected coastal areas of Aegean Sea presence of chlorophyll b and pheopigments. Limnol. Oceanogr. 39, 1985–1992. 121
56 (East Mediterranean Sea). J. Environ. Sci. Eng. 1 (Α1), 78–88. Xu, J., Ho, A.Y.T., Yin, K., Yuan, X., Anderson, D.M., Lee, J.H.W., Harrison, P.J., 2008. 122
Pavlidou, A., Souvermezoglou, E., 2006. Nutrients in the Mediterranean Sea. In: Temporal and spatial variations in nutrient stoichiometry and regulation of
57 123
Karageorgis, A. (Ed.), Pollution in the Mediterranean Sea: An overview from phytoplankton biomass in Hong Kong waters: influence of the Pearl River
58 IASON SSA. HCMR, pp. 5–12. outflow and sewage inputs. Mar. Pollut. Bull. 57, 335–348. 124
59 Pavlidou, A., Kontoyiannis, H., Psyllidou-Giouranovits, R., 2004. Trophic conditions 125
and stoichiometric nutrient balance in the Inner Saronikos Gulf (Central
60 126
61 127
62 128
63 129
64 130
65 131
66 132

Continental Shelf Research

Uploaded by

Copyright:

Available Formats

Continental Shelf Research

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Continental Shelf Research

Uploaded by

Copyright:

Available Formats

Continental Shelf Research ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 Contents lists available at ScienceDirect

A. Pavlidou et al. / Continental Shelf Research ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 sewage discharges. According to those correlations, heavy in- 67

1 Res. 54, 58–69. Aegean). Fresen. Environ. Bull. 13, 1509–1514. 67

You might also like