Marine Pollution Bulletin 93 (2015) 144–152

Contents lists available at ScienceDirect

Marine Pollution Bulletin

journal homepage: www.elsevier.com/locate/marpolbul

Biodegradation of dispersed Macondo oil in seawater at low temperature

and different oil droplet sizes
Odd G. Brakstad ⇑, Trond Nordtug, Mimmi Throne-Holst
SINTEF Materials and Chemistry, Environmental Technology, Dept. Applied Environmental Biology and Chemistry, N-7465 Trondheim, Norway

a r t i c l e i n f o a b s t r a c t

Article history: During the Deepwater Horizon (DWH) accident in 2010 a dispersant (Corexit 9500) was applied at the
Available online 5 March 2015 wellhead to disperse the Macondo oil and reduce the formation of surface slicks. A subsurface plume
of small oil droplets was generated near the leaking well at 900–1300 m depth. A novel laboratory system
Keywords: was established to investigate biodegradation of small droplet oil dispersions (10 lm or 30 lm droplet
Oil sizes) of the Macondo oil premixed with Corexit 9500, using coastal Norwegian seawater at a
Biodegradation temperature similar to the DWH plume (4–5 °C). Biotransformation of volatile and semivolatile hydrocar-
Dispersions
bons and oil compound groups was generally faster in the 10 lm than in the 30 lm dispersions, showing
Seawater
Low temperature
the importance of oil droplet size for biodegradation. These data therefore indicated that dispersant
Droplet size treatment to reduce the oil droplet size may increase the biodegradation rates of oil compounds in the
deepwater oil droplets.
Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction et al., 2010; Redmond and Valentine, 2012; Dubinsky et al.,

2013), and data from the field and from laboratory experiments
During the Deepwater Horizon (DWH) incident some of the oil indicated short biodegradation half-lives for nC13–nC26 alkanes,
was dispersed as small droplets in the deep ocean at depths of ranging from 1 to 8 days (Hazen et al., 2010). A sequential and
900–1300 m near the leaking well, and at water temperatures pulsed propagation of bacteria degrading specific hydrocarbons
of 4–6 °C (Camilli et al., 2010; Reddy et al., 2012). Simulations of was suggested, including n-alkanes, aromatics and C1–C3 gas com-
the plume trajectories predicted that oil droplets with diameters ponents (Valentine et al., 2012; Dubinsky et al., 2013). Results from
between 10 and 50 lm would form in a distinct subsurface layer stable isotope analyses reported that close to 70% of the oxygen
and be transported horizontally, while droplets larger than consumption in the deepwater plume was related to bacterial
90 lm rapidly rose to the surface (North et al., 2011). Samples from respiration of propane and ethane (Valentine et al., 2010), while
the deepwater plume showed enrichment of water-soluble compo- methane oxidation was delayed (Valentine et al., 2010; Kessler
nents (Camilli et al., 2010). Measurements of PAH in water samples et al., 2011).
during the release showed rapid decrease with distance from the Biodegradation rates of components in dispersed oil are nor-
release point. The responders began injecting the chemical disper- mally expected to increase when oil droplet size is reduced, for
sant Corexit 9500A at the wellhead in early May 2010 to reduce the instance by treatment of the oil with chemical dispersants. Smaller
amount of oil that formed surface slicks and reached the Gulf of droplets will generate larger oil–water interphase areas for the oil-
Mexico (GoM) coastlines (Atlas and Hazen, 2011; Kujawinski degrading bacteria to attach to, and dissolution of water-soluble
et al., 2011). In addition, surface operations included mechanical compounds will increase. Volatile oil compounds will rapidly dis-
recovery techniques and application of the chemical dispersants solve during subsurface spills and thereby become subject to
Corexit 9527 and 9500A (Coastal Response Research Center, 2010). biodegradation. In contrast volatiles will largely evaporate after
It was early observed that the deepwater plume stimulated surface spills. Several laboratory studies have shown that disper-
bacteria associated with hydrocarbon biodegradation in cold sant treatment of dispersible oils increase oil compound biodegra-
seawater (Oceanospirillaceae, Pseudomonas, Cycloclasticus, Colwellia, dation (Siron et al., 1995; Venosa and Holder, 2007; Prince et al.,
Pseudoalteromonas, Methylophaga) (Valentine et al., 2010; Hazen 2013; McFarlin et al., 2014; Brakstad et al., 2014).
To our knowledge, no laboratory studies have attempted to
relate the biodegradation of oil compounds to measured droplet
⇑ Corresponding author. Tel.: +47 98243447; fax: +47 73597043. sizes in the dispersions. We therefore established a novel
E-mail address: odd.brakstad@sintef.no (O.G. Brakstad).

http://dx.doi.org/10.1016/j.marpolbul.2015.02.006
0025-326X/Ó 2015 Elsevier Ltd. All rights reserved.
 O.G. Brakstad et al. / Marine Pollution Bulletin 93 (2015) 144–152 145

experimental system for comparing oil dispersions of different seawater (pre-incubated over night at 5 °C) in 2-L flasks complete-
small droplet sizes in natural seawater. This system was based ly filled (no air bubbles) with the dispersions, with incubation in
on combining an oil dispersion generator (Nordtug et al., 2011) the carousel system at 5 °C for 5 days (rotation 0.75 r.p.m.). The
and a slowly rotating carousel system to generate small droplet experiment was then started by generating dispersions in the
dispersions and avoid droplets rising to the surface. The primary pre-adapted seawater with median oil droplet sizes of 10 lm or
aim of our study was therefore to quantify biodegradation of 30 lm and nominal oil concentration of 2 mg/L from a stock oil dis-
hydrocarbons in dispersions with defined oil droplet sizes relevant persion (200 mg/L oil) in natural unfiltered seawater. Dispersions
for chemically dispersed oil, and at low seawater temperatures, of 10 lm or 30 lm median oil droplet sizes and 2 mg/L nominal
corresponding to deepwater discharges. Biodegradation of concentrations were also prepared in sterile-filtered (0.2 lm) sea-
chemically dispersed Macondo oil at two median oil droplet sizes water (incubated at 4–5 °C overnight before use) supplemented
(10 and 30 lm) in natural seawater at 4–5 °C was investigated with HgCl2 (100 mg/L), to minimize bacterial growth. Immediately
with this system. Norwegian coastal seawater was used in the after generation of dispersions flask were mounted on the carousel
study, but experimental conditions (oil, oil droplet size distribu- at slow rotation (0.75 r.p.m.). Flasks with experimental blank sam-
tions and seawater temperature) were relevant to the DWH deep- ples (unfiltered seawater without oil) were also included in the
water plume. experiment for background analyses. Experiments were conducted
at 5 °C for up to 64 days. Flasks (triplicate samples; 2.3 L sample) of
dispersions in natural seawater were sacrificed for analyses after
2. Materials and methods
15 min on the carousel (0-day samples) and after 2, 4, 8, 16, 32,
or 64 days of incubation on the carousels. One sterilized control
2.1. Oil, dispersant and seawater used in experiments
flask and one experimental blank flask were also sacrificed for ana-
lyses at the each sampling point.
Unweathered Macondo (MASS) oil was collected from a riser
pipe connected to the well on 22 May 2010. COREXIT 9500A (Nal-
2.4. Analyses
co, Sugar Land, TX) dispersant was used at a dispersant-to-oil ratio
(DOR) of 1:100. Seawater was collected from a depth of 80 m (be-
2.4.1. Oil droplet analyses
low thermocline) in a Norwegian fjord (Trondheimsfjord; 63°260 N,
Oil droplet concentrations and size distributions were
10°230 E), outside the harbor area of Trondheim, the water is sup-
determined by Coulter Counter measure (Beckman Multisizer 4;
plied via a pipeline system to our laboratories. The water source
Beckman Coulter Inc., Brea, CA, USA) fitted with either 100 lm or
is considered to be non-polluted and not influenced by seasonal
280 lm apertures, for measurement of droplets within a diameter
variations. The salinity of the seawater was 34‰, with a water
range of 2–60 lm or 5.6–100 lm, respectively. Filtered (0.22 lm)
temperature of 6–8 °C and dissolved oxygen (DO) of approximately
seawater was used as electrolyte. All median droplet sizes reported
8 mg/L when reaching our laboratory (Brakstad et al., 2004).
here are expressed as median droplets diameter on droplet volume
if not otherwise mentioned.
2.2. Oil dispersion generator and carousel system
2.4.2. Chemical analyses
An oil droplet generator was used for preparing oil dispersions Samples of dispersions and seawater were solvent–solvent
with defined droplet size distributions, as previously described extracted with dichloromethane (DCM) for measurements of
(Nordtug et al., 2011). The system consists of 3 chambers with semivolatile organic compounds (SVOC) by gas chromatographic
an inner diameter of 8 mm connected by nozzles with a diameter methods. The flask glass walls were also rinsed with DCM after
of 0.5 mm. Oil dispersions with defined oil droplet size distribu- removal of dispersions to extract material attached to the glass
tions and concentrations are generated by injecting oil via a capil- walls. All extracts were pooled before analyses. A gas chro-
lary into a flow of seawater which moves through the chambers matograph coupled to a flame ionization detector (GC-FID; Agilent
(Nordtug et al., 2011). Oil droplet concentrations and size distribu- 6890 N with 30 mDB1 column; Agilent Technologies) was used for
tion were determined by Coulter Counter analyses (see below). quantification of extracted C10–C36 saturates separated according to
Flasks (2 L; Schott) were baked (450 °C; 3 h) and soaked overnight boiling points, termed as total petroleum hydrocarbons extractable
(4% Deconex 11 Universal, Borer, Switzerland), washed with deter- organic carbon (TEOC), while 96 individual targeted compounds
gent (Neodisher N, Dr. Weigert, Germany), and thoroughly rinsed (C10–C36 n-alkanes, decalines, phenols, 2- to 5-ring polyaromatic
and autoclaved (121 °C, 20 min) before use. hydrocarbons (PAH) and 17a(H),21b(H)-Hopane (30ab Hopane)
The flasks were completely filled with oil dispersions (no head- were analyzed in a gas chromatograph coupled to a mass spec-
space or air bubbles) and mounted on a carousel system. During trometer (GC–MS; Agilent 6890 plus GC coupled with an Agilent
mounting of flasks and biodegradation experiments the wheels 5973 MSD detector, operated in Selected Ion Monitoring [SIM]
were slowly rotated around the carousel axis (0.75 r.p.m.) by a gear modus; Agilent Technologies), as recently described (Brakstad
motor (SEW Eurodrive, Moss, Norway). In the main biodegradation et al., 2014). Dispersions were acidified (pH < 2) for analyses of 87
experiment described here, four carousels with double sets of targeted volatile organic carbon (VOC) compounds in a Purge & Trap
wheels, holding up to 16 flasks each, were used. unit (Teledyne Tekmar Atomx; Mason OH, USA.) coupled to a GC–
MS (Agilent 6890 N GC and an Agilent 5975B MSD detector; Agilent
2.3. Biodegradation experiment Technologies)(P&T GC–MS). The VOC and SVOC target compounds
included in GC–MS analyses are shown in Table S1. The response
The MASS oil was heated at 50 °C for 30 min, mixed by shaking values for individual target analytes were determined, with a sig-
for 3 min, and cooled to room temperature, after which it was nal-to-noise ratio of 10 as the lower detection limit, and a lower
mixed with the dispersant Corexit 9500A (DOR 1:100) at room limit of detection (LOD) of 0.01 lg/L was defined for individual oil
temperature. This dispersant was shown to effectively disperse compounds.
the Macondo oil (Daling et al., 2014). Pre-adaption of the seawater
prior to the biodegradation experiments was conducted by diluting 2.4.3. Temperature, dissolved oxygen, nutrients, and biomass
a stock dispersion (200 mg/L oil; oil droplet size 10 lm) to 0.2 mg/L Air and seawater temperatures were logged through the
Macondo oil (measured by Coulter Counter) in natural unfiltered experimental period. Dissolved oxygen was measured by a
146 O.G. Brakstad et al. / Marine Pollution Bulletin 93 (2015) 144–152

dissolved oxygen meter (YSI, Inc., Yellow Springs, OH). Nutrient sions over time. All experiments in the carousels were conducted
analyses included NO3–N + NO2–N (ISO 13395), NH4–N (ISO in air-tight flasks without headspace to avoid evaporative deple-
11732), o-PO4–P (ISO 15681-2), and total Phosphorous (ISO tion. The system was able to handle dispersions of fresh Macondo
15681-2), analyzed by Eurofins Environment Testing Norway, oil with small droplets, in the diameter size range of 10–30 lm,
Bergen, Norway. Microbial biomass concentrations were deter- with rising velocities of 1.6–14.4 cm/h, according to Stoke’s law.
mined by epifluorescence microcopy analyses of samples stained We believe the ability of keeping the droplets in suspension is
by the nucleic acid stain 40 ,6-diamidino-2-phenylindol (Porter due to the combined effect of the turbulence created in the flasks
and Feig, 1980). and the fact that the oil droplet rising velocities will make them
constantly change direction in relation to any fixed point at the
2.5. Calculations and statistics flasks inner surface. Larger than 30 lm oil droplets resulted in
inhomogeneous dispersions during transfer from dispersion
Non-linear regression analyses were performed by a 1st order generator to flasks in carousel system because of elevated rising
rate approach with determination of lag-phases included using velocities. In initial experiments with 10 lm median volumetric
the option ‘‘plateau followed by one-phase decay’’ in GraphPad droplet dispersions in a sterilized system (filtered seawater poi-
Prism vs. 6.0 (GraphPad Software Inc., La Jolla, CA, USA). Rate coef- soned with HgCl2), the droplet sizes were maintained for at least
ficients (k1) were determined for the decay-period, the plateau 28 days in the carousel system, while measured droplet concentra-
period defined the lag-phase, and half-lives were determined from tion decreased (see Supplementary Information, Fig. S1).
rate coefficients and plateau periods (t1/2 = plateau period + 0.693/
k1). Rate data were determined for both single oil components and 3.2. Biodegradation experiment with two dispersions of different oil
for 22 oil compound groups described in the Oil-Spill Contingency droplet size distributions
and Response model OSCAR (Reed et al., 2000). The group compo-
sition is described in Table S1. An oil dispersion biodegradation experiment was performed at
Column statistics were compared by Wilcoxon matched paired 5 °C (64 days) with nominal oil concentrations of 2 mg/L generated
test in GraphPad Prism vs. 6. in the dispersion generator at median oil droplet sizes on volume
of 10 lm and 30 lm in diameter, and with natural seawater
3. Results and discussion pre-adapted to the oil and temperature (see Section 2). The low
oil concentration used in this experiment is consistent with recent
3.1. Generation and maintenance of oil dispersions over time recommendations to avoid too high oil loading during biodegrada-
tion studies of dispersed oil (Lee et al., 2013). Seawater pre-adap-
Dispersions of Macondo oil mixed with Corexit 9500A (DOR of tion was included since many offshore oil production areas may
1:100) were generated by different seawater flow-rates through be exposed to hydrocarbons through natural seepage or produced
the oil droplet generator system, from 50 to 150 ml/min. This water releases. For instance, a recent hydrocarbon seepage flux of
resulted in oil dispersions with median droplet size distributions 2.2–18  107 tons per year has been estimated for the GoM region,
of approximately 10 lm to 30 lm in diameter (Fig. 1). The oil corresponding to 14–120% of the DWH discharge rate (Smith et al.,
pump of the dispersion generator was set to generate fixed nom- 2014). Analyses of macronutrients (NO3 + NO2–N, NH4–N and
inal oil concentrations during the dispersion generation (0.2– o-PO4–P) and dissolved oxygen showed no apparent limitations
200 mg oil per L of water), and good agreement between nominal in oxygen or inorganic nutrient levels during the biodegradation
oil concentrations and measured oil droplet concentrations was experiment (Supplementary Information, Figs. S2 and S3).
recorded (data not shown). Recent studies have shown that nearly
the same oil droplet size distribution is generated in this system 3.2.1. Oil droplet sizes and concentrations
both with and without premixing of oil with chemical dispersant The oil droplet size distributions and concentrations were
(Nordtug et al., 2011). followed in the dispersions during the biodegradation experiment
The carousel system was developed for maintaining the droplet (Fig. 2). Results from Coulter Counter analyses showed that the ini-
size distribution in the oil dispersions while incubating the disper- tial median oil droplet size distributions in the 10 lm and 30 lm
dispersions were 11.0 ± 0.04 lm and 29.4 ± 3.5 lm, and that the
10 lm and 30 lm dispersions represented oil surfaces of
Droplet size distribution 10.9 ± 0.2  105 and 5.4 ± 0.9  105 lm2/ml, respectively. The dis-
1.5 persions with the smaller droplets therefore contained
approximately twice the surface area of the dispersions with the
larger droplet. Assuming a microbial size of 1 lm in diameter
and the maximum microbial density of 1 cell/lm2 of oil droplet
150 ml/min
Diff. volume (%)

1.0 surface area (Horowitz et al., 1975; Vilcaez et al., 2013) the initial
125 ml/min oil droplet concentrations represented sufficient surface areas for
colonization of up 10  105 cells/ml in the 10 lm dispersions and
100 ml/min
5  105 cells/ml in the 30 lm dispersions. Initial cell concentra-
75 ml/min tions determined by epifluorescence microscopy were 8  105
0.5
50 ml/min and 12  105 cells/ml in the 10 lm and 30 lm dispersions, respec-
tively. The oil droplets therefore represented surface areas for
attachment of the entire bacterial population in the 10 lm disper-
sions and approximately 40% of the population in the 30 lm dis-
0.0
persions at the start of the experiment.
10 20 30 40 50 60
The oil droplet size distributions during the biodegradation
Droplet size (µm) period (Fig. 2A) were maintained in the 10 lm dispersions (aver-
Fig. 1. Average oil droplet size distributions generated in the oil dispersion
age median droplet size of 3 replicates ranging from 10.7 to
generator at different seawater flow-rates after premixing of the oil with Corexit 13.4 lm in the natural unfiltered seawater and 8.5–11.1 lm in
9500A. sterile dispersions). However, the 30 lm dispersions in natural
 O.G. Brakstad et al. / Marine Pollution Bulletin 93 (2015) 144–152 147

Median oil drplet size (µm) A Median oil droplet sizes B Oil droplet concentrations

Median oil drplet size (µm)

2.5
40
2.0
30
1.5
20 10 µm - unfiltered SW
10 µm - unfiltered SW 1.0
10 µm - sterilized SW 10 µm - sterlized SW
10 30 µm - unfiltered SW 0.5 30 µm - unfiltered SW
30 µm - sterilized SW 30 µm - sterlized SW
0 0.0
0 20 40 60 80 0 20 40 60 80
Time (days) Time (days)

Fig. 2. Median oil droplet sizes (A) and concentrations (B) in oil dispersions generated during a biodegradation period of 64 days at 5 °C in natural unfiltered seawater
(unfiltered SW), or in filtered (0.1 lm) and sterilized (HgCl2) seawater (sterilized SW). The dispersions were generated at nominal particle sizes of 10 and 30 lm diameter. The
dispersions in unfiltered seawater were analyzed in triplicate (error bars show standard deviation), while sterile samples were analyzed as single samples.

seawater experienced a rapid decline in droplet size between days phenanthrenes, dibenzothiophenes, fluoranthenes and chrysenes
0 and 8, from 29.4 to 10.4 lm in average diameter (3 replicates), were examined in the dispersions after normalization against the
with a subsequent particle size between 10.2 and 13.4 lm median persistent biomarker 17a(H),21b(H)-Hopane (30ab Hopane), com-
in the period from day 8 to 64. In contrast, the median oil droplet monly used in biodegradation studies (Prince et al., 1994; Page
diameter was maintained in the sterile 30 lm dispersions, ranging et al., 1995; Wang et al., 1998). The depletions were determined
from 27.2 to 34.9 lm during the experimental period (Fig. 2A). as percentages of normalized data at the start of the experiment
When oil droplet concentrations were determined by Coulter (day 0) for each target oil compound (Fig. 3). In average, the oil
Counter analyses the concentrations declined more rapidly in nat- compound groups showed low to negligible depletion in sterilized
ural unfiltered seawater than in the sterilized dispersions. In the dispersions during the first 32 days (average <1%), although
10 lm dispersions the concentrations were decreased from days decaline concentrations were reduced by more than 20%. After
0 to 8 by 41.9% (from 1.60 to 0.93 mg/L average values) in natural 64 days the oil compounds were further depleted in the sterilized
unfiltered seawater and by 29.2% (from 1.44 to 1.02 mg/L) in the dispersions (average reduction of 19%). However, compounds in
sterilized dispersions, while the reductions in the 30 lm disper- natural unfiltered seawater were mostly reduced to near comple-
sions were 86.9% (from 1.72 mg/L to 0.22 mg/L) and 35.0% (from tion before they were depleted in the sterilized dispersions. The
1.43 mg/L to 0.93 mg/L) in unfiltered and sterilized seawater, oil compound depletion in natural unfiltered seawater was
respectively (Fig. 2B). In the sterilized systems with 30 lm oil dro- therefore mainly the result of biotransformation.
plets it was evident that the larger droplets were removed faster Low-range (nC10–nC15) and medium-range (nC16–nC21) mole-
than the smaller droplets (Supplementary Information, Fig. S4D). cular weight (MW) n-alkanes were rapidly biotransformed after
Reduced concentrations of droplets may therefore be caused by 4 days in the 10 lm dispersions (76.3 ± 2.7 and 74.5 ± 1.9% deple-
removal of the larger droplets with the fastest rising velocities or tion, respectively), while biotransformation of these n-alkanes
by droplet coalescence, generating droplets of sizes outside the started later in the 30 lm dispersions and did not become evident
measuring range of the Coulter Counter instrument (5.6– until 8 days (84.2 ± 4.6 and 82.2 ± 3.4% degradation, respectively)
100 lm), as shown in oil droplet coalescence models (Sterling in the 30 lm dispersions (Fig. 3A). Faster biotransformation of
et al., 2004). The 10 lm dispersions also included oil droplets the low-range and medium-range MW n-alkanes was therefore
smaller than the lower instrumental measuring size of 5.6 lm measured in the 10 lm than the 30 lm dispersions. Also decaline
(Supplementary Information, Figs. S4A and S4B). In addition, oil biotransformation started faster in the smaller droplet dispersions,
droplets could be removed from the system by attachment to the between days 8 and 16 in 10 lm dispersions (60.6 ± 3.6% degrada-
glass walls of the flasks, as measured by chemical analyses (see tion), and between days 16 and 32 (88.7 ± 9.5% degradation at
below). The higher reduction in oil droplet concentrations in unfil- 32 days) in the 30 lm dispersions (Fig. 3A). However, high-range
tered than sterilized dispersions may be caused by integration of MW n-alkanes (nC22–nC36) shower slower and more similar bio-
droplets in ‘‘clusters’’ or ‘‘flocs’’ in natural seawater, which were transformation in the two dispersions (73.0 ± 4.8 and 63.5 ± 0.5%
visually observed in the experiment. These aggregates have been degradation in the 10 lm and 30 lm dispersions after
observed in several biodegradation studies of dispersed oil (e.g. 16 days)(Fig. 3A).
Swannell et al., 1997; Harris et al., 2002; MacNaughton et al., Low-range and medium-range MW PAH-compounds (2- to 3-
2003; Bælum et al., 2012). ring PAH) were biotransformed between day 8 and 16 in both
dispersions, with 99.7 ± 0.2 and 88.9 ± 9.0% degradation after
3.2.2. Biodegradation of targeted oil compounds 16 days in the 10 and 30 lm dispersions, respectively (Fig. 3B).
Concentrations of 183 targeted semivolatile or volatile oil com- Also for these PAH compounds faster biodegradation in the
pound measured by GC–MS analyses showed that 81.4% of these 10 lm than the 30 lm oil droplet dispersions was indicated. How-
were detected in the dispersions above the LOD (0.01 lg/L) at ever, larger PAH (4- to 5-ring) were only biotransformed between
the start of the experiment. Compounds not detected above LOD 32 and 64 days, with 95.7 ± 5.2 and 88.7 ± 2.1% decay in the 10 and
included C0–C4 phenols and benzothiophenes (Supplementary 30 lm dispersions after 64 days.
Information, Fig. S5), while additional 13 compounds (7.1%) with VOC compounds determined by P&T GC–MS analyses could not
concentrations close to LOD (60.1 lg/L) were excluded for be normalized against a biomarker, since analyses of these did not
biodegradation determination (Supplementary Information, include any recalcitrant biomarker. Determination of % concentra-
Fig. S5). tions relative to 0-day concentrations for the alkane and aromatic
Depletion of semivolatile nC10–nC36 alkanes and non-alkylated VOC compound groups showed biotransformation mainly between
and alkyl-substituted decalines, naphthalenes, fluorenes, 8 and 16 days, resulting in 99.6 ± 0.05 and 96.5 ± 5.0% depletion of
148 O.G. Brakstad et al. / Marine Pollution Bulletin 93 (2015) 144–152

10 µm dispersion NSW

(normalized against 30ab Hopane)

30 µm dispersions-NSW
A
10 µm dispersion StSW
120 30 µm dispersions-StSW
100
% of day 0

80

60

40

20

0
2 4 8 16 32 64 2 4 8 16 32 64 2 4 8 16 32 64 2 4 8 16 32 64
nC10-nC15 alkanes C0-C4 Decalines nC16-nC21 alkanes nC22-nC36 alkanes

10 µm dispersion NSW
30 µm dispersions-NSW
(normalized against 30ab Hopane)

B 10 µm dispersion StSW
>120 >120
120 30 µm dispersions-StSW

100
% of day 0

80

60

40

20

0
2 4 8 16 32 64 2 4 8 16 32 64 2 4 8 16 32 64
2-ring PAH 3-ring PAH/DBT 4- to 5-ring PAH

10 µm dispersion NSW
30 µm dispersions-NSW
C
>120 >120 >120 10 µm dispersion StSW
120 30 µm dispersions-StSW
100
% of day 0

80

60

40

20

0
2 4 8 16 32 64 2 4 8 16 32 64
C5-C9 Saturates C0- to C4/5 Benzenes

Fig. 3. Percentage concentrations compared to day 0 of semivolatile n-alkanes and C0–C4 alkylated decalines (A), semivolatile 2–5 ring PAH (B), and volatile alkanes and
aromatic hydrocarbons (C) in 10 lm and 30 lm dispersions in natural seawater (NSW) and sterilized seawater (StSW) during an incubation period of 64 days, with samples
collected after 2, 4, 8, 16, 32 and 64 days. The percentages of semivolatile compounds were determined from GC–MS analyses (triplicate) normalized against the biomarker
30ab Hopane with standard deviation error bars of 3 replicates. Volatile compounds were determined without normalization. The broken line shows the 0-day 100% for each
target compound.

C5–C9 alkanes in 10 lm and 30 lm dispersions, respectively, while during the biodegradation period (average cell numbers differed
corresponding data for C0- to C4/C5 benzenes were 96.8 ± 0.5 and with a factor of 1.02 between days 0 and 64) argued for 1st order
94.4 ± 2.3% depletion (Fig. 3C). Although sterilized dispersions also kinetics to determine biotransformation coefficients and half-lives
showed decreased concentrations of volatiles after 16 days (39.7– (Battersby, 1990). Non-responsive periods (lag-phases) were
57.7% and 25.9–33.5% depletion of saturates and benzenes, respec- observed for most of the targeted compounds (Fig. 3). During the
tively) most of the depletion in natural unfiltered seawater may DWH spill it was suggested that deepwater with bacteria
therefore be accounted for by biodegradation. re-entered the spill site and caused accelerated hydrocarbon
Non-linear regression analyses of single oil compounds (nor- biodegradation (Valentine et al., 2012), resulting in a situation in
malized against 30ab Hopane) were performed by 1st order rate which hydrocarbon-stimulated bacterial populations eventually
kinetics not including the determined lag-phases, and rate coeffi- could degrade oil compounds without a preceding lag-phase.
cients and half-lives for targeted oil compounds were determined First-order rate calculations of targeted semivolatile n-alkanes
in the dispersions. Low oil concentrations with no significant and 2- to 4-ring PAH normalized against 30ab Hopane were
increase in microbial concentrations (epifluorescence microscopy) performed (Fig. 4, Table S2). Half-lives calculated from the rate
 O.G. Brakstad et al. / Marine Pollution Bulletin 93 (2015) 144–152 149

coefficients varied from 0.6–5.6 days (median 1.1 days) and 1.1– et al., in preparation). And importantly, the oil characteristics will
11.1 days (median 2.5 days) for nC10–nC30 alkanes in 10 lm and be different at the two temperatures. For instance some of the larg-
30 lm dispersions, respectively, while corresponding data for 2- er n-alkanes may generate wax crystals at low temperature, there-
to 4-ring PAH were 0.9–31.9 (median 5.3 days) and 0.9–54.9 days by becoming less susceptible to biodegradation.
(median 10.2 days). The half-lives for most of the targeted com- First-order rates are often calculated without accounting for the
pounds (44 of 48 compounds) were lower in the 10 lm than the lag-phases. In these studies the lag-periods become a part of the
30 lm dispersions, and the differences between the dispersions half-live determination. In Table S2 (Supplementary Information)
were significant (P < 0.05; Wilcoxon non-parametric t-test). The the half-lives are also determined as the sum of the lag-periods
n-alkane half-lives were increased with longer carbon-chains in and the half-lives determined from the rate coefficients. By these
both dispersions, primarily in the high-boiling-point range calculations the n-alkane half-lives varied from 2–17.6 days in
(PnC21). PAH half-lives increased in both dispersions with increas- the 10 lm dispersions and 3.1–20.8 days in the 30 lm dispersions.
ing numbers of aromatic rings and with degree of alkyl-substitu- PAH half-lives varied correspondently from 8.8–49.9 and 8.7–
tion, except for the chrysenes (Fig. 4). Only small differences in 79.9 days in the two dispersions. The same trends for n-alkanes
half-lives were determined for chrysenes, which appeared in low and PAH were observed with these half-life calculations as
concentrations in the dispersions (sum of C0–C4 chrysenes of described above, with generally shorter half-lives of the same com-
0.59 and 0.79 lg/L in 10 lm and 30 lm dispersions, respectively; pounds in the 10 lm than in the 30 lm dispersions. The sums of
supplementary information Fig. S5). Small oil droplets therefore half-lives and lag-phases also differed significantly between the
had a positive impact on biotransformation of both n-alkanes 10 lm and the 30 lm dispersions (P < 0.05; Wilcoxon non-para-
and PAH. In a recent study of Macondo oil biodegradation under metric t-test).
a near-surface temperature relevant for the GoM, comparing Our n-alkane and PAH biodegradation data, using coastal Nor-
physically and chemically dispersed oil, we observed higher wegian seawater from 80 m depth, mainly coincided with results
biodegradation differences related to droplet size for n-alkanes from studies with chemically dispersed oils incubated at low tem-
(nC10–nC30) than 2- to 4-ring PAH (Brakstad et al., 2014). However, perature (1 to +8 °C), in which n-alkane depletion corresponded
in the near-surface oil biodegradation study both temperature (30– to half-lives of 2 to approximately 10 days, while 2- to 4-ring
32 °C), experimental system and oil droplet size distributions dif- half-lives ranged between 2 and 37 days (Siron et al., 1995; Venosa
fered from the current experiment. The different temperatures will and Holder, 2007; Prince et al., 2013; McFarlin et al., 2014). Also
also stimulate different taxa of oil-degrading bacteria, as shown results from the DWH deepwater plume samples and laboratory
from field analyses of the DWH incident, in which none of the studies with microbial consortia or seawater from GoM indicated
predominant bacteria in the deepwater plume (Oceanospirillales, average n-alkane (nC13-nC26) half-lives of 1.2–6.1 days (Hazen
Colwellia and Cycloclasticus) were abundant in surface slicks et al., 2010), showing good agreement with our results. In a recent
(Redmond and Valentine, 2012). Detailed results from charac- study with small droplet (10–30 lm) dispersions of Macondo oil in
terization of cell counts and bacterial communities from the cur- GoM deepwater, biotransformation of 2–3 ring PAH appeared
rent study will be presented in a separate manuscript (Brakstad mainly between 9 and 39 days of incubation, while 4–6 ring PAH

A 15
10 µm dispersions
30 µm dispersions
Half-lives (days)

10

0
12

14

16

17

18

20

22

23

24

25

26

27

28

29

30
10

11

15

19

21
1
nC

nC

nC

nC

nC

nC

nC

nC

nC

nC

nC

nC

nC

nC

nC

nC

nC

nC
nC

nC

nC

B 60
10 µm dispersions
30 µm dispersions
Half-lives (days)

40

20

0
F
N

D
P

FL

3
C
F1

F2

F3

4
1

3
2

FL

FL

FL
P

P
N

Naphthalenes Fluorenes Phenanthrenes Dibenzothiophenes Fluoranthenes Chrysenes

Fig. 4. Half-lives of nC10–nC30 alkanes (A) and naphthalenes, fluorenes, phenanthrenes, dibenzothiophenes, fluoranthenes and chrysenes (B) in 10 lm and 30 lm dispersions
with standard deviation errors bars of 3 replicates. Half-lives were determined from rate coefficients without lag-periods included. The numbering of the aromatic groups
describes the degree of alkyl-substitution.
150 O.G. Brakstad et al. / Marine Pollution Bulletin 93 (2015) 144–152

First-order rate coefficient

0.3
10 µm dispersions
30 µm dispersions
0.2

k1 0.1

0.0 s

ze atu e
(to e s

tu )
s
es

C enz es

tu s
en es

tu s

- 1 Na s

t u -1

- 1 Na s

tu 2

PA s
at -1

-2 atu s

tu s

PA s
O Sat -2

es

s
en s

Sa ne
te

te

Sa ne

e
te

te

Sa ph-

te

e
Sa ate

te

up
te

Sa ph

H
t

en

C 4 B rat
2 zen

C 0 S rat

at
ze
ra

ne ra

ra

3 ura

ra

ra

ra

ra
e
ra

ro
r

ur
lu

C nz
tu

u
tu

G
t
Sa

Sa

Sa
B

e
Sa

R
S

C 8S
B

A
2-
5

10

+
6

SC
C

-1

-1

-2

SU 2 5
C

C
C
C

11

13

15

17

19

21

C
en

M
B
1
C

Fig. 5. First-order rate coefficients for oil compound groups of the Macondo oil in 10 and 30 lm dispersions.

transformation had started after 39 days of incubation (Wang targeted compounds (n-alkenes, 2–5 ring PAH, volatile C5–C9 satu-
et al., 2014), consistent with PAH results from our study. rates and C0–C5 benzenes) constituted 33.5% and 40.4% of the sum
Oil compound glass wall attachment is well known, and com- of TEOC and VOC concentration at the start of the experiment in
parison of oil concentrations (TEOC analyses) in water and on glass the 10 lm and 30 lm dispersions, respectively (Supplementary
walls rinsed with solvent (days 0, 8 and 32) showed a gradual Information, Fig. S6). The TEOC concentrations determined by
increase of wall-attached material, from <3% of total TEOC (day GC-FID analyses represented 72% and 81% of the oil droplet con-
0), to up to 45% (30 lm dispersions at day 32). Targeted analyses centrations measured by the Coulter Counter analyses for the
of compounds normalized against 30ab Hopane showed relative two dispersions, and these results corresponded well with the
preference of 2-ring and non-alkylated 3-ring PAH for the water- assumption from true boiling point curves, assuming that C5–C36
phase, while larger alkylated 3-ring and 4- to 5-ring PAH, decalines saturates (boiling points of 27 to approximately 500 °C) represent
and n-alkenes gradually distributed between the water- and glass- 70–80% of typical crude oil (Pasquini and Bueno, 2007).
phase (results not shown). These data emphasize that degradation Biotransformation rate coefficients and half-lives of 22 volatile
of dispersed oil in laboratory systems is a result of processes in and semivolatile oil compound groups, representing the boiling
both water and glass wall surfaces. point range described above (Table S1) were calculated, including
C5–C36 saturates and 1- to 5-ring aromatic compounds (Reed
3.2.3. Biodegradation of oil compound groups et al., 2000). The results of Fig. 5 and Table S3 showed rate coeffi-
The well-known targeted compounds described above only cov- cients of 0.0096–0.2665 (median 0.1692) in the 10 lm dispersions
er a fraction of the oil. Comparison of TEOC concentrations (GC-FID and 0.0054–0.2475 (median 0.0995) in the 30 lm dispersions.
analyses) with the targeted SVOC and VOC saturate and aromatic These rates resulted in half-lives of 2.8 to 72.2 days (median
oil compound (GC–MS analyses) showed that the sum of the 4.1 days) and 2.8 to 128.3 days (median 7.2 days) in the 10 and

A B
UCM UCM
Volatiles Volatiles
n-alkanes n-alkanes
PAH PAH

Total = 1155.61 Total = 445.047

C
UCM
n-alkanes
PAH
Volatiles

Total = 682.196

Fig. 6. Distributions of targeted oil compounds and UCM at the start of the biodegradation (A) and after 32 days of incubation in unfiltered (B) and sterilized (C) dispersions.
The total concentrations of TEOC (lg/L) are shown below each pie chart.
 O.G. Brakstad et al. / Marine Pollution Bulletin 93 (2015) 144–152 151

30 lm dispersions, respectively. The rate coefficients were higher may therefore compensate for possible negative effects of disper-
in the 10 lm than the 30 lm dispersions for all oil compounds, sant components in the marine environment.
except for C15–C16 saturates. Comparison of the rate coefficients Operationally, dispersants have mainly been used for surface oil
in the two dispersions showed significant differences (P < 0.05; spills to remove the slick from the seawater surface by generation
Wilcoxon non-parametric t-test). The rate coefficients were higher of small droplet dispersions in the water column. We have previous-
in favor of the 10 lm dispersions by factors of 1.21 and 1.67 (me- ly shown that oil compounds in surface oil emulsions are poorly
dian values) for the saturate and aromatic compound groups, biodegradable compared to chemically dispersed oil (Brakstad
respectively. The rate coefficients were generally higher for the et al., 2014). The dispersant application during the DWH incident
aromatic (average of 0.175039) than the saturate (0.099304) oil aimed at reducing surface slick generation and may also have con-
compound groups when data from both dispersions were com- tributed to the generation of the small oil droplets in the Macondo
piled. This is probably associated with the water-solubility of the deepwater plume. If deepwater injection of dispersant during an
aromatic hydrocarbons, as well as the presence of poorly oil spill is able to reduce oil droplet size and avoid oil surfacing, this
biodegradable branched and cyclic alkanes in the semivolatile may enhance oil compound biodegradation and thereby additionally
saturate groups. reduce the impacts of the spill. However, it is important to empha-
In previous studies we have compared dissolution and size the lack of knowledge related to the many unidentifiable oil
biodegradation of oil compounds from thin oil films (10–20 lm compounds, both with respect to biodegradability and toxic effects.
thickness) immobilized on hydrophobic adsorbents, showing that Although the water used in the current study was Norwegian
fast dissolution of naphthalenes and non-alkylated 3-ring PAH coastal seawater, the results may be related to the Macondo deep-
resulted in biodegradation mainly of dissolved compounds, while water plume. Both oil, oil droplet size distribution (North et al.,
negligible dissolution of n-alkanes (nC12–nC36) resulted in 2011) and seawater temperature was similar in the current experi-
biodegradation of these compounds exclusively on the oil–water ment and the DWH deepwater plume. Our experiments were per-
interphases. For the larger PAH-compounds dissolution decreased formed at atmospheric pressure, but laboratory experimental
with increasing numbers of aromatic rings and alkyl substitution, studies have shown that bacteria associated with deepwater plume
with biodegradation of these both as dissolved and as oil-associat- biodegradation of the Macondo oil also biodegraded the oil at
ed compounds (Brakstad et al., 2004; Brakstad and Bonaunet, atmospheric pressure (Hazen et al., 2010; Bælum et al., 2012).
2006). Oil compound biodegradation is therefore a complex pro-
cess, involving bacterial communities in the water-phases as well
4. Conclusions
as bacteria attaching the oil surfaces.
The rate for the sum of the 22 oil compound groups in this study
The study presented here compares, for the first time to our
was also faster in the 10 lm (k1 = 0.03984) than the 30 lm
knowledge, biodegradation rates in oil dispersions with pre-defined
(k1 = 0.02004) dispersions, resulting in half-lives of 17.4 and
droplet sizes, and under temperatures relevant for a deepwater
34.6 days in the two dispersions (Fig. 5; Table S3). Since the sum
discharge. The results of this study showed that biodegradation
of these groups may represent most of the oil, these numbers
of alkane and aromatic hydrocarbons in a dispersible oil (Macondo)
may be used as biotransformation rates for the oil. The oil droplet
was faster in 10 lm than 30 lm dispersions for targeted oil com-
size was therefore important both for the biotransformation rates
pounds and oil compound groups, as well as the sum of the groups
of the oil compound groups and the sum of these groups used in
representing most of the oil. These data indicate that dispersant
this study and presenting a boiling point range of 27–>500 °C.
treatment of deepwater oil spills will be beneficial for the oil com-
The saturates representing the C10 to C25+ oil compound groups
pound biodegradation, if oil droplet sizes are reduced. Our data
are mostly unidentifiable oil compounds, often termed as the unre-
showed good agreement with results from other biodegradation
solved complex mixture (UCM) (Gough and Rowland, 1990; Killops
studies performed with chemically dispersed oil at low seawater
and Al-Juboori, 1990), and accounts for 60–70% of the groups. Com-
temperatures. Although the current study was performed in sea-
parison of the sum of TEOC and VOC concentrations with targeted
water from a Norwegian fjord and under atmospheric pressure,
compound groups of n-alkanes, PAH and volatile hydrocarbons
these data may also have relevance for the DWH deepwater oil
showed that the UCM became the predominant oil fraction during
plume. We used the same oil, similar temperatures (4–5 °C), and
the biodegradation period (Fig. 6; Fig. S6, Supplementary Informa-
oil droplet size distributions were comparable to the DWH deep-
tion). Several studies have shown that the UCM fraction includes
water plume. Published oil compound degradation data with
compounds that are not readily susceptible to biodegradation
GoM seawater or enrichment cultures also showed agreement with
(Reddy et al., 2002; Booth et al., 2007). The UCM was predominant
our results. Comparison of fate model predictions, laboratory
in each of the saturate groups from C11–C12 to C25+ (Table S1). The
results and field data may further substantiate the fate of oil dis-
degradation of these compounds decreased by increasing boiling
persions in deepwater spills and improve the information concern-
point ranges in both 10 lm and 30 lm dispersions (Table S3), and
ing the impacted deepwater area of a spill. This information is
the rate calculations showed that their degradation was mainly fas-
important to include in risk assessment analyses.
ter in the 10 lm than the 30 lm dispersions (Fig. 5; Table S3).
The result from the current study show that reduced oil droplet
Acknowledgments
sizes resulted in increased biodegradation rates. The use of chemi-
cal dispersants will generate small droplet oil dispersions, and the
This study was supported by BP Exploration & Production Inc.,
result from this and other studies (e.g. Siron et al., 1995; Venosa
and the BP Gulf Coast Restoration Organization. We will thank Bror
and Holder, 2007.; Prince et al., 2013; McFarlin et al., 2014;
Johansen, Inger Steinsvik, Marianne Unaas Rønsberg and Inger K.
Brakstad et al., 2014) demonstrate that the use of dispersants to
Almås for their technical assistance with the chemical analyses.
generate small droplets will promote biodegradation in seawater
at different environmental conditions. Dispersants like Corexit
9500 are considered to have moderate to low acute toxicity (e.g. Appendix A. Supplementary material
Mitchell and Holdway, 2000; Judson et al., 2010), and surfactants
like alkyl sulfosuccinates may be biodegradable (Garcia et al., Supplementary data associated with this article can be found, in
2009; Bælum et al., 2012). The positive influence of chemical dis- the online version, at http://dx.doi.org/10.1016/j.marpolbul.2015.
persants on the biodegradation of potentially toxic oil compounds 02.006.
152 O.G. Brakstad et al. / Marine Pollution Bulletin 93 (2015) 144–152

References McFarlin, K.M., Prince, R.C., Perkins, R., Leigh, M.B., 2014. Biodegradation of
dispersed oil in Arctic seawater at – 1°C. PLoS one 9, e84297.
Mitchell, F.M., Holdway, D.A., 2000. The acute and chronic toxicity of the
Atlas, R.M., Hazen, T.C., 2011. Oil biodegradation and bioremediation: A tale of the
dispersants Corexit 9527 and 9500, water accommodated fraction (WAF) of
two worst spills in U.S. history. Env. Sci. Technol. 1 (45), 6709–6715.
crude oil, and dispersant enhanced WAF (DEWAF) to Hydra viridissima (green
Battersby, N.S., 1990. A review of biodegradation kinetics in the aquatic
hydra). Water Res. 34, 343–348.
environment. Chemosphere 21, 1243–1284.
Nordtug, T., Olsen, A.J., Altin, D., Meier, S., Overrein, I., Hansen, B.H., Johansen, Ø.,
Booth, A.M., Sutton, P.A., Lewis, C.A., Lewis, A.C., Scarlett, A., Chau, W., Widdows, J.,
2011. Method for generating parameterized ecotoxicity data of dispersed oil for
Rowland, S.J., 2007. Unresolved complex mixtures of aromatic hydrocarbons:
use in environmental modelling. Mar. Pollut. Bull. 62, 2106–2113.
thousands of overlooked persistent, bioaccumulative, and toxic contaminants in
North, E.W., Adams, E.E., Schlag, Z., Sherwood, C.R., He, R., Hyun, K.H., Socolofsky,
mussels. Environ. Sci. Technol. 41, 457–464.
C.A., 2011. Simulating oil droplet dispersal from the Deepwater Horizon spill
Brakstad, O.G., Bonaunet, K., Nordtug, T., Johansen, Ø., 2004. Biotransformation of
with a Lagrangian approach. Geophys. Monograph Ser. 195, 217–226.
petroleum hydrocarbons in natural flowing seawater at low temperature.
Page, D.S., Boehm, P.D., Douglas, G.S., Bence, E.A., 1995. Identification of
Biodegradation 15, 337–346.
hydrocarbon sources in the benthic sediments of Prince William Sound and
Brakstad, O.G., Bonaunet, K., 2006. Biodegradation of petroleum hydrocarbons in
the Gulf of Alaska following the Exxon Valdez oil spill. ASTM Special Technical
seawater at low temperatures (0–5 °C) and bacterial communities associated
Publication 1219, 41–83.
with degradation. Biodegradation 17, 71–82.
Pasquini, C., Bueno, A.F., 2007. Characterization of petroleum using near-infrared
Brakstad, O.G., Daling, P.S., Faksness, L.-G., Almås, I.K., Vang, S.-H., Syslak, S., Leirvik,
spectroscopy: quantitative modeling for the true boiling point curve and
F., 2014. Depletion and biodegradation of the Macondo 252 oil in dispersions
specific gravity. Fuel 86, 1927–1934.
and emulsions generated in an oil-on-seawater mesocosm flume basin. Mar.
Porter, K.G., Feig, Y.S., 1980. The use of DAPI for identifying and counting aquatic
Poll. Bull. 84, 125–134.
microflora. Limnol. Oceanogr. 25, 943–948.
Bælum, J., Borglin, S., Chakraborty, R., Fortney, J.L., Lamendella, R., Mason, O.U., Auer,
Prince, R.C., Elmendorf, D.L., Lute, J.R., Hsu, C.H., Haith, C.E., Senius, J.D., Dechert, G.J.,
M., Zemla, M., Bill, M., Conrad, M.E., Malfatti, S.A., Tringe, S.G., Holman, H.-Y.,
Douglas, G.S., Butier, E.L., 1994. 17a(H),21@(H)-Hopane as a conserved internal
Hazen1, T.C., Jansson, J.K., 2012. Deep-sea bacteria enriched by oil and
marker for estimating the biodegradation of crude oil. Env. Sci. Technol. 28,
dispersant from the Deepwater Horizon spill. Environ. Microbiol. 14, 2405–
142–145.
2416.
Prince, R.C., McFarlin, K.M., Butler, J.D., Febbo, E.J., Wang, F.-C.Y., Nedwed, T.J., 2013.
Camilli, R., Reddy, C.M., Yoerger, D.R., Van Mooy, B.A.S., Jakuba, M.V., Kinsey, J.C.,
The primary biodegradation of dispersed crude oil in the sea. Chemosphere 90,
McIntyre, C.P., Sylva, S.P., Maloney, J.V., 2010. Tracking hydrocarbon plume
521–526.
transport and biodegradation at Deepwater Horizon. Science 330, 201–204.
Reddy, C.M., Eglinton, T.I., Hounshell, A., White, H.K., Xu, L., Gaines, R.B., Frysinger,
Coastal Response Research Center. Deepwater Horizon Dispersant Use. Meeting
G.S., 2002. The West Falmouth oil spill after thirty years: the persistence of
Report. May 26–27, 2010. <www.crrc.unh.edu/dwg/dwh_dispersants_use_
petroleum hydrocarbons in marsh Sediments. Environ. Sci. Technol 36, 4754–
meeting_report.pdf>.
4760.
Daling, P.S., Leirvik, F., Almås, I.K., Brandvik, P.J., Hansen, B.H., Lewis, A., Reed, R.,
Reddy, C.M., Arey, J.S., Seewald, J.S., Sylva, S.P., Lemkau, K.L., Nelson, R.K.,
2014. Surface weathering and dispersibility of MC252 crude oil. Mar. Poll. Bull.
Carmichael, C.A., McIntyre, C.P., Fenwick, J., Ventura, G.Y., Van Mooy, B.A.S.,
87, 300–310.
Camilli, R., 2012. Composition and fate of gas and oil released to the water
Dubinsky, E.A., Conrad, M.E., Chakraborty, R., Bill, M., Borglin, S.E., Hollibaugh, J.T.,
column during the Deepwater Horizon oil spill. PNAS 109, 20229–20234.
Mason, O.U., Piceno, Y.M., Reid, F.C., Stringfellow, W.T., Tom, L.M., Hazen, T.C.,
Redmond, M.C., Valentine, D.L., 2012. Natural gas and temperature structured a
Andersen, G.L., 2013. Succession of hydrocarbon-degrading bacteria in the
microbial community response to the Deepwater Horizon oil spill. PNAS 109,
aftermath of the Deepwater Horizon oil spill in the Gulf of Mexico. Environ. Sci.
20292–20297.
Technol. 47, 10860–10867.
Reed, M., Daling, P.S., Brakstad, O.G., Singsaas, I., Faksness, L.-G., Hetland, B., Ekrol,
Garcia, M.T., Campos, E., Marsal, A., Ribosa, I., 2009. Biodegradability and toxicity of
N., 2000. OSCAR2000: A multi-component 3-dimensional oil-spill contingency
sulphonate-based surfactants in aerobic and anaerobic aquatic environments.
and response model. In: Proceedings of the 2000 Arctic and Marine Oil spill
Water Res. 43, 295–302.
Program Technical Seminar, June 14–16, 2000, Vancouver, Canada.
Gough, M.A., Rowland, S.J., 1990. Characterisation of unresolved complex mixtures
Siron, R., Pelletier, E., Brochu, C., 1995. Environmental factors influencing the
of hydrocarbons in petroleum. Nature 344, 648–650.
biodegradation of petroleum hydrocarbons in cold seawater. Arch. Environ.
Harris, B.C., Bonner, J.S., McDonald, T.J., Fuller, C.B., Page, C.A., Dimitriou-Christidis,
Contam. Toxicol. 28, 406–416.
P., Sterling, M.C., Autenrieth, R.L., 2002. Bioavailability of chemically-dispersed
Smith, A.J., Fleming, P.B., Fulton, P.M., 2014. Hydrocarbon flux from natural
crude oil. In: Proceedings of the 25th Arctic and Marine Oilspill Program,
deepwater Gulf of Mexico vents. Earth Planet. Sci. Lett. 395, 241–253.
Technical Seminar, June 11–13, Calgary, AB, Canada, pp 895–905.
Sterling Jr., M.C., Bonner, J.S., m Ernest, A.N.S., Page, C.A., Autenrieth, R.L., 2004.
Hazen, T., Eric Dubinsky, A., DeSantis, T.Z., Andersen, G.L., Piceno, Y.M., Singh, N.,
Hemical dispersant effectiveness testing: influence of droplet coalescence. Mar.
Jansson, J.K., Probst, A., Borglin, S.E., Fortney, J.L., Stringfellow, W.T., Bill, M.,
Poll. Bull. 48, 969–977.
Conrad, M.E., Tom, L.M., Chavarria, K.L., Alusi, T.R., Lamendella, R., Joyner, D.C.,
Swannell, R.P.J., Daniel, F., Croft B.C., Engelhardt, M.C., Wilson, S., Mitchell D.J., Lunel,
Spier, C., Baelum, J., Auer, M., Zemla, M.L., Chakraborty, R., Sonnenthal, E.L.,
T. Influence of physical and chemical dispersion of the biodegradation of oil
D’haeseleer, P., Holman, H.-Y.N., Osman, S., Lu, Z., Van Nostrand, J.D., Deng, Y.,
under simulated marine conditions. In: Proceedings of the Twentieth Arctic and
Zhou, J., Mason, O.U., 2010. Deep-sea oil plume enriches indigenous oil-
Marine Oilspill Program (AMOP) Technical Seminar, 1997, Vancouver, Alberta,
degrading bacteria. Science 330, 204–208.
Canada, pp. 617–641.
Horowitz, A., Gutnick, D., Rosenberg, E., 1975. Sequential growth of bacteria on
Valentine, D.L., Kessler, J.D., Redmond, M.C., Mendes, S.D., Heintz, M.B., Farwell, C.,
crude oil. Appl. Microbiol. 30, 10–19.
Hu, L., Kinnaman, F.S., Yvon-Lewis, S., Du, M., Chan, E.W., Tigreros, F.G.,
Judson, R.S., Martin, M.T., Reif, D.M., Houck, K.A., Knudsen, T.B., Rotroff, D.M., Xia, M.,
Villanueva, C.V., 2010. Propane respiration jump-starts microbial response to a
Sakamuru, S., Huang, R., Shinn, P., Austin, C.P., Kavlock, R.J., Dix, D.J., 2010.
Deep Oil Spill. Science 330, 208–211.
Analysis of eight oil spill dispersants using rapid, in vitro tests for endocrine and
Valentine, D.L., Mezic, I., Macešic, S., Crnjaric-Žic, N., Ivic, S., Hogan, P.J., Fonoberov,
other biological activity. Environ. Sci. Technol. 44, 5979–5985.
V.A., Loire, S., 2012. Dynamic autoinoculation and the microbial ecology of a
Kessler, J.D., Valentine, D.L., Redmond, M.C., Du, M., Chan, E.W., Mendes, S.D.,
deep water hydrocarbon irruption. PNAS 109, 20286–20291.
Quirez, E.W., Villanueva, C.J., Shusta, S.S., Werra, L.M., Yvon-Lewis, S.A., Weber,
Venosa, A.D., Holder, E.L., 2007. Biodegradability of dispersed crude oil at two
T.C., 2011. A persistent oxygen anomaly reveals the fate of spilled methane in
different temperatures. Mar. Poll. Bull. 54, 545–553.
the deep Gulf of Mexico. Science 331, 312–315.
Vilcáez, J., Li, L., Hubbard, S.S., 2013. A new model for the biodegradation kinetics of
Killops, S.D., Al-Juboori, M.A.H.A., 1990. Characterisations of unresolved complex
oil droplets: application to the Deepwater Horizon oil spill in the Gulf
mixtures in the gas chromatograms of biodegraded petroleums. Organ.
of Mexico. Geochem. Trans.14:4. <http://www.geochemicaltransactions.com/
Geochem. 15, 147–160.
content/14/1/4>.
Kujawinski, E.B., Kido Soule, M.C., Valentine, D.L., Boysen, A.K., Longnecker, K.,
Wang, J., Sandoval, K., Ding, Y., Atlas, R., Gardinali, P.R. Biodegradation of dispersed
Redmond, M.C., 2011. Fate if dispersants associated with the Deepwater
MC252 crude oil by indigenous Gulf of Mexico microbial communities. In:
Horizon oil spill. Env. Sci. Technol. 45, 1298–1306.
Abstract of the 2014 SETAC North America 35th Annual Meeting, November 10,
Lee, K., Nedwed, T., Prince, R.C., Palandro, D., 2013. Lab tests on the biodegradation
2014, Vancouver, Canada.
of chemically dispersed oil should consider the rapid dilution that occurs at sea.
Wang, Z., Fingas, M., Blenkinsopp, S., Sergy, G., Landriault, M., Sigouin, L., Lambert,
Mar. Poll. Bull. 73, 314–318.
P., 1998. Study of 25-year old Nipisi oil spill: persistence of oil residues and
MacNaughton, S.J., Swannell, R., Daniel, F., Bristow, L., 2003. Biodegradation of
comparisons between surface and subsurface sediments. Environ. Sci. Technol.
dispersed Forties crude and Alaskan North Slope oils in microcosms under
32, 2222–2232.
stimulated marine conditions. Spill Sci. Technol. 8, 179–186.