Characterization of Sea Surface Chemical Contamination after

Feb 13, 2008 - Cefas Burnham Laboratory, Remembrance Avenue, ... contamination in the sea surface during emergency response. Introduction. The English...
0 downloads 0 Views 2MB Size
Download PDF
Research Characterization of Sea Surface Chemical Contamination after Shipping Accidents C A R L O S G U I T A R T , * ,† PATRICIA FRICKERS,† JOSE HORRILLO-CARABALLO,‡ ROBIN J. LAW,§ AND JAMES W. READMAN† Plymouth Marine Laboratory, Prospect Place, The Hoe, Plymouth, PL1 3DH, U.K., University of Plymouth, Reynolds Building, Drake Circus, Plymouth, PL4 8AA, U.K., and The Centre for Environment, Fisheries and Aquaculture Science, Cefas Burnham Laboratory, Remembrance Avenue, Burnham-on-Crouch, Essex, CM0 8HA, U.K.

Received August 8, 2007. Revised manuscript received January 21, 2008. Accepted January 22, 2008.

A contamination survey was conducted after the beaching of the stricken cargo ship MSC Napoli in Lyme Bay on the south coast of Devon (UK). A grid of 22 coastal and offshore stations was sampled to investigate the extent of spilled oil and to screen for chemical contamination, as well as to evaluate the behavior of the oil at the air-sea interface. Samples were collected from the sea surface microlayer (SML) and from subsurface waters (SSW) at each station. The fuel oil spilled (IFO 380) was also analyzed. The determination of oil-related hydrocarbons (aliphatic hydrocarbons, polycyclic aromatic hydrocarbons (PAHs), terpanes, and steranes) and the screening for other harmful chemicals on the inventory of the MSC Napoli in the seawater samples, was performed by PTV-GC/ MS using large volume injection (LVI) techniques. Screening did not reveal the presence of any harmful chemicals other than petroleum-related compounds. Results afforded investigation of oil sources and spatial distributions of total PAH concentrations and enrichments in the sea surface microlayer (SML). Rather than a single source, oil fingerprinting analyses of the samples revealed a mixture of three types of oil: heavy fuel oil, lubricating oil, and a lighter oil (probably diesel oil). Enrichment factors (EF) in the SML (EF ) CSML/CSSW) were calculated and, in the vicinity of the ship, approached 2000, declining with distance away from the wreck. These factors represent approximately a 1000-fold enrichment over typical coastal total PAH enrichments in the SML and reflected a clear petrogenic origin of the contamination (as demonstrated, for example, by a Fl/Py ratio 10,000 shipwrecks) and major oil spills (e.g., Torrey Canyon in 1967). On January 20, 2007, the 53,000 tonne container ship MSC Napoli was beached in Lyme Bay on the south Devon coast (UK) after suffering a serious fracture in the hull due to severe weather conditions prevailing on January 18 in the English Channel. The ship carried more than 1,600 tonnes of chemical products classified by the International Maritime Organisation (IMO) as dangerous goods (e.g., bisphenol A, epichlorohydrin-epoxy resin, alkylphenols, isophoronediamine nonylphenol, propaquizafop, profenofos, carbendazim, hexamethylindanopyran, and dibutyltinoxide, as well as organic solvents, acids, and corrosive materials). In addition, 3,780 tonnes of heavy fuel oil (IFO 380) and ca. 45 tonnes of diesel oil were carried as fuel. During the initial stages of the incident approximately 200 tonnes of fuel oil were lost, of which 50 tonnes directly affected the Lyme Bay area, with most visually apparent damage to sea birds inhabiting the area. The remainder of the spilled fuel oil was dispersed into the open sea carried by the prevailing currents (in a SSE direction). Approximately 100 of the containers were also lost overboard, half of which beached rapidly on the shoreline, with others presumed sunk. These events provoked fears of major environmental damage to the surrounding area from toxic chemical spillages. Although over the last 40 years the number of oil tanker accidents and the amount of oil spilled at sea (worldwide) have been reduced (1), the impact of chemical spillages (i.e., other than oil spills) has rarely caught the attention of policy makers and society. Following the beaching of the Napoli, a survey was organized to assess the extent of chemical and oil spill contamination in the water column. Additionally, the transport processes and contamination at the sea surface microlayer were investigated. The physical and chemical processes undergone by oil spilt at sea (e.g., advection, spreading, dispersion, emulsification, evaporation, dissolution, oxidation, biodegradation, sedimentation) have been intensely studied (2–4) and models have been constructed to predict behavior (5, 6). However, after oil has been released into the marine environment, it contributes to the sea surface microlayer (SML), changing its physicochemical properties and composition. The sea surface microlayer is the boundary compartment (defined as the top millimeter of the water column) and is a key research target of ongoing international research projects (7) due to its unique compositional characteristics and fundamental role in the biogeochemistry cycles of natural and anthropogenic substances between the atmosphere and oceans (8, 9), and because it has also been demonstrated to accumulate organic contaminants (10). However, the beVOL. 42, NO. 7, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2275

FIGURE 1. Coastal and offshore sampling stations and location of the cargo ship MSC Napoli in Lyme Bay. havior and impact of minor oil contributions to the SML remains poorly investigated, although detrimental biological effects have been speculated (11, 12). Oil fingerprinting by GC/MS is a well-known and established methodology (13, 14) to investigate oil composition and to elucidate weathering processes, and has been applied to assess sources and fates, and to predict the long-term impact of spilled oils (15, 16). The laboratory methods (17, 18), however, must be also applicable at the low analyte concentrations associated with marine environmental samples. Recently, increased sample injection volumes in gas chromatography [large volume injection techniques (LVI)], have improved analytical precision and accuracy at trace levels, in particular for environmental water analysis (19). In the present study, the oil-related hydrocarbons were determined by LVI-GC/MS, using a programmable temperature volume (PTV) inlet, in extracts from the sea surface microlayer (SML) and subsurface waters (SSW) surrounding the Napoli and across Lyme Bay. In the area affected by the oil spill, the sources, enrichment, processes, and fates of petroleum hydrocarbons in the sea surface were investigated.

Experimental Procedures Strategy and Sampling. The cargo ship was beached approximately 1/2 mile from the shoreline (50° 40.62 N, 3° 09.89 W). The sampling program, undertaken during January 30 and 31, was designed to assess levels of the oil contamination and the potential occurrence of other harmful chemicals throughout Lyme Bay after the initial oil spill on January 20. Nearshore and offshore stations were selected along perpendicular transects from the coast at approximately 10, 20, and 40 m depth (see Figure 1). In total, 21 stations were sampled for seawater. Samples from the top sea surface (1 cm depth, which included the sea surface microlayer) (0.5 L) were obtained using a surface slick sampler (20), a custommade design using a SCHOTT Duran glass bottle and manually operated from the deck of the ship. Weather conditions were calm with a wind speed under 2.5 m/s. An additional sea surface sample was collected close to the ship (station 4′). The subsurface water samples were obtained at 2 m depth with 2.5 L Winchester amber glass bottles using a custom-made stainless steel and Teflon sampling device (21, 22). All the samples were preserved immediately after collection by the addition of dichloromethane to prevent biodegradation. In addition, a sample from the original heavy oil carried aboard the MSC Napoli was supplied by the UK Maritime and Coastguard Agency. Variability in thickness and composition of microlayers render sampling and definitive quantification of components difficult (9). By definition, the sea surface microlayer compartment is assumed to comprise the uppermost 1 mm (1000 2276

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 7, 2008

µm) (8). Each sampling technique of the sea surface has a methodological influence on the results obtained. To characterize the performance of the selected sampling protocol, the device was calibrated using laboratory tests designed to evaluate the dilution factor. An oil slick (1 mm thick) was prepared in a water tank using heavy oil mixed with hexane. The surface sampling device was repetitively tested (n ) 5). Results gave a dilution ratio of the surface microlayer with subsurface water of 21 ( 7. The variability is due to the inherent difficulty of consistently sampling the sea surface interface (20). An arbitrary concentration factor of 10 (in the lower range, to avoid overestimations when collecting slicks) was selected to be applied to the surface samples to compensate for the dilution. The software Surfer Version 8.0 (Surface Mapping System) was used to interpolate and to plot the analytical data sets using Kriging as a gridding method. Analytical Methodologies. The unfiltered seawater samples were spiked, for quantification purposes, with each surrogate solution (acenaphthene-d10, phenanthrene-d10 and chrysene-d12) to a concentration of 100 ng/L and 500 ng/L, for the 2.5 L (SSW) and 0.5 L (SML) samples, respectively. The samples were liquid–liquid extracted (LLE) with dichloromethane (DCM). For the 2.5 L samples, a triple extraction was used with 2 × 50 mL followed by 1 × 25 mL DCM. For the 0.5 L samples (SML), 2 × 25 mL and 1 × 15 mL of DCM were used. The extracts were combined for each sample and were rotary-evaporated down to a few mL, dried with anhydrous sodium sulfate (Na2SO4), and transferred to vials to provide a final volume of 500 µL. Triphenylamine was also added to the vials prior to injection. The heavy oil sample from the original oil spill was weighed into a vial and was dissolved in DCM. Appropriate dilutions of the extract were made prior to analysis by GC/MS. Instrumentation and Quality Assurance. Instrumental analyses were performed using a gas chromatograph interfaced with a mass spectrometer (Agilent Technologies 6890 + 5973N model) using a large volume injection (LVI) technique by means of a programmable temperature volume (PTV) inlet. Quantification was performed by the relative internal standard method using the closest isotopically labeled PAH for each parent and alkylated compound. Matrix spike recovery experiments at different trace levels, standard calibration curves, and quality controls (i.e., method blank, QC standards, analytical grade solvents, etc.) were performed for quality assurance purposes (see details in Supporting Information).

Results and Discussion Chemical Fingerprinting and Oil Identification. A selection of specific target analytes, including individual aliphatic hydrocarbons from C9 to C36 and isoprenoid hydrocarbons (pristane and phytane), the EPA priority parent polycyclic aromatic hydrocarbons (PAHs), the petroleum-specific alkylated homologues of PAHs (i.e., C1-C2 naphthalenes, dibenzothiophenes, and phenanthrenes) were chosen to characterize the samples. In addition, other chemical biomarkers (i.e., terpanes and steranes) were investigated to fingerprint the source(s). Figure 2a shows the aliphatic fraction of the heavy fuel oil sample (IFO 380), a commonly used fuel for ships’ engines. The pristane/phytane ratio (Pr/Ph) in the fuel oil was found to be 0.94 and is comparable to ratios reported for previous spills of heavy oil, such as that from the Prestige (4). Although the chemical composition of heavy fuel oils can vary widely as they are produced largely to physical rather than chemical criteria, many are produced by blending residual oils with diesel fuels or other lighter fuels. This results in a bimodal distribution of n-alkanes which is characteristic of such heavy fuel oils (Figure 2a). The first mode is centered at C14 with

FIGURE 2. (a) GC/MS ion chromatogram profile (m/z ) 57) for aliphatic hydrocarbons of the IFO 380 heavy oil spilled and (b) the sample taken in the sea surface microlayer at station 5 (near the ship wreck) where lubricating oil and diesel signatures can be observed. the second, typically in the carbon range C18-C26, around C22. Figure 2a is consistent with a blending of a lighter fuel (such as diesel). In contrast, Figure 2b shows the aliphatic profile for the extract from the sample taken in the SML near the wreck where an oil sheen was visible. The distribution includes an unresolved complex mixture (UCM) that is characteristic of lubricating oils (lube oil) together with resolved n-alkanes in the C14-C26 range, indicating a mixture with a lighter oil (such as diesel). Associated with the UCM is the presence of biomarkers (terpanes and steranes) which is a characteristic of lubricating oils (23, 24) such as transmission or hydraulic oil. The PAH distribution patterns in the SML and SSW samples (at Station 5) also differ from those of the IFO 380 oil (Figure 3). The heavy oil sample is characterized by a high percentage of naphthalene and its alkylated homologues (>60% of the total PAH) with phenanthrene and dibenzothiophene series comprising 18% and 12%, respectively. It should also be noted that the heavy oil contains a similar proportion of the parent compound naphthalene compared to its alkyl homologues (i.e., C0 ∼ ΣC1 ∼ ΣC2). This is not the case for the rest of the PAH series, where the proportion of alkylated PAH increases with the degree of alkylation (i.e., C0 < C1 < C2). As lubricating oils are characterized by a low proportion of PAHs (13), the PAHs in both the sea surface microlayer and the subsurface samples (station 5) are likely

to originate from the water accommodated fractions (WAF, complex mixtures of the most soluble compounds of the oil which rapidly dissolve in the water) of the heavy fuel oil or from the lighter oil source (i.e., diesel). PAHs in the SML and SSW extracts exhibit similar profiles for the naphthalene, phenanthrene, and dibenzothiophene series with a characteristic C0 < C1 < C2 distribution, indicating that they originate basically from the same source. Profiles of terpanes (m/z ) 191) and steranes (m/z ) 217) from the SML station 5 extract (Figure 4a and c) were clearly different from the profiles obtained for the fuel oil, in which terpane concentrations were very low (Figure 4b) and steranes were not detected. Chemical biomarkers such as steranes and terpanes (see full compound identification in Supporting Information) can be removed or concentrated during the refining process and have been used to differentiate lube oils from refined products (13). The presence of the tricyclic and pentacyclic terpanes in the microlayer extract is consistent with diffuse lubricating oil contamination (24) rather than fuel oil. The source of the contamination is more likely, therefore, to be the engine room/bilge water from the wreck, washed out through the fractured hull and containing engine and hydraulic oils. From these analyses, the contaminated microlayer samples can therefore be linked to three sources of hydrocarbons: heavy fuel oil, lubricating oil, and a lighter oil. VOL. 42, NO. 7, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2277

FIGURE 3. Comparison of the polycyclic aromatic hydrocarbon (PAH) profiles between the IFO 380, the SML, and the SSW samples at Station 5. N ) naphthalene, C1-N ) methylnaphthalene, C2-N ) dimethylnaphthalene, F ) fluorene, DBT ) dibenzothiophene, C1-DBT ) methyldibenzothiophene, C2-DBT ) dimethyldibenzothiophene, P ) phenanthrene, C1-P ) methylphenanthrene, C2-P ) dimethylphenanthrene, A ) anthracene, Fl ) fluoranthene, Py ) Pyrene, BaA ) benz[a]anthracene, C ) chrysene. Peak area normalization of each sample was used for comparison. Dashed lines indicate the trends in parent to alkyl homologues for SML and SSW samples. Oil Distribution in the Water Column. Figure 5a and b show spatial interpolation plots of the analytical data for the total PAH concentrations in the SML and the SSW, respectively. The chemical signature of the heavy fuel oil was not apparent in the coastal sample extracts. The SML and SSW extracts exhibited a contamination pattern which could correspond to the WAF (i.e., soluble fraction) subject to selective losses by volatilization that could have occurred during the first days after the spill. The lubricating oil signature was restricted to samples taken immediately around the beached ship. The total PAH concentration gradient in the sea surface microlayer ranged from very high levels (∼57 µg/L) in the vicinity of the ship, down to trace levels at a distance of approximately 15 km radius from the wreck. Spatial contamination in the SML appears split into two branches and dispersed irregularly based on the main current flow and the smaller scale coastal dynamics. Further from the MSC Napoli, in the western part of the bay, concentrations ranged from 12.1 to 87.2 ng/L; contrasting with higher concentrations of total PAHs that were found irregularly in the eastern part of the bay (120.2 and 245.3 ng/L at stations 8 and 15, respectively). Offshore stations in the vicinity of the wreck exhibit higher concentrations (up to 642.2 ng/L at station 18). These data reflect the dispersion of the WAF from the sea surface microlayer (Figure 5a) with distance from the ship. In contrast, the water column (at 2 m depth) appears relatively clean with concentrations from 0.7 to 9.1 ng/L, which are typical of offshore coastal seawaters. For example, Law et al. (25) have published background levels for coastal waters from the region, with total PAH concentrations ranging up to ∼9 ng/L (in Plymouth Sound). Higher concentrations in the SSW (up to 30.8 ng/L) were recorded close to the MSC Napoli where the mixture of lubricating oil and WAF was found in the surface microlayer. Slightly elevated concentrations were also found in isolated areas of the eastern part of the bay (ranging from 13.8 to 14.6 ng/L at stations 9 and 10, respectively). These can not be related directly to the oil spill 2278

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 7, 2008

from MSC Napoli and might relate to different inputs of PAH from the costal area (e.g., runoff, harbor contamination). The highest concentration found in the surface microlayer (∼ 57 µg/L) is approximately double the dissolved concentrations of total PAHs reported during other oil spills which generally range from 0.02 to 29.3 µg/L in subsurface waters ((26) and references therein). Clearly, the concentrations of PAHs in SSW reported in the present survey are much lower than these. For comparison purposes, concentrations within the range of 2.0-8.5 µg/L have been reported for the produced formation water (PFW) discharged from an offshore oil production platform (27). Reddy and Quinn (28, 29) observed higher concentrations of total PAHs ranging up to 49.7 µg/L in surface waters 4 days after the North Cape oil spill, which decreased to background levels (0.3 µg/L) after 32 days. However, it should be noted that this oil tanker carried No. 2 fuel oil (i.e., diesel oil), a refined product that contains a higher proportion of soluble petroleum hydrocarbons than the heavy fuel oil associated with the MSC Napoli, which could explain these differences. Enrichment and Processes at the SML. The enrichment factor (EF ) CSML/CSSW) is a relative measure of the SML contamination in comparison to the subsurface water column concentrations. Figure 5c shows the spatial distribution of enrichments in the SML throughout Lyme Bay. As previously observed with total PAH concentrations (Figure 5a), the EF confirms dispersion of the oil with distance from the vessel, with highest SML enrichments located adjacent to the ship. Further from the wreck, the EF contours indicate contamination occurring in patches, which probably correspond to the local surface hydrodynamics. The patches in both the eastern sector and offshore relate to differing degrees of SML enrichment, as well as SML depletion around station 6. Such patchiness has been observed for numerous physical, chemical, and biological parameters (e.g., temperature, turbulence, salinity, nutrients, phytoplankton) in the oceans (30). In this study, considering the calm weather conditions during the sampling which benefited the formation of oil

FIGURE 4. GC/MS ion chromatograms showing the biomarkers profile in the SML and the IFO 380 samples. Panels (a) and (b) correspond to terpanes (m/z ) 191) in the SML and heavy oil samples, respectively. Panel (c) corresponds to steranes (m/z ) 217) in the SML sample alone, since steranes were not found in the heavy oil. slicks (i.e., wind speed lower than 2.5 m/s), as well as the surface hydrodynamics, the nonuniform concentration distributions can be observed in both surface microlayer and subsurface waters. The calculated enrichment factors of up to 1,900 (at the station closest to the wreck) are approximately one thousand times greater than typical values for total PAHs in coastal

waters, which vary from 1-2 up to approximately a maximum of 50 for dissolved PAHs (31–33) in clean and polluted coastal waters, respectively. Different analytical methods, locations, and choices of individual PAHs to sum in different studies may also contribute to the difference. In our case, the samples were extracted without filtration, thus including PAHs from both the dissolved and particulate phases. As a reference VOL. 42, NO. 7, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2279

FIGURE 5. Spatial plots of total PAH and EF in Lyme Bay: (a) concentrations (ng/L) in the sea surface microlayer (SML), (b) concentrations (ng/L) in the subsurface water (SSW), and (c) enrichment factors (EF). value for high EF factors, Burns and Codi (27) report enrichment from 7 to 510 times for hydrocarbons in the surface microlayer for samples in the proximity of oil platforms discharging produced formation water (PFW, raw water originated from drilling activities in oil platforms, which is discharged into the sea with elevated quantities of oil and other chemical compounds). These authors point out that the higher EF values are indicative of a thin surface oil slick, with concentrations up to 111 µg/L when a slick was visible and up to 9.4 µg/L when the slick was not visible. The results found in our study correlate with these findings with a visible slick associated with a SML concentration of 57 µg/L. Naphthalene contributes largely to the SML enrichment (Table 3 in Supporting Information). High concentrations of naphthalene in the surface microlayer were found at stations located near the shoreline (i.e., 3, 4, 4′, 5, 6, 7, 8, and 9), as well as at the offshore stations in the middle of Lyme Bay (i.e., 15, 16, and 18). This compound, despite having a high solubility, is often not detected in seawater due to its volatility. In this study, the ship is confirmed as major source oil (i.e., heavy fuel oil, lubricating oil, and diesel), with the spilled heavy oil having the highest content of naphthalene and alkyl-naphthalenes (Figure 3). However, the volatility decreases as the degree of alkylation increases. This accounts for the lower proportions of naphthalene found in the SML and SSW samples relative to the alkyl-homologues. Similar profiles for the naphthalene series have been reported for surface microlayers associated with oil platforms (27). It is well-known that evaporation is an initial and important 2280

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 7, 2008

weathering process in oil spills. In particular, 40% of the volume of light petroleum products and up to 10% of the volume of heavy oils can be rapidly lost (34). Nevertheless, in the present study the higher proportion of naphthalene in the subsurface water sample compared to the sea surface microlayer in the vicinity of the ship (Station 5) should be noted (Figure 3). This might be explained by the surface slick (damping effect) (35) at the interface acting as a boundary layer and reducing sea-air transport (i.e., volatilization) of the dissolved naphthalene from the subsurface water to the atmosphere, thus accumulating naphthalene in the SSW. In contrast, the volatilization of naphthalene directly from the oil sheen to the atmosphere decreases naphthalene concentrations in the SML. In addition, dissolution of naphthalene from the oil slick into the subsurface water would also be likely to contribute to the observed distribution largely depending on its relative solubility between the water phase and the oil phase. To elucidate the behavior of naphthalene it is therefore necessary to consider two boundary layers: first the air/oil interface and, second, the oil/water interface. Moreover, when a chemical compound (e.g., naphthalene) concentrates at the sea surface microlayer, it can act as a source to both the atmosphere and the water column interfering with the air-sea transport contaminant fluxes. This can be described as a “flux capping effect”, in addition to the damping effect. This transport phenomenon, as far as we are aware, has not been previously described and has toxicological implications as the water column could contain higher concentrations of toxic volatile hydrocarbons. Inter-

Literature Cited

FIGURE 6. Spatial plots of Fl/Py ratios at the SML and SSW in Lyme Bay. facial absorption (Gibbs absorption law), surface tension, viscosity, diffusion coefficients, and turbulence will play an important role in both controlling the air-sea (or sea-air) transport of volatile hydrocarbons (30), and the dissipation of the oil from the sea surface microlayer (i.e., slick thickness). Oil Spill Impact Assessment. It can be assumed from the spatial plots (Figure 5) that the elevated concentrations of total PAHs in the sea surface microlayer close to the ship result from oils spilled from the MSC Napoli. Elevated concentrations at the same stations in the subsurface waters can be attributed to the dissolution of the lighter compounds from the surface slick (i.e., WAF). Consideration of Fl/Py ratios (Figure 6a and b) can differentiate between petrogenic (Fl/ Py < 1) and pyrolytic (Fl/Py < 1) inputs. Pyrolytic inputs of PAH are shown to be dominant in Lyme Bay. The petrogenic input due to the oil spill is, however, evident, especially in the surface microlayer in the vicinity of the MSC Napoli. This is shown to contaminate subsurface waters close to the ship. Throughout the remainder of the bay, Fl/Py ratios are consistent with pyrolytic sources, with major inputs from the Exe estuary and Brixham harbor on the western side, and from Portland (i.e., runoff) on the eastern side. These are seen as patches with Fl/Py ratios >1. It is noteworthy that, outside of the localized area of the spill, pyrolytic sources are more dominant in the subsurface waters. In summary, oil spilled following the beaching of the MSC Napoli was primarily present as a localized surface sheen. Volatile PAH transfer might be impeded due to both damping and flux capping effects. The distributions of the PAHs between the SML and the SSW, as discussed, will have substantial implications for aquatic toxicology. Thus, sea surface microlayer sampling should be considered for both the accurate determination of the chemical composition of oil spilled, as well as to evaluate the extent of the potential toxicity.

Acknowledgments We thank the Maritime and Coastguard Agency (MCA) who provided the oil sample and aerial photographs. As well we thank the crew of R/V Plymouth Quest which participated in the contamination survey. This work was supported by a Defra/Cefas contract (PO007355). C.G. acknowledges a postdoctoral EIF Fellowship Contract “PPCPs-Transwater” (CT2005-023699) through the European Marie Curie Program.

Supporting Information Available Additional details from chemical analyses, recovery experiments, quality assurance procedures, as well as both the full compound identification in the chromatograms and analytical data sets. This information is available free of charge via the Internet at http://pubs.acs.org.

(1) ITOPF. International Tanker Owners Pollution Federation Limited; http://www.itopf.com. (2) Ehrhardt, M. G.; Burns, K. A.; Bicego, M. C. Sunlight-induced compositional alterations in the seawater-soluble fraction of crude oil. Mar. Chem. 1992, 37, 53–64. (3) Ezra, S.; Feinstein, S.; Pelly, I.; Bauman, D.; Miloslavsky, I. Weathering of fuel oil spill on the east Mediterranean coast, Ashdod, Israel. Org. Geochem. 2000, 1733–1741. (4) Díez, S.; Jover, E.; Bayona, J. M.; Albaigés, J. Prestige oil spill. III. Fate of a heavy oil in the marine environment. Environ. Sci. Technol. 2007, 41, 3075–3082. (5) Reed, M.; Johansen, O.; Brandvik, P. J.; Daling, P.; Lewis, A.; Fiocco, R.; Mackay, D.; Prentki, R. Oil spill modelling towards the close of the 20th century: Overview of the state of the art. Spill Sci. Technol. Bull. 1999, 5, 3–16. (6) Fingas, M.; Fieldhouse, B. Formation of water-in-oil emulsions and application to oil spill modelling. J. Hazard. Mater. 2004, 107, 37–50. (7) SOLAS. Surface Ocean-Lower Atmosphere Study; IGBP/SCOR/ WCRP; http://www.solas.org. (8) Hardy, J. T. The sea-surface microlayer: biology, chemistry and anthrophogenic enrichment. Prog. Oceanogr. 1982, 11, 307– 328. (9) Liss, P. S.; Duce, R. A. The Sea Surface and the Global Change; Cambridge University Press: U.K., 1997. (10) Wurl, O.; Obbard, J. P. A review of pollutants in the sea surface microlayer (SML): a unique habitat for marine organisms. Mar. Pollut. Bull. 2004, 48, 1016–1030. (11) Brown, E. D.; Norcross, B. L.; Short, J. W. An introduction to studies on the effects of the Exxon Valdez oil spill on early life history stages of Pacific herring, Clupea pallasi, in Prince William Sound, Alaska. Can. J. Fish. Aquat. Sci. 1996, 53, 2337–2342. (12) Siron, R.; Pelletier, E.; Brochu, C. Environmental-factors influencing the biodegradation of petroleum-hydrocarbons in cold seawater. Arch. Environ. Contam. Toxicol. 1995, 28, 406–416. (13) Wang, Z.; Fingas, M. F. Development of oil hydrocarbon fingerprinting and identification techniques. Mar. Pollut. Bull. 2003, 47, 423–452. (14) Wang, Z. D.; Yang, C.; Hollebone, B.; Fingas, M. Forensic fingerprinting of diamondoids for correlation and differentiation of spilled oil and petroleum products. Environ. Sci. Technol. 2006, 40, 5636–5646. (15) Stout, S. Applications of petroleum fingerprinting in known and suspected pipeline releases—two case studies. Appl. Geochem. 2003, 18, 915–926. (16) Barakat, A.; Mostafa, A. R.; Qian, Y.; Kennicutt, M. C. Application of petroleum hydrocarbon chemical fingerprinting in oil spill investigations-Gulf of Suez, Egypt. Spill Sci. Technol. Bull. 2002, 7, 229–239. (17) Alzaga, R.; Montouri, P.; Ortiz, L.; Bayona, J. M.; Albaigés, J. Fast solid-phase extraction-gas chromatography-mass spectrometry procedure for oil fingerprinting. Application to the Prestige oil spill. J. Chromatogr. A 2004, 133–138. (18) Douglas, G. S.; Burns, W. A.; Bence, A. E.; Page, D. S.; Boehm, P. Optimizing detection limits for the analysis of petroleum hydrocarbons in complex environmental samples. Environ. Sci. Technol. 2004, 38, 3958–3964. (19) Malleret, L.; Bruchet, A.; Hennion, M. C. Picogram determination of “Earthy-Musty” odorous compounds in water using modified closed loop stripping analysis and large volume injection GC/ MS. Anal. Chem. 2001, 1485–1490. (20) Guitart, C.; Garcia-Flor, N.; Dachs, J.; Bayona, J. M.; Albaigés, J. Evaluation of sampling devices for the determination of polycyclic aromatic hydrocarbons in surface microlayer coastal waters. Mar. Pollut. Bull. 2004, 961–968. (21) Law, R. J.; Fileman, T. W.; Portmann, J. E. Methods of Analysis for Hydrocarbons in Marine Samples; Aquatic Environment Protection: Analytical Methods (2); MAFF Directorate of Fisheries Research: Lowestoft, 1988; 25 pp. (22) Kelly, C. A.; Law, R. J.; Emerson, H. S. Methods of Analysing Hydrocarbons and Polycyclic Aromatic Hydrocarbons (PAH) in Marine Samples; Science Series, Aquatic Environment Protection: Analytical Methods (12); CEFAS: Lowestoft, 2000; 18 pp. (23) Wang, Z.; Fingas, M.; Lambert, P.; Zeng, G.; Yang, C.; Hollebone, B. Characterization and identification of the Detroit River mystery oil spill (2002). J. Chromatogr. A 2004, 201–214. (24) Bieger, T.; Hellou, J.; Abrajano, T. A., Jr. Petroleum biomarkers as tracers of lubricating oil contamination. Mar. Pollut. Bull. 1996, 32, 270–274. VOL. 42, NO. 7, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2281

(25) Law, R.; Dawes, V. J.; Woodhead, R. J.; Matthiessen, P. Polycyclic aromatic hydrocarbons (PAH) in seawater around England and Wales. Mar. Pollut. Bull. 1997, 34, 306–322. (26) Gonzalez, J. J.; Vinas, L.; Franco, M. A.; Fumega, J.; Soriano, J. A.; Grueiro, G.; Muniategui, S.; Lopez-Mahia, P.; Prada, D.; Bayona, J. M.; Alzaga, R.; Albaigés, J. Spatial and temporal distribution of dissolved/dispersed aromatic hydrocarbons in seawater in the area affected by the Prestige oil spill. Mar. Pollut. Bull. 2006, 53, 250–259. (27) Burns, K. A.; Codi, S. Non-volatile hydrocarbon chemistry studies around a production platform on Austrlia’s northwest shelf. Estuarine, Coastal Shelf Sci. 1999, 49, 853–876. (28) Reddy, C. M.; Quinn, J. G. The North Cape oil spill: hydrocarbons in Rhode Island coastal waters and Point Judith Pond. Mar. Environ. Res. 2001, 52, 445–461. (29) Reddy, C. M.; Quinn, J. G. GC-MS analysis of total petroleum hydrocarbons and polycyclic aromatic hydrocarbons in seawater samples after the North Cape oil spill. Mar. Pollut. Bull. 1999, 38, 126–135. (30) Powell, T. M.; Okubo, A. Turbulence, diffusion and patchiness in the sea. Philos. Trans. R. Soc. London 1994, 343B, 11–18.

2282

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 7, 2008

(31) Manodori, L.; Gambaro, A.; Piazza, R.; Ferrari, S.; Stortini, A. M.; Moret, I.; Capodaglio, G. PCBs and PAHs in the sea surface microlayer and subsurface water samples of the Venice Lagoon (Italy). Mar. Pollut. Bull. 2006, 52, 184–192. (32) Guitart, C.; Garcia-Flor, N.; Bayona, J. M.; Albaigés, J. Ocurrence and fate of polycyclic aromatic hydrocarbons in the coastal surface microlayer. Mar. Pollut. Bull. 2007, 54, 186–194. (33) Cincinelli, A.; Stortini, A. M.; Perugini, M.; Checchini, L.; Lepri, L. Organic pollutants in the sea surface microlayer and aerosol in the coastal environment of Leghorn (Tyrrhenian Sea). Mar. Chem. 2001, 76, 77–98. (34) Fingas, M. A literature review of the physics and predictive modelling of oil spill evaporation. J. Hazard. Mater. 1995, 42, 157–175. (35) Asher, W. E.; Pankow, J. F. The interaction of mechanically generated turbulence and interfacial films with a liquid phase controlled gas/liquid transport process. Tellus 1986, 38B, 305–318.

ES703125E