ABSTRACT
Following the sinking of a derelict fishing vessel near a commercial mussel farm in Penn Cove, Washington, mussels ((Mytilus trossulus), suspended 1-3 m below the sea surface, were sampled to determine if they were contaminated with diesel fuel and if so, how long it took for the contamination to depurate. Composite mussel samples were collected from 6 culture floats and two intertidal beach stations on four occasions and analyzed for 43 polycyclic aromatic hydrocarbons (PAH43), and a suite of alkanes and petroleum biomarkers. Mussel mean total PAH43 concentrations in μ/kg (parts per billion, ppb) dry weight (dw), declined from 6462 on day 5; to 4574 on day 51; 2615 on day 186; and 1097 one year later. A market sample from the same commercial operation collected January, 2012 and analyzed by the Battelle Pacific Northwest National Laboratory had a value of 839 ppb dw for the sum of 42 PAHs. The affected mussels required between 6 and 12 months to return to this total PAH concentration putative background. The whole-soft tissue total PAH43 half-life was about 5 months, far longer than the days or few weeks expected based on past literature. During the incident the commercial mussel grower voluntarily shut-down harvest, and on May 15, the Washington Department of Health (WA DOH) closed all of Penn Cove to both commercial and recreational harvest. Although individual PAH concentrations never exceeded thresholds for human consumption, samples did not initially pass sensory testing. The Penn Cove commercial and recreational harvesting was subsequently re-opened in stages to give the shellfish time to depurate and return to an untainted state. Additionally, all the concentrations of total PAHs measured in Penn Cove float and shoreline mussels during this study were within the range of concentrations reported in mussels from other monitoring sites in Puget Sound and other marine coastal areas of Washington. This relatively simple but sustained monitoring effort showed that submerged as well as inter-tidal bivalves can be contaminated by a nearby diesel spill, and that recovery may take longer than generally expected.
INTRODUCTION:
Diesel and No. 2 fuel oil spills are common in US coastal waters. Typically, responses focus on containing leaks and booming, skimming, and sorbing surface sheens. The response is generally terminated when the source is controlled and the sheen “dissipates”, as diesel readily evaporates and disperses, and surface sheens are relatively short-lived once the leak is secured. The fate and effects of the naturally-dispersed portion of spilled diesel are rarely monitored or modeled, as the ability to do so is both not readily available and difficult to implement real time during a spill. Nevertheless, when fish or shellfish encounter the naturally-dispersed fuel, it is often assumed that their exposure is brief; and, if contaminated, that they will depurate the hydrocarbons within a few weeks.
In May 2012, aerial photos showed silver sheen passing through commercial mussel culture floats in Penn Cove, Whidbey Island, Washington (Figure 1). The sheen resulted from release of a diesel blend from the 128-foot derelict fishing vessel F/V Deep Sea, which caught fire on May 13 and sank in 20 meters of water on May 14. Mussels were growing on hundreds of ropes suspended 1 to 3 meters below the floats. The mussel farm, the oldest and largest in the U.S, voluntarily ceased harvesting after the F/V Deep Sea sank.
Overview of Penn Cove, May 15, 2012, looking east and showing location of the source (F/V Deep Sea sinking location), geometry and labels of the rope-mussel floats sampled, two beach collecting sites (Madrona Beach, upper right) and the sheen passing through and around the mussel floats. Courtesy of US Coast Guard.
Overview of Penn Cove, May 15, 2012, looking east and showing location of the source (F/V Deep Sea sinking location), geometry and labels of the rope-mussel floats sampled, two beach collecting sites (Madrona Beach, upper right) and the sheen passing through and around the mussel floats. Courtesy of US Coast Guard.
The question arose: did the sub-surface mussels were exposed to, and accumulated any of, the naturally-dispersed diesel fuel? To assess this a “small science” monitoring project was initiated by the National Oceanic and Atmospheric Administration's (NOAA) Office of Response and Restoration (OR&R) Emergency Response Department (ERD). On May 18, 2012, with the cooperation and assistance of the shellfish farm, ERD scientists collected mussels from six floats and two intertidal beach locations, and shipped them to Louisiana State University for analysis of petroleum hydrocarbons. The analysis showed that the sampled mussels had elevated levels of accumulated petroleum hydrocarbons in their tissues. The analysis also showed that despite concentrations within seafood safety thresholds (WDOH, 2012), the mussels did not pass the first round of sensory testing instituted as part of the harvest reopening that ultimately occurred on June 5th.
This led to a second question of how long it would take for the mussels to depurate, or cleanse themselves, of the diesel contamination. Based on existing literature (e.g., Pruell et al., 1986; Pevin et al, 1996; Sericano et al., 1996; Thorsen et al., 2004), the expectation was that hydrocarbon concentrations in the exposed commercial mussels would decline rapidly, approaching background for similar Puget Sound locations within weeks. To test this hypothesis, the same floats and shoreline sites were resampled on July 2, 2012, and continued to be re-sampled until concentrations approached background, or pre-spill, total PAH concentrations. This paper summarizes the adaptive evolution of this study and highlights some of the challenges of conducting small science (Mearns and Simecek-Beatty, 1999) at spills of opportunity.
BACKGROUND:
Penn Cove
Penn Cove is a small (12 sq km, 3,955 surface acres) bay open to Saratoga Passage, located between Whidbey Island and Camano Island in northern Puget Sound, Washington Figure 2). The Cove is located within the Ebey's Landing National Historical Reserve, the first historical reserve in the National Park System (Klinger et al., 2007) and is home to Penn Cove Shellfish, LLC, the oldest (est. 1975) and largest mussel farm in the US, with over 60 employees and distribution of live mussels throughout the US and in Asia (Carvalho et al, 2010). Mussels are grown on hundreds of weighted ropes suspended 1 to 3 meters beneath 42 anchored floats (Figure 1). The culture operation is dependent on annual natural recruitment of mussel larvae, which settle onto rope systems during spring and summer. Harvesting occurs throughout the year.
Map of western Washington State, USA, showing locations of NOAA National Mussel Watch sites (circles) mentioned in text an in upper right corner, Penn Cove.
Map of western Washington State, USA, showing locations of NOAA National Mussel Watch sites (circles) mentioned in text an in upper right corner, Penn Cove.
Details of the Response and Amount Spilled
In December, 2011, the 128-foot fishing vessel F/V Deep Sea anchored in Penn Cove on Washington state-owned aquatic lands, 200 meters north of the outer edge of a mussel float array owned by Penn Cove Shellfish Inc. (I. Jefferds, PCS, Inc., personal comm). By Washington law, a vessel owner must remove an anchored vessel anchored on state-owned aquatic lands after 30 days unless they have received permission to moor longer. Washington State Department of Natural Resources took measures to have the owner remove the vessel to no avail. At 23:45 on Saturday May 12, 2012, the vessel caught fire, and at 1800 on May 13 the Deep Sea sank to the bottom with an unknown amount of diesel fuel on board. A cracked vent allowed fuel to leak at an estimated rate of 2 gallons per minute. Penn Cove Shellfish Inc. suspended harvesting before fuel surface sheens reached the mussel floats. The volume of diesel discharge grew through Monday, May 14, resulting in oil sheen drifting over the commercial operation (visible in Figure 1), as well as contacting shorelines, leading the Washington Department of Health to close the area to both recreational and commercial shellfish harvesting. By noon Friday, May 18, 3,100 gallons of fuel had been recovered from the vessel tanks and 500 gallons from absorbent material. Small amounts of oil periodically floated up into the boom area and into surrounding waters through May 31. By May 31, from a combination of recovering operations and leaks, a total of over 5000 gallons of fuel had been removed from the sunken vessel's tanks.
On June 3 the F/V Deep Sea was raised, and on June 6 it was towed out from Penn Cove, but the shellfish harvest was not fully opened even after the vessel was removed. Part of the incident specific re-opening procedures developed jointly between WA DOH and NOAA Seafood Inspection Program (SIP) was for the shellfish to pass sensory testing. Even when seafood samples from spill areas pass the standard chemical-analytical tests (the levels of polycyclic aromatic hydrocarbons are below the limits permitted as determined by human health risk assessment), flavor or odor may still be affected, and they are then “tainted”. Taint in seafood renders it adulterated and unfit for human consumption according to U.S. law (Federal, Food, Drug and Cosmetics Act, US Code 21, Chapter IV, Sec. 402 [342], a.3). After the initial sensory test on June 5 only the northern parts of Penn Cove were opened to recreational harvest and the commercial rafts and southern beaches remained closed due to detected taint. Following the second sensory testing on June 8, the commercial mussel raft samples passed and commercial harvest re-opened, while the southern beaches remained closed for recreational harvest. Finally on June 22 all shellfisheries were open again (WA DOH. September 2012).
The Washington Department of Ecology Oil Spills Program estimated that the potential cargo on board was 8,655 gallons; pumped from tanks while underwater was 3,100 gallons; on- water recovery was 4,166 gallons (David Byers, pers. comm., June 14, 2013). By this account, 1,389 gallons presumably dispersed and evaporated, the largest portion likely during the first few days following the sinking.
METHODS:
Sampling Design
The layout of the 42 mussel floats at the shellfish growing operation provided an ideal grid for replicate sampling of the submerged mussels. The anchored floats were distributed across a rectangle area of 55 acres. Sufficient replication to characterize the mean and range of concentrations of polycyclic aromatic hydrocarbons (PAHs) in mussels across the entire grid was desired. Mussels were sampled from six floats, four sample sites at approximately the four corners of the grid (floats A1, A7, F4, and F8, Figure 1) and two in the center (C1 and C8, Figure 1). Mussels were also sampled from two intertidal shorelines. A search for pre-spill mussels failed to produce material because the mussel farm switched the source of its retail product to an unaffected production site on the Washington coast. Fortuitously, this gap was filled with the discovery of a sample of Penn Cove mussels harvested in February, 2012 that had been analyzed for baseline PAHs by the Battelle Pacific Northwest National Lab for a U.S. Navy monitoring program (see below).
In order to compare the initial May 18, 2012, samples with subsequent samples, collection was continued under the original fixed-station, 8-sample strategy described above. This allowed for comparing means and ranges of contaminants over time.
Field Sampling
The initial sampling occurred during a low tide on May 18, 2013, five days after the sinking of the F/V Deep Sea. Biologists with NOAA's Emergency Response Division (ERD), an inspector from the Washington Seattle Department of Health (DOH) and the Penn Cove Shellfish, Inc. owner and field manager, met and quickly developed the sampling strategy (above) and then visited the eight sites by boat. The NOAA-contract chemistry laboratory at Louisiana State University was immediately notified to prepare for receiving and analyzing mussel samples.
At each float site, a rope was selected at random, pulled onto the float and groups of 30 to 50 4-6 cm mussels removed with gloved hands, wrapped in aluminum foil and placed in Ziploc bags labeled with the date and site. The bags were returned to the shellfish farm operation center, packed in an insulated box with ice packs, sealed and shipped overnight to the Louisiana State University Response and Chemical Assessment Team (LSU-RCAT) in Baton Rouge for extraction and chemical analysis.
Source oil samples were collected by the Washington Department of Ecology from two bow tanks and one crab hold aboard the Deep Sea provided to NOAA, and submitted on June 11, 2012 to LSU-RCAT for analysis.
Mussels were again sampled on July 2, 2012, 27 days after the removal of the Deep Sea. Based on the available literature (reviewed below), the assumption was that the levels of contaminants would rapidly decline to approximately half their original concentrations. When the analytical results were completed several weeks later, it was apparent that this had not occurred. As a consequence, another round of sampling was coordinated with Penn Cove Shellfish and took place on November 13, 2012. While tissue concentrations were lower, the rate of decline continued to be slower than anticipated, based on the literature. After review of those results and the passage of a longer period of time, one last sampling occurred, on May 21, 2013, 350 days after removal of the vessel and more than a year after it had sunk.
Chemical Analysis.
All samples (mussel soft tissue, fuel and settlement line) were analyzed by gas chromatography/mass spectrometry (GC/MS) for concentrations of 43 PAHs (Table 1), as well as nC10-nC35 saturate hydrocarbons and isoprenoids, steranes and diasteranes, and triaromatic steroids (not shown here). Mussel tissue composites were also analyzed for percent moisture and lipid content.
Source Oil
Approximately 0.20-grams of each of the three source diesel sample were extracted with 30 milliliters of dichloromethane (DCM). One milliliter of each extract was then transferred with a clean graduated, gas-tight syringe into an autosampler vial.
Mussel Tissue Preparation
At the LSU-RCAT laboratory, all mussels from each sampling site and for each sampling date had the soft tissue removed, composited, homogenized, and extracted for petroleum compounds. Approximately 50-80 grams of homogenized tissue from each site and sampling event were weighed, freeze-dried for a minimum of 12-hours, and approximately 5-6-g of freeze dried tissue was extracted using DCM and ultra-sonication. The extracted samples and methods blanks were solvent exchanged into hexane, concentrated to 2 ml, and fractionated on activated alumina/silica gel columns (modified EPA SW-846 methods 3610 and 3630 (U.S. EPA, 1996)). The alkane and PAH fractions were collected in one flat bottom flask and concentrated to 1-mL by rotary evaporation and nitrogen blowdown.
GC/MS Methodology
The GC/MS was an Agilent 6890A GC system configured with a 5% diphenyl/95% dimethyl polysiloxane high resolution capillary column (30 meter, 0.25 mm ID, 0.25 micron film) directly interfaced to an Agilent 5973 MS detector system. An Agilent 7693 Auto Injector was used for sample introduction into the GC/MS system. The injection temperature was set at 280°C and only high-temperature, low thermal-bleed septa were used in the GC inlet. The GC was operated in the temperature program mode with an initial column temperature of 60°C for 3 minutes then increased to 280°C at a rate of 5°C/minute and held for 3 minutes. The oven was heated from 280°C to 300°C at a rate of 1.5°C/min and held at 300°C for two minutes. Total run time was 65.33 minutes per sample. The interface to the MS was maintained at 300°C. The MS was operated in the Selective Ion Monitoring (SIM) to maximize the detection of the target analytes in Table 1.
Quantitative Analysis
The concentration of specific target oil analytes was determined by a 5-point calibration and internal standard method. Internal standards include naphthalene-d8, acenaphthylene-d10, chrysene-d12, and perylene-d12. Internal standards were always added to the samples just prior to GC/MS analysis. A commercially available oil analysis calibration standard containing parent (non-alkylated) hydrocarbons was used to establish the five point calibration curve that results in an average response factor for each analyte in the standard mixture. Alkylated homologs were quantified using the response factor of the parent, and were, therefore, only semiquantitative. This is a standard procedure since alkylated standards are not readily available. Recovery of all analytes was estimated using two hydrocarbon surrogate standards: 5 alpha androstane (alkanes) and phenanthrene-d10 (PAHs). Surrogate standards were added to each sample prior to all extraction procedures. Method extraction blanks were prepared and analyzed with each group of sample extractions (methods blanks averaged 6.2 μ/kg Total PAHs and 0.49 mg/kg for total alkanes; pump oil was free of PAHs). All samples were analyzed in exclusive analytical batches, on the same GC/MS instrument, and in the same analytical sequence.
Data and Statistical Analysis
PAH concentrations reported here are based on the dry weight, were surrogate corrected, and have the ∑PAHs in the method blank subtracted out. The focus of this paper is mainly total PAH43 concentrations reported as micrograms of PAHs per kilogram of tissue (μ/kg) dry weight (dw), equivalent to parts per billion (ppb). Additional analyses, not reported here, included a modified Fossil Fuel Pollution Index (FFPI, Boehm and Farrington, 1984), source fingerprinting indices and diagnostic biomarker ratio analysis.
RESULTS:
Source Oil
The three source oils had similar PAH43 group profiles (example in Figure 3), the naphthalenes were the dominant PAHs, followed by the phenanthrenes, the fluorenes, the dibenzothiophenes, the pyrenes, the naphthobenzothiophenes, and, finally, the chrysenes. The source oil profile was typical of diesel-blend type petroleum product since chrysenes and the four groups of biomarker compounds were also present. These groups are often absent from a straight diesel or marine diesel due to the refining process.
Average PAH profile of 43 PAHs in F/V Deep Sea source diesel collected from the “bow” fuel tank.
Average PAH profile of 43 PAHs in F/V Deep Sea source diesel collected from the “bow” fuel tank.
Not reported here are detail analyses of fuel source fingerprinting indexes and diagnostic biomarker ratios, which further confirmed the identity of the fuel oil.
Mussels
Moisture content for all mussel samples ranged from 75.6 to 88.0 percent, with higher content in November 2012 and May 2013 than in May or July 2012. Lipid content ranged from 0.82 to 4.56 percent.
Total PAH43 concentrations in mussels ranged from 114 μ/kg dw in the sample from float C-1 in May, 2013, to 8,780 μ /kg dw in the sample from float C-8 in May, 2012 (Table 2). Concentrations of Total PAH43 decreased over time. For the six float sites, mean Total PAH43 concentrations were 5825 μ /kg dw for May 2012, 4626 μ /kg dw for July 2012, 2585 μ /kg dw for November 2012, and 1311 μ /kg dw for May 2013 (Table 2.
Total PAH43 concentrations (sum of 43 analytes, μ /kg dw) in whole soft tissue of mussels six rope float sites and two intertidal beach sites in Penn Cove, WA, for four sampling events.

PAH histogram plots for all the mussel tissues showed a characteristic bell-shaped distribution between parent PAHs and their alkyl homologs indicating a petrogenic (i.e., oil-related) source of PAHs and whole-oil (non-selective) accumulation. Figure 4 shows individual PAH histograms for one site typical site.
An alkane profile (not shown) for the recreational harvest site at Madrona Beach suggested that there was contamination from a heavier-than-diesel petroleum product that was not in the May 2012 tissue samples but was present in the July 2012 tissue samples.
PAH histogram profiles for site A-1, a typical site, May 2012 to May 2013. Total of 43 PAHs in ug/kg dry weight. Note differences in concentration axis scales.
PAH histogram profiles for site A-1, a typical site, May 2012 to May 2013. Total of 43 PAHs in ug/kg dry weight. Note differences in concentration axis scales.
DISCUSSION AND IMPLICATIONS:
Initial Contamination
The initial mussel sampling, five days after the sinking, confirmed that both suspended and intertidal shoreline mussels were contaminated with diesel-derived petroleum hydrocarbons. The mean concentration of 43 Total PAHs (PAH43) was 5825 μ /kg dw (range 2777 to 8780 μ /kg dw). Although it was not possible to secure Penn Cove mussels collected directly prior to the incident, the principal investigator of the US Naval Environmental Investment (ENVVEST) monitoring program in Bremerton, Washington had obtained Penn Cove mussels in February, 2012 (R. Johnston, pers. comm.). They were analyzed for 41 PAHs by Battelle Pacific Northwest National Lab in Sequim, Washington, and the results made available to us (J. Brandenberger, pers. comm.). The total PAH41 concentration was 839 μ /kg dw. Presuming the comparability of analytical methods and results, this suggests that the samples from May 18, 2012 contained Total PAH concentrations about 7.5 times higher than the February 2012 samples.
Biological Half-Life of PAHs
Total PAH43 concentrations declined during the course of the monitoring, and approached the February 2012 default (ENVVEST) baseline Total PAH41 concentration of 839 μ /kg dw on or before the May 2013 sampling. Figure 5 shows a plot of the median Total PAH43 concentrations for the six floats over correct time scale. The data fit an exponential decline with an R2 of 0.995, resulting in a biological or tissue half-life for Total PAH43 of about 4 months (120 days), although the start time for depuration may have been delayed until June 5, 2012, given that sheens were visible until this date. The next sampling following the removal of the vessel was on July 2, 2012. Calculation of a linear rate of loss of Total PAH43 between July 2 and November 13, 2012 yields a Total PAH43 half-life of 158 days, or about 5.1 months.
Median concentrations of Total PAH43 for float mussels at the four sampling times, together with first-order decay curve.
Median concentrations of Total PAH43 for float mussels at the four sampling times, together with first-order decay curve.
A four- to five-month half-life of total PAH43 in mussels at this site was unexpected. Research conducted during the 1970s and 1980s subjected mussels and oysters to multi-day elevations of diesel, PAH-contaminated sediments or PAH mixtures, and then allowed the shellfish to depurate in clean water. Pruell et al (1986) exposed mussels in tanks containing PAH-contaminated sediments and then returned them to clean water. The mussels depurated PAHs with half-lives ranging from 14 to 30 days. In a reciprocal transplant study in Massachusetts, Peven et al (1996) transferred mussels from a PAH-polluted site near Boston to a relatively clean site, and documented an overall total PAH half-life on the order of 7 days. Likewise, Sericano et al. (1996) transferred oysters (Crassostrea virginica) from a PAH-contaminated site in the Houston Ship Channel, Texas, to a relatively clean site in southern Galveston Bay, monitoring depuration and reporting biological half-lives of 8 PAHs ranging from 10 to 16 days—in rough agreement with Pruell et al (1986). Sericano et al. (1996) also pointed out that the depuration of PAHs in chronically-contaminated oysters was about double the rates in newly contaminated oysters, citing similar conclusion from Jackim and Wilson (1977; cited in Jackim and Lake, 1978), who exposed three hundred soft-shell clams (Mya arenaria) in sediment in a continuous flow (14° C) tank dosed with labeled benzo[α]pyrene (B(a)p) in No. 2 fuel oil daily. Animals were removed weekly and assayed for accumulation of the radio-labeled B(a)p. Depuration biological half-life for the clams ranged between 5.5 and 9 days. The depuration rate appeared to be somewhat slower upon prolonged accumulation. More recently, Thorsen et al (2004) loaded freshwater mussels, Elliptio complanta, with 46 PAHs, measured depuration rates, and reported individual PAH half-lives ranging from 2-3 days for naphthalenes to 16 days for perylene. It was these studies that lead us to expect a total PAH half-life of between several to 30 days.
An interesting exception to these apparently short biological half-lives is in an experiment conducted McIntosh et al. (2004). Here, mussels from a mussel farm site contaminated with PAHs, were transferred to laboratory flow-thru tanks and the decline of PAHs was measured on a weekly basis. McIntosh et al. concluded that “In the experimental system used, the total concentration of PAH compounds in contaminated mussels fell by <50% in a period of 122 days.” They also cautioned that their flow- or flushing rate was very low compared to what aquaculture enterprises use to depurate bacterial contamination.
Therefore, a possible explanation for the slow loss of total PAHs in Penn Cove mussels may be that the water mass in Penn Cove has a low exchange rate with water from the Saratoga Basin and Puget Sound. The Cove is located 42 km south of the Deception Pass entrance to the Strait of Juan de Fuca, and 55 km north of the entrance to Puget Sound (Figure 2), large water masses and with very active currents. Presumably, it would take many tidal cycles for clean Puget Sound or Strait water to reach and exchange with Penn Cove water. A slow water mass exchange may be why the Cove is a prime location for mussel rearing, both indigenous and farmed, as it may be an excellent larval retention zone, ensuring that planktonic eggs and larvae produced locally would remain local. At the time of this writing specific current or water mass exchange data to confirm this idea had not been identified.
In addition, the intertidal shore site mussels indicated the possibility of shoreline sediment contamination, possibly directly from sheen contact. If sediments were contaminated, diesel would be slowly released back into the water. Unfortunately additional measurements to confirm sediment contamination were not made.
Comparison to Other Regional Mussel Watch Data
The concentrations of total PAHs measure in Penn Cove float and shoreline mussels during this study were within the range of concentrations reported in mussels from other monitoring sites in Puget Sound and other marine coastal areas of Washington.
The most recent regional Mussel Watch survey with comparable and available data was conducted in the winter of 2010 (December 2009 to April 2010), two years prior to the Deep Sea incident. The survey, “Washington State 2009/10 Mussel Watch Pilot Project” involved 24 sites sampled by state and NOAA biologists and volunteers from local organizations (http://wdfw.wa.gov/publications/01127/). NOAA's National Mussel Watch contractor, TDI Brooks, analyzed single composites of 25 to 50 mussels for 52 PAHs: 38 of the PAHs were common to the target list used in this project. For these, the average total PAH38 concentration was 2363 μ /kg (ppb) dw with a range of 229 μ /kg (ppb) dw at Gray's Harbor West Jetty to 15,300 μ /kg (ppb) dw in a composite from Myrtle Edwards Park in Seattle. For visual comparison with the 2012 and 2013 Penn Cove float mussel data, data from the 24 sites were pooled into five sub-areas: the Outer Coast (3 sites), the Straits of Juan de Fuca and Georgia (5 sites), northern Puget Sound (basically the Whidbey Basin, 9 sites), Central Puget Sound (5 sites) and South Puget Sound (3 sites).
As shown in Figure 6, the means and ranges of Total PAH38 from the 2012/13 Penn Cove float samples are comparable to the means and ranges of several 2010 regional study sub-areas. For example, the average Total PAH38 concentrations for the North Puget Sound region, mainly site samples in Snohomish County and on southern Whidbey Island and east Camano Island, Island County, was 865 μ /kg (ppb) dw, ranging from 394 ppb at Possession Point, Whidbey Island, to 1475 ppb at the Mukilteo Ferry, Snohomish County. These were lower, but comparable to, the May 2013 Penn Cove float Total PAH38 samples (mean, 1097, range 114 to 2217 ppb). Concentrations were much higher and more variable at five sites in Central Puget Sound, averaging 5399 ppb and with a range of 436 ppb at Waterman Point, Sinclair Inlet to 15,300 ppb at the Myrtle Edwards site. The 2010 Central Puget Sound mean was comparable to the May 2012 Penn Cove float mussel mean of 6405 ppb (within the first week of the Deep Sea spill.) In looking at the PAH38 levels note that the Central Puget Sound mussel watch sites are near densely populated, high use urban areas with additional sources of PAH's (e.g., marinas, urban runoff) and petroleum hydrocarbons entering the water, while Penn Cove is a relatively rural and isolated bay on an island in Washington's northern Puget Sound. The concentration in the 2010 Myrtle Edwards Park (Seattle) sample, 15300 ppb, was nearly twice as high as the most contaminated Penn Cove float sample (8,780 ppb dw) taken during the first week of the Deep Sea spill. Thus the spill elevated total PAH38 concentrations to within the range of 2010 Central Puget Sound concentrations during the first week of the spill, followed by a decline that by May 2013 overlapped with 2010 North Puget Sound samples.
Box plots showing means (columns), and maxima and minima (horizontal marks) of concentrations (μ /kg or ppb, dw) of total PAHs in mussels from (A) multiple sites sampled during the 2010 Washington Mussel Watch Pilot Project compared to (B) February 2012 ENVVEST market mussels and the four time series sampling events from Penn Cove floats sampled during this study. The charts compare the sums of 38 PAHs common to the three data sets. Sites with the highest concentrations sampled during the 2010 Mussel Watch (Nahcotta, Willapa Bay and Myrtle Edwards Park) noted. Number of sites in each sub-region noted at bottom of chart (A). Number of samples for each Penn Cove sampling event was 6.
Box plots showing means (columns), and maxima and minima (horizontal marks) of concentrations (μ /kg or ppb, dw) of total PAHs in mussels from (A) multiple sites sampled during the 2010 Washington Mussel Watch Pilot Project compared to (B) February 2012 ENVVEST market mussels and the four time series sampling events from Penn Cove floats sampled during this study. The charts compare the sums of 38 PAHs common to the three data sets. Sites with the highest concentrations sampled during the 2010 Mussel Watch (Nahcotta, Willapa Bay and Myrtle Edwards Park) noted. Number of sites in each sub-region noted at bottom of chart (A). Number of samples for each Penn Cove sampling event was 6.
While the spill did not push PAH concentrations above the range of values for the region as a whole, the composition of the Penn Cove PAHs clearly reflected a petrogenic (or petroleum) source throughout the monitoring event whereas most the 2010 region wide Mussel Watch samples reflect mainly pyrogenic, or combustion sources.
Implications for Response and Assessment
This small study provides new insights into the fate and effects of relatively small near shore diesel fuel spills. It suggests that the effects of these incidents can be characterized and measured for a period of time after the response itself is complete. In the case of the F/V Deep Sea incident, shellfish retained the chemical signature of the diesel release for many months after the initial release and removal of the wreck. The incident occurred in a highly productive body of water with oceanographic characteristics that not only make it an ideal mussel culture area, but also may have contributed to persistence of the spill signature. Finally, it led to a revision of State closure and re-opening protocols (Drury, 2014).
Since diesel readily disperses, it may be prudent for response or resource agencies to monitor the water column for hydrocarbons around an active small vessel response, particularly when such response is taking place in relatively confined waters and areas of special biological or economic significance. One way to do this is through fluorometry or other in situ water column sampling
This study provided an interesting and potentially useful perspective on the fate of diesel releases in the marine environment. Hopefully it also provides an incentive for others to take advantage of spills of opportunity to investigate basic questions about the fate and effects of oil spills.
ACKNOWLEDGEMENTS:
The authors greatly appreciate the collaboration of Penn Cove Shellfish CEO Mr. Ian Jefferds and site manager Mr. Tim Jones, who provided both access and logistical support for the project. Washington Department of Health inspector Mark Toy conferred with us and helped sample during the first sampling. The authors also thank Drs. Robert Johnston (US Naval Environmental Investment (ENVVEST) monitoring program) and Jill Brandenberger (Battelle PNNL) for providing their data on January 2012 Penn Cove mussel samples. Finally, thanks to ERD Director Debbie Payton for supporting small science at spills of opportunity and at this incident in particular. The NOAA Emergency Response Division, Office of Response and Restoration, Seattle, WA provided funding for this project.