A review of seafood safety after the Deepwater Horizon blowout.pdf

7/30/2019 A review of seafood safety after the Deepwater Horizon blowout.pdf

1/8

1062 volume 119 |number 8 | August 2011 Environmental Health Perspectives

Review

Current estimates suggest that the DeepwaterHorizon (DH) blowout resulted in the releaseo approximately 4.4 million barrels 20%

(7.0 105 m3) into the northern Gul oMexico over a 3month period during thesummer o 2010 (Crone and olstoy 2010).Te leak was a result o a deepwater rig explosion on 20 April 2010 due to methane gasrelease ater drilling an exploratory well. Anattempt to activate a saety eature to preventa blowout ailed. Ater burning or 36 hr, theentire platorm sank to the seaoor.

Because o concerns over seaood saety,on 2 May 2010, the National Oceanic andAtmospheric Administration (NOAA) initiated closures o ederal waters to commercialand recreational shing; Louisiana, Mississippi,

Alabama, and eventually Florida subsequentlyinstituted sheries closures in state waters, incoordination with the U.S. Food and DrugAdministration (FDA) (Figure 1). By 21 June,closures covered approximately 37% o theGul o Mexico (225,290 km2), extending eastrom Atchaalaya Bay, Louisiana, to PanamaCity, Florida (NOAA 2010b). Te well wascapped on 15 July, and on 19 September,relie wells were completed that permanentlydisabled the well. Reopening o closed areasto shing began on 23 June, although NOAAreclosed 10,911 km2 to deepwater shing or

royal red shrimp northwest o the wellheadon 24 November 2010 (NOAA 2010a). On21 January 2011, only 0.4% o ederal waters

(1,041 mi2; 2,697 km2) immediately surrounding the well remained closed to sheries.As o 19 April 2011 all Gul o Mexico ederalwaters are open to sheries.

Te ecological and human health impactso the DH oil blowout may dier rom previous oil spills because o the depth at whichthe oil leak occurred, the large volume o oilreleased, and the unprecedented volume odispersants (Corexit 9500 and Corexit 9527)used at the wellhead. Contamination o Gulseaood has been a central concern or government agencies, shing businesses, and consumers. An Institute o Medicine (IOM) report

and a report rom the Oil Spill Commissionrecommended continued analysis o Gul seaood as an important component to determinepotential longterm health impacts (IOM2010) and restore consumer conidence inGul sheries (Oil Spill Commission 2011).he goal o the present review is to inormrisk assessment and monitoring eorts bya) synthesizing existing inormation on theduration and extent o seaood contamination ater oil spills; b) evaluating the currentregulatory process used to determine when itis appropriate to reopen sheries in the Gul;

and c) providing recommendations or uture

testing and monitoring o seaood.

Methods

We began by gathering inormation rom scientiic publications identiied using webbasedtools including Google Scholar (Google 2011),PubMed (National Center or BiotechnologyInormation 2011), and Web o Science(Tomson Reuters 2011), and combinations othe ollowing search terms: oil spill (including individual names o previous and presentoil spills), seaood saety, heavy metals(including individual metals), polycyclic aromatic hydrocarbons (PAHs; including speciic individual PAHs), isheries, seaoodconsumption, nsh, shellsh, oysters,crustaceans, lobsters, and mollusks. Wealso searched or these terms in ederal and international organization technical reports anddatabases rom NOAA, the FDA, the U.S.Environmental Protection Agency (EPA), theNational oxicology Program, the InternationalAgency For Research on Cancer, and the WorldHealth Organization (WHO). In addition, weconsulted with government oicials and academic experts involved in the development andimplementation o the testing strategy or evaluating seaood saety ater the DH oil spill.

ResultsSeaood contamination ater previous oil spills.Studies ater previous oil spills have shownthat seaood contamination is determined bynumerous actors, including the type and quality o the oil, the proximity o the spill to shing grounds, ambient temperature and weatherconditions, and species and ecosystemspecicparameters that determine metabolism andthe potential or bioaccumulation at dierentlevels o the ood chain. Most studies haveocused on levels o PAHs in seaood ater oilspills, and only a ew have evaluated metal

Address corre sponden ce to J.M. Goh lke , 153 03rd Ave. North, RHB 530B, Department oEnvironmental Health Sciences, University o Alabamaat Birmingham, Birmingham, AL 35294 USA.elephone: (205) 9347060. Fax: (205) 9756340.Email: [email protected]

Supplemental Material is available online (doi:10.1289/ehp.1103507 via http://dx.doi.org/).

his review was generously supported by theWalton Family Foundation.

.F. is employed by Environmental Deense Fund,a national nonproft organization. Te authors declarethey have no other actual or potential competingfnancial interests.

Received 31 January 2011; accepted 29 April 2011.

A Review of Seafood Safety after the Deepwater HorizonBlowoutJulia M Gohlke,1 Dzigbodi Doke,1 Meghan Tipre,2Mark Leader,2and Timothy Fitzgerald3

1Department of Environmental Health, and 2Department of Epidemiology, School of Public Health, University of Alabama atBirmingham, Birmingham, Alabama, USA; 3Environmental Defense Fund, Washington, District of Columbia, USA

Background: Te Deepwater Horizon (DH) blowout resulted in fsheries closings across the Gulo Mexico. Federal agencies, in collaboration with impacted Gul states, developed a protocol todetermine when it is sae to reopen fsheries based on sensory and chemical analyses o seaood. Allederal waters have been reopened, yet concerns have been raised regarding the robustness o theprotocol to identiy all potential harmul exposures and protect the most sensitive populations.

oBjectives: We aimed to assess this protocol based on comparisons with previous oil spills, pub-lished testing results, and current knowledge regarding chemicals released during the DH oil spill.

Methods: We perormed a comprehensive review o relevant scientifc journal articles and govern-ment documents concerning seaood contamination and oil spills and consulted with academic andgovernment experts.

results: Protocols to evaluate seaood saety beore reopening fsheries have relied on risk assessmento health impacts rom polycyclic aromatic hydrocarbon (PAH) exposures, but metal contaminationmay also be a concern. Assumptions used to determine levels o concern (LOCs) ater oil spills havenot been consistent across risk assessments perormed ater oil spills. Chemical testing results aterthe DH oil spill suggest PAH levels are at or below levels reported ater previous oil spills, and well

below LOCs, even when more conservative parameters are used to estimate risk.conclusions: We recommend use o a range o plausible risk parameters to set bounds aroundLOCs, comparisons o post-spill measurements with baseline levels, and the development andimplementation o long-term monitoring strategies or metals as well as PAHs and dispersant com-ponents. In addition, the methods, results, and uncertainties associated with estimating seaoodsaety ater oil spills should be communicated in a transparent and timely manner, and stakeholdersshould be actively involved in developing a long-term monitoring strategy.

keywords:Deepwater Horizon, dispersants, Gul o Mexico, heavy metals, oil spill, poly-cyclic aromatic hydrocarbons, risk assessment, seaood. Environ Health Perspect119:10621069(2011). doi:10.1289/ehp.1103507[Online 12 May 2011]


2/8

Seafood safety after the DH blowout

Environmental Health Perspectives volume 119 |number 8 | August 2011 1063

contamination. Very little is known about seaood contamination rom dispersants, but toxicological studies o crude oil components anddispersants have been undertaken by international and national research bodies.

PAHs. PAHs are widespread organicpollutants containing two or more aromaticrings, and their toxicity varies widely among

individual PAH compounds [summarized inSupplemental Material, able 1 (doi:10.1289/ehp.1103507) or cancer end points andSupplemental Material, able 2 or noncancerend points]. PAHs have been shown to accumulate in ish and shellish, and concentrations o particular PAHs considered unsae,particularly in shellish, have been recordedater previous oil spills.

Overall PAH levels provide an indication o the extent o potential contamination.However, because toxicity varies greatly amongPAHs, inormation on individual PAH concentrations is needed to assess potential humanhealth implications rom these types o data. Atoxic equivalency approach is oten used to estimate risk based on the relative potency o individual PAHs relative to benzo[a]pyrene (BaP),but this approach does not take into accountdierent modes o action, and toxic equivalencyactors have been derived or only a small subseto PAHs ound in oil (Reeves et al. 2001).

Te duration o elevated PAH levels in seaood ater previous oil spills has varied romseveral weeks to several years. Highmolecularweight PAHs (those with our or more benzenerings) are more likely to accumulate in tissuesand persist or longer periods than lowmolecularweight PAHs (Meador et al. 1995). PAH

levels ater previous oil spills have been consistently lower in vertebrates compared withinvertebrates (Law and Hellou 1999). hisphenomenon is explained at least partly by therelatively welldeveloped cytochrome P450monooxygenase enzyme system in inish

(Hellou et al. 1995; Stegeman and Lech 1991),a system that results in increased metabolicclearance o PAH relative to bivalve mollusks,which have limited PAH metabolic capability,and crustaceans (Yender et al. 2002).

Several studies have reported evidenceo increased PAH concentrations in seaoodater oil spills compared with baseline levels.

For example, the concentration o two tosixring PAHs in scallops ater the Braeroilspill o the coast o the Shetland Islands wasound to be > 1,300 ppb versus 1290 ppb inreerence samples (Kingston 1999); ater theSea Empressoil spill o the coast o Wales,concentrations ranged rom 12 to 186 ppbin salmon samples compared with reerencesamples ranging rom 9 to 86 ppb (Lawet al. 1997) [Supplemental Material, able 3(doi:10.1289/ehp.1103507)]. Law et al.(2002) examined PAH levels in reerence datasets and postoil spill seaood across 19 studiesand ound average total PAH concentrations between 20 and 1,600 ppb in baselinemonitoring studies and between 104 and27,400 ppb ater various oil spills.

imeseries data sets rom the Sea Empressoil spill showed that elevated PAH levels inmost o the impacted area returned to baseline levels within 4 months in inish andcrustaceans, and within 6 months or bivalvemollusks (Law et al. 1997). PAH levels wereincreased in several species o sh and shellshor 3 weeks ater the Braeroil spill, but ell tobaseline levels within 2 months in wild ish(Kingston 1999). However, salmon arms inthe area were contaminated or 6 months, crabsand bivalves had elevated levels or 12 months,

and the Norway lobster, which burrows insediment, had elevated PAH levels or > 6 years(Kingston 1999). In contrast, PAH concentrations in mussels ater the /VDubai Staroil spill in San Francisco Bay, Caliornia, ellto baseline levels ater only 4 weeks (Klasing

and Brodberg 2010). Findings based on theAmoco Cadiz, Exxon Valdez, Braer, and Erikaoil spills suggest that contamination is particularly longlasting when oil becomes entrainedin sediment, where it can be remobilized bybenthic organisms or storm events (Law et al.2002). PAH levels in sediment and bivalveswere elevated in many areas or approximately

34 years ater the Exxon Valdezspill (Tomaset al. 1999), and PAH levels in mussels wereelevated or 6 months ater the much smaller1997 M/VKuroshimaoil spill (NOAA 2002).Because o longterm PAH contamination,restrictions on oyster and lobster harvestingwere in place or 7 years ater theAmoco Cadizand Braeroil spills (Law et al. 2002). Overall,studies o previous spills suggest that severalactors may play a role in determining theduration o PAH contamination, includingthe amount o sedimentation and likelihoodo subsequent resuspension o the oil, the composition o the oil, the rate o biodegradation(which tends to be higher in warmer climates),and the particular species o interest.

Metals. Metals are normal constituents ocrude oil and drilling uids used in the oil production process and bioaccumulation in seaood has been raised as a potential longtermhealth concern rom the DH oil spill (Solomonand Janssen 2010). Metals including zinc (Zn),manganese (Mn), arsenic (As), cobalt (Co),chromium (Cr), selenium (Se), mercury (Hg),cadmium (Cd), copper (Cu), lead (Pb), nickel(Ni), tin (Sn), antimony (Sb), and vanadium(V) have been ound to accumulate in sediments and marine organisms harvested romoil spill zones. In particular, Ni and V are

ound at signicant levels in crude oil, and BP(2010) has reported concentrations o these elements ranging between 13 and 29 ppm or Niand 10 and 106 ppm or V in Gul o Mexicocrude oils. In addition, Wainipee et al. (2010)recently suggested that there is potential or

Figure 1. Timeline o sheries closures and reopenings in the Gul o Mexico due to the DH blowout in 2010. Abbreviations: AL, Alabama; MS, Mississippi; LA,Louisiana; FL, Florida. Data rom NOAA (2010b) and FDA (2010). On 13 January 2011, 4,213 mi 2 o ederal waters around the well and parts o area 12 in LA Statecoastal waters remained closed. On 1 February 2011, NOAA repoened the ederal waters that had been reclosed to royal red shrimp shing. On 19 April 2011,NOAA reopened all remaining ederal waters. Approximately 1.5% o LA coastal waters remain closed.

29 April

MS announces

disaster declaration

15 July

Well

shut-in

29 and 30 July

FL, MS, and

LA begin

reopening

state waters

19 September

Well permanently

disabled

29 October

NOAA and FDA

announce results

of dispersant

tests on seafood

in the Gulf

24 November

NOAA closes

4,000 mi2 of

previously open

waters

7 August

AL begins

reopening

state waters

7 May12 July

NOAA expands

fisheries

closures to

total 83,927 mi2

23 June

Reopening

of closed

area begins

22 July

NOAA Reopening

of one-third

of closed areas

10 and 27 August

NOAA Reopening

of 10,000 mi2

of closed areas

2, 3, and 21 September

NOAA Reopening

of 16,000 mi2

of closed areas

1, 5, 15, and 22 October

NOAA Reopening

of 22,000 mi2

of closed areas

15 November

NOAA Reopening

of 8,000 mi2

2 May

NOAA

initiates

first

fisheries

closures

20 April

DH rig

explosion

April May June July August September October November

2010


3/8

Gohlke et al.


increased concentrations o As ater an oil spillby creating a physical barrier and altering thechemistry o goethite, which normally bindsAs in ocean sediments; another study reportedincreased elemental Hg in air immediately aterthe M Hebei Spiritoil spill o the coast oKorea (Pandey et al. 2009). As with PAHs, itis difcult to attribute metal contamination to

a particular source (such as an oil spill), becausemetals also may accumulate in seaood as aresult o normal geological phenomena such asore ormation, weathering o rocks, degassing,or leaching, or because o anthropogenic activities such as smelting, burning o ossil uels,and discharges o industrial, agricultural, anddomestic wastes (Osuji and Onojake 2004).However, good baseline data exist or concentrations o several metals in seaood romthe Gul o Mexico (Kimbrough et al. 2008;NOAA 2011; U.S. EPA 2009) that could beused to determine whether concentrationsincreased ater the DH blowout.

wo studies looking at the occurrence anddistribution o metals in mussels 24 years aterthe Prestigeoil spill ound that V was increasedin postoil spill samples when comparing withmussels rom oiled with nonoiled sites and historical samples (Bartolom et al. 2010; Villareset al. 2007). Another study conducted aterthe Jiyeh oil spill in the Eastern MediterraneanSea ound that heavy metals including Pb, Ni,and V were increased in oysters 305 days versus 72 days ater the oil spill, whereas PAHlevels had decreased (Barbour et al. 2009a)and levels had high positive correlation withsize and length o the oysters (Barbour et al.2009b). In a study o metal accumulation in

marine blue crabs almost 10 years ater the1991 Gul War oil spill, AlMohanna andSubrahmanyam (2001) ound high concentrations o Zn and Cu in crabs at one samplinglocation and high concentrations o As, Pb,Mn, magnesium (Mg), Se, and V in crabs romanother sampling site. Similarly, a study o thetemporal variation and bioavailability o metalsdue to the oil industry in the Coatzacoalcosestuary in the southwestern Gul o Mexicoound average sediment metal concentrationsthat were highest or Zn, ollowed by Cu, Cd,and Pb (RuelasInzunza et al. 2009). Resultssuggested that accumulation in sediment and

marine organisms varied seasonally and thatcrabs and oysters bioaccumulated Cd, Zn, andPb. In the Niger Delta, where several oil spillshave occurred, Nduka et al. (2006) ound highconcentrations o metals including Pb, Zn, Cu,Ni, Cd, Co, Cr, Fe, and Mn in brains and gillso dierent reshwater and marine sh species.

Seaood tissue samples were not analyzed or metals ater the Cosco Busan oil spillbecause levels o Cd, Cr, Cu, Pb, Zn, Hg,and Sn were below detection limits in sampleso source oil; although Ni was detected, it waswell below a level o health concern (Brodberg

2007). Seaood samples were analyzed or Vater the /VDubai Staroil spill, but levelswere reported to be well below the set healthconcern (43.75 ppm) and were not published(Klasing and Brodberg 2010).

In summary, previous studies indicatethat metals present in oil can bioaccumulatein marine organisms and that higher levels

o some metals have been documented ateroil spills relative to baseline levels. In addition, contamination was noted 2 monthsater the Prestige oil spill (Villares et al. 2007)and may continue or > 10 years, as notedater the Gul War oil spill (AlMohanna andSubrahmanyam 2001). However, the extentand persistence o seaood contaminationwith metals is likely to vary depending on thesource oil, the type o organisms, and environmental conditions, and the lack o robusttimeseries data sets and inconsistent ndingsrom previous studies make it difcult to predict the likelihood o metal contamination oseaood ater the DH oil spill.

Dispersants. Dispersants are a mix osolvents and suractants used to acilitate thebreakup o oil into tiny droplets that are moreeasily broken down by natural processes. Tetwo dispersants used in the DH oil spillCorexit 9500 and 9527Acontain 2butoxyethanol, propylene glycol, and dioctyl sodiumsulosuccinate (DOSS). Dispersants may oersignicant protection to wildlie during a surace oil spill, but little is known about theireects ater deepwater application, as usedor the DH oil spill. Kujawinski et al. (2011)reported that DOSS underwent negligible biodegradation and was sequestered in deepwater

hydrocarbon plumes 64 days ater dispersantapplications had ceased.

he approval o dispersants is contingent upon demon strating that mortality inbrown shrimp and mussels is not signicantlygreater in an oildispersant mixture thanwith an oilonly exposure (U.S. EPA 2010).ests conducted by the U.S. EPA using themysid shrimp, Americamysis bahia, and theinland silverside,Menidia beryllinastandardtest organisms used in acute toxicity testsrevealed that the DH dispersants alone wereless toxic than a dispersantLouisiana sweetcrude oil mixture, and that the toxicity o

the oildispersant mixture was comparablewith that o Louisiana sweet crude oil alone(Hemmer et al. 2010). Additional in vitroanalyses suggested that cytotoxicity values oCorexit 9500 were indistinguishable romthose o three other dispersants tested, but notused, during the DH spill, whereas two otherdispersants on the market were signicantlyless cytotoxic (Judson et al. 2010).

Role o regulatory agencies in seaoodsaety ater the DH oil spill. Responsibilityor seaood saety management during andater the DH blowout is divided between

state regulators and two ederal agencies. Instate waters, which extend rom 0 to 3 milesoshore, responsibility lies primarily withstate health agencies. Local sh consumptionadvisories or harvest closures are oten issuedbased on waterquality monitoring programs,which may include analys is o inish andshellish tissues or contamination, and can

be based on a variety o ederal or state guidelines. he Food, Drug, and Cosmetic Act(2004) mandates that the FDA keep adulterated ood o the market and gives the FDAjurisdiction over seaood in interstate commerce. Finally, the MagnusonStevens FisheryConservation and Management Act (1996)gives NOAA the authority to regulate shingin ederal waters (3200 miles rom shore)and allows emergency action sheries closuresin the event o an oil spill.

It is important to stress that, ater an oilspill, ederal and state regulators must makerapid decisions, oten based on limited data.Speciically, regulators must determinea) whether seaood harvesting in the spill areashould be closed or restricted, b) what criteriashould be met beore reopening sheries, andc) how to communicate these decisions andpotential health risks to the public (Yenderet al. 2002).

Te rst step o the decisionmaking process entails collection and evaluation o thecharacteristics, ate, and transport o the oil,and the identiication o seaood resourcesat risk o exposure. I it is determined thatoil will likely contaminate seaood, as wasthe case during the DH blowout, closuresare issued. Here we ocus on the criteria or

reopening areas previously closed to sheries,and hence, assessing seaood saety.

Method of testing. Decisions to reopenoilimpacted areas to seaood harvesting aterthe DH blowout are based on the NOAA/FDA protocol or the interpretation and useo sensory testing and analytical chemistryresults (NOAA/FDA 2010). his protocoldraws extensively rom the NOAA publicationManaging Seaood Saety ater an Oil Spill(Yender et al. 2002), which details decisionmaking guidelines or state and ederal regulators based on NOAA/FDA experiences duringeight previous oil spills during 19891999.

Reopening criteria include an assessmentbased on monitoring or oil in harvest areas,collecting seaood samples rom spill and reerence areas, conducting sensory testing and/orchemical testing to determine whether seaoodis unadulterated and it or interstate commerce, and estimating the human health riskrom consumption. Sensory testing, based onthe smell o raw and cooked seaood samples,is conducted by a panel o 10 experts usingthe protocol outlined in a NOAA echnicalMemorandum (Reilly and York 2001).Chemical analysis is perormed using liquid


4/8



chromatography (LC)/uorescence detectionor gas chromatography/mass spectrometry(GC/MS) to determine whether target PAHlevels exceed the level o concern (LOC) set bythe FDA risk assessment. esting or metals isnot part o the reopening protocol. Subsequentto reopening, NOAA developed and presentedresults o chemical analysis or DOSS in sea

ood tissue, which is a component o the dispersants used during the cleanup eort.In general, many more samples have been

assessed via sensory testing (i.e., the snitest) than by chemical analysis. For example, 285 samples rom nsh and 55 samplesrom shrimp taken rom an area reopened on15 November 2010 were assessed or contamination by sensory testing, whereas chemical analyses were perormed on 33 compositesamples o nsh (rom a total o 207 sh)and 9 composite samples o shrimp (rom50 shrimp). We are unaware o studies examining the relationship between sensory testingand measured values o PAHs.

Sampling strategy. Te NOAA Strategyor Future Reopenings web site (NOAA2010d) indicates that the overall samplingscheme has been to work rom the lesseroiledouter boundaries o the originally closed areastoward the more heavily oiled areas immediately surrounding the DH wellhead, withlarger numbers o samples collected romheavily oiled areas. For ederal waters, closedareas were divided into 30Nm2 grids, and atleast six samples each o nsh, crab, and/orshrimp were acquired rom three to ive

sampling points within each grid. Compositesamples ormed by mixing several individualsamples together are then tested or PAHs.Composite nsh samples may consist o oneor two species or several species but includeonly sh ound at the same depth (e.g., bottom eeders such as grouper would be groupedwith other nsh species ound at that depth)

(NOAA, personal communication). Shellshsamples generally contain only one species.Requirements or sampling over time are notclear, but samples tested to support reopening o previously closed waters were collectedwithin 2day time rames prior to reopening,thus a timeseries analysis was not conductedbeore reopening.

Determination of LOC. Te LOC is thecalculated concentration o chemical that,i ound in seaood samples, would be considered unsae or human consumption. TeNOAA/FDA risk assessment or determining the LOC included a subset o 13 PAHsand their alkylated homologs [naphthalene,uorene, anthracene, phenanthrene, pyrene,uoranthene, chrysene, benzo[b]uoranthene,benzo[k]luoranthene, benz[a]anthracene,indeno[1,2,3cd]pyrene, dibenz[a,h]anthracene, and BaP] based on known toxic potentials and previous oil spill risk assessments.PAH LOCs are determined using a cancerrisk calculation based on benzo[a]pyreneequivalents (BaPE) or 7 PAHs [chrysene,benzo[b]luoranthene, benzo[k]luoranthene, benz[a]anthracene, indeno[1,2,3cd]pyrene, dibenz[a,h]anthracene, and BaP]. Te

ollowing equation is used to determine theLOC or carcinogenic PAH compounds:

LOC(BaPE) = (RL BW AT CF)

(CSF CR ED). [1]

Parameters used to estimate the LOC orthe DH spill and previous spills are shown

in able 1. For the DH spill, the risk level(RL) is set at 1 105; body weight (BW) is setat 80 kg based on average adult body weight(McDowell et al. 2008); averaging time (AT) isset at 78 years based on average lie expectancy(Heron et al. 2009); the unit conversion actor(CF) is 1,000 g/mg; the cancer slope actor(CSF) is set at 7.3 mg/kg/day based on theU.S. EPA BaP risk assessment or oral exposure (U.S. EPA 1994); and the seaood consumption rate (CR), which is set at 13 g/dayor shrimp and crab, 12 g/day or oysters, and49 g/day or nsh, based on 90th percentileseaood consumers in 20052006 NationalHealth and Nutrition Examination Survey(NHANES) study (CDC 2007). Te exposureduration (ED) is 5 years, which is the estimatedretention period o PAH contamination in seaood. Te calculated LOCs or the DH spillare thereore 35 ng/g (ppb) BaPE or nsh,132 ng/g (ppb) BaPE or shrimp/crab, and143 ng/g (ppb) BaPE or oysters.

Determination o noncancer risk or theDH spill was based on U.S. EPA reerence dosecalculations (RD; an estimate o a daily exposure o each chemical that likely has no signicant risk during a lietime) or the six additional

Table 1. Human health risk calculation parameters or reopening o sheries waters ater the DH blowout and previous U.S. oil spills.

Oil spillMagnitude ospill (gallons) Fisheries closures RLa BW (kg)

PAH CSF(mg/kg/day)b CR (g/day)

ED(years) LOC (ppb BaPE)c Reerence

DH, Gul o Mexico,2010

206,000,000 112 months, areaaround platormopened 19 April 2011

1:100,000 80 7.3 Finfsh: 49Shrimp/crab: 13Oysters: 12

5 Finfsh: 35Shrimp/crab: 132Oysters: 143

NOAA/FDA2010

T/V Dubai Star, SanFrancisco Bay,Caliornia, 2009

400800 1 month 1:10,000 70 11.5 32.5 (one8-ounce mealper week)

30 Fish and shel lfsh: 44 Klasing andBrodberg 2010

Cosco Busan, SanFrancisco Bay,Caliornia, 2007

58,000 1 month 1:10,000 70 11.5 32.5 (one8-ounce mealper week)

30 Fish and shel lfsh: 44 Brodberg 2007

M/V New Carissa,Oregon, 1999

70,000 Bivalves: 21 days 1:1,000,000 70 7.3 Shellfsh:7.5 (average)32.5 (high)

2 Shellfsh:10 (high)45 (average)

Gilroy 2000

M/V Kure, Caliornia1997

4,537 Oyster, crab: 49 days 1:1,000,000 70 9.5 Shellfsh:7.5 (average)50 (high)

2 Shellfsh:5 (high)34 (average)

Challenger andMauseth 1998

T/V Julie N, Maine,1996

180,000 Shellfsh: 15 days 1:100,000 70 7.3 Lobster: 13.6 10, 30 Lobster: 50 (16)(or 30-year ED)

Mauseth et al.1997

T/B North Cape,Rhode Island, 1996

828,000 Finfsh and bivalves:73 days

Lobsters: 75155 days

1:1,000,000 70 7.3 30 5 Lobster: 20 Mauseth et al.1997

T/V Braer, ShetlandIslands, 1993

25,000,000 Finfsh: 2 monthsFarmed salmon: 2 yearsLobster: > 6 years

Reach backgroundlevels o PAHs

Kingston 1999

T/V Exxon Valdez,Alaska, 1989

11,000,000 Herring/salmon: entireseason

Bivalves: advisoriesin our subsistenceharvest areas

1:1,000,000 70 7.3 Salmon: 89Other fnfsh: 52Crustaceans: 21Bivalves: 2

10, 70 Salmon: 3 (0.3)Finfsh: 5 (0.5)Crustaceans: 11 (1.1)Bivalves: 120 (12)

(or 70-year ED)

NOAA 2002

aAcceptable level o risk set or determining the LOC. bRate o increase in cancer risk per unit dose. cCalculated LOC. See Equation 1 or urther details.


5/8

Gohlke et al.


PAHs (naphthalene, luorene, anthracene,phenanthrene, pyrene, and uoranthene):

LOC = (RfD)(BW)(CF) CR. [2]

RfDs used are rom the U.S. EPA IntegratedRisk Inormation System (IRIS): 0.02 mg/kg/day naphthalene, 0.04 mg/kg/day luo

rene, 0.30 mg/kg/day anthracene/phenanthrene, 0.03 mg/kg/day pyrene, and 0.04 mg/kg/day luoranthene). BW, CF, and CR aredened as above or the cancer risk assessment.Calculated LOCs range between 49.0 g/g(pyrene in nsh) and 2,000 g/g (anthracene/phenanthrene in oysters). Note that theseLOCs are much higher than those calculatedor carcinogenic PAHs.

NOAA also developed and implementeda test to detect levels o DOSS, the dispersantcomponent considered most likely to bioaccumulate, in seaood (NOAA 2010c). Teacceptable daily intake (ADI) or DOSS is setat 0.1 mg/kg/day, based on the no observedadverse eect level o 50 mg/day or reproductive toxicity in rats and pulmonary circulatoryeects in rabbits and dogs, with saety/uncertainty actors o 500 [Food and AgricultureOrganization o the United Nations (FAO)/WHO 1995].

Comparison o current protocol to seaoodcontamination analyses ater previous oil spills.able 1 outlines sheries closures and providesa comparison o the PAH cancer risk calculation parameters used or sheries reopeningsater previous oil spills; it also provides acontext or evaluating the risk calculationsused or the DH blowout. For example, the

Caliornia Environmental Protection Agency(CalEPA) developed a protocol in the wakeo the 2007 Cosco Busan and /VDubai Staroil spills [CalEPA Oice o EnvironmentalHealth Hazard Assessment (OEHHA) 2007;Klasing and Brodberg 2010] that uses updatedtoxicity inormation based on another CalEPArisk assessment or PAHs, including anupdated CSF and toxic equivalency actors orPAHs (CalEPA OEHHA 2005). Te protocol(Klasing and Brodberg 2010) also takes intoaccount noncancer end points or V. Aterthe New Carissaoil spill, LOCs were determined based on the average shellsh consumer

CR (45 ppb BaPE) and highend CR (10 ppbBaPE) (Gilroy 2000). Shellsh was consideredsae i analyses showed BaPE below the highend consumer LOC (10 ppb BaPE).

As shown in able 1, the LOCs calculatedor the DH oil blowout are higher than orany previous oil spill risk assessment, particularly or shrimp and oysters, based on severalactors. Te /VDubai Starand Cosco Busanassessments used a 1 in 10,000 RL, whereasthe /VJulie Noil spill is consistent with thecurrent NOAA/FDA protocol or the DH(NOAA/FDA 2010) using an RL o 1 in

100,000. Previous risk assessments, includingthose or the Exxon Valdez, have used an RL o1 in 1 million. Te RL sets the tolerable number o cancer cases attributable to the exposure.So, or example, an RL o 1 in 1 million meansan exposure at the stated LOC is estimated tocause, at most, one additional cancer case in apopulation o 1 million people. All o these all

within the acceptable range o risks (1 104

to1 106) used by the U.S. EPA in regulatorycriteria or drinking water; 1 104 is providedas an example o a maximum acceptable RL inthe U.S. EPA Guidance or Assessing ChemicalContaminant Data or Use in Fish Advisories(U.S. EPA 2000). he CalEPA OEHHAconsidered an RL o 1 104 appropriate oruse in ish consumption advisories ater theDubai Starand Cosco Busan oil spills and as acounterbalance recognizing the health benetso sh consumption (Brodberg 2007; Klasingand Brodberg 2010). In addition, the LOCor the DH blowout assumes a higher bodyweight (80 kg vs. the 70 kg used previously)based on recent NHANES data or averagemale weight (McDowell et al. 2008). However,in a recent survey o Gul Coast residents, theNatural Resources Deense Council (NRDC2010) reported that 60% o the adults weighed< 80 kg and 44% o the residents have childrenwho eat seaood; this suggests that the U.S.average male weight may not be appropriate orpredicting risk in all populations. Assumptionsregarding seaood intakes also vary substantiallyamong risk assessments. For example, the riskassessment or the Exxon Valdezspill speciically calculated risk based on subsistence levelso consumption seen in local populations (Fall

et al. 1999), whereas other assessments usedestimates or average and highlevel consumers.Although the DH consumption levels arebased on a 90th percentile consumer, theCR or shellish consumption (13 g/day orshrimp and crab and 12 g/day or oysters) isstill well below estimates or highend shellish consumers used in recent risk assessments (New Carissa, Cosco Busan, and DubaiStarat 32.5 g/day and Kureat 50 g/day); arecent survey o Gul residents (n = 547) suggests that shrimp CRs are actually 3.612.1times higher than FDA estimates used orthe DH risk assessment (NRDC 2010). he

CalEPA used a more conservative CSF (11.5 vs.7.3 mg/kg/day) based on recent toxicity analyses (CalEPA OEHHA 2005) in risk assessmentsor the Kure, Cosco Busan, and Dubai Staroilspills. ED estimates also vary considerably, withthe Exxon Valdezassessment assuming a 10 or70year duration, the Dubai Starand CoscoBusan using 30 years, and the New CarissaandKureassessments assuming an ED o 2 years.

Federal and state seaood testing results romDH oil spill.Federal seaood testing resultsreleased to date have demonstrated PAH levels at least two orders o magnitude below the

LOCs established or the DH assessment (FDA2010; NOAA 2010b). Supplemental Material,able 3 (doi:10.1289/ehp.1103507) summarizes results in terms o total estimated PAH,when available, or BaPE, when available, toallow or comparison with previous studies,including studies ater previous oil spills andstudies o baseline levels. As mentioned above,

estimated total PAH levels are not useul indetermining risk to human health, because toxicity varies widely across specic PAHs; however, estimated total PAH levels do providea useul metric or comparing across studies.Compared with estimated total PAH levelsrom previous spills, samples collected ater theDH spill have PAH concentrations similar to orbelow levels reported in oyster, shrimp, and nsh samples ater previous oil spills. However,this comparison can be made only with the current data released by the FDA (state water testing), as available NOAA data or ederal watersdo not include estimated total PAH levels. It isalso o note that the seaood samples taken romcoastal state waters were analyzed by the FDAusing LCuorescence, and those rom ederalwaters were analyzed by GC/MS according tothe NOAA protocol (Sloan et al. 2004). Inaddition, when compared with studies lookingat PAH concentrations rom baseline samplesaround the world, samples collected ater theDH spill have PAH levels within ranges oundin these baseline monitoring data sets (seeSupplemental Material, able 3).

As o 9 January 2011, only 4 o 185 composite tissue samples acquired between Juneand September 2010 showed trace amounts oDOSS (2 tuna, 1 wahoo, and a mixedspecies

sample), all o which were ar below the LOCvalue o 100 ppm or nsh and 500 ppm orshrimp, crabs, and oysters (NOAA 2010c).DOSS was detected in 6 o 18 red snapper andgrouper composite samples collected at dockside between 8 July and 30 August at levels wellbelow the LOC (mean, 0.12 ppm) (NOAA2010c). One hundred twenty composite samples rom state waters were also analyzed, andeight were ound to have detectable levels oDOSS (six crab samples in Louisiana, one crabsample in Alabama, and one brown shrimpsample in Louisiana). Again, all were at leastthree orders o magnitude below the LOC.

Discussion

We reviewed the protocol used or reopeningsheries and the results o seaood testing aterthe DH and previous oil spills to a) provide aoundation or recommendations on additional data collection and analysis that canacilitate the development o a more comprehensive assessment o seaood saety and b) tohelp restore consumer condence in seaoodrom the Gul o Mexico. We also make theollowing recommendations or addressinguncertainty and risk communication.


6/8



1. Continue monitoring o PAHs.Althoughlevels o PAHs measured thus ar in seaoodater the DH blowout are well below levelsthat would be o concern or human health,continued monitoring should be in place toensure that concentrations stay below LOCs,especially in light o potential recontamination rom disturbance and redistribution o

sedimented oil. Studies have shown signicantsedimentation o the highmolecularweightPAHs (Bouloubassi et al. 2005) that may beincreased with the addition o dispersants ina coastal setting (Khelia et al. 2008), whereasother reports suggest dispersants may decreasePAH sedimentation but increase PAH concentrations in the water column (Stakeniene andJokas 2004; Yamada et al. 2003). However,eects o large quantities o dispersant appliedat considerable depths, as used ater the DH oilspill, are unknown. Continued monitoring willensure PAHs that may be present in sedimentare not being redistributed and accumulatingthrough the ood chain.

2. Begin testing or metals. Tere is considerable uncertainty regarding the potentialor increased metal contamination o seaood.In addition to trace levels o metals in oil,drilling luids containing metals were alsoreleased into the Gul ater ailed attemptsto cap the oil well. Recent research suggeststhat oil spills may alter natural sedimentation processes o As, raising the concern thatincreased exposure via seaood consumptionmay be an issue (Wainipee et al. 2010). Evenlowdose exposure to certain metals is o concern in pregnant women and children becauseo wellknown impacts on neurodevelopment

(Grandjean and Landrigan 2006).We recommend additional testing ocused

on metals or which the U.S. EPA has developed riskbased consumption limits and orwhich there is reason or concern based onprevious oil spills, including As, methylatedHg, and Cd, in addition to Ni and V, whichare clearly associated with oil contaminationand have known impacts on human health. Inaddition, based on inormation rom previousoil spills, longterm monitoring over at leastthe next 57 years is necessary to ensure that

any potential or bioaccumulation through theood chain is captured by the supplementaltesting scheme.

3. Co nt inu e mo ni tori ng o DO SS .Although current evidence suggests minimaldirect toxicity risk, there are still uncertaintiesregarding the longterm implications o theuse o Corexit 9500 and Corexit 9527 dis

persants. For example, it has been postulatedthat the unprecedented use o dispersants atthe wellhead could increase bio accumulationo oil components by increasing bioavailabilityand oil sedimentation, and a recent study conrmed that dispersants were ound in deepwater plumes at least 2 months ater the wellwas capped (Kujawinski et al. 2011).

Based on the NOAA dispersant tests odockside seaood samples taken in July andAugust 2010, there is evidence o dispersantcomponents in tissue, albeit at very low concentrations (NOAA 2010c). Continued monitoring o DOSS in inish or at least 1 yearis recommended, at which point monitoringcould be implemented in areas where seaoodwith detectable levels o DOSS was previouslyharvested. Current inormation suggests thatDOSS concentrations are unlikely to pose arisk to human health, but longterm monitoring is needed to determine whether DOSSpersists in the Gul ecosystem, particularlybecause this is the rst time that large volumeso dispersants have been applied directly to adeepwater wellhead.

4. Defne and continuously reassess optimalshort- and long-term sampling strategies.Basedon evidence o longterm contamination romprevious oil spills, seaood monitoring should

continue or several years. Te requency andspatial distribution o sample collection shouldbe well dened and should include areas predicted to have exposure based on monitoringand modeling o subsurace plumes, as well asareas known to be oiled.

he ull range o inish and shellishspecies should be sampled rom all relevantsites using adequate sample sizes. he useo composite samples is necessary to limitcosts, but the species included in composite samples should be clearly identiied and

communicated. Optimally, composite samples should only include one species rom asingle site, as diets, habitats, and behaviorsvary widely, especially across nsh species.

Continuous reassessment o the samplingstrategy based on testing results will ensurethat eorts are ocused on the most important contaminants, species, and areas. Ater an

initial period o sampling the entire Gul oMexico, more ocused and requent samplingo hotspots (where detectable levels have beenound in seaood and/or sediments) wouldthen be appropriate.

5. Develop guidelines to estimate variabil-ity and uncertainty in risk parameters. As isevident rom able 1, estimated LOCs canvary widely based on underlying assumptionsused in calculations. hereore, we proposethat guidelines be developed to estimate arange o LOC values according to age, sex,and seaood consumption parameters usedater previous oil spills, as outlined below.An example o age, sex, and consumptionspeciic LOC rangesassuming a 10yearED, the 50th percentile body weight o awoman (70 kg) and average body weight oa 6yearold child (22.68 kg), and one or twoseaood meals per weekis shown in able 2.We recommend the ollowing: Deine alternate LOCs based on a 1 in

1 million RL [traditionally used in U.S.EPA risk assessments and used as therisk LOC ater the Exxon Valdezoil spill(NOAA 2002)] and 1 in 10,000 RL [usedby CalEPA to take into account the healthbeneits o eating seaood (Klasing andBrodberg 2010)]. Resulting LOCs will di

er by an order o magnitude. Estimate LOCs using EDs ranging rom

5 to 30 years. Estimate LOCs across a variety o bodyweights appropriate or women and children. Estimate LOCs based on the CalEPA CSF

(11.3 mg/kg/day) that uses updated toxicityparameters (Brodberg 2007).

Estimate LOCs using a range o seaood eating patterns. Seaood consumption patternsvary widely across populations, but determining seaood consumption patterns in

Table 2. Estimated LOCs or children, women, and men, based on 12 seaood meals per week.

Young children(16 years o age)

Older children(712 years o age)

Adult women( 20 years o age)

Adult men( 20 years o age)

Contaminant 2 meals/week 1 meal/week 2 meals/week 1 meal/week 2 meals/week 1 meal/week 2 meals/week 1 meal/week

PAH (BaPE) LOC (ppb)a 0.6600 1116 0.986 2173 1119 2239 1108 2217MeHg LOC (ppm)b 0.07 0.15 0.11 0.21 0.15 0.29 0.13 0.26As (inorganic) LOC (ppm)b,c 0.00383.79 0.00747.36 0.00555.46 0.01110.94 0.0767.56 0.01515.13 0.0696.87 0.01413.74Cd LOC (ppm)b 0.73 1.46 1.05 2.1 1.46 2.9 1.32 2.64DOSS LOC (ppm)d 72.84 145.70 105.05 210.11 145.56 291.12 132.10 264.20

MeHg, methylmercury. Body weights (portion sizes) or the populations were as ollows: young children, 17.7 kg (3 oz); older children, 38.3 kg (4.5 oz); adult women, 70.7 kg (6 oz); adultmen, 85.6 kg (8 oz). Body weights were set at the 50th percentile reported by McDowell et al. (2008).aUsing a 1 in 1 million or 1 in 10,000 RL (range indicated), CalEPA CSF (11.5 mg/kg/day; CalEPA OEHHA 2005), and a 10-year ED. bBased on U.S. EPA IRIS RDs (Cd, 0.001 mg/kg/day;MeHg, 0.0001 mg/kg/day) or CSF (As, 1.5 mg/kg/day) as reported by the U.S. EPA (2000) using the ollowing ormulas to calculate the LOCs: or Cd or MeHg, LOC = ( RfD)(BW)(CF)/CR; orAs, LOC = (RL BW AT CF)/(CSF CR ED). cA small percentage (< 13%) o As ound in sh and shellsh is inorganic As (Food Standards Agency 2005). dBased on FDA LOC (FDA2010) use o the WHO/FAO ADI (0.1 mg/kg/day) (FAO/WHO 1995) and using the ollowing ormula to calculate the LOC: LOC = (ADI)(BW)(CF)/CR]. Note that the FDA uses a body weighto 60 kg instead o 80 kg (used or PAH LOC calculation) to calculate the DOSS LOC, based on WHO/FAO guidelines.


7/8

Gohlke et al.


particularly vulnerable populations is critical or determining risk associated with seaood contamination. Te current reliance onNHANES data sets may not adequately represent Gul Coast populations, as suggestedby a recent survey o Gul residents (NRDC2010). Additional survey work is necessary todene upper limits o seaood consumption

along the Gul Coast.he lower ranges o PAH LOC valuesin able 2, which are consistent with LOCsused or the Exxon Valdezand Kureoil spills(able 1), suggest there is approximately anorder o magnitude margin o saety based onFDA and NOAA chemical analyses o seaoodater the DH oil spill. For example, concentrations o the individual carcinogenic PAHstested to date have been below the limit odetection (LOD) (LODs between 0.07 and0.28 ppb or individual PAHs across dierent BaPE values) and are well below the lowest LOC estimate or BaPE (0.60 ppb BaPEor young children who consume two seaood meals per week). Age and consumption patternspecic LOCs or methylmercury(MeHg), As, and Cd [based on U.S.EPA IRISRDs (U.S. EPA 2000)] and the dispersantcomponent DOSS also are lowest or children,consistent with higher predicted exposuresrelative to body weight in children who eatsh requently. Tis also highlights the importance o children as a vulnerable population.

6. Communicate risk and uncertainty.Since the reopening o onethird o previously closed Gul waters to isheries in lateJuly 2010, numerous groups, including members o Congress, nongovernmental organiza

tions, scientists, local isherman, processors,and ches, have raised concerns over the adequacy o the NOAA and FDA protocol orensuring the saety o seaood caught in theGul. Issues raised include heavy reliance onan initial smell test, chemical testing or onlya select number o PAH components o oil,small sample sizes or chemical analysis, theuse o composite samples, and divergencerom previous calculations used to determinethe LOC or PAHs. imely communicationstrategies would improve public understandingand trust in the ederal process and shouldinclude the ollowing.

Direct community engagement in therisk assessment process. Engagement shouldinclude discussion o current risk assessmentand input on development o parameters usedto calculate LOCs, as well as sensitivity anduncertainty analyses.

Timely and effective communication oftesting results in the context of previous oilspills and baseline data. Te presentation anduse o baseline data, data rom previous oilspills, and risks posed in relation to previoustechnological disasters and alternative routeso PAH exposure are critical to convey levels

o risk in the context o previous disasters aswell as in the context o other routes o exposure people may encounter. For example,PAH levels currently ound in Gul seaoodare within ranges seen in routine surveillancedata sets o seaood and a variety o other oodstus including meats and oils [SupplementalMaterial, able 3 (doi:10.1289/ehp.1103507);

Kim et al. 2008].Presentation of testing results and sam-pling strategy in clear and simple formats.Using dierent ormats to present testingresults makes it diicult to convey overalltrends in the data. Coordination o the presentation o results across ederal and stateagencies is critical or eective communication o seaood saety.

7. Engage local communities in seaoodsaety testing. Providing demonstrations and/or training or perorming the FDAapprovedsmell test to interested community membersat sentinel sites along the Gul Coast mayhelp to reassure consumers o the validity othe approach and create an efcient mechanism or ongoing monitoring, particularly atsites most heavily oiled or most in danger obeing contaminated by other sources. In addition, continuous engagement o local community members in methods to determineseaood saety and making additional undingand technical guidance available to groupsthat want to conduct longterm chemical testing will provide a mechanism or timely andeective inormation exchange that may helpto alleviate concerns over seaood saety ateruture disasters along the Gul Coast.

8. Increase knowledge base on toxicity o

mixtures o contaminants.Very little is knownabout the potential or interactions betweenvarious contaminants rom an oil spill. Forexample, are eects o contaminants additive,or could they act synergistically? Alternatively,could eects o one contaminant reduce theimpact o other contaminants? Mixturesor cumulative toxicity research is still in itsinancy, but there is some evidence that concentration addition o toxicants impactingthe same cellular pathway can predict toxicity o mixtures, at least in model systems(Syberg et al. 2008); a recent study suggested that synergism does exist or criteria

air pollutants such as ozone (Mauderly andSamet 2009). For carcinogenicity o PAHs,the BaP equivalency is used to model additive eects across PAHs. Studies have shownneurodevelopmental impacts attributed toPAHs rom combustion sources (ang et al.2006), and neurodevelopmental impacts arealso related to some metals; thereore, interactions may exist based on similar end pointso concern. Mixtures research o contaminants associated with oil spills should be apriority to more accurately dene health risksater uture oil spills.

ConclusionsIn the short term, a detailed monitoring andtesting strategy that includes assessment ometals should be designed and coupled withan eective risk communication campaignto present the current results and longtermmonitoring plans ater the DH oil spill. Anydecisions made based on the current protocol

should be transparent, and inherent uncertainties should be ully discussed. We believe thatimplementation o these recommendations willhelp ensure the saety o seaood and restoreconsumer condence in seaood rom the GulCoast. In addition, a robust longterm monitoring and exposure assessment program iscritical to improve our understanding o seaood saety ater oil spills, which will provideinvaluable inormation or greater preparednessin response to uture oil spills.

RefeRences

Al-Mohanna SY, Subrahmanyam MN. 2001. Flux o heavy metal

accumulation in various organs o the intertidal marineblue crab, Portunus pelagicus (L.) rom the Kuwait coast

ater the Gul War. Environ Int 27(4):321326.

Barbour EK, Sabra AH, Bianu EG, Jaber LS, Shaib HA. 2009a.

Oppositional dynamics o organic versus inorganic

contaminants in oysters ollowing an oil spill. J Coastal

Res 25(4):864869.

Barbour EK, Shaib HA, Yaghi RH, Sabra AH. 2009b. Regression

o the level o dierent heavy metals to size o marine

organisms harvested rom the Jiyeh oil spill zone o the

eastern Mediterranean Sea. Bull Environ Contam Toxicol

83(2):219222.

Bartolom L, Navarro P, Raposo JC, Arana G, Zuloaga O,

Etxebarria N, et al. 2010. Occurrence and distribution o

metals in mussels rom the Cantabrian coast. Arch Environ

Contam Toxicol 59(2):235243.

Bouloubassi I, Mejanelle L, Pete R, Fillaux J, Lorre A, Point V.

2005. PAH transport by sinking particles in the open

Mediterranean Sea: a 1 year sediment trap study. Mar

Pollut Bull 52(5):560571.BP. 2010. BP Has Extensive Production Operations in North

America Focused on Oshore Gul o Mexico and Alaska.

Available: http://www.bp.com/productamily.do?categoryI

d=16002776&contentId=7020157 [accessed 22 June 2011].

CalEPA OEHHA (Caliornia Environmental Protection Agency

Oce o Environmental Health Hazard Assessment). 2005.

Air Toxics Hot Spots Program Risk Assessment Guidelines.

Part II. Technical Support Document or Describing

Available Cancer Potency Factors. Sacramento, CA:CalEPA.

Available: http://www.oehha.ca.gov/air/hot_spots/pd/

May2005Hotspots.pd [accessed 15 March 2011].

CalEPA OEHHA (Caliornia Environmental Protection Agency Oce

o Environmental Health Hazard Assessment). 2007. Report

on the Saety o Consuming Fish and Shellish rom Areas

Impacted by the M/V Cosco BusanOil Spill in San Francisco

Bay, Caliornia. Sacramento, CA:Caliornia Environmental

Protection Agency. Available: http://oehha.ca.gov/ish/pd/

SF%20BayFishShell112907.pd [accessed 21 June 2011].

CDC (Centers or Disease Control and Prevention). 2007.

NHANES 20052006. Available: http://www.cdc.gov/nchs/

nhanes/nhanes2005-2006/nhanes05_06.htm [accessed

13 December 2010].

Challenger GE, Mauseth GS. 1998. Closing and opening

sheries ollowing oil spills; a case study in Humboldt Bay,

Caliornia. In: Proceedings o the 21st Arctic and Marine

Oilspill Program Technical Seminar, Vol. 1. Ottawa, Ontario,

Canada:Environment Canada, 167179.

Crone TJ, Tolstoy M. 2010. Magnitude o the 2010 Gul o

Mexico oil leak. Science 330(6004):634.

Fall JA, Field LJ, Nighswander TS, Peacock N, Varanasi U.

1999. Evaluating and Communicating Subsistence Seaood

Saety in a Cross-Cultural Context: Lessons Learned

rom the Exxon Valdez Oil Spill. Pensacola, FL:Society or

Environmental Toxicology and Chemistry.


8/8



FAO/WHO (Food and Agriculture Organization o the United

Nations/World Health Organization). 1995. Evaluation o

Certain Food Additives. Forty-ourth Report o the Joint FDA/

WHO Expert Committee on Food Additives. Technical Report

Series 859. Geneva:WHO. Available: http://whqlibdoc.who.

int/trs/WHO_TRS_859.pd [accessed 20 June 2011].

FDA (Food and Drug Administration). 2010. Gul o Mexico Oil Spill

Update. Available: http://www.da.gov/Food/FoodSaety/

Product-SpeciicInormation/Seaood/ucm210970.

htm#background_testing [accessed 13 December 2010].

Food, Drug, and Cosmetic Act. 2004. 21USC301. Available:

http://www.da.gov/RegulatoryInormation/Legislation/

FederalFoodDrugandCosmeticActFDCAct/ [accessed

20 June 2011].

Food Standards Agency. 2005. Survey o Arsenic in Fish and

Shellish. FSIS 82/05. London:Food Standards Agency.

Available: http://www.ood.gov.uk/science/surveillance/

sis2005/sis8205 [accessed 15 December 2010].

Gilroy DJ. 2000. Derivation o shellish harvest reopening

criteria ollowing the New Carissa oil spill in Coos Bay,

Oregon. J Toxicol Environ Health A 60(5):317329.

Google. 2011. Google S cholar. Available: http://scholar.google.

com/ [accessed 15 March 2011].

Grandjean P, Landrigan PJ. 2006. Developmental neurotoxicity

o industrial chemicals. Lancet 368(9553):21672178.

Hellou J, Warren WG, Mercer G. 1995. Organochlorines in

Pleuronectidae: comparison between three tissues o three

species inhabiting the Northwest Atlantic. Arch Environ

Contam Toxicol 29(3):302308.

Hemmer MJ, Barron MG, Greene RM. 2010. Comparative

Toxicity o Louisiana Sweet Crude Oil (LSC) and Chemically

Dispersed LSC to Two Gul o Mexico Aquatic Test Species.

Washington, DC:U.S. Environmental Protection Agency.

Available: http://www.epa.gov/bpspill/reports/updated-

phase2dispersant-toxtest.pd [accessed 15 December 2010].

Heron MP, Hoyert DL, Murphy SL, Xu JQ, Kochanek KD,

Tejada-Vera B. 2009. Deaths: Final Data or 2006. Natl Vital

Stat Rep 57(14):1134.

IOM (Institute o Medicine). 2010. Review o the Proposal or

the Gul Long- Term Foll ow-u p Stud y: High ligh ts rom

the Sep tem ber 201 0 Wor ksh op Wor ksh op Rep ort .

Washington, DC:Institute o Medicine.

Judson RS, Martin MT, Rei DM, Houck KA, Knudsen TB,

Rotro DM, et al. 2010. Analysis o eight oil spill dispersants

using rapid, in vitro tests or endocrine and other biological

activity. Environ Sci Technol 44(15):59795985.

Khelia A, Fieldhouse B, Wang Z, Yang C, Laudriault M,

Brown CE, et al. 2008. Eects o chemical dispersant on

oil sedimentation due to oil-SPM focculation: experimentswith the NIST standard reerence material 1941B. In:

Proceedings o the 2008 International Oil Spill Conerence.

Washington, DC:American Petroleum Institute, 627631.

Kim Y, Powell EN, Wade TL, Presley BJ. 2008. Relationship o

parasites and pathologies to contaminant body burden

in sentinel bivalves: NOAA Status and Trends Mussel

Watch Program. Mar Environ Res 65(2):101127.

Kimbrough K, Johnson WE, Lauenstein GG, Christensen JD,

Apeti DA. 2008. An Assessment o Two Decades o

Contaminant Monitoring in the Nations Coastal Zone. NOAA

Technical Memorandum NOS NCCOS 74. Silver Spring,

MD:National Oceanic and Atmospheric Administration.

Kingston P. 1999. Recovery o the marine environment ollowing

the Braer spill, S hetland. In: Proceedings o the 1999 Oil

Spill Conerence. Washington, DC:American Petroleum

Institute, 103109.

Klasing S, Brodberg R. 2010. Report on the Saety o Consuming

Fish and Shellsh rom Areas Impacted by the T/V Dubai

Star Oil Spill in San Francisco Bay, Caliornia. Sacramento,CA:Caliornia Environmental Protection Agency. Available:

http://oehha.ca.gov/sh/pd/DubaiStar2010.pd [accessed

21 June 2011].

Kujawinski EB, Kido Soule MC, Valentine DL, Boysen AK,

Longnecker K, Redmond MC. 2011. Fate o dispersants

associated with the Deepwater Horizon Oil Spill. Environ

Sci Technol 45(4):12981306; doi:10.1021/es103838p [Online

26 January 2011].

Law RJ, Hellou J. 1999. Contamination o ish and shellish

ollowing oil spill incidents. Environ Geosci 6:9098.

Law RJ, Kelly C, Baker K, Jones J, McIntosh AD, Moat CF.

2002. Toxic equivalency actors or PAH and their

applicability in shellish pollution monitoring studies.

J Environ Monitor 4(3):383388.

Law RJ, Kelly CA, Graham KL, Woodhead RJ. 1997.

Hydrocarbons and PAH in sh and shellsh rom southwest

Wales ollowing the Sea Empress oil spill in 1996. In:

Proceedings o the 1997 International Oil Spill Conerence.

Washington, DC:American Petroleum Institute, 205211.

Magnuson-Stevens Fishery Conservation and Management

Act. 1996. Public Law 94-265. Available: http://www.nms.

noaa.gov/sa/magact/ [accessed 20 June 2011].

Mauderly JL, Samet JM. 2009. Is there evidence or synergy

among air pollutants in causing health eects? Environ

Health Perspect 117:16.

Mauseth GS, Martin CA, Whittle K. 1997. Closing and reopening

isheries ollowing oil spills; three dierent cases with

similar problems. In: Proceedings o the 21st Arctic and

Marine Oil Spill Program Technical Seminar, 1113 June

1997, Vancouver, British Columbia, Canada (Vol 2). Ottawa,

Ontario, Canada:Environment Canada, 12831303.

McDowell MA, Fryar CD, Ogden CL, Flegal KM. 2008.

Anthropometric Reerence Data or Children and Adults:

United States, 20032006. National Health Statistics Reports

No. 10. Hyattsville, MD:National Center or Health Statistics.

Meador JP, Stein JE, Reichert WL, Varanasi U. 1995.

Bioaccumulation o polycyclic aromatic hydrocarbons by

marine organisms. Rev Environ Contam 143:79165.

National Center or Biotechnology Inormation. 2011.

PubMed. Available: http://www.ncbi.nlm.nih.gov/sites/

entrez?db=pubmed [accessed 15 March 2011].

Nduka JK, Constance E, Obiakor E. 2006. Selective bioaccumu-

lation o met als by di erent parts o s ome i sh s pecies

rom crude oil polluted water. Bull Environ Contam Toxicol

77(6):846853.

NOAA (National Oceanic and Atmospheric Administration).

2002. Final Restoration Plan and Environmental Assessment

or the M/V Kuroshima Oil Spill, Summer Bay, Unalaska,

Alaska. Available: http://www.darrp.noaa.gov/northwest/

kuro/admin.html [accessed 23 Jun 2011].


2010a. BP Oil Spill: NOAA Closes Federal Waters to Royal

Red Shrimp Fishing. Southeast Fishery Bulletin 24 November.

Available: http://sero.nms.noaa.gov/s/deepwater_horizon/

FB10-102_112410.pd [accessed 15 December 2010].


2010b. Deepwater Horizon/BP Oil Spill Inormation. Available:

http://sero.nms.noaa.gov/deepwater_horizon_oil_spill.htm

[accessed 13 December 2010].


2010c. NOAA and FDA Announce Chemical Test or

Dispersant in Gul Seaood. All Samples Test within

Saety Threshold. Available: http://www.noaanews.noaa.gov/stories2010/20101029_seaood.html [accessed

10 December 2010].


2010d. NOAA Strategy or Future Reopenings. Available:

http://www.noaanews.noaa.gov/stories2010/20100927_

reopening.html [accessed 20 December 2010].

NOAA (National Oceanic and Atmospheric Administration). 2011.

The Mussel Watch Program, National Status and Trends

Database. Available: http://ccma.nos.noaa.gov/about/coast/

nsandt/download.aspx [accessed 21 June 2011].

NOAA/FDA (National Oceanic and Atmospheric Administration/

Food and Drug Administration). 2010. Protocol or

Interpretation and Use o Sensory Testing and Analytical

Chemistry Results or Re-opening Oil-impacted Areas Closed

to Seaood Harvesting. Available: sero.nms.noaa.gov/s/

deepwater_horizon/attachment%201%20(3).doc [accessed

27 June 2011].

NRDC (Natural Resources Deense Council). 2010. Gul Coast

Seaood Consumption Survey. Available: http://docs.nrdc.org/health/les/hea_10120702a.pd [accessed 13 December

2010].

Oil Spill Commission. 2011. Rebuilding an Appetite or Gul

Seaood ater Deepwater Horizon. Sta Working Paper

No. 16. Washington, DC:National Commission on the

Deepwater Horizon Oil Spill and Oil Drilling.

Osuji LC, Onojake CM. 2004. Trace heavy metals associated with

crude oil: a case study o Ebocha-8 oil-spill-polluted site in

Niger Delta, Nigeria. Chem Biodivers 1(11):17081715.

Pandey SK, Kim KH, Yim UH, Jung MC, Kang CH. 2009. Airborne

mercury pollution rom a large oil spill accident on the

west coast o Korea. J Hazard Mater 164(1):380384.

Reeves WR, Barhoumi R, Burghardt RC, Lemke SL, Mayura K,

McDonald TJ, et al. 2001. Evaluation o methods or

predicting the toxicity o polycyclic aromatic hydrocarbon

mixtures. Environ Sci Technol 35(8):16301636.

Reilly TI, York RK. 2001. Guidance on Sensory Testing and

Monitoring o Seaood or Presence o Petroleum Taint

ollowing an Oil Spill. Technical Memorandum NOS

OR&R 9. Seattle, WA:National Oceanic and Atmospheric

Administration. Available: http://response.restoration.noaa.

gov/book_shel/964_seaood.pd [accessed 20 June 2011].

Ruelas-Inzunza J, Spanopoulos-Zarco P, Paez-Osuna F.

2009. Cd, Cu, Pb and Zn in clams and sediments rom an

impacted estuary by the oil industry in the southwestern

Gul o Mexico: concentrations and bioaccumulation

actors. J Environ Sci Health A Tox Hazard Subst Environ

Eng 44(14):15031511.

Sloan CA, Brown DW, Pearce RW, Boyer RH, Bolton JL,

Burrows DG, et al. 2004. Extraction, Cleanup, and Gas

Chromatography/Mass Spectrometry Analysis o Sediments

and Tissues or Organic Contaminants. NMFS-NWFSC-59.

Washington, DC:U.S. Department o Commerce.

Solomon GM, Janssen S. 2010. Health eects o the Gul Oil

Spill. JAMA 304(10):11181119.

Stakeniene R, Jokas K. 2004. Dispersants and their infuence

on oil spread in water bodies. Environ Res Eng Manage

4(34):6167.

Stegeman JJ, Lech JJ. 1991. Cytochrome-P-450 monooxygenase

systems in aquatic species: carcinogen metabolism and

biomarkers or carcinogen and pollutant exposure. Environ

Health Perspect 90:101109.

Syberg K, Elleby A, Pedersen H, Cedergreen N, Forbes VE.

2008. Mixture toxicity o three toxicants with similar and

dissimilar modes o action to Daphnia magna. Ecotox

Environ Sae 69(3):428436.

Tang D, Li TY, Liu JJ, Chen YH, Qu L, Perera F. 2006. PAHDNA

adducts in cord blood and etal and child development in a

Chinese cohort. Environ Health Perspect 114:12971300.

Thomas RE, Brodersen C, Carls MG, Babcock M, Rice SD.

1999. Lack o physiological responses to hydrocarbon

accumulation by Mytilus trossulusater 3-4 years chronic

exposure to spilled Exxon Valdez crude oil in Prince William

Sound. Comp Biochem Physiol C Pharmacol Toxicol

Endocrinol 122(1):153163.

Thomson Reuters. 2011. Web o Science. Available: http://

apps.isiknowledge.com/WOS_GeneralSearch_input.

do?highlighted_tab=WOS&product=WOS&last_pr

od=WOS&SID=4DjBF2O%40okebEJCdh6&search_

mode=GeneralSearch [accessed 15 March 2011].

U.S. EPA (U.S. Environmental Protection Agency). 1994.

Benzo[a]pyrene (BaP) (CASRN 50-32-8). Available: http://

www.epa.gov/iris/subst/0136.htm [accessed 15 November2010].


Guidance or Assessing Chemical Contaminant Data or

Use in Fish Advisories, Vol 2. Risk Assessment and Fish

Consumption Limits. 3rd ed. EPA 823/B/00/008. Washington,

DC:Oce o Science and Technology, U.S. EPA.

U.S. EPA (U.S. Environmental Protection Agency). 2009. The

National Listing o Fish Advisories: Release o 2008 Data.

Available: http://water.epa.gov/scitech/swguidance/

shshellsh/shadvisories/ [accessed 14 December 2010].


National Contingency Plan Product Schedule Toxicity

and Eectiveness Summaries. Available: http://www.epa.

gov/emergencies/content/ncp/tox_tables.htm [accessed

19 March 2011].

Villares R, Real C, Fernandez JA, Aboal J, Carballeira A. 2007.

Use o an environmental specimen bank or evaluating

the impact o the Prestige oil spill on the levels o trace

elements in two species o Fucuson the coast o Galicia(NW Spain). Sci Total Environ 374(23):379387.

Wainipee W, Weiss DJ, Sephton MA, Coles BJ, Unsworth C,

Court R. 2010. The eect o crude oil on arsenate adsorption

on goethite. Water Res 44(19):56735683.

Yamada M, Takada H, Toyoda K, Yoshida A, Shibata A,

Nomura H, et al. 2003. Study on the ate o petroleum-

derived polycyclic aromatic hydrocarbons (PAHs) and

the eec t o chem ical disp ersa nt usin g an encl osed

ecosystem, mesocosm. Mar Pollut Bull 47(16):105113.

Yender R, Michel J, Lord C. 2002. Managing Seaood Saety

ater an Oil Spill. Seattle, WA:Hazardous Materials

Response Division, Oce o Response and Restoration,

National Oceanic and Atmospheric Administration.

A review of seafood safety after the Deepwater Horizon blowout.pdf

Documents

Transcript of A review of seafood safety after the Deepwater Horizon blowout.pdf