OIL SPILL SCIENCE chapter 27 – effects of oil in the environment

OIL SPILL SCIENCE chapter 27 – effects of oil in the environment OIL SPILL SCIENCE chapter 27 – effects of oil in the environment OIL SPILL SCIENCE chapter 27 – effects of oil in the environment OIL SPILL SCIENCE chapter 27 – effects of oil in the environment OIL SPILL SCIENCE chapter 27 – effects of oil in the environment OIL SPILL SCIENCE chapter 27 – effects of oil in the environment OIL SPILL SCIENCE chapter 27 – effects of oil in the environment

Effects of Oil in the EnvironmentGary Shigenaka

Convey Toxic Impact

99127.5 Route of Exposure:

The Anthrax Example

99927.6 Route of Exposure:

Oil

100027.7 Oil Chemistry,

Physical Behavior,

and Oil Effects

1003

27.8 Freshwater/SaltwaterDifferences

100827.9 Tropical Environments 101027.10 Arctic Environments 101327.11 Ecological Effects of

Oil Spills

101427.12 The Future of Oil

Effects Science

101727.13 Summary and

it all out yet But, we seem to be getting closer

A common perception of the effects of oil in the environment is captured inthe evocative photographs and footage of oiled wildlife struggling afterOil Spill Science and Technology DOI: 10.1016/B978-1-85617-943-0.10027-9

exposure to a spill While the impacts represented by this kind of outcomeare only a few of the many possible, they are visually compelling, memorable,and by default assumed by many to be diagnostic for what occurs when oilintersects with the environment With these kinds of images burned into thecollective consciousness and id, nearly any elementary school student orviewer of the evening news can tell us, with some certainty, that oil is indeedharmful.

The recent (2010) Deepwater Horizon oil spill in the Gulf of Mexico, and itsattendant media coverage, have done little to dispel the notion that these eventsare ecological and human disasters of the first order The early images of brightred oil emulsions stretching for miles on the water, thousands of response andrecovery vessels near the accident site, massive applications of chemicaldispersants, plumes of smoke from burning of oil slicks, the inevitablephotographs of oiled pelicans and sea turtles, and stories about impacted fishingcommunities and local and regional economies only serve to underscore theperception that oil spill IMPACTS are profound and far-reaching

Similarly, the scientific literature contains countless studies that documenttoxicity of petroleum products and their chemical constituents to virtually theentire spectra of plants and animals But not every oil spill is an environmentalcatastrophedand even the same spill incident variably affects different org-anisms in different ways, at a given moment and over time The empirical viewfrom 40,000 feet suggests that generalizations about impacts may not beappropriatedand we begin to sense the challenges of understanding the effects

of oil Some of these challenges have been articulated by others undertaking thetask of rendering at least a portion of the oil impacts universe into somethingthat we can understand:

l Determining the effects of oil is complex, and generalizing about effects

is difficult One must remember that specific impacts are very species andsituation dependent.2

l Because many physical and biological processes in the marine and coastalenvironment are poorly understood, it is difficult for scientists to measurethe full impacts of an oil spill, and sometimes the results appearcontradictory.3

l Oil spills will have different environmental effects.the environmentaleffects will depend on factors such as type of oil, different oceanographicconditions, latitude, season, and type of ecosystem.this complicatesextrapolation of data in even the most general terms.4

l Many spill impacts have been documented in the scientific and technicalliterature, and although not all the effects of oil pollution are completelyunderstood, an indication of the likely scale and duration of damage canusually be deduced from the information available However, it can be diffi-cult to present a balanced view of the realities of spill effects.the simplereality is that sometimes significant damage occurs, sometimes not.5

l Oil can kill marine organisms, reduce their fitness through sublethaleffects, and disrupt the structure and function of marine communities andecosystems While such effects have been unambiguously established inlaboratory studies.and after well-studied spills determining the subtler.effects on populations, communities, and ecosystems.poses significantscientific challenges.6

Petroleum is, of course, a natural material that is extracted like otherminerals and then processed and refined into thousands of products applied to

a myriad of general and specialized uses Oil spills are one of the unfortunateconsequences of accidents that occur during the extraction, production, ortransportation processes, or during end uses

However, oil is also released naturally into both terrestrial and marineenvironments through seeps, where oil deposits close to the surface of the soil

or sediment are exposed Rather than reflect severe biological impacts from thischronic localized exposure, these natural oil seep areas can be remarkable intheir lack of apparent effects That is, despite the presence of what amounts to

be large, continuous oil spills, oil seep areas do not show impacts surate with our perceptions of oil as a poison and the results of research thatconfirm those perceptions

commen-How do we reconcile these disparate notions and observations in order tounderstand the effects of oil in the environment?

In this chapter, we will discuss the characteristics of the substances weconsider to be “oil” and describe how these form the basis for a complexformula to be solved in order to understand how oil affects organisms exposedduring oil spills We will summarize the toxicology and research studies thathave helped us to understand if and how oil is harmful Finally, we will considerthe implications for oil spill response

27.2 SOME DEFINITIONS

The derivation of the word, “oil” dates to 13th-century Middle English (oile),and stems from Latin (oleum) and Greek (elaion) terms relating, not surpris-ingly, to olive oil and olives The official Merriam-Webster definition for oil is

“any of numerous unctuous combustible substances that are liquid or can beliquefied easily on warming, are soluble in ether but not in water, and leave

a greasy stain on paper or cloth.”

The particular oils with which we are concerned for the purposes of thisbook are those derived from petroleum Referring again to Merriam-Webster,this takes us to a more focused definition: “an oily flammable bituminous liquidthat may vary from almost colorless to black, occurs in many places in theupper strata of the earth, is a complex mixture of hydrocarbons with smallamounts of other substances, and is prepared for use as gasoline, naphtha, orother products by various refining processes.”

A part of this lengthy definitiond“a complex mixture of hydrocarbons”dreflects one of the key concepts relevant to the environmental impacts ofpetroleum oil spills: oil, specifically petroleum oil, is not a singular substancethat is the same from place-to-place, time-to-time, and, for our purposes, fromspill-to-spill As we shall learn, this fact immensely complicates our task ofunderstanding the behavior and effects of oil.

Toxicity is another concept about which all of us have some intuitiveunderstanding A simple but broadly applicable definition for toxicity is “thedegree to which a substance is able to damage an exposed organism.” In thecase of oil, however, the simplicity of this definition begins to elude us as wedissect it into its component parts and consider it in the context of petroleum;

we can use it to preview the actual difficulty of discussing the toxicity of oil.For example:

.the degree to which a substance is able to damage an exposedorganism

As we noted in the preceding definition, the oil we are considering is not

a single substance, but a complex mixture of many substances, or cals This means that our assessment of oil toxicity begins with as manyoil constituents or constituent groups as we care to evaluate

chemi-.the degree to which a substance is able to damage an exposedorganism

What do we mean by the term, “damage?” Does this means only a lethalendpoint? Or does it include sublethal injury from which an organism mightrecover? Does cellular injury that we cannot link to some observableimpairment constitute damage? What about behavioral shifts resultingfrom exposure? Or shifts in community structure? Given the rather dynamicand elusive elixir that we suspect petroleum to be, we can begin to anticipatethe many permutations of impact that will result if we consider multiplesubstances and multiple endpoints for damage

.the degree to which a substance is able to damage an exposed organism.The definition infers that exposure is a precondition for damage If

a poisonous material does not come into contact with an organism ofconcern, is it still toxic? This may sound a little bit like the old philosophicalriddle, “If a tree falls in the woods and nobody hears it, does it make

a sound?” But this is a highly relevant question from the perspective of spillresponse, because if we accept that reducing exposure reduces impact, wecan attempt to implement measures for reducing or preventing oil fromcoming into contact with a resource of concern

.the degree to which a substance is able to damage an exposedorganism

It is a fact of toxicology that there are differences in the way a given toxin ortoxic mixture affects different organisms and even different life stages ofthe same organism Therefore, consideration of multiple organisms or life

stages in an assessment of oil effect complicates our already challengingtask by an additional possibly daunting factor.

We will use the definitions above as a template for discussing the basics ofoil effects in the environment and will expand on the challenges they represent

to us when we assess and respond to an oil spill

27.3 SIZE MATTERS: SEEPS VS SPILLS

In its raw form, oil is a natural material that can readily be found, in many parts

of the world, above ground or on the water, and is quite visible The NationalResearch Council estimated that 45% of the oil entering the world’s oceansderives from natural seepage from geologic formations, and Hoefler counted

200 natural underwater oil seeps that have been identified around the world.5,7One of the best-known of these areas lies offshore from Santa Barbara,California, near Coal Oil Point These seeps release 20e25 tons of oil each day,ultimately resulting in a degree of nearshore sediment oiling equivalent to8e80 Exxon Valdez-size spills, as well as countless tarballs on the beaches ofthe central California coast.8By any measure, this is a considerable amount ofoil in a relatively small portion of the marine environment

This same area is also known as the site of the first major oil spill disaster inU.S history, the Santa Barbara spill of 1969 However, the seeps and the iconichistorical spill are only marginally related That is, although the oil spilled in

1969 derived from the same source that feeds the seeps (an oil-rich geologicfeature called the Venture Avenue Anticline), the seeps themselves were notresponsible for the spill Rather, it was a blowout at a production platformtapped into the submarine oil reservoir: uncompensated pressure increases in

a 3500-foot well drilled under Union Oil Platform Alpha split the well casingand then fractured the seafloor around it, allowing oil to leak uncontrollablyinto the water column directly from the reservoir By the time the source of thespill had been contained (11 days later), approximately 3 million gallons of oilhad been released.9

News reports showed beaches with oil pooled as deeply as 6 inches, alongwith oiled, dead seabirds and marine mammals Photographs, film footage, andwritten accounts of these and other spill-related impacts not only stoked publicresentment against oil companies, they also played an important role infostering the beginnings of the American environmental movement thatculminated in the first Earth Day celebration and the passage of landmark U.S.legislation to strengthen environmental protection As noted by U.S PresidentRichard Nixon, “The Santa Barbara incident has frankly touched theconscience of the American people.”

The Santa Barbara oil seeps and the 1969 oil spill are extreme examples ofthe same-source crude oil released into the same marine environment, butresulting in very different perceived and documented impacts The oil seeps

have been and continue to be generally considered as an inconvenience ornuisance, requiring tar to be cleaned off the feet of beachgoers and blankets; theSanta Barbara spill, on the other hand, was a seminal event in U.S environ-mental history whose impacts were seen as devastating and ultimately far-reaching The continuous inputs from the Coal Oil Point seeps and the lack ofadverse environmental impact related to them suggest that the marine envi-ronment can tolerate some level of exposure to oil; however, the dramaticimpacts from the Santa Barbara spill illustrate that an effects threshold can beand was exceeded How do we account for the range of impacts (or nonim-pacts), and how can we apply this insight to other oil spill situations? What arethe lessons for oil spill response?

More narrowly focused studies of the biology of oil seeps have revealedrelatively moderate effects attributable to oil exposure in this setting Forexample, Helix summarized the studies of benthic communities around theseeps and did not find the areas to be substantially affected.10In fact, proximity

to natural oil and gas seeps actually increased overall productivity andenhanced fecundity in species like copepods, which are generally considered to

be sensitive to hydrocarbon exposure Spies et al studied benthic organisms aswell as fish around the California seeps and found few indications of adverseimpactdalthough fish sampled near the oil sources showed physiologicalevidence of aromatic hydrocarbon exposure (i.e., enzyme activation) and hadhigher incidences of gill and liver lesions.11,12 In this case, documented oilexposure did not translate into documented oil effect

Helix attributed the modest incidence of adverse effects to a number offactors:

l Seeps are patchy in distribution, and amounts released are quite variable

l The oil to which communities are exposed varies in degree of “freshness,”which substantially influences toxicity

l Different biological communities have differing tolerances to the differentlevels of exposure

l Some organisms are capable of adapting to the presence of oil, either byaccommodating or simply avoiding accumulations of oil

We can make the same kinds of analytical observations for human-causedoil spills to suggest why more adverse impacts may occur, even when the sameoil source as relatively benign seeps is involved:

l Oil spills are not uniform in distribution of product released (i.e., patchy), sothe amounts to which organisms are exposed can vary widely

l Spilled oils vary widely in chemical composition, from highly refined fuels

to unrefined crude oils and remnants of the refining process Adding to thiscomplexity is the fact that spilled oil changes, or weathers, once it isreleased into the environment, and so the same oil days or weeks after a spilloccurs is likely to differ substantially from the fresh oil initially released It

is essentially a different kind of oil spill, even with a single original sourceoil These differences considerably affect the native toxicity of the product,

as well as our assessment of that toxicity

l Different biological communities have differing tolerances to the differentproducts and different levels of exposure

l In an oil spill situation, the sheer volume of oil released into the environmentover a relatively short period of time tends to overwhelm any inherent capa-bility of the affected environment (also referred to as “net assimilativecapacity” by Overton et al.13) to tolerate or accommodate exposure to the oil

In the seep versus Santa Barbara spill example, all of the considerationsplayed a role in very different effects profiles However, the differences involumes spilled over time probably were the most significant Although 20-25tons of oil per day released from seeps is a considerable amount of oildequivalent to around 7000 gallons per daydthe total for an 11-day period (thelength of the uncontrolled Santa Barbara spill) would amount to a maximum ofaround 85,000 gallons This is far less oil than the estimated 3 million gallonsreleased by the Platform Alpha blowout

Based on the empirical qualitative and quantitative information we have atour disposal, we can make some crude assignments of impact for the ongoingSanta Barbara seep release and the Platform Alpha release:

85,000 gallons/11 days (seep release)¼ not so bad;

3,000,000 gallons/11 days (Platform Alpha)¼ bad

It should be abundantly apparent that at this scale, the impact assessment isnot quantum mechanics, nor are the results transferable to stone tablets It couldalso be argued that a metric of “bad” overstates the broader population orecological impact represented by the hundreds to thousands of bird and marinemammal mortalitiesdalthough the social/political/cultural/historical signifi-cance of the 1969 spill cannot be denied Also, less debatable is the dubiouswisdom of then-Union Oil President Fred L Hartley’s stated opinion (Clarkeand Hemphill9) concerning the singular lens through which he interpreted theimpacts of the Santa Barbara spill:

“I don’t like to call it a disaster, because there has been no loss of humanlife I am amazed at the publicity for the loss of a few birds.”

27.4 AN “EQUATION” TO CONVEY TOXIC IMPACT

Our task here does not include factoring in the cultural or social contexts of oilspills in determining overall effect or impact (though we should note thatcompletely ignoring those other contexts would be very foolish indeeddtheysimply are to be contemplated elsewhere) That reprieve, however, does notnecessarily simplify the challenge of assessing or predicting our more narrowlyconstrued biological impacts It remains a daunting task

How, then, can we begin to sort through the facts of what is known about oiltoxicity and the documented effects of a given spill event to make some sense of

it all, to extract some pearls of wisdom to take with us to the next oil spillresponse?

First of all, although we suggested above that size (of a release) matters, it isuseful to think beyond only bulk amounts of oil introduced into a habitat ofconcern when assessing effects of that oil Yes, the amount of oil inflicted onbiological communities is important But additional considerations enter intothe calculus of impact The amount of oil is equivalent to the dose But theoil also has a unique toxicity signature that results from the complex chemistry

we have discussed Finally, the extent and the characteristics of exposure toorganisms and communities of concern is itself the result of a complex series ofphysical interactions dependent on the chemistry of the oil, the physics of itsinteraction with receiving waters, and the dynamics of how it is moved anddistributed in the environment We can reduce this to a deceptively simple

“equation”:

Oil Impact¼ Dose  Toxicity  Exposure or some function thereof.This can be applied to specific oils and to individual organisms as well as toportray potential overall effect of entire spills We can work through severalpermutations of this function, with information from actual incidents tounderstand how it works A few examples follow

Case 1: High Dose High Toxicity  High Exposure (North Cape, 1996)Dose¼ 828,000 gallons, high dose

Toxicity¼ Home heating oil, similar to diesel, higher acute toxicityExposure¼ Storm conditions in nearshore zone of Rhode Island mixedoil throughout water column

Impact¼ High, widespread highly visible mortalities of benthic isms, some with high intrinsic and cultural value (lobsters)

organ-Case 2: High Dose Medium Toxicity  Low Exposure (Odyssey, 1988)Dose¼ 40 million gallons, very high

Toxicity¼ Crude oil, medium toxicity

Exposure ¼ Low, tanker broke apart 900 miles off the coast ofNewfoundland

Impact¼ Low, limited landfall for spilled oil

Case 3: High Dose Low Toxicity  Low Exposure (Barge DM932, 2008)Dose¼ 420,000 gallons, medium

Toxicity¼ #6 fuel oil (“bunker”), lower acute toxicity but higher densityand viscosity, potential for submergence and wildlife fouling impactsExposure¼ Low, despite downstream transport and stranding along 200miles of Mississippi River banks

Impact¼ Extensive shoreline fouling, some recreational use impacts inNew Orleans, minimal resource impacts

Case 4: Low Dose Medium Toxicity  Low Exposure (Cosco Busan, 2007)

Dose¼ 53,000 gallons, low

Toxicity¼ Intermediate fuel oil, medium acute toxicity

to differences in environmental and geographic conditions at spill sites, and todifferences in the spilled oils themselves

In the preceding examples, we have shown only a few different tions of the many (3 3  3 ¼ 27, if we accept that there are three qualitativelevels of low, medium, and high for each of the three components of the impactequation) If we choose to refine the levels to include more nuance beyond high,medium, and low, then correspondingly more combinations are generated Thepoint of this is the impact and the factors that enter into generating it There aremany paths to the same destinationdwhich may sound like cheap philosophy,but it simply indicates that very different inputs into the impact equation cangenerate similar results Or the flip side of the mixed metaphorical coin is thatthe presence or absence of one or more factors can amplify or negate anotherfactor Finally, it is impossible to anticipate and account for everything, as in theCosco Busan example The metrics of our simple equation alone would havecalculated that this spill would be minor in its impact, but public opinion, mediacoverage, and political interest elevated perceived impact to a much higherlevel

permuta-The dose/toxicity/exposure equation incorporates the information inputs

we consider during spill response, that is, how much spilled, what spilled, whatcan be done to protect valued resources? Of those three considerations in

a response, two are relatively easily addressed: how much spilled (dose), andwhat can we do to contain/divert/collect the spilled material (exposure)? Themost difficult piece for us to determine, all other things being equal, is the

“what spilled question,” along with its implications

A narrower focus on impacts leads us to pose more questions about detailsand brings to the fore the questions of what the composition of the spilledmaterial is and what is its toxicity In the case of oil products, the questions can

be slightly rephrased to ask, how do we expect the oil to harm? For a spillresponder, the answer to that question then doubles back to considerations of oilbehavior, exposure, and the differences implicit in routes of exposure

The petroleum industry often characterizes crude oils according to theirgeographic source location, for example, Alaska North Slope crude However,this designation by itself does not provide any insight into fate and effects if theoil is spilled, and it is not very useful for response personnel That is, oil toxicity,

physical state, and the changes that occur with time and weathering are notconveyed or distinguished by geographic source names The U.S Environ-mental Protection Agency (EPA) uses the physical characteristics of petroleumoils as a way to consider the many types of oils from a response-orientedperspective; that is, how will the oil behave in the environment, and how willexposed organisms respond? The four U.S EPA categories are defined as:15Light, Volatile Oils.These oils are highly fluid, often clear, spread rapidly

on solid or water surfaces, have a strong odor, a high evaporation rate, andare usually flammable They penetrate porous surfaces, such as dirt andsand, and may be persistent in such a matrix They do not tend to adhere

to surfaces; flushing with water generally removes them Light, volatileoils may be highly toxic to humans, fish, and other biota Most refined prod-ucts and many of the highest quality light crudes can be included in thisclass

Nonsticky Oils.These oils have a waxy or oily feel They are less toxic andadhere more firmly to surfaces than light, volatile oils, although they can beremoved from surfaces by vigorous flushing As temperatures rise, theirtendency to penetrate porous substrates increases and they can be persistent.Evaporation of volatiles may lead to a heavier and more persistent residueoil Medium-to-heavy paraffin-based oils fall into this class

Heavy, Sticky Oils.These oils are characteristically viscous, sticky or tarry,and brown or black Flushing with water will not readily remove this mate-rial from surfaces, but the oil does not readily penetrate porous surfaces.The density of heavy, sticky oils may be near that of water, and they oftensink Weathering or evaporation of volatiles may produce solid or tarry non-fluid oil Toxicity is low, but wildlife can be smothered or drowned whencontaminated This class includes residual fuel oils and medium to heavycrudes

Nonfluid Oils.These oils are relatively nontoxic, do not penetrate poroussubstrates, and are usually black or dark brown in color When heated, non-fluid oils may melt and coat surfaces that become very difficult to clean.Residual oils, heavy crude oils, some high-paraffin oils, and some weath-ered oils fall into this class

During an oil spill, these classifications are dynamic and may change for

a given product released into the environment, dependent on the effects ofweathering or more transient changes such as ambient temperature As we havenoted, these influence the state and behavior of crude oil and refined petroleumproducts For example, as volatiles evaporate from a nonsticky oil, it maybecome a heavier product with different physical characteristics If a significanttemperature drop occurs (e.g., at night), a heavy, but still fluid, oil may solidify.Upon warming, however, it may revert back to its original state

A general rule of thumb for oil spill responders has long been that therefined fractions of crude oil, like gasoline or jet fuels, were the most toxic of

the oil products we might encounter in an aquatic or a marine incident Thisjudgment was based on the assumption that the lighter “ends,” as the fractionsare called, are more soluble in water and thus water-borne organisms are moreinclined to be exposed Certainly, a multitude of comparative studies supportedthe notion that the solubility of different oils and petroleum distillates in waterwas correlated with toxicity (primarily in the form of acute lethality) to targetorganisms.

Tagatz, for example, was representative of toxicological results that clearlysupported this conceptual framework.16Using acute lethality as an endpoint, hetested the toxicity of gasoline, diesel, and heavy (bunker) fuel oil to Americanshad (Alosa sapidissima) and found that gasoline was most toxic, diesel less butsimilarly so, and Bunker C much less toxic

With these kinds of results in hand, spill responders distilled them intogeneral rules of thumb related to oil fate and effects:

Light or refined products:short environmental residence time due to tilization; high acute toxicity due to high-water solubility and ability topenetrate cellular membranes

vola-Heavy or residual products:long environmental residence time and hightendency to physically foul feathers, fur, and orifices due to high viscosity;but low acute toxicity due to low-water solubility

Crude oils:somewhere in between light and heavy products, depending onthe chemistry of the specific crudes

We still mostly adhere to these generalizations today, especially withrespect to acute (lethal) toxicity: we expect mortalities of water column orbenthic organisms if a light product spills and is subjected to mixing energy;while heavy fuel oils are ugly and persistent, we usually don’t worry muchabout direct toxicity except that resulting from physical fouling of wildlife.This sounds simple enough; but complications begin to arise when the spilledproduct does not fit neatly into one of our predesigned categories For example,many of the heavy fuel products that are used as bunker or boilers fuels are

“cut” with a lighter distillate to facilitate pumping and transfer And, todaymany of the most commonly transported oils are intermediate fuel oils that fallsomewhere between light and heavy

Things get even messier as we stray beyond consideration of acute lethalityonly and also consider chronic and sublethal endpoints And a sense ofconfused exasperation or resignation settles in if we look beyond impacts toonly organisms, and consider populations and ecosystemsdor focus in theother direction and examine suborganismal effects: physiological, develop-mental, molecular Once we commit to moving beyond the unrealisticallysimplistic, our quest to understand the effects of oil may begin to look hope-lessly incomprehensible: it is not There are, however, many pieces to be sortedthrough and assembled in order to build a perspective on oil toxicity and effectthat takes into account its many complex components

Oil effects research has had its shares of peaks and valleys Like most otherscience, the research is driven by the availability of funding In the case of oil,the funding tends to be correlated with leasing activities or the occurrence ofmajor disasters The leadup to major oil development on the continental shelf ofAlaska included a major new multidisciplinary scientific effort and providedsupport for new research Oil spills in various parts of the world spiked interestand support for effects and monitoring studies.

The idea that oil toxicity is more complicated than its water solubility hadits origins, as is the case with many other shifts in beliefs and practices related

to oil spills, in the wake of the Exxon Valdez This 1989 oil spill in PrinceWilliam Sound, Alaska (which remains the largest U.S spill in history)provided the impetus for more oil spill effects studies than for any other spill inhistory, many of them funded by the Exxon Valdez Oil Spill Trustee Council(the research and restoration oversight body formed with settlement moniespaid by Exxon)

After the spill occurred but prior to the creation of the Exxon Valdez OilSpill Trustee Council, federal agencies undertook their own efforts to charac-terize early damages from the spill Fisheries biologists and pathologists fromthe Montlake Laboratory of National Oceanic and Atmospheric Admin-istration’s (NOAA’s) National Marine Fisheries Service in Seattle sampled fishthroughout the spill-affected regions and used two different methods (analysis

of biliary fluorescent aromatic compounds and measurement of hepatic chrome P4501A) to quantify exposures to Exxon Valdez oil.17High values forboth markers were encountered in fish during the first year of the spill (1989),especially in Dolly Varden char which frequent shallow waters close to theshoreline (where large quantities of oil stranded) Values declined substantially

cyto-in 1990, cyto-indicatcyto-ing lower exposure Results for bottomfish were lower, but stillindicated an elevated degree of exposure above background levels The positiveresults for these biochemical markers importantly suggested that organisms not

in the immediate vicinity of oil can still be exposeddand thus, at risk.While confirming exposure and quantifying what that exposure is, we arestill faced with the formidable task of answering the so-called so what?question: what is the significance of a given degree of exposure to oil?For the spill in Prince William Sound, the Auke Bay Laboratory of NOAAFisheries in Juneau, Alaska, became a focal point for U.S federally sponsoredExxon Valdez oil spill studies, and one of the areas of interest was the effect ofvery low levels of weathered crude oil on important fisheries species in Alaska,pink salmon (Onchorhynchus gorbuscha) and Pacific herring (Clupea pallasi).Carls et al found that exposure of Pacific herring eggs to total PAH (poly-aromatic hydrocarbons) levels rarely contemplated in a toxicological settingd

<1 part per billion (ppb)dresulted in a profound suite of impacts to larval fish:

at 0.7e7.6 ppb, malformations, genetic damage, decreased size, inhibitedswimming, and mortality; at 0.4 ppb total PAH, yolk sac edema and immaturityconsistent with premature hatching.18Interestingly, fresher oil was less toxic

(i.e., higher effects concentrations of 9e34 ppb) than the weathered product.This was attributed to a relatively higher proportion of higher-molecula-weightPAHs in the weathered product.

Heintz et al came to similar conclusions in examining the effects of weatheredNorth Slope crude oil on pink salmon.19 Their toxicity endpoint of embryomortality was found to be significantly higher at 1.0 ppb total PAH derived fromvery weathered oil and was similar to that determined by Carls et al.18

Heintz et al simulated exposures of pink salmon to Exxon Valdez oil byincubating eggs in water that had been percolated through gravel coated withartificially weathered Alaska North Slope crude (i.e., same source oil from theExxon Valdez spill).20 The “preweathering” skewed the PAH compositiontoward alkyl-substituted naphthalenes and larger hydrocarbons The salmonsmolts that survived the initial exposure were tagged (coded wire tags) andreleased, to return two years later

Salmon that had been exposed to total PAH concentrations of 5.4 ppbexperienced 15% reduced survival compared to unexposed salmon Thisconcept of delayed mortality due to early exposure was a new concept; the ideathat sublethal exposure resulted in compromise of fitness later had not beenpreviously considered Heintz et al concluded that evaluation of oil toxicity byshort-term effects alonedfor example, in early exposures in salmon streamsdunderestimated long-term effect The take-home lesson was well-articulated bythe authors: “Reliance on toxicity tests that fail to realistically simulate exp-osure conditions is likely to misguide water-quality managers.”

The fact that a number of earlier oil-related toxicity studies took place in thesame Auke Bay Lab with very different conclusions (e.g., effects levels at muchhigher concentrations, in the range of 1000e1400 ppb; relative resistance ofsalmonid eggs to oil exposure versus juveniles) was recognized by Carls et al.18They noted that the water-soluble fraction of the test oil for the earlier study,

a Cook Inlet crude oil, was comprised of mono- and di-aromatic hydrocarbons;

in contrast, the test oil for the more recent study, the North Slope crude that hadbeen carried by the Exxon Valdez, had higher proportions of multi-ring and alkyl-substituted PAHs The take-home message for us is that composition of thespilled oils, and their weathering state, are key determinants of toxicity Theother take-home messages are that the more recent literature on oil toxicityreflects a much lower threshold for effects than we had previously accepted andthat oil exposures to early life stages may have delayed consequences longafterward

The other key fishery in Prince William Sound at the time of the ExxonValdez oil spill was that for Pacific herring (Clupea pallasi), or more accuratelyPacific herring roe A series of herring-related studies took place after the spillwith support from the Exxon Valdez Oil Spill Trustee Council to determine andcharacterize oil-related injuries; these included McGurk and Brown; Hose

et al.; Kocan et al.; and Norcross et al The coordinated investigations involvedboth field and laboratory efforts.21-26

The reader is referred to the source papers for details of each of thestudies, but a summary of results from the five investigations includes thefollowing.

l Mean eggelarval mortality (defined as the ratio of larval density at hatch tomean egg density divided by the number of elapsed days between the twoestimates) was twice as great in oiled areas compared to unoiled areas, sup-porting the hypothesis of oil injury to herring embryos and larvae

l Between 1989 and 1991, herring egg masses were collected from oiledand unoiled beaches and incubated to hatch Larvae were assessed formorphological deformities, cytogenetic abnormalities, and histopatholog-ical lesions In 1989, oiled areas had significantly more morphologicaldeformities and cytogenetic abnormalities than did the unoiled reference

In 1990 and 1991, there were no significant differences between oiledand unoiled

l Herring embryos were exposed to oilewater dispersions of PrudhoeBay crude oil in seawater Genetic damage was the most sensitivebiomarker for exposure, followed by physical deformities, reducedmitotic activity, lower hatch weight, and premature hatching Embryosplaced at oiled sites in Prince William Sound three years after the spillyielded a greater proportion of abnormal and lower weight larvae thandid unoiled sites

l Herring larvae collected throughout Prince William Sound two to fourmonths after the spill occurred Many exhibited morphological malforma-tions, genetic damage, and small size consistent with oil exposure damageobserved in laboratory experiments Collections in 1995 showed muchdifferent, normal parameters

l Adult herring collected three years after the spill at a site oiled by the ExxonValdez showed lower percent hatch and fewer morphologically normallarvae than fish from an unoiled site

The sum of these studies indicate that early (i.e., 1989) exposure toExxon Valdez oil resulted in a number of toxicological endpoints, fromgenetic damage to abnormal development The impacts appeared to betransient, with some fading by the following year However, in-situdeployments of embryos indicated potential lingering effects from oiledbeaches up to three years after the spill Although several different endpointswere used to assess damage to the herring, the mechanisms of toxicity werenot investigated

From an operational perspective, research results of this type force us toreconsider cleanup endpoints and the relationships between response optionsand the amounts of oil left in the environment once endpoints are met.Subparts-per-billion effects levels are orders of magnitude beyond what wehave tacitly accepted in the past, and discussion of their implications for thefuture will likely occur

27.5 ROUTE OF EXPOSURE: THE ANTHRAX EXAMPLE

Anthrax, of course, has absolutely nothing to do with the effects of oil It does,however, provide some interesting, and hopefully relevant, insights into routes

of exposure and the differences in toxicity effect that can result We will usethese observations to consider the implications for oil

Shortly after the horrific events of September 11, 2001, heightened levels ofalert, paranoia, and sophisticated criminal pathology intersected in a series ofseemingly random incidents involving anthrax Anthrax is a serious infectiousdisease caused by exposure to the bacteria Bacillus anthracis Anthraxcommonly affects hoofed animals such as sheep and goats, but humans whocome into contact with infected animals can be infected as well In the past, thepeople at greatest risk for anthrax were farm workers, veterinarians, andtannery and wool workers However, the virulence and lethality of anthraxbacteria inevitably brought it under scrutiny as a potentially effective biologicalagent of aggressiondthereby introducing us all to the startling adjective

“weaponized,” and the bizarre concept of “weapons-grade” anthrax to describehighly refined cultures maximized for human lethality by secret biologicalwarfare labs around the world

In the weeks and months following the 9/11 terrorist attacks in the UnitedStates, a series of what appeared to be random incidents occurred around thecountry, mostly on the East Coast These involved intentional exposures toanthrax spores, usually packaged in ordinary letters mailed to media outlets andpoliticians At least 22 infections and exposures were confirmed, and fivepeople died, including two employees of a U.S Postal Service processingfacility in Washington, D.C

The 2001 anthrax attacks, of course, have nothing to do with oil toxicity.They are, however, instructive and representative for showing how route ofexposure to a toxin can make a significant, and sometimes life-or-death,difference in effect of that exposure

In the case of anthrax, we learned that cutaneous (skin) exposure to thebacteria was obviously a concern, but considerably less deadly than inhalation.For example, an infant who was cutaneously infected at ABC News in NewYork (unknown source) survived; in Washington D.C., several postal workerswere exposed to inhalation anthrax, and two died

The Centers for Disease Control (CDC) issued guidance and fact sheetsabout exposures to anthrax.26This brings us to the relevant point about route

of exposuredthat it makes a difference (For anthrax, italicized emphasis isadded.)26

l Cutaneous:most (about 95%) anthrax infections occur when the bacteriumenters a cut or an abrasion on the skin, such as when handling contaminatedwool, hides, leather, or hair products (especially goat hair) of infectedanimals Skin infection begins as a raised itchy bump that resembles an

insect bite but within 1e2 days develops into a vesicle and then a painlessulcer, usually 1e3 cm in diameter, with a characteristic black necrotic(dying) area in the center Lymph glands in the adjacent area may swell.About 20% of untreated cases of cutaneous anthrax will result in death.Deaths are rare with appropriate antimicrobial therapy.

l Inhalation:initial symptoms may resemble a common colddsore throat,mild fever, muscle aches, and malaise After several days, the symptomsmay progress to severe breathing problems and shock Inhalation anthrax

is usually fatal

l Gastrointestinal: the intestinal disease form of anthrax may follow theconsumption of contaminated meat and is characterized by an acuteinflammation of the intestinal tract Initial signs of nausea, loss of appetite,vomiting, and fever are followed by abdominal pain, vomiting of blood,and severe diarrhea Intestinal anthrax results in death in 25 to 60% ofcases

27.6 ROUTE OF EXPOSURE: OIL

As was the case with the preceding anthrax example, route of exposure alsomakes a difference for oil However, the results of those different exposures arenot nearly as consistent or predictable for oil as for anthrax We can nonethelessuse the same template for exposure to examine the distinct differences inresultant effect

l Cutaneous, fur, feathers:this is the most familiar route of exposure during

an oil spill, almost exclusively because it is the most visible As a result, it ishighly “mediagenic,” and both print and video news coverage rely onimages of oil-fouled birds, mammals, and other animals to link a spill tobiological impacts The nature of those impacts can range from cutaneousirritation to serious impairment of fur and feather function

Many, if not most, petroleum oils irritate unprotected skin, human or wise This was confirmed for wildlife by Frost, who examined largenumbers of dead, oiled harbor seals after the Exxon Valdez spill.27 Sheobserved that some of these animals had severely irritated skin and inflamedeyes, which would be expected with direct contact of those tissues withfresh oil

other-In a raredpossibly the onlyddirect study of the effects of oil on sea turtles,Lutcavage et al investigated the consequences of oil exposure to seaturtles.28 Contact of sea turtle skin with oil, particularly the soft pliableareas of the neck and flippers, caused skin to slough off in layers

However, it is through the fouling of fur and feathers that we see the directand nearly immediate consequences of oil exposure For all birds and manymammals, the physical structure of fur and feathers permits avian flight and

most importantly, thermal insulation: disruption of the physical integrity offeather and fur by oil contamination substantially decreases or eliminatesaltogether the animal’s ability to retain its warmth As a result, wildlifeexposed to oil in this way can suffer from hypothermia and ultimatelymay perish.

Wildlife rehabilitation efforts in the wake of an oil spill (another favoritemedia image) focus not only on removal of the oil contamination, butalso on restoring the structure and integrity of fouled fur or feathers sothat the affected animals can effectively thermoregulate

l Inhalation/respiratory:most of the petroleum products spilled in the ronment have a volatile component, which is apparent to anyone who hasworked in the vicinity of spilled oil, pumped gas, or filled a kerosenelamp: it has a distinct odor Oil vapors can be harmful both to responsepersonnel and to exposed organisms

envi-The Material Safety Data Sheet (MSDS) for Alaska North Slope crude oilsummarizes the occupational inhalation hazard from that particular product:

“May cause respiratory tract irritation.29Initially, high concentrations willcause central nervous system depression and symptom such as headache,drowsiness, dizziness, nausea, lack of coordination If exposure continues,convulsions, coma, and death may result Inhalation of high concentrations

of a mist may lead to a pneumonia.”

It is reasonable to predict that oil vapor concentrations just above a floatingoil slick, at the airewater interface, would be high, particularly in conditions

of low winds and elevated temperatures For those organisms frequenting thisportion of the open-water habitatdbirds, marine mammals, sea turtlesdinhalation exposure may well represent a significant risk factor Confirmation

of that risk and exposure, however, is difficult to quantify The linkagebetween oil vapor exposure and actual damage or impairment has beenlargely indirect or circumstantial, but some of these links are compelling.Frost observed external signs of irritation and behavioral abnormalities

in oiled harbor seals (Phoca vitulina richardsi) following the 1989 ExxonValdez oil spill.27She also found mild to severe neurological lesions in oiledseals, consistent with hydrocarbon toxicity, and suggested that these mayhave played a role in the disoriented and lethargic behavior observed in theanimals immediately after the spill

Stronger indirect evidence linking the Exxon Valdez spill to population-levelimpacts in another marine mammal in Prince William Sound, the killer whaleOrcinus orca, was provided by Matkin et al.30Prior to that incident, manymarine mammal experts felt that spill impacts to large cetaceans would likely

be minimal For example, Geraci wrote that, “On the whole, it is quiteimprobable that a species or population of cetaceans will be disabled by a spill

at sea, whatever the likelihood that one or a few animals might be affected oreven killed.” However, long-term studies in the aftermath of the Exxon Valdezsuggest exactly this.31

Matkin et al not only documented nearly immediate and substantial spill declines in two long-studied orca pods in Prince William Sound (33%and 41%), but also followed population recovery for the next 16 years.30Their results showed that the resident AB pod had still not attained pre-spillnumbers, and the transient AT1 pod continued to decline The populationtrajectory and structure (e.g., loss of reproductive age females) for the latterwas such that the authors predicted eventual extinction for the transientgroup.

post-The links to putative Exxon Valdez oil vapor exposure for the orcas are graphic, but undeniable: several photographs show groups of whales adjacent

photo-to or cutting through oil slicks, and one by Los Angeles Times phophoto-tographerRosemary Kaul shows members of AT1 at the stern of the stricken tanker(this photo is incorporated into Matkin et al.30)

l Gastrointestinal:observations of surface-oiled birds and mammals showthat contaminated animals will preen feathers and fur in an effort to removethe source of fouling, and thus, considerable quantities of oil can potentially

be ingested Hartung and Hunt examined the effects of this kind of intestinal exposure in ducks for a range of different oils and found that allinduced gastrointestinal irritation, pneumonia, fatty livers, and adrenalcortical hyperplasia.32

gastro-Lutcavage et al found sea turtles incidentally ingested oil, and as a result,oil was observed clinging to the nares, eyes, and upper esophagus, and wasfound in the feces.28 Oiled turtles had up to a fourfold increase in whiteblood cell counts, a 50% reduction in red blood cell counts, and red bloodcell polychromasia Most serum blood chemistries (e.g., BUN, protein)were within normal ranges, although glucose returned more slowly to base-line values than in the controls Gross and histologic changes were present

in the skin and mucosal surfaces of oiled turtles, including acute tory cell infiltrates, dysplasia of epidermal epithelium, and a loss of cellulararchitectural organization of the skin layers The cellular changes in theepidermis are of particular concern because they may increase susceptibility

inflamma-to infection Although many of the observed physiological insults resolvedwithin a 21-day recovery period, the long-term biological effects of oil onsea turtles remain completely unknown

This discussion of routes of exposure has, for reasons of simplicity, usedbirds and mammals as illustrative examples The same concepts are applicable

to water column or benthic inhabitants in the marine environment, but arecomplicated by the aqueous medium and the distinct differences in both oilphysical behavior and the resultant exposure scenarios we can anticipate tooccur Most of our remaining discussion about oil toxicity will focus on fish andother aquatic and marine organisms In order to lay the foundation for ourdiscussion on these points, it is necessary to rewind the tape somewhat to takeanother look at the physical chemistry of oil in water

27.7 OIL CHEMISTRY, PHYSICAL BEHAVIOR, AND OIL

EFFECTS

Having discussed routes of exposure for oils, and determined that (a) they domake a difference and (b) adverse effects of those exposures are apparent, wecan begin to delve more deeply into the subject of oil effects to learn about thecomponents of oil that are harmful and the mechanisms by which they conferthat harm This brings us to oil chemistry, and what it is we focus on incontemplating oil effects It is not our intent here to repeat the precedingmaterial on basic oil chemistry However, oil chemistry is unavoidably relevant

to understanding exposure in the water and issues of comparative and relativetoxicity, and we would be remiss if we failed to discuss it at some rudimentarylevel in the context of oil effect

As discussed in the chemistry segment of the book (Chapter 5) and asalluded to earlier in this chapter, the chemical composition of different oilsvaries considerablydwhich by itself translates into a variable range ofpotential impacts to exposed organisms and communities The mixtures wegenerically consider as “oil” are comprised of several classes of compounds, aswell as thousands of individual hydrocarbon and nonhydrocarbon chemicals Inthis discussion related to toxicity, we will focus on hydrocarbons and one class

of hydrocarbons; but we also need to recognize that oils contain many classes

of compounds that are not hydrocarbons; we will reserve a discussion of thetoxicity of those compounds for some future volume

The most commonly found hydrocarbon molecules in petroleum arealkanes (linear or branched), cycloalkanes, aromatic hydrocarbons, orlarger, more complex chemicals such as asphaltenes As we have empha-sized, each oil we will encounter during a spill event has a unique mix ofcompounds that contributes to or even defines its physical and chemicalproperties, like color and viscosity, as well as its toxic impact to exposedorganisms

The highly variable chemistry of petroleum products means that the range

of products in the complex mixture spilled at the outset of an incident can begreat Adding to this chemical complexity is the fact that the composition is notstatic, that the compositional characteristics of the oil begin changing imme-diately in response to its surroundings The sum of the changes brought about

by contact with the environment, called weathering, is discussed in detail inChapter 5 A number of processes are included under the umbrella of weath-ering: evaporation, emulsification, dissolution, and physical and biologicaloxidation All of these processes begin transforming the original source productinto a mixture with different physical and chemical characteristics, and, itwould follow, different toxicity Neff et al., for example, found that artificiallyweathering different oils changed their chemical composition and their relativetoxicities.33 The study identified the mononuclear aromatic hydrocarbons asthe most acutely toxic fraction of the oils tested, but with weathering, the

proportion of monoaromatic hydrocarbons decreased and the contribution ofPAHs to oil toxicity increased Oil toxicity is, then, a moving target.

The idea that different chemicals have different toxicities or modes ofaction would seem to be intuitively obvious Defining how this works, andhow we might apply it, is (to substantially understate the magnitude of thechallenge) difficult A well-known, and generally well-accepted, approach tomodeling how the structure of a chemical affects its biological activity,including toxicity, is called structure activity relationships (SARs)dalso calledquantitative structure activity relationships (QSARs) QSAR is defined as theprocess by which chemical structure is quantitatively correlated with a bio-logical activity or chemical reactivity

Frequently used surrogates for chemical structure are certain physicalbehaviors, such as lipophilicity A standard measure of lipophilicity is theoctanolewater partitioning coefficient, or Kow Octanol (straight-chain fattyalcohol with eight carbon atoms and the molecular formula CH3(CH2)7OH) andwater do not mix The distribution of a compound between water and octanol isused to calculate the equilibrium partition coefficient ‘P’ of that molecule(often expressed as its logarithm to the base 10, log P) Water/octanol parti-tioning is thought to be a relatively good approximation of the partitioningbetween the cytosol and lipid membranes of living systems

Di Toro et al use Kow as the basis for a model to estimate the toxicity

of weathered and unweathered crude oils.34,35 They defined toxicity of oilmixtures as the toxicity of each individual component of the oil at the watersolubility of that component and termed the approach the target lipid model DiToro et al demonstrated that components with lower log Kowhave greater toxicpotential than those with higher log Kow Weathering removes the lower log Kowchemicals with calculated greater toxic potential, leaving the higher log Kowchemicals with lower calculated toxic potential The replacement of moretoxically potent compounds with less toxically potent compounds lowers thetoxicity of the aqueous phase in equilibrium with the oil, which is consistentwith many studies concluding that weathering lowers the toxicity of oil Theauthors asserted that the contrary ideadthat weathering increases toxicitydwas based on the erroneous use of the total petroleum hydrocarbons or the totalPAHs concentration as if either were a single chemical that could be used togauge the toxicity of a mixture, regardless of its makeup This, as we shall see,continues to be debated

The importance of compositional chemistry in determining physicalbehavior and fate, together with toxic effects, is not a new concept: Anderson

et al published a remarkably complex and robust examination of differential oilbehaviors, organism toxic thresholds, and the challenges of appropriatelydocumenting the relationships between the two, and we review the results inthe next few paragraphs.14 They chemically characterized four oils (SouthLouisiana crude, Kuwaiti crude, No 2 fuel oil, and No 6 fuel oil) and deter-mined how each oil mixed into water Differences between calculated and