Eng.Kelvin

oil_spills.pdf

Articles

Oil Impacts on Coastal Wetlands: Implications for the Mississippi River Delta Ecosystem after the Deepwater Horizon Oil Spiii

IRVING A. MENDELSSOHN, GARY L. ANDERSEN, DONALD M. BALTZ, REX H. CAFFEY, KEVIN R. CARMAN,

JOHN W. FLEEGER, SAMANTHA B. JOYE, QIANXIN LIN, EDWARD MALTBY, EDWARD B. OVERTON, AND

LAWRENCE R ROZAS

On 20 April 2010, the Deepwater Horizon explosion, which released a US government-estimated 4.9 million barrels of crude oil into the Gulf of

Mexico, was responsible for the death of 11 oil workers and, possibly, for an environmental disaster unparalleled in US history. For 87 consecutive

days, the Macondo well continuously released crude oil into the Gulf of Mexico. Many kilometers of shoreline in the northern Gulf of Mexico were

affected, including the fragile and ecologically important wetlands of Louisiana's Mississippi River Delta ecosystem. These wetlands are responsible

for a third of the nation's fish production and, ironically, help to protect an energy infrastructure that provides a third of the nation's oil and gas

supply. Here, we provide a basic overview of the chemistry and biology of oil spills in coastal wetlands and an assessment of the potential and

realized effects on the ecological condition of the Mississippi River Delta and its associated flora and fauna.

Keywords: wetlands, environmental science, ecology, coastal ecosystems, microbiology

The uncontrolled blowout of the Macondo wellhead, locatedat Mississippi Canyon Block 252, which occurred on 20 April 2010 during the completion of drilling by the Deepwater Horizon (DWH) drilling platform, is potentially one of the largest environmental disasters ever experienced in the United States, and, without question, the largest marine oil discharge. The release of a US government- estimated 4.9 million barrels of oil exposed the nation's largest and most productive wetland-estuarine environ- ment to an unprecedented level of environmental impact (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling 2011). The coastal wetlands of the Mississippi River Delta ecosystem, which constitute almost 40% of the coastal wetlands of the 48 conterminous United States, is of special concern because of the multitude of environmentally and economically important services that they supply to the northern Gulf of Mexico (GOM) and to the entire United States. These wetlands provide the base for such ecosystem services as storm protection, water qual- ity enhancement, faunal support, and carbon sequestration. Approximately 30% of the United States' commercial fishery

production is dependent on these wetlands, and ironically, they protect an oil and gas infrastructure that provides one-third of the nation's oil and gas supply and 50% of the nation's refining capacity.

Of course, the DWH spill is not the first to affect coastal wetlands. In the United States alone, multiple smaller spills occur each year. However, large spills that result in significant coastal wetland impacts do occur periodically (table 1). Two of the earliest global spills of note are the West Falmouth release of approximately 4400 barrels of fuel oil from the barge Florida into Buzzards Bay, Massachusetts, and the spill from the tanker Amoco Cadiz, which discharged 1.6 million barrels of crude along the shoreline of Brittany, France. Both spills affected coastal marshes. In the case of the Amoco Cadiz event, animal and plant recovery occurred within 8 years where the soil had not been removed during oil cleanup (Baca et al. 1987). In contrast, 40 years after the West Falmouth spill, impacts were still evident (Culbertson et al. 2008). Marsh recovery after an oil spill can vary greatly (table 1), depending on a variety of factors, which are described further below. The aim of this overview is to

permission to photocopy or reproduce article content at the University of California Press's Rights and Permissions Web site at www.ucpressjournals.com/

p. doi:10.1525/bio.2012.62.6.7

562 BioScience • June 2012 / Yol 62 No. 6 www. biosciencemag. org

Articles

0 \>

oÜ QJ ^ , C

3 C d CO

Ji m

Û . CO

en CD

in -O)

SI ID C3

C CD c

•iií CD N CD

1 1

O o OJ o " in

li CN CD vi S

I . o CTJ"

g 3 -g

11.1

o o o Ö o CNI

s s

a l.

o CO en CD

" O

O •a O

CO CN

CO o

X ÍG . eu

O -O

O CN

10 OJ

c oen CD

• _ r i o

CD OJ ^ •

o QJ -Í-S ' ^í ^ AL ^ r>

00 01

P !5

0) 2? eu m

— CD C

— '5 E ° ' .1 P P .2

CD " > S2

o o ^ ' o o '^ • H o c !

"c CÎ) i p c o lO en "̂

^ 1 o o ID í=

=-°.E.Íc

E ä -

i ? 2 CD

not: ï .E o

5Í

ü : en en ^ 4.J jz CD eD "a in •^ S E Sí en ^ en 2 c::" c

, 4- ^ JD tS ro ' S 'en 0) Sí -c:

-d Q

i .2 p S S

= S ÎTn— CD

ï "*ft f^ 7\ ^"^ fn '̂ -v

^ CD ' " CD o ^ s

Li_ en C/) CO . b CO C

CO t ^ E <l> CD il=i

.i.sl CO C/l Q C/i CO

I to

o o

Q.CO O) 01 (/) -H

1 3

C/)

s p

ill ,

OJ o c CO

ne r

u p

CD Q .

, L

o u

c 'cD

Û. 2

CO CÛ

F a lm

o a

c h

u s

en en CO

= o en npfi

t n " CD

" - CM j r

^ ^ ^ 'en en O c i¿ en "Q

" o CO CO en •— _ Ü . CO P

E ÍE ü̂ F ^

CO . _ ^ _

"S CO

S en

"03

ip tu

_) Q) C

"oj Q .

lîr e e

o en

CO 5

Û_ C/)

•o c CO >

fe ^

c CD

CO 0) ~ CD en 73 "c

0 1 CD • H CD

tm CD ' 5 ' £ S- ^ "¡D

Uli en o > . . £ : £ •

If CN CD tn .E .5Í

~ JD (/) > o •cto"o "D 2 tn • - c c tLO CD ' ^ CO CO c CD

— -n 't- fÇ m

o c if) o

._ ^ H (D

< D • - Í2 ^ oj ^ en (0 .2 E i3 CO CD en = CO CD >i o y

c -ti

^ i OÍ o

CD . a ë

1 = 2II.g • ^ et io

(û

z ^ ro

CD CO 3 ^

'S eD .»̂ o

8 8'

"1.5 eD CQ

0.5° 0.2

^ c § t i S z o I o S o

1 "CD (il i r .Í-J

i| f i t ' -î;:; t;i t/î := ai .

I ^ CO

i "D üb

í" CO

III}}III} ^ ^ CLC/j

8 o"

g CM

S o

¡II o s., t

=: 5 c .g CD â 4_r m

s oî

WWW biosciencemag. org June 2012 / Vol. 62 No. 6 • BioScience 563

Articles

provide a context from which the public and the scientific communities can better understand the realized and poten- tial impacts and prospects for future recovery related to the DWH spill on coastal wetlands and their associated natural resources. Below, we summarize the general knowledge of oil spill impacts on coastal marshes—and, when possible, the impacts specific to the DWH spill—relative to the chem- istry of oil and its related toxicity; the capacity of microbial processes to degrade oil and reduce toxicity; the responses of wetland vegetation, benthic biota, and marsh-dependent fishery resources to oil; and impacts on ecosystem services on which humans rely (figure 1).

Why oil is toxic: Its chemical composition and how it changes over time The toxicity of oil and its impact on biota are primarily determined by its chemical composition. Oil and natural gas are derived from biological materials whose composi- tion has been modified by diagenesis over millions of years to produce the complex mixture of hydrocarbon and non- hydrocarbon compounds that constitute these fossil fuels. Typically, crude oils are all made up of the same types of molecular hydrocarbons. However, the proportion of spe- cific molecules in crude oil from a given reservoir depends on the reservoir's location, depth, and age.

The four classes of hydrocarbons in crude oil are satu- rates, aromatics, asphaltenes, and resins (Leahy and Colwell 1990); saturates and aromatics generally dominate. Saturate hydrocarbons, which contain straight-chain, branched, and cyclic structures, constitute the greatest percentage of crude oil. The majority of crude oils encountered in oil spills con- tain straight-chain hydrocarbon molecules, ranging from single-carbon methane to molecules that contain in excess of 35 carbons, with associated branched and cyclic hydro- carbon structures. Some of the saturate C19-C30 cyclic hydrocarbons are particularly resistant to biodégradation and serve as crude-oil biomarkers. Asphaltenes—large- molecular-weight hydrocarbons containing trace amounts of nickel and vanadium—are even more resistant to micro- bial degradation and are commonly used as roofing tar and road asphalt.

In contrast to saturates, the aromatic hydrocarbons of crude oils consist of relatively simple single-ring struc- tures, such as benzene, and more-complex cyclic aromatic structures with multiple condensed rings, called polycyclic aromatic hydrocarbons (PAHs). The unique structure and bonding of PAHs increase their solubility and, therefore, their ability to influence various enzyme-mediated reactions in biota. Because of their toxic and mutagenic effects, aro- matic compounds are the most environmentally significant of all compounds in crude oils.

When oil enters the environment from spills, ruptures, or blowouts, it undergoes continuous compositional changes associated with weathering (figure 2). Weathering processes include evaporation, dissolution, emulsification, sedimentation, microbial oxidation, and photooxidation.

Weathering changes the oil's physical and toxic properties. Fresh oil is more volatile, contains more water-soluble com- ponents, floats, is not very viscous, and easily disperses from the source. Therefore, freshly spilled oil is the most environ- mentally significant type of oil. Weathered oil initially loses volatile components, which are also the most water-soluble components, and the oil becomes more viscous and more likely to coagulate as opposed to spreading out in a thin film. Over time, weathering continues to change the composition of oil until it degrades in the environment, leaving behind only small quantities of residue (e.g., tar balls; figure 2). Typically, during weathering, much of the oil (especially heavier oil) will mix with water and emulsify, forming a viscous mixture that is resistant to rapid weathering and more difficult to remediate.

Oil can cause environmental damage through several mechanisms, including the toxicity associated with inges- tion or absorption through the biota's respiratory structures or skin; coating or smothering, which affects gas exchange, temperature regulation, or other life-supporting processes; and oxygen depletion by microbial processes associated with oil degradation. Weathering changes the effectiveness of these mechanisms (toxicity, routes of exposure, bioavail- ability) for causing environmental impacts and, in general, lessens the opportunity for damage.

Most of the oil from the DWH event that reached coastal marshes had been extensively weathered (Reddy et al. 2011). As such, the initial considerations of the oil impact were focused on injury from coating of the biota and, to a lesser extent, oil-induced oxygen-deficiency stress in already hypoxic areas. Because weathered oil does not have a signifi- cant route of exposure from dissolution, its potential for toxic impacts is generally lessened compared with the impacts of fi"esh oil. Nonetheless, the toxic effects of the DWH oil on one fish species have been documented (Whitehead et al. 2011). Benthic animals that are bulk-deposit feeders could also con- sume oil in this weathered state from the sediment and be exposed via ingestion (Muijs and Jonker 2010). Regardless, direct oil toxicity to ñora and fauna cannot be ruled out, given the variable extent of oil weathering.

Microbial processes: Key to oil degradation and reduced toxicity The chemical composition of oil and its toxicity are not stable over time but change, in part, because of microbial processes, the primary biological means by which oil is degraded in wetlands. Microbial degradation activity in wetlands depends primarily on the type and concentration of petroleum hydrocarbons and environmental factors such as oxygen, nutrients such as nitrogen and phosphorus, salin- ity, and pH. Surface waters and marsh sediments contain a high diversity of microorganisms. This rich diversity allows for maximum efficiency in resource (especially carbon) utilization and degradation—whether it is petrogenic or not—under changing environmental and nutrient-input conditions.

564 BioScience • June 2012 / Vol. 62 No. 6 www. biosciencemag. org

Articles

Major drivers of

environmental

impact

Ecosystem structure

and function

Ecosystem services

Provisioning Fisheries Livestock Timber Crops Water supply

Regulatory Climate Storm protection Water quality Air quality

Supporting Habitat Soil formation Nutrient cycling

Cultural Meaningful places Heritage Aesthetics

Figure 1. The drivers of environmental impacts during an oil spill include the oil, which changes in composition and toxicity over time, and other Stressors and disturbances acting on the ecosystem. The environmental drivers may influence coastal wetlands directly by affecting the habitat or indirectly by way of various interacting ecosystem components. The final outcome may be expressed at higher trophic levels, such as fisheries, and in the numerous wetland-derived ecosystem services on which society relies. We provide an overview of the impacts of the Deepwater Horizon oil spill from this theoretical context.

The effects of weathering on spilled oil

Toxic and Sticky Floating

Sticky Floating

Possible Sinking

Globular, Nonsticky Floating., Sinking

Tarball Nuisance

Figure 2. Stages in the weathering of spilled oil from the Macondo well. Photographs: National Oceanic and Atmospheric Administration.

Microbes generally degrade the complex mixture constitut- ing crude oil by first metaboliz- ing the linear alkanes, followed by the branched alkanes, small aromatics, and cyclic alkanes and, finally, by performing a limited degradation of the high- molecular-weight PAHs. The relative abundance of various components in complex petro- leum hydrocarbon mixtures shapes the microbial community structure affer contamination. Changes in the types of organ- isms present in sediments or surface waters occur as a result of the sequential degradation of petroleum constituents and the abundance of compounds that are toxic to microbial popula- tions (e.g., heterocyclics, naph- thenic acids).

PAHs are the most toxic con- taminants and are quite per- sistent in marshes. The highest levels generally occur below the sediment surface, where there is limited oxygen and a concomi- tant shift from aerobic to anaero- bic bacterial taxa. In a heavily contaminated mangrove sWamp, PAH concentrations increased with increasing substrate depth and decreasing oxygen content (Li et al. 2009). PAH-degrading anaerobic bacteria have been identified in contaminated wet- land sediments, but given their low oil-degradation capability, it is unlikely that anaerobic bac- teria could greatly reduce PAH contamination, except at low concentrations of PAH.

Alternatively, fungi and yeasts play a lesser but still potentially significant role in the biodég- radation of hydrocarbons. The advantages of fungal degradation in salt marshes include the ability of hyphae to penetrate through anoxic sediment aggregates or hydrophobic environments. Many fungi can transform hydro- carbons into oxidized deriva- tives, and a few strains are able

www. biosciencemag. org June 2012 / Vol. 62 No. 6 • BioScience 565

Articles

to cleave aromatic rings. Several fungal genera are able to remediate soil contaminated with oil. Yeast can also degrade mixtures of both short- and long-chained alkanes found in crude oil (Ijah 1998).

Of the extrinsic environmental factors regulating crude- oil degradation in coastal wetlands, nutrient availability (particularly of nitrogen; Jackson WA and Pardue 1999) and flooding regimes (Shin et al. 2000) are critical controls. For example, the flooding regime of an area exerts a major influence on the biodégradation of crude oil in salt marsh sediments: significant degradation of crude oil through both sulfate reduction and aerobic respiration, the latter of which occurs when the surface of the salt marsh is exposed to the atmosphere (Shin et al. 2000). Flooding, which promotes anaerobic conditions in sediments, hinders hydrocarbon breakdown. However, other studies have demonstrated the breakdown (though often slower) of crude-oil components under anaerobic conditions (Widdel and Rabus 2001). Hydrologie regimes could therefore be expected to influence the rates of crude-oil degradation, and one might antici- pate the degradation of aromatic components, which break down faster under aerobic conditions, to be more affected by flooding regimes than are the saturate components.

Nutrients are known to strongly regulate crude-oil or hydrocarbon degradation because petroleum hydrocarbons provide little regenerated nitrogen or phosphorus when they mineralize. Therefore, their mineralization increases carbon:nitrogen and carbon:phosphorus ratios, which hin- ders further microbial production (Leahy and Colwell 1990). The addition of crude oil to salt marshes can stimulate nitro- gen fixation and can drive the system to phosphorus limita- tion (Griffiths et al. 1982, Gaur and Singh 1990). Although nutrient addition could accelerate microbially mediated hydrocarbon breakdown, the resulting long-term effects on the marsh could be negative (altering nutrient cycling and plant species composition and potentially resulting in eutro- phication and oxygen depletion).

Crude-oil deposition can be expected to alter wetland microbial processes and, hence, marsh biogeochemistry (Leahy and Colwell 1990, Shin et al. 2000). A complex suite of microorganisms can use a range of electron acceptors (e.g., oxygen, nitrate, sulfate, iron oxyhydroxides) to degrade saturated and aromatic hydrocarbons (Widdel and Rabus 2001). Typically, crude-oil exposure reduces microbial diver- sity and increases the abundance of taxa that are able to uti- lize these unique carbon sources. As such, greater availability of crude-oil-derived carbon and its subsequent oxidation will increase microbial production (So and Young 2001) and the flux of carbon dioxide from marshes. Increased produc- tion rates of methane, a potent greenhouse gas, are also pos- sible, because some linear alkanes (e.g., hexadecane) can be converted to methane under anaerobic conditions (Widdel and Rabus 2001). The effects of spilled oil on important microbial processes such as nitrogen fixation, coupled nitrification-denitrification, and sulfate reduction could modify nutrient cycling in oil-affected marshes. Although

the experimental addition of crude oil has a negative impact on the overall microbial community composition, it stimulates sulfate-reduction rates (Suárez-Suárez et al. 2011) and enhances the diversity of microorganisms that degrade linear and aromatic ou components (Abed et al. 2011). Nitrogen fixation and denitriflcation rates, however, can be strongly inhibited by long-term exposure to crude oil (Griffiths et al. 1982), even though some denitrifiers are known to metabolize oil-derived substrates like toluene or ethylbenzene (Champion et al. 1999). By affecting microbial processes, petroleum hydrocarbons can modify nutrient cycling and oil detoxification and can thereby influence veg- etation responses.

Wetland vegetation: The responses of foundation species to oil Wetland vegetation is the ecosystem component providing the foundation for wetland structure and function on which many important ecosystem services rely. Therefore, an understand- ing of the impact of ou on wetland vegetation is particularly important. Vegetation responses to petroleum hydrocarbons and the vegetation's capacity to recover are dependent on a variety of factors both intrinsic to the plant species and specific to the spill event. Because spills occur under different chemical, environmental, and biotic conditions, impacts and recovery trajectories can vary greatly and can be difficult to predict. The primary determinants of vegetation responses to petroleum hydrocarbons are (a) the toxicity of the oil, which is itself dependent on the type of oil, the amount of weathering, and the extent of plant coverage; (b) the oil's amount of contact with and penetration of the sou; (c) plant species composition; (d) oiling fî equency; (e) the season of the spill; and (f) cleanup activities (Lin and Mendelssohn 1996, Hester and Mendelssohn 2000, Pezeshki et al. 2000).

The type of oil is a primary determinant of toxicity. Heavy crude oils, such as San Joaquin or Venezuela crude, which are composed of small concentrations of low-molecular-weight alkanes and aromatics, have a small amount of direct toxic- ity to plants (figure 3a), whereas light crudes, such as South Louisiana crude, which has relatively high concentrations of lighter-weight hydrocarbons, can cause necrosis and plant mortality on contact (figure 3b). Of course, even highly toxic refined products such as diesel, if they are weathered enough, will eventually lose toxicity, but the less-toxic resid- uals can still coat vegetation (figure 3c). This condition pre- vents photosynthesis, thereby impairing the assimilation of carbon used for growth and transpiration, which promotes evaporative cooling. The frequency of repetitive oiling of vegetation is also an important determinant of the ultimate injury; repetitive oiling depletes the underground nutrient reserves used to generate new shoots after successive reoil- ings. The time of the year in which an oil spill occurs also influences the spill's impacts on plants. Spills during colder periods, when the plants have a lower metabolism or are dormant, have a reduced impact relative to oil exposure dur- ing warmer seasons (Alexander and Webb 1985). However,

566 BioScience • June2012/Vol. 62No. 6 www. biosciencemag. org

Articles

Figure 3. (a) Distichlis spicata growing in low-toxicity, heavy San Joaquin crude oil in Suisun Bay, California, (bj South Louisiana crude oil causes direct toxicity to Salicornia bigelovii along a barrier shoreline, (c) San Joaquin crude smothers the stems o/Schoenoplectus californicus along Suisun Bay, causing plant mortality, (d) When oil penetrates marsh soil, such as in this case in a floating marsh in south Louisiana, crude oil comes in direct contact with plant roots and rhizomes, killing the plant. Oil degrades slowly in these anoxic wetland soils. Photographs: Irving A. Mendelssohn.

arguably the most important determinant of severity is whether the oil penetrates the soil and comes into con- tact with nutrient-absorbing roots and shoot-regenerating rhizomes; this scenario can cause plant death (figure 3d). Perennial marsh plants, which regenerate new aboveground shoots each spring, usually recover from stem and leaf oiling, but oiling of belowground plant organs more often results in plant death. Although one might expect that all plant species are similarly susceptible to oil, this is not the case. Species- specific differences in responses to oil can be dramatic (Lin and Mendelssohn 1996). Oil cleanup is another important controller of oil spill impacts. Manual removal of oil by spill- response personnel can break plant shoots, which may do more harm to the vegetation than the oil itself, or they may push oil farther into the soil (Hoff et al. 1993, Hester and Mendelssohn 2000). Therefore, first responders to oil spills try to minimize such damaging impacts.

The extent of recovery of marsh vegetation after an acute oil spill impact can be just as variable as the initial effect of

the spill on vegetation. In situations in which the oil has an impact only on aboveground shoots and leaves, recovery can be relatively rapid, occurring the following growing season or earlier, depending on the presence of viable propagules, the prevalence of residual oil, the extent of shoreline erosion, and the impacts of the cleanup (Hester and Mendelssohn 2000). However, when oil penetrates the soil and the initial mortality of the vegetation is extensive, recovery to reference conditions may take 3-4 years (Hester and Mendelssohn 2000) or even longer (Bergen et al. 2000, Michel et al. 2009). In extreme cases, recovery may never occur if certain intru- sive remediation actions are performed (e.g., soil removal, as in the Amoco Cadiz spill; Baca et al. 1987, Gilfillan et al. 1995), or if erosion due to wave energy or subsidence is accelerated after plant mortality. In contrast, some remedia- tion actions, such as in situ burning, can provide postspill conditions that enable initial recovery within days or weeks (Baustian et al. 2010). Therefore, although a number of fac- tors infiuence the degree of impact and the speed of recovery

www. biosciencemag. org June 2012 / Yol 62 No. 6 • BioScience 567

Articles

from oiling, vegetation recovery is more the rule than the exception.

Although a US government-estimated 4.9 million barrels of oil were released into the GOM during the DWH oil spill, marsh shorelines and generally not the marsh interior were primarily exposed to the weathered oil (figure 4), a situation that limited the extent of environmental damage. Nonetheless, approximately 430 miles of marsh shore- lines were oiled (Zengel and Michel 2011). Of those marsh shorelines. that were oiled, 41% (176 miles) were either heavily or moderately oiled (Zengel and Michel 2011). The primary marsh types affected were salt marshes dominated by Spartina alterniflora and Juncus roemerianus; mangroves, dominated by the black mangrove {Avicennia germinans), which were located on small islands and shorelines and as scattered stands within salt marshes; and low- to intermediate-salinity marshes, dominated by Phragmites australis, the common reed, along the margin of the Mississippi River Birdfoot Delta {http://gomex.erma. noaa.gov/erma). Although few quantitative data are yet available on the extent of vegetation impacts throughout the oil-affected areas of the northern GOM (Mishra et al. 2012), recent findings for the salt marshes in the Bay Jimmy area of northern Barataria Bay, Louisiana docu- mented variable impacts depending on oiling intensity (Lin and Mendelssohn 2012). Along heavily oiled shorelines (figure 5), near-complete mortality of the two dominant species, S. alternifiora and /. roemerianus, occurred. In con- trast, moderate oiling had no significant effect on Spartina,

Figure 4. Although the Deepwater Horizon spill caused extensive oiling of marsh shorelines, the oil did not generally penetrate into the vast interior of these marshes (29 July 2010). Photograph: Irving A. Mendelssohn.

despite significantly lowering live aboveground biomass and stem density of Juncus. Since the spill, some recovery has been noted for oiled marshes (Mendelssohn et al. 2011). For example, in the oiled delta marshes at the mouth of the Mississippi River, P. australis produced new shoots from oiled nodes on the stems (figure 6a), and S. alterniflora regenerated from rhizomes along moderately and some heavily oiled shorelines throughout Louisiana (figure 6b). However, as of the fall of 2011, many of the most heavily oiled shorelines had minimal to no recovery, and only time will tell whether these shorelines will revegetate naturally before shoreline erosion occurs.

The unseen responses of the benthic biota: Links to higher trophic levels Although injury to the vegetated aboveground habitat is often the most visible impact, the effects of oil spills on ben- thic organisms may be especially important because of the multiple ecological processes and ecosystem functions that these organisms support. The primary productivity of ben- thic microalgae may rival that of macrophytes, and benthic microalgae serve as the principal food resource for much of the wetland food web. As ecological engineers, infauna bioturbate and aerate sediments, facilitate the decomposi- tion of detritus, and enhance the fiux of nutrients between sediments and the water column. They also serve as food for nekton (fishes and other natant organisms) foraging on flooded wetlands. Epifauna (e.g., the marsh periwinkle) may contribute to trophic cascades that regulate macrophyte

abundance, and reef-building sus- pension feeders (mussels and oys- ters) create nursery habitat for many species. The biomass, species com- position, and availability of benthos therefore affect higher trophic levels in food webs that include humans, and they even contribute to the per- sistence of wetlands.

Oil spill impacts on benthos can be variable and therefore difficult to predict. The toxic effects of oil can cause an acute reduction in the abun- dance of benthic invertebrates due to mortality or avoidance (Sanders et al. 1980), although minor or sub- tle changes in abundance have also been documented (Lee et al. 1981, DeLaune et al. 1984). Differences in responses are likely due to many of the same factors discussed for marsh vegetation. For benthic fauna, the indirect effects of oil on oxygen availability may also be especially significant. If high rates of mortality occur, opportutiist species colonize and reach high population densities.

568 BioScience • June 2012 / Vol. 62 No. 6 www. biosciencemag.org

Articles

These disturbance specialists are typified by small-bodied, deposit-feeding annelids, such as Capitella species. Wetlands that have been chronically exposed to low-level hydrocarbon

Figure 5. Along heavily oiled shorelines, the oil caused plant mortality, leaving an unvegetated marsh platform that can be subject to wave erosion. Vegetation recovery has thus far been minimal in the most heavily oiled seaward marsh zones (3 September 2010). Photograph: Irving A. Mendelssohn.

contamination may have communities that are relatively resistant to oil spills (Carman et al. 2000a). Toxicity is also influenced strongly by the chemical composition of the oil,

as was discussed previously, and by the presence of other Stressors, such as metals or high nutrient concentrations (Sundbäcketal. 2010).

Even if an oil spill does not cause a large reduction in abundance, the relative abundances of benthic species typically change, increasing or decreas- ing after exposure to hydrocarbons (Carman et al. 1997). Although the broader impacts of such changes in species composition are not well under- stood, an increasing body of literature indicates that indirect effects can ripple through communities and ecosystems, and such effects can be more significant than those of direct oil toxicity (Fleeger et al. 2003). For example, there is sub- stantial evidence from microcosm- and small-scale field experiments that toxic effects on benthic consum- ers lead to dramatic changes in the composition of benthic microalgal communities (primarily diatoms, and secondarily cyanobacteria). A reduc- tion in the abundance of oil-sensitive consumers leads to high growth and

Figure 6. Vegetative regrowthfrom heavily oiled plants occurred in some marshes during and after the spill, (a) Green shoots produced from the nodes of oiled stems o/Phragmites australis in the Mississippi River Birdfoot Delta (14 July 2010). Although the stems appeared dead (necrotic), the plants were viable and produced new shoots, (b) Green tillers produced from belowground stems (rhizomes) of heavily oiled Spartina alterniflora plants in northern Barataria Bay, Louisiana (3 September 2010). Photographs: Irving A. Mendelssohn.

www. biosciencemag. org June 2012 / Vol. 62 No. 6 • BioScience 569

Articles

altered species composition of benthic microalgae. These changes also lead to the formation of dense microalgal biofilms, which dramatically alter the flux of nutrients and oxygen between sediments and the overlying water (Carman et al. 2000b).

The effects of oil cleanup on benthos are also highly vari- able. Dispersants and oil together can have a greater impact on benthic communities than oil alone, and techniques that chp and remove macrophytes affect the benthic community much more than would an absence of such intervention in some marshes (DeLaune et al. 1984). Techniques causing additional disturbances (e.g., foot traffic) to the wetland itself may also harm the marsh benthic community. In fact, it has frequently been concluded that the best approach for benthic communities may be no cleanup at all.

The ecological services provided by benthic invertebrates in Louisiana wetlands may have experienced a broad range of effects from the DWH spill, and these effects probably varied, depending on many factors. Because of their high biomass and ecosystem-engineering qualities, large epi- fauna, such as fiddler crabs, oysters, periwinkles, and ribbed mussels, have a high potential to alter marsh ecological func- tion. Fiddler crabs are sensitive to the toxic effects of oil, and therefore, declines in abundance could cause indirect effects that are significant but difficuh to predict. The marsh peri- winkle [Littoraria irrorata) has a significant grazing impact on Spartina (Silliman et al. 2005). Toxic effects on L. irrorata would presumably decrease grazing pressure on Spartina and may aid its postspill recovery. However, oil toxicity expressed on Ceukensia demissa (the ribbed mussel) may reduce Spartina production, because mussels increase soil nitrogen content (Bertness 1984). Oysters are the principal benthic suspension feeder in the northern GOM and appear to be sensitive to oil—particularly to oil suspended in the water by dispersants. They may also have suffered from the effects of reduced salinity affer fireshwater diversions were opened to increase outflow from coastal bays in an attempt to prevent the entry of oil into marshes. Oyster reefs provide important habitat for many estuarine species, and therefore, the indirect effects of oyster toxicity could be substantial. Benthic infauna in Mississippi River Delta marshes is typi- cally composed of small surface-deposit-feeding macro fauna (mostly small annelids) and meiofauna (mostly nematodes and copepods) (Carman et al. 1997) that are consumed in high numbers by juvenile fishes and by crustaceans, such as brown shrimp. Oil-induced changes in species composition in Louisiana will likely not greatly affect the body size of infauna, and therefore prédation rates by nekton may be less affected than in other wetlands.

Potential impacts on fishery resources: The transfer to higher trophic levels Impacts on both vegetation and benthos can translate to higher trophic levels and may impact fishery resources. Given that the spill occurred offshore of Mississippi River Delta wetlands, the focal point of the Fertile Fisheries

Crescent, which extends east and west into Mississippi and Texas, the potential impact on fisheries is of national con- cern (Gunter 1963). Over 90% of coastal fishery landings in the region are derived from species using estuaries at some point in their life cycles (Gunter 1967). Most fishery spe- cies spawn in nearshore or offshore waters of the northern GOM. Species in their early life stages enter estuaries and use the shallow wetlands therein as nursery and rearing areas. Estuaries within the Mississippi River Delta provide habitat for estuary-dependent species populating coastal systems across the northern GOM.

Nektonic species may be directly exposed to oil when they swim through concentrations of dissolved or suspended petroleum constituents. Gill-breathing animals, such as fishes, exchange gases and solutes with their environment across gill surfaces. Gill damage imperils respiration, and gill uptake results in a body load of toxins that may have lethal or sublethal effects (Whitehead et al. 2011). Individual contaminants may be at low concentrations and may have only minor or sublethal effects, but the individual and combined effects of other petroleum constituents, disper- sants, and other contaminants may greatly lower ecological performance and may be synergistic (Stead et al. 2005). The sublethal effects of the DWH spill on a marsh fish {Fundulus grandis) in northern GOM salt marshes were shown to include effects on genes, enzymes, and gills, even though tissue loads were low or undetectable (Wliitehead et al. 2011); therefore, low levels of exposure can lead to large and delayed effects on individuals and modeled populations (Rose et al. 2003).

Coastal fishery populations are vulnerable to oil spills both in their spawning grounds in the GOM and in estua- rine nursery areas. Eggs and individuals at early life stages are particularly susceptible to the toxic effects of petroleum hydrocarbons (Whitehead et al. 2011), and spills near vital spawning areas that coincide with spawning events may have large impacts on recruitment and, eventually, on adult populations. For example. Pacific herring spawned in Prince William Sound, Alaska, shortly after the 1989 Exxon Valdez oil spill. Much of the spawning habitat and all age classes of herring were exposed to oil. Although a clear link between the spill and the Pacific herring population in Prince William Sound was never established, the herring's decline was coincident with the spill, and the population has still not recovered (Thorne and Thomas 2008); however, Pearson and colleagues (2012) see the connection as unlikely.

In the case of the DWH event, the spill overlapped with peak spawning periods for several important fishery spe- cies spawning in the GOM and in coastal passes, including brown shrimp {Farfantepenaeus aztecus), white shrimp {Litopenaeus setiferus), blue crab {Callinectes sapidus), and spotted seatrout {Cynoscion nebulosus). Although the loca- tion of the spill was in deep water, currents carried oil into the shallow spawning areas of these species. The short- and long-term effects of this oil (and dispersants) on eggs and larvae are not yet known, but Fodrie and Heck (2011) did

570 BioScience • June 2012 / Vol. 62 No. 6 www. biosciencemag. org

Articles

not find short-term negative impacts of the spill on juvenile fishes associated with inshore seagrass beds. Many of the species included in their study spawn offshore, where their eggs and larvae could have been exposed to DWH oil. Their comparisons of the pre- (2006-2009) and postspill (2010) abundances of newly settled juvenile fishes in seagrass beds from the Chandeleur Islands, Louisiana, to Saint Joseph Bay, Florida, revealed no losses of 2010 cohorts or shifts in the composition of the fish assemblages of seagrass beds.

The response of nekton populations to an oil spill may depend on species differences in longevity. Populations of short-lived (1-3 year) fishery species (e.g., gulf menhaden, penaeid shrimps, blue crab) are quite volatile, following good and bad years of spawning success. Additional mortal- ity from a large oil spill could accentuate this volatility and could cause an immediate decline in the populations and fisheries of these species. If the effects of the spill are brief and if damage to essential habitats is neither extensive nor long lasting, recovery for these species can also be rapid— within 1 or 2 years (Tunnel 2011). In contrast, the effects of high-mortality years or low recruitment on a long-lived species (e.g., red drum [Sciaenops ocellatus]) may not be seen immediately but could leave gaps in the age structure of the population, which could affect future reproductive output.

The Mississippi River Delta's most important fishery asset is its expansive coastal wetland system, with an extensive marsh-edge shoreline that provides feeding and shelter sites for fishery and forage species in their early life history stages. This marsh-edge shoreline is also the wetland habitat most vulnerable to oiling, because shoreline vegetation is the first to come into contact with oil driven in from adjacent water- ways (figures 4 and 5). Although oil from the DWH spill had undergone considerable weathering before making landfall, F. grandis, which is closely associated with coastal marshes of the northern GOM, showed significant sublethal effects from exposure to weathered DWH oil in the marshes of Barataria Bay, Louisiana (Whitehead et al. 2011).

Fish are mobile organisms and, therefore, may avoid heav- ily oiled shoreline habitat (Roth and Baltz 2009), but if this primary habitat is limited, these individuals will be forced to use less-favorable alternatives. Where oiling is less severe, hydrocarbon concentrations may be too low to be detected and avoided. For example, background levels of weathered oil in marsh sediments do not diminish the use of shoreline marsh by most estuarine organisms (Rozas et al. 2000). In some cases, fish and other animals may not avoid lightly to moderately oiled habitat, in spite of its toxicity, and the use of these areas may induce long-term chronic health (sub- lethal) effects (Whitehead et al. 2011). The long-term use of oiled sites and the avoidance of primary habitat may both negatively affect fishery populations. Therefore, the relative availability of unoiled shorelines, which is extensive (Zengel and Michel 2011), as alternative habitat will be an important mediator of long-term impacts.

Notwithstanding any avoidance strategy, the indirect exposure to petroleum hydrocarbons by fishery species

may result from ingestion of exposed prey. Many prey taxa (e.g., annelids, meiofauna) have a high petroleum tolerance (Peterson et al. 1996, Carman et al. 2000a), and some may harbor relatively high hydrocarbon concentrations ( DeLaune et al. 1984). Predators feeding on contaminated prey can accumulate an additional body load to that acquired from direct exposure. Exposure to petroleum hydrocarbons may also change the species composition of food webs (Carman et al. 2000a). Sensitive species will diminish in population size and reduce prey availability to higher trophic levels. As was noted above, the indirect effects of oiling events often spread through the food web in unforeseen ways.

Clearly, then, fishery species responses to large-scale dis- turbances such as the DWH spill may be difficult to tease out from the high interannual variability in the populations of most species (Rose 2000) and from a long history of human and natural perturbations and overfishing (Chesney et al. 2000, Jackson JBC et al. 2001). The responses of these popu- lations to the 2011 Mississippi River flood may, in particular, obscure any potential spill effects. Fisheries closures, such as those implemented following the DWH spill, may also be confounded with the spill effects and may make those effects difficult to isolate (Fodrie and Heck 2011). Moreover, fish populations and recruitment are likely to exhibit com- pensatory responses (e.g., better growth and lower mortality of survivors) to an episodic stress or mortality event from a spill; therefore, a spill may not necessarily affect adult population numbers and biomass if the exposure is not sus- tained or widespread (Tunnell 2011). However, a spill dur- ing critical life stages could reduce recruitment and could add an additional stress to populations already burdened by overfishing, reduced habitat quality, and contaminants (Rose 2000). The uncertainties associated with the poten- tial impacts of the DWH spill on fishery populations and fisheries may be greater than that for most other biotic com- ponents of the ecosystem and will probably be difficult to resolve (Tunnell 2011). A final conclusion on the assessment of spill effects may not be possible for years.

Integrating impacts through ecosystem services The impacts of oil on the ecological structure and function of wetland ecosystems may alter the resulting benefits to human well-being, which are often referred to as ecosys- tem services (Maltby 2009). The extensive literature on the functioning of coastal wetland ecosystems both worldwide and adjacent to the GOM demonstrates the importance of healthy, functioning coastal wetlands to local, regional, and national economies. Coastal wetland functioning is under- pinned by fundamental hydrogeomorphic, biogeochemical, and ecological processes that are susceptible to changes from environmental as well as human-induced perturba- tions. Oil spills, like the DWH event, are one such human input that can alter important ecosystem services, which are derived from ecological functions, and these inputs may have adverse consequences for the flow of benefits to society. For example, coastal marshes can act as storm

v\rww. biosciencemag. org June2012/Vol. 62No. 6 • BioScience 571

Articles

barriers, providing the services of flood-risk reduction and prevention of property damage; they also promote nutrient cycling and assimilation, which helps to prevent the eutro- phication of adjacent estuaries; and they provide the habitat and food-web support underpinning both commercial and recreational fisheries. Directly or indirectly, oil contamina- tion may alter these ecosystem services.

Interest in ecosystem services has significantly gained prominence since its formalization by the United Nations Millennium Ecosystem Assessment (MA) in 2005. The MA placed ecosystem services into four categories: provisioning, regulating, supporting, and cultural (figure 1). Ecosystem services in each category contribute variously to societal well-being, economics, and culture, as well as less directly through the maintenance of environmental quality. Coastal wetlands can perform many of the services listed in figure 1 (see UK National Ecosystem Assessment 2011 for the most recent review). The formal recognition of the wide range of services and the resulting benefits provides the framework to link the condition of (and impacts on) the coastal environ- ment with the human economy in practical terms. These can include both nonmarket and more traditionally accounted market values, as well as important metrics such as jobs and particular community values.

Research has long established links between simple eco- system descriptors, such as coastal wetland area (which is indicative of their nursery functions) and coastal fisher- ies production. However, considerable uncertainty over the precise relationships among specific coastal ecosystem processes, functioning, and the quantity and quality of the resulting services still exists. This is a significant and chal- lenging research area in which focus is necessary in order to strengthen the evidence base for a full assessment of environmental impacts such as oil spills. The disproportion- ate importance of the natural capital of wetlands and the adjacent coastlines to the nation's renewable resources can- not be overemphasized. For example, a recent study in the United Kingdom has revealed that although coastal habitat constitutes only 0.6% of its land area, the total value of the ecosystem services provided by it is 3.5% of gross national income, or more than $72 billion (adjusted to 2003 values; UK National Ecosystem Assessment 2011).

Of all the ecosystem services that coastal wetlands sup- port, the two most easily quantified are arguably fishery support and production. Ports in Louisiana, Mississippi, and Alabama typically account for 90% of the commercial fisheries biomass harvested from the five US states bordering the northern GOM and for approximately 30% of the US fisheries production. In 2009, revenue from fisheries land- ings in the northern GOM exceeded $697 million, with more than half of that value derived ffom the harvest of penaeid shrimps (NMES 2010). However, a comprehensive assessment of direct economic fisheries losses in the wake of the DWH ou spiU requires the inclusion of both dockside losses and losses to seafood dealers, processors, and retailers. Moreover, economic losses to the recreational fishing sector could prove

equal to or greater than commercial fisheries losses. In 2009, 6 million anglers took 22.4 million recreational fishing trips in the GOM region. The trip-related expenditures from those anglers provided more than $2 billion in revenue for fishing- related businesses (NMFS 2010). Unlike their commercial counterparts, revenues in recreational fisheries are primarily a function of human population. Correspondingly, recreational losses are expected to be greatest in spill-affected coastal regions with large angler populations. These apparent eco- nomic losses to commercial and recreational fisheries thus far are due to fishery closures, vessels leaving the fishery for spOl cleanup work, and consumer perception and choice, rather than to effects on stocks. As with other spills, the full extent of the impacts of the oil, itself, on fisheries wiU only become apparent after some years.

Conclusions Although more than two years have passed since the DWH blowout, considerable uncertainty still exists concerning both the short- and long-term impacts of the oil spill on wetland ecosystems (DeLaune and Wright 2011). With the exception of the obvious impacts on shoreline marsh veg- etation, the short-term effects on microbial communities, benthic and epibenthic biota, and marsh-dependent fishes and fisheries are still being investigated. Few published stud- ies specifically identifying impacts on the wetlands from the DWH spill are available at this time. Furthermore, informa- tion on the impacts on wetland biogeochemical processes and the resulting ecosystem services together with their ulti- mate link to human well-being is absent. Investigations being done through the Natural Resource Damage Assessment and by independent scientists will certainly fill in some of these gaps over time. Regardless, the long-term impacts will be even more difficult to quantify. This is especially relevant to the Mississippi River Delta ecosystem, where wetland degradation has been a chronic problem due to a variety of natural and human-induced factors, such as the levee- ing of the Mississippi River and dam construction upriver, the river's present fixed position, land subsidence from oil and gas extraction and natural geologic processes, extreme weather conditions such as hurricanes and droughts, inva- sive species (e.g., the nutria), and canal construction for oil and gas activities and for navigation. These impacts caused a 5400-square-kilometer loss of wetlands between 1930 and 2010 (Couvillion et al. 2011), the greatest wetland loss in the United States. Whether the DWH event has accelerated this loss is presently unknown.

A major concern relative to wetlands is whether their sus- tainability and resistance to future disturbances have been degraded by the DWH spill. The Mississippi River Delta wetlands remain viable only if their rate of elevation change keeps pace with the rate of relative sea-level rise (a result of both land subsidence and global sea-level rise). In addition, marsh shorelines that have become unvegetated because of oil impacts may experience increased shoreline erosion, which may contribute to wetland loss. Similar to those of

572 BioScience • June2012/Vol. 62No. 6 www.biosciencemag.org

Articles

the Exxon Valdez oil spill, the impacts on coastal ecosystems, including wetlands, from the DWH spill may not be immedi- ately evident and will require future evaluations to enable us to more fully understand the spill's ecological consequences. Given that the wetlands associated with the Mississippi River Delta system were in a state of deterioration prior to the DWH event, it will be all the more difficult to differentiate between future impacts caused by the DWH event and those resulting from the ongoing processes of wetland degradation (Boesch et al. 1994). This problem of separating DWH oil- related impacts from the multitude of other perturbations to the Mississippi River Delta ecosystem will be the ultimate challenge for scientists and resource managers attempting to accurately assess long-term injury from the spill.

Acknowledgments '* This review was made possible through support from the Gulf Research Initiative to Louisiana State University (to IAM, QL, KRC, and JWF); fi-om the Northern Gulf Institute (to IAM and QL); from the Louisiana Sea Grant Program, a part of the National Sea Grant Program maintained by the National Oceanic and Atmospheric Administration (NOAA) of the US Department of Commerce (to IAM, QL, and EM, who is a Sea Grant Laborde Endowed Chair); from the National Science Foundation (to SBJ); and from the Lawrence Berkeley National Laboratory under contract DE-AC02-05CH11231 with the US Department of Energy (to GLA). The findings and conclusions in this article are those of the authors and do not necessarily represent the views of NOAA or any other federal, state, or private agency.

References cited Abed RMM, Muscat N, Musat F, Mussmann M. 2011. Structure of microbial

communities and hydrocarbon-dependent sulfate reduction in the anoxic layer of a polluted microbial mat. Marine Pollution Bulletin 62: 539-546.

Alexander SK, Webb IW. 1985. Seasonal response of Spartina alterniflora to oil. Pages 355-357 in the Proceedings of the 1985 Oil Spill Conference. American Petroleum Institute.

Baca BJ, Lankford TE, Gundlach ER. 1987. Recovery of Brittany coastal marshes in the eight years following the Amoco Cadiz incident. Pages 459-464 in the Proceedings of the 1987 Oil Spill Conference. American Petroleum Institute.

Baustian J, Mendelssohn I, Lin Q, Rapp ). 2010. In situ burning restores the ecological function and structure of an oil-impacted coastal marsh. Environmental Management 46: 781-789.

Bergen A, Alderson C, Bergfors R, Aquila C, Matsil MA. 2000. Restoration of a Spartina alterniflora salt marsh following a fuel oil spill. New York City, NY. Wetlands Ecology and Management 8: 185-195.

Bertness MD. 1984. Ribbed mussels and Spartina alterniflora production in a New England salt-marsh. Ecology 65: 1794-1807.

Boesch DE, losselyn MN, Mehta AI, Morris IT, Nuttle WK, Simenstad CA, Swift DIP. 1994. Scientific assessment of coastal wetland loss, restoration and management in Louisiana. lournal of Coastal Research, Special Issue 20: 1-103.

Carman KR, Fleeger IW, Pomarico S. 1997. Response of a benthic food web to hydrocarbon contamination. Limnology and Oceanography 42: 561-571.

. 2000a. Does historical exposure to hydrocarbon contamination alter the response of benthic communities to diesel contamination? Marine Environmental Research 49: 255-278.

Carman KR, Bianchi TS, Kloep E. 2000b. Infiuence of grazing and nitrogen on benthic algal blooms in diesel fuel-contaminated saltmarsh sedi- ments. Environmental Science and Technology 34: 107-111.

Champion KM, Zengler K, Rabus R. 1999. Anaerobic degradation of ethyl- benzene and toluene in denitrifying strain BbNl proceeds via in depen- dent substrate-induced pathways. lournal of Molecular Microbiology and Biotechnology 1: 157-164.

Chesney El, Baltz DM, Thomas RG. 2000. Louisiana estuarine and coastal fisheries and habitats: Perspective from a fish's eye view. Ecological Applications 10: 350-366.

Couvillion BR, Barras lA, Steyer GD, Sleavin W, Eischer M, Beck H, Trahan N, Griffin B, Heckman D. 2011. Land Area Change in Coastal Louisiana fi-om 1932 to 2010. US Geological Survey. Scientific Investigations Map no. 3164.

Culbertson IB, Valiela I, Pickart M, Peacock EE, Reddy CM. 2008. Long- term consequences of residual petroleum on salt marsh grass. lournal of Applied Ecology 45: 1284-1292.

DeLaune RD, Wright AL. 2011. Potential impact of Deepwater Horizon oil spill on gulf coast wetlands. Soil Science Society of America lournal 75: 1-11.

DeLaune RD, Smith CL Patrick WH Jr, Eleeger IW, ToUey MD. 1984. Effect of oil on salt marsh hiota: Methods for restoration. Environmental Pol- lution 36: 207-227.

Duke NC, Pinzón ZS, Prada MC. 1997. Large-scale damage to mangrove forests following two large oil spills in Panama. Biotropica 29: 2-14.

Eleeger IW, Carman KR, Nishet RM. 2003. Indirect effects of contaminants in aquatic ecosystems. Science of the Total Environment 317: 207-233.

Eodrie FI, Heck KL Ir. 2011. Response of coastal fishes to the Gulf of Mexico oil disaster. PLoS ONE 6: e21609. doi:10.1371/journal.pone.0021609

Gaur IP, Singh AK. 1990. Growth, photosynthesis and nitrogen fixation of Anabaena doliolum exposed to Assam crude extract. Bulletin of Envi- ronmental Contaminate Toxicology 44: 494—500.

GiJfillan ES, Mäher NP, Krejsa CM, Lanphear ME, Ball CD, Meltzer IB, Page DS. 1995. Use of remote sensing to document changes in marsh vegeta- tion following the Amoco Cadiz oil spill (Brittany, Erance, 1978). Marine Pollution Bulletin 30: 780-787.

Griffiths RP, Caldwell BA, Broich WA, Morita RY. 1982. The long-term effects of crude oil on microbial processes in subarctic marine sedi- ments. Estuarine, Coastal and Shelf Science 15: 183-198.

Gunter G. 1963. The fertile fisheries crescent. lournal of Mississippi Acad- emy of Science 9: 286-290.

. 1967. Some relationships of estuaries to the fisheries of the Gulf of Mexico. Pages 621-638 in Lauff GH, ed. Estuaries. American Associa- tion for the Advancement of Science. Publication no. 83.

Hester MW, Mendelssohn IA. 2000. Long-term recovery of a Louisiana hrackish marsh plant community from oil spill impact: Vegetation response and mitigating effects of marsh surface elevation. Marine Environmental Research 49: 233-254.

Hoff RZ, Shigenaka G, Henry CB. 1993. Salt marsh recovery from a crude oil spill: Vegetation, ou weathering, and response. Pages 307-311 in the Pro- ceedings of the 1993 Oil Spill Conference. American Petroleum Institute.

Ijah UIJ. 1998. Studies on relative capabilities of bacterial and yeast iso- lates fi-om tropical soil in degrading crude oil. Waste Management 18: 293-299.

Jackson JBC, et al. 2001. Historical overfishing and the recent collapse of coastal ecosystems. Science 293: 629-638.

Jackson WA, Pardue JH. 1999. Potential for enhancement of biodégrada- tion of crude oil in Louisiana salt marshes using nutrient amendments. Water, Air, and Soil Pollution 109: 343-355.

Leahy JG, Colwell RR. 1990. Microbial degradation of hydrocarbons in the environment. Microbiological Reviews 54: 305-315.

Lee RE, Dornseif B, Gonsoulin E Tenore K, Hanson R. 1981. Eate and effects of a heavy fuel oil spill on a Georgia salt marsh. Marine Environmental Research 5: 125-143.

Levings SC, Garrity SD. 1995. Oiling of mangrove keys in the 1993 Tampa Bay oil spUI. Pages 421-428 in the Proceedings of the 1995 Oil Spill Conference. American Petroleum Institute.

www.biosciencemag.org June 2012 / Vol. 62 No. 6 • BioScience 573

Articles

. 1997. Sublethal injury to red mangroves two years after oiling. Pages 1040-1041 in the Proceedings of the 1997 Oil Spill Conference. American Petroleum Institute.

Li CH, Zhou HW, Wong YS, Tam NFY. 2009. Vertical distribution and anaerobic biodégradation of polycyclic aromatic hydrocarbons in mangrove sediments in Hong Kong, South China. Science of the Total Environment 407: 5772-5779.

Lin Q, Mendelssohn I A. 1996. A comparative investigation of the effects of Louisiana crude oil on the vegetation of fresh, brackish, and salt marsh. Marine Pollution Bulletin 32: 202-209.

. 2012. Impacts and recovery of the Deepwater Horizon oil spill on vegetation structure and function of coastal salt marshes in the Northern Gulf of Mexico. Environmental Science and Technology 46: 3737-3743.

Maltby E, ed. 2009. Functional Assessment of Wetlands: Towards Evaluation of Ecosystem Services. Woodhead.

Mendelssohn IA, Hester MW, Sasser C, Fischel M. 1990. The effect of a Louisiana crude ou discharge fi'om a pipeline break on the vegetation of a southeast Louisiana brackish marsh. Oil and Chemical Pollution 7: 1-15.

Mendelssohn IA, Lin Q, Carman KR, Fleeger J, Hou A, Michel J, Zengel S. 2011. The good news-bad news story of the Deepwater Horizon blowout for coastal marshes: A preliminary overview. 21 st Biennial Conference of the Coastal and Estuarine Research Federation, 6-10 November, 2011, Day tona, Florida. Coastal and Estuarine Research Federation. (6 April 2012; www.sgmeetcom/cerf2011/files/cerf20U_abstract_book.pdf)

Michel J, Nixon Z, Dahlin J, Betenbaugh D, White M, Burton D, Turley S. 2009. Recovery of interior brackish marshes seven years after the Chalk Point oil spill. Marine Pollution Bulletin 58: 995-1006.

Mishra DR, Cho HJ, Ghosh S, Fox A, Downs C, Merani PBT, Kirui P, lackson N, Mishra S. 2012. Post-spill state of the marsh: Remote estima- tion of the ecological impact of the Gulf of Mexico oil spill on Louisiana salt marshes. Remote Sensing of the Environment 18: 176-185.

Muijs B, Jonker MTO. 2010. A closer look at bioaccumulation of petroleum hydrocarbon mixtures in aquatic worms. Environmental Toxicology and Chemistry 29: 1943-1949.

National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling. 2011. Deep Water: The Gulf Oil Disaster and the Future of Offshore Drilling. Government Printing Office.

[NMFS] National Marine Fisheries Service. 2010. Annual Commercial and Recreational Landing Statistics. NOAA Fisheries, Office of Science and Technology, Fisheries Statistics Division, NMFS. (6 April 2012; www. st.nmfs.noaa.gov/stl/commercial/landings/annual_landings.html)

Pearson WH, Deriso RB, Elston RA, Hook SE, Parker KR, Anderson JW. 2012. Hypotheses concerning the decline and poor recovery of Pacific herring in Prince William Sound, Alaska. Reviews in Fish Biology and Fisheries 22: 95-135.

Peterson CH, Kennicutt MC II, Green RH, Montagna P, Harper DE Jr, Powell EN, Roscigno PF. 1996. Ecological consequences of environmen- tal perturbations associated with offshore hydrocarbon production: A perspective on long-term exposures in the Gulf of Mexico. Canadian Journal of Fisheries and Aquatic Sciences 53: 2637-2654.

Pezeshki SR, Hester M, Lin Q, Nyman JA. 2000. The effects of oil spill and clean-up on dominant US Gulf coast marsh macrophytes: A review. Environmental Pollution 108: 129-139.

Reddy CM, et al. 2011. Science applications in the Deepwater Horizon oil spill special feature: Composition and fate of gas and oil released to the water column during the Deepwater Horizon oil spill. Proceedings of the National Academy of Sciences. (6 April 2012; http://dx.doi.org/10.1073/ pnas.U01242108)

Rose KA. 2000. Why are quantitative relationships between environmental quality and fish populations so elusive? Ecological Applications 10: 367-385.

Rose KA, Murphy CA, Diamond SL, Fuiman LA, Thomas P. 2003. Using nested models and laboratory data for predicting population effects of contaminants on fish: A step toward a bottom-up approach for estab- lishing causality in field studies. Human and Ecological Risk Assessment 9:231-257.

Roth A-MF, Baltz DM. 2009. Short-term effects of an oil spill on marsh-edge fishes and decapod crustaceans. Estuaries and Coasts 32: 565-572.

Rozas LP, Minello TJ, Henry CB. 2000. An assessment of potential oil spill damage to salt marsh habitats and fishery resources in Galveston Bay, Texas. Marine Pollution Bulletin 40: 1148-1160.

Sanders HL, Grassle JF, Hampson GR, Morse LS, Garner-Price S, Jones CC. 1980. Anatomy of an oil spill: Long-term effects from the grounding of the barge Florida off West Falmouth, Massachusetts. Journal of Marine Research 38: 265-380.

Shin WS, Pardue JH, Jackson WA. 2000. Oxygen demand and sulfate reduc- tion in petroleum hydrocarbon contaminated salt marsh soils. Water Research 34: 1345-1353.

Silliman BR, van de Koppel J, Bertness MD, Stanton LE, Mendelssohn IA. 12005. Drought, snails, and large-scale die-oft" of southern U.S. salt marshes. Science 310: 1803-1806.

So CM, Young LY. 2001. Anaerobic biodégradation of alkanes by enriched consortia under four different reducing conditions. Environmental Toxicology and Chemistry 20: 473-478.

Stead MA, Baltz DM, Chesney EJ, Tarr MA, Kolok AS, Marx BD. 2005. Swimming performance of juvenile Florida pompano after sublethal exposure to ethylene glycol and methanol: Synergistic effects. Transac- tions of the American Fisheries Society 134: 1438-1447.

Suárez-Suárez A, et al. 2011. Response of sulfate-reducing bacteria to an artificial oil-spill in a coastal marine sediment. Environmental Micro- biology 13: 1488-1499.

Sundbäck K, Alsterberg C, Larson F. 2010. Effects of multiple Stressors on marine shallow-water sediments: Response of microalgae and meio- fauna to nutrient-toxicant exposure. Journal of Experimental Marine Biology and Ecology 388: 39-50.

Thorne RE, Thomas GL. 2008. Herring and the "Exxon Valdez" oil spill: An investigation into historical data conflicts. ICES Journal of Marine Science 65: 40-50.

Tunnell JW Jr. 2011. An Expert Opinion of When the Gulf of Mexico Will Return to Pre-spill Harvest Status Following the BP Deepwater Horizon MC 252 Oil Spill. Gulf Coast Claims Facihty, Feinberg Rozen, LLR

UK National Ecosystem Assessment. 2011. UK National Ecosystem Assess- ment: Synthesis of the key findings. United Nations Environmental Programme and World Conservation Monitoring Centre.

Wliitehead A, et al. 2011. Genomic and physiological footprint of the Deepwater Horizon oil spill on resident marsh fishes. Proceedings of the National Academy of Sciences. (19 March 2012; www.pnas.org/cgi/ doi/10.1073/pnas. 1109545108)

Widdel E, Rabus R. 2001. Anaerobic biodégradation of saturated and aro- matic hydrocarbons. Current Opinion in Biotechnology 12: 259-276.

Zengel SA, Michel J. 2011. Testing and implementation of treatment methods for marshes heavily oiled during the Deepwater Horizon spill. Gulf Oil Spill SETAC Focused Meeting, 26-28 April 2011, Pensa- cola, Florida. (6 April 2012; http://gulfoilspill.setac.org/sites/default/files/ abstract-book-1 .pdf)

Irving A. Mendelssohn ([email protected]), Donald M. Baltz, and Qianxin Lin are affiliated with the Department of Oceanography and Coastal Sciences, at Louisiana State University, in Baton Rouge. Gary L. Andersen is affili- ated with the Center for Environmental Biotechnology, Lawrence Berkeley National Laboratory, in Berkeley, California. Rex H. Caffey is affiliated with the Department of Agricultural Economics and Agribusiness at Louisiana State University, in Baton Rouge. Kevin R. Carman and John W. Fleeger are affili- ated with the Department of Biological Sciences at Louisiana State University, in Baton Rouge. Samantha B. Joye is affiliated with the Department of Marine Science at the University of Georgia, in Athens. Edward Maltby is affiliated with the School of Environmental Science at the University of Liverpool, in England. Edward B. Overton is affiliated with the Department of Environmental Sciences at Louisiana State University, in Baton Rouge. Lawrence P. Rozas is affiliated with the National Oceanic and Atmospheric Administration Fisheries Service's Southeast Fisheries Science Center and Estuarine Habitats and Coastal Fisheries Center, in Lafayette, Louisiana.

574 BioScience • June 2012 / Yol 62 No. 6 www. biosciencemag. org

copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written

permission. However, users may print, download, or email articles for individual use.