Abstract

In addition to external oiling, marine oil spills may affect vertebrate animals through degradation of habitat; alterations in food web structure; and contamination of resources by toxic compounds, including polycyclic aromatic hydrocarbons. These processes are not well understood for vertebrates breeding and foraging in terrestrial ecosystems affected by oil, such as coastal marshes that were heavily oiled following the 2010 Macondo oil spill. Here, we review what is known about the ecological and physiological effects of oil exposure on vertebrates in general. We then apply these concepts to salt-marsh vertebrates, with special reference to our ongoing monitoring of impacts and recovery in the seaside sparrow (Ammodramus maritimus) and marsh rice rat (Oryzomys palustris) in Louisiana following the Macondo spill.

The 2010 BP Deepwater Horizon (DWH) oil spill at the Macondo well was one of the worst environmental disasters to affect the United States and one of the largest marine oil spills in history. The spill released an estimated 4.9 million barrels (nearly 80,000,000 liters) over 87 days into the Gulf of Mexico (GOM). The combination of depth (approximately 1600 meters [m]) and distance from shore (approximately 70 kilometers [km] south of the Louisiana coast) allowed crude oil from the Macondo well to disperse widely, with vast amounts reaching shorelines along the northern GOM. Macondo oil in various states of degradation was detected on shorelines of all five US states bordering the GOM for many months after the well had been capped. The most extreme oiling, however, occurred in the coastal marshes of southeastern Louisiana, causing justifiable concern for species inhabiting these areas (Mendelssohn et al. 2012, Michel et al. 2013).

The initial acute and often lethal effects of marine oil spills on vertebrate organisms result from direct contact with oil. Following oil spills, attention is naturally focused on charismatic species, such as visibly oiled brown pelicans (Pelecanus occidentalis). External oiling of seabird plumage can result in reduced water-repellant properties, inadequate thermoregulation, and death (Jenssen 1994). Mammals, particularly those with fur, can also suffer from external oiling (Garshelis 1997). Other effects of acute oil exposure include a compromised ability to deal with natural stressors, such as compensation for oxygen deficiency by fish in hypoxic areas (Whitehead 2013). These acute immediate effects of oil spills on vertebrates are often devastating to individuals and populations and are widely reported to the public, but they may be relatively short lived.

The nonlethal effects associated with crude oil exposure may not be immediately apparent, but they have the potential to persist for many years. Toxicity from chronic exposure to weathering oil (e.g., via ingestion of contaminated items) can induce physiological responses—for example, immunosuppression (Malcolm and Shore 2003). In addition, because many vertebrates are top consumers, they are particularly vulnerable to habitat and trophic-level alterations arising from damage to habitat structure and prey communities (Velando et al. 2005). Both the physiological and ecological effects of oil on vertebrates can have important consequences for fitness. As with acute effects, research on chronic effects in vertebrates has largely been focused on marine species, because oil spills of large magnitude generally occur in marine systems. In the case of the Macondo blowout, however, approximately 1700 km of coastline was oiled across the GOM, with 44.9% of the oiling occurring on marsh habitat (50.8% on beaches, 4.3% on other shoreline types) and 94.8% of all marsh oiling occurring in Louisiana (Michel et al. 2013), posing risks for species inhabiting these areas. Moreover, high organic carbon content and often-anoxic conditions in marsh sediments can limit microbial degradation, which can facilitate the persistence of oil contamination (Reddy et al. 2002). The extent of oiling and the potential for persistence of oil in marshes highlight the need to document the long-term responses of salt-marsh terrestrial vertebrate species.

Here, we provide a broad overview of known chronic physiological and ecological effects of oil on vertebrates, focusing on mammals, birds, reptiles, and amphibians. Given the paucity of data from terrestrial systems, we provide a framework for identifying such effects in terrestrial vertebrate species in particular, especially those with a high risk of exposure to oiled habitats following the DWH spill. We introduce our ongoing studies on two abundant species that reside in salt marshes along the northern GOM year round: the seaside sparrow (Ammodramus maritimus) and the marsh rice rat (Oryzomys palustris).

Physiological effects of oil on terrestrial vertebrates

There are several classes of molecular hydrocarbons present in crude oil. One group, the aromatics (including polycyclic aromatic hydrocarbons [PAHs]), poses a significant threat to wildlife because of toxic and mutagenic effects (Akcha et al. 2003). Both the chemical composition and the toxicity of oil can change as it degrades in the environment (e.g., via weathering by microbes). However, contaminants such as PAHs are some of the last components of oil to degrade and can persist in the environment for many years, even where oil is no longer visually apparent (Mendelssohn et al. 2012).

Uptake and metabolism

In order for oil metabolites to have a direct biological effect on terrestrial vertebrates, they must enter the individual, typically via ingestion, inhalation, or absorption (Smith et al. 2007). For most organisms, the primary route of PAH exposure in oil-affected habitats is through the ingestion of contaminated soils, sediments, and diet items. Consequently, species that feed heavily on sediment-associated invertebrates tend to be at greater risk of PAH exposure relative to higher-order consumers (Brooks et al. 2009). However, PAHs seldom exhibit food web bioaccumulation and biomagnification; therefore, their potential for transfer up the food chain is limited (Neff 1979). This is primarily associated with the increased capacity of vertebrates, including birds and mammals, to metabolize and subsequently eliminate PAH residues.

PAHs can be detected soon after exposure across a wide range of vertebrate organisms and tissues. For example, field studies have identified PAHs in the blood of birds (e.g., Alonso-Alvarez et al. 2007) and in turtle eggs (e.g., Holliday et al. 2007), and lab work has detected PAHs in snake skins (Jones et al. 2009). Following their uptake, PAHs are metabolized by hepatic cytochrome P450 (CYP) oxygenase or mixed-function oxygenase enzymes (Oris and Roberts 2013). Metabolism can also occur in ovo (Malcolm and Shore 2003). Because of this biotransformation, direct measurement of oil components such as total PAH in tissues is not always an accurate reflection of exposure. Rather, the various isoforms of CYP (e.g., CYP1A) or CYP-related enzymes (e.g., ethoxyresorufin-O-deethylase [EROD]) that are upregulated in the presence of PAH are often used as indirect biomarkers of crude oil or PAH exposure (e.g., Head and Kennedy 2007). For example, captive rats exposed to crude oil showed a dose-dependent increase in several hepatic CYP-linked enzymes (Khan et al. 2001). Field studies of sea ducks potentially exposed to crude oil from the Exxon Valdez spill indicated elevated levels of these biomarkers in oiled areas even decades later (e.g., Esler et al. 2010).

Toxicity in terrestrial vertebrates

Although molecular biomarkers such as CYP1A can be indicative of relative PAH exposure, they alone may not imply harm or biological significance (Oris and Roberts 2013). Adverse health effects associated with PAH exposure often result from the formation of PAH metabolites, which have been demonstrated to be genotoxic (Neff 1979). Specifically, these metabolites can bind to and damage DNA, forming DNA adducts (i.e., the binding of DNA to a chemical contaminant). For example, captive rats exposed to naturally contaminated soils with a wide range of PAHs were found to have a subset of these PAHs in the liver and significant upregulation of EROD, and induction of DNA adducts resulted (Fouchécourt et al. 1999). If the DNA adduct is not repaired, otherwise normal cells can malfunction, leading to mutations and cancer (Akcha et al. 2003). Other recognized toxic effects of PAH on vertebrates include reproductive dysfunction, immunosuppression, and edema (Malcolm and Shore 2003). However, most of what is known about PAH metabolism comes from captive studies, in which dosing may not reflect natural levels. There are relatively few field studies of toxicity that link physiological consequences with vertebrate exposure to PAHs. This is particularly true for terrestrial species, and most of this work has been conducted on birds. For instance, a study of yellow-legged gulls (Larus michahellis) following the Prestige oil spill near Spain found blood parameters indicative of hepatic and renal damage in adults from oiled colonies, some of which correlated with total PAH present in blood (Alonso-Alvarez et al. 2007). Similar work was conducted on marine species following the Exxon Valdez oil spill (e.g., Golet et al. 2002). Such effects may be expected in terrestrial amphibians, reptiles, and mammals but remain poorly studied.

Developmental effects

The detrimental effects of PAH and crude oil exposure on developing avian embryos have long been recognized (e.g., Grau et al. 1977). Exposure may result from maternal transfer or topical exposure (e.g., oil being passed from incubating birds’ feathers to eggshells). Much of the experimental work to date has been conducted in model bird organisms (e.g., chicken [Gallus gallus domesticus], mallard [Anas platyrhynchos]) and has involved either the injection of toxicants into various components of the egg or the application of crude oil to egg shells. The consequences include damage to cells, developmental abnormalities, a reduction in body measurements, and reduced survival (Albers 2006). A recent study showed decreased hatching success of mallards when weathered oil collected from the GOM following the DWH spill was applied to egg shells, although the rate of deformities in hatchlings did not differ significantly from controls (Finch et al. 2011). Eggs of free-living terrestrial vertebrates are frequently assayed for PAHs and used as bioindicators of environmental contamination (e.g., Holliday et al. 2007), but less work has been done on the potential biological effects of exposure. In addition to effects in utero or in ovo, juveniles may experience toxicity from contaminated prey items (e.g., Alonso-Alvarez et al. 2007). Depending on the species, adults may or may not consume the same prey items that they deliver to their offspring; therefore, independent consideration of these groups is warranted.

Taxon-specific considerations

Few generalizations can be made about the effect of oil on particular terrestrial vertebrate taxa. Indeed, even two rodent species sharing the same environment may differ in their sensitivity to pollutants such as PAHs, presumably because of differences in life history or physiology (Smith et al. 2002). However, some broad patterns have been suggested: Secondary poisoning (e.g., by ingesting contaminated prey items) is thought to be more common than poisoning from the original source (e.g., oil in the sediment) in birds. Inhalation as a route of exposure may be more relevant to animals spending significant time in or near the contaminated sediment or water (e.g., rodents) than other taxa. Uptake through the skin is particularly important in amphibians (Smith et al. 2007), especially in the presence of ultraviolet light, which may increase PAH toxicity (Malcolm and Shore 2003). Following exposure, shorter-lived species (e.g., some rodents) may be less affected by the carcinogenic effects of PAHs than are longer-lived species (e.g., turtles), because they are likely to die before tumors develop or significantly affect their fitness (Malcolm and Shore 2003). Lactational or placental transfer of toxins is an important potential route of maternal transfer in mammals, whereas developmental exposure to toxins occurs during egg formation in other groups (Smith et al. 2007). Finally, humans living and working in coastal areas may be exposed via routes similar to those of other vertebrates in close and consistent contact with the oiled sediment (e.g., prey ingestion, inhalation), although perhaps to a lesser extent. Research on free-living terrestrial vertebrate responses to contamination can therefore be useful as a sentinel for potential human effects, but to our knowledge, little work has specifically compared the effects of contamination in free-living human and nonhuman vertebrates in this way (but see Espinosa-Reyes et al. 2010).

Ecological impacts of oil on vertebrates

In addition to the detrimental effects of PAH exposure, degradation of preferred habitats and foraging resources may alter demographic processes and, ultimately, population persistence. Quantifying the effect of oil spills on terrestrial vertebrate populations is challenging, however, given a frequent lack of baseline data on population sizes, habitat use, and foraging strategies of species residing in affected areas (see Henkel et al. 2012).

The availability of suitable, uncontaminated habitat is a prerequisite for population recovery from oil-spill-related impacts. Some mobile fauna simply avoid contaminated areas. For example, a number of marine birds reduced their occupancy of oil-contaminated intertidal shoreline habitats for up to 2.5 years following the Exxon Valdez oil spill (Wiens et al. 2004). However, some species are less capable of abandoning preferred habitats, particularly those with small home ranges, high site fidelity, or reliance on specific nesting habitats. In these cases, behavior or ecological interactions may be altered. In Alaska, river otters (Lontra canadensis), whose coastal habitat was heavily oiled following the Exxon Valdez oil spill, selected different habitat characteristics and maintained larger home ranges in oiled habitats for more than 1 year following the oil spill (Bowyer et al. 1995). Aside from coping with reduced habitat quality, individuals may be faced with increased intra- and interspecific competition in new habitats. For example, a West African black turtle species (Pelusios niger) that changed its habitat use following an oil spill in the Niger Delta experienced increased competition with a congener (Pelusios castaneus) already resident in the new habitat (Luiselli et al. 2006).

Decreased availability of preferred prey resources (e.g., as a result of PAH contamination) may cause species to alter their diet or forage in potentially lower-quality habitats (Henkel et al. 2012). This can lead to species-, population-, and community-level effects through changes in trophic interactions that propagate up or down trophic levels, and reduced growth, survival, or reproductive fitness can carry over into subsequent seasons. Following the 2003 Prestige oil spill, the reproductive success of European shag (Phalacrocorax aristotelis) declined: Nestling survival was significantly lower; chick growth and overall productivity also tended to be lower after the oil spill than before it. These changes were, in part, attributed to a decrease in the percentage of preferred forage fish prey (e.g., sand eel, Gymnammodytes semisquamatus; Velando et al. 2005). Similarly, Hebert and colleagues (2006) demonstrated changes in seabird foraging ecology from an aquatic- to terrestrial-based diet, attributable to declining populations of preferred prey fish species associated with human stocking of predatory fishes. Crucially, such changes in seabird foraging ecology have been correlated with declining egg energy content and reproductive potential in these populations (Paterson et al. 2014). An alteration in the assemblages of prey species or a reduced abundance of preferred prey following the DWH spill may similarly have negative effects on terrestrial salt-marsh vertebrate species.

Consequences of the Macondo blowout for resident terrestrial salt-marsh vertebrates

Despite many potential effects, there is not yet a consensus regarding long-term responses of terrestrial vertebrates to the Macondo blowout. For a more thorough understanding of the chronic effects of the DWH spill on terrestrial vertebrates breeding in coastal Louisiana salt marshes, we must first identify the animals at risk and consider exposure pathways in these species. It is then essential to examine both the physiological burden of toxicity and the interacting ecological effects that result from habitat alterations.

Study organisms

Salt marshes are often characterized by high productivity of a few dominant plant species (e.g., Spartina alterniflora and Juncus roemerianus), whereas isolation, high salinity, flooding, and low structural heterogeneity may be responsible for a relatively low species richness of terrestrially foraging vertebrates (Greenberg et al. 2006). For example, amphibians are absent from GOM coastal salt marshes. Some generalist reptiles and mammals may be present (e.g., American alligator, Alligator mississippiensis; coyote, Canis latrans; raccoon, Procyon lotor; white-tailed deer, Odocoileus virginianus), but these become increasingly rare with distance away from freshwater, which reduces their vulnerability to oil in marshes directly adjacent to the GOM (Lowry and Pratt 1974, Dundee and Rossman 1996). A variety of migratory landbirds can be found using the marsh in winter; Nelson's sparrow (Ammodramus nelsoni) winters almost exclusively in coastal marshes (Shriver et al. 2011) and may be vulnerable to oil spills. However, the terrestrial vertebrate species most vulnerable to population-level consequences of the DWH spill are the small number of salt-marsh specialists and year-round residents. A few reptiles (e.g., the salt-marsh snake, Nerodia clarkii; the diamondback terrapin, Malaclemys terrapin) and birds (e.g., the seaside sparrow; the clapper rail, Rallus longirostris) are almost entirely restricted to coastal marshes. Of these, the seaside sparrow is by far the numerically dominant bird species. Marsh rice rats are not restricted to salt marsh, but this species is exceptional among mammals in its abundance and success in coastal GOM salt marshes.

Selecting representative sentinel organisms for assessing contamination risk can often be difficult, because both biotic and abiotic factors can affect the degree of exposure, and the effects can be quite species specific (Smith et al. 2007). Given how few terrestrial vertebrates are resident at high densities in GOM salt marshes, seaside sparrows and marsh rice rats are obvious focal species because of their abundance and strong association with marsh environments across the Gulf coast. These species are likely to play an important ecological role as some of the top consumers in this ecosystem. Combined with their high potential for exposure, these characteristics rank them quite high relative to other salt-marsh vertebrates as crude oil biomonitors (sensu Golden and Rattner 2002). Previous studies in seaside sparrows and marsh rice rats have demonstrated the utility of the species as bioindicators of mercury and polychlorinated biphenyl exposure, respectively (Smith et al. 2002, Warner et al. 2010). Our group began studying the seaside sparrow late in 2011 and the marsh rice rat in spring of 2013 (figure 1). These studies are ongoing; some of the primary research questions being addressed are outlined below.

Figure 1.

Searching for seaside sparrow nests in 2012 in Plaquemines Parish, Louisiana. This plot had been exposed to oil in 2011. Abundance and reproductive success can be compared among plots with varying degrees of exposure to oil. (a) Ear-tagged marsh rice rat captured on the marsh in 2013. (b) Seaside sparrow perched above an oiled marsh in 2011. (c) The oiled marsh surface in 2011. Photographs: Philip C Stouffer.

The seaside sparrow (figure 2) is of interest to conservation biologists for a number of reasons. The species has a tendency to form localized, morphologically distinct populations recognized as subspecies. Two of these (Ammodramus maritima pelonota and Ammodramus maritima nigrescens) have gone extinct in the past 30 years; the latter (the dusky seaside sparrow) went extinct shortly after being listed as endangered under the Endangered Species Act (McDonald 1988, Post and Greenlaw 2009). A third population (the Cape Sable seaside sparrow, Ammodramus maritima mirabilis) is currently on the Endangered Species List. Along the coast of the GOM, four of the five recognized subspecies occupy small geographic ranges, and seaside sparrows are considered a species of conservation concern in all five Gulf Coast states. Despite the loss of vast areas of marsh in southeast Louisiana over the past 100 years, the species remains relatively abundant and is amenable to both collection and mark–recapture studies. This small (around 20-gram) songbird feeds on invertebrates and seeds and spends a considerable amount of time foraging on the ground (Post and Greenlaw 2009).

Figure 2.

A seaside sparrow captured on the salt marsh in Plaquemines Parish, Louisiana, in 2012. This bird was captured on a plot that was not visibly oiled in 2011. In seaside sparrows and marsh rice rats, individual-level condition and cytochrome P450 (e.g., CYP1A) enzyme biomarkers can be compared for animals living on plots with varying levels of exposure to oil. Photograph: Philip C Stouffer.

The marsh rice rat (figure 1a) inhabits salt marshes along the GOM and Atlantic coasts but, unlike the seaside sparrow, can also be found in freshwater marshes as far inland as Illinois (Eubanks et al. 2011). Several subspecies are recognized, but recent molecular work has suggested that there are two clades that may represent distinct genetic species, with a boundary located somewhere near the Mississippi–Alabama border (Hanson et al. 2010). It has been suggested that this species may be susceptible to local population extinction (Kruchek 2004), and an isolated subspecies endemic to the Everglades of Florida, the silver rice rat (Oryzomys palustris natator), is federally endangered. Population density of marsh rice rats has been linked to vegetation structure, with rats appearing to prefer areas with thicker cover (Eubanks et al. 2011). When upland habitat is present, it may serve as a refuge during wetland flooding (Kruchek 2004); however, these semiaquatic rodents are heavily dependent on wetland habitat and are good swimmers. The species is known to eat wetland vegetation, aquatic invertebrates, and bird eggs (Post 1981, Kruchek 2004). Marsh rice rats are the only rodents that we have captured on our study plots and are present in large numbers, permitting both mark–recapture work and collection.

Exposure potential

The degree and duration of exposure to oil or its components in terrestrial vertebrates will affect the magnitude of any physiological response and is highly dependent on the pattern of oiling in the environment. Following the DWH spill, the pattern of oiling in coastal salt marsh was complex: Some shorelines showed no evidence of oiling, whereas others were heavily oiled. On the oiled sites, the marsh behind the shoreline often consisted of a patchwork of oiled and unoiled areas, determined on the basis of the Shoreline Cleanup Assessment Technique (SCAT) program (Michel et al. 2013). We established replicate study areas on the basis of SCAT data in order to compare unoiled areas with sites that experienced moderate-to-heavy oiling. In the absence of prespill data, a multiplot design provides an opportunity to examine the magnitude, reproducibility, and variance in population- and individual-level responses to oiling (Skalski 2000). In addition to our long-term study sites, plots were added in spring 2013 because of concerns about sample size and potential oil redistribution following Hurricane Isaac in August 2012. Collection of sediment samples for PAH analysis on all plots is ongoing and allows the interpretation of vertebrate data with respect to complex spatial and temporal patterns of oiling among locations.

Exposure to oil and our interpretation of response variables may also depend on animal movements, both daily and seasonal. For daily movements, we may not expect to see differences between individuals captured on oiled sites and those from unoiled sites if they use both areas (e.g., while foraging). Radiotelemetry and mark–recapture are being used to examine these movements in our study species. Among seaside sparrows, our data indicate that individuals generally have restricted movements (within a few hundred meters of their nest) during the breeding season, which suggests that we may indeed expect differences among plots, because individuals are consistently experiencing these local environments. However, we do not yet understand the scale of movements outside of the breeding season. Although seaside sparrows are typically considered nonmigratory, fewer marked birds are present on Mississippi breeding grounds in winter than in summer (Mark Woodrey, Mississippi State University Coastal Research and Extension Center, Grand Bay National Estuarine Research Reserve, Moss Point, Mississippi, personal communication, 2 December 2013). Preliminary population genetic analyses suggest considerable gene flow among seaside sparrows from the mid-Texas coast to Mississippi. Both migration and dispersal suggest that the physiological effects of oil have the potential to be observed well outside of oiled areas (e.g., Henkel et al. 2012).

Assessing the physiological impacts of Macondo blowout in resident terrestrial vertebrates

PAHs contained in crude oil can persist in the environment for many years, potentially creating long-term physiological effects in terrestrial vertebrates (Mendelssohn et al. 2012). For example, Exxon Valdez oil persisted in sediment for at least 16 years (Short et al. 2007), and harlequin ducks (Histrionicus histrionicus) showed evidence of PAH exposure even 20 years later (Esler et al. 2010). On the basis of what is known about the foraging and breeding ecology of seaside sparrows and marsh rice rats, these species both face a high likelihood of PAH exposure over the longer term following the DWH spill—for example, via contact with oiled sediment or through the ingestion of contaminated prey. Therefore, we are collecting individuals of both species for long-term study of hepatic CYP1A gene expression, which will be used as a proxy of PAH exposure, with upregulation predicted in individuals living in oiled environments relative to those living in reference areas, particularly in the several years immediately following the spill. Ecological work linking PAH exposure across trophic levels also has the potential to provide insights into ingestion as a potential route of exposure and may be an avenue worth exploring.

However, we must connect exposure metrics with individual- and population-level effects (Hinton et al. 2005). In short-lived species such as seaside sparrows and marsh rice rats, cancerous effects of PAH are less likely than short-term consequences. Developmental malformations in offspring or shifting life-history characteristics (e.g., life span or fecundity) may be more relevant, although detecting such effects is not trivial. We are beginning to explore circulating stress hormones (e.g., corticosterone) as a potential link between oil exposure and health consequences in seaside sparrows, following similar work on other species (e.g., Romero and Wikelski 2002). Hormones are of interest because they can act systemically to simultaneously adjust multiple behavioral, morphological, and physiological traits to cope with environmental challenges but may also be vulnerable to endocrine disruption by environmental contaminants (Wikelski and Cooke 2006).

Perhaps most crucial to understanding impacts and recovery, researchers must track the population size and fitness measures of free-living animals exposed to contaminants over time. Our ongoing work on terrestrial vertebrate responses to the Macondo blowout is addressing the abundance and reproductive success of these species in our study areas. Estimates of bird density are generated on each plot twice each breeding season with standardized point counts. Our first 2 years of data indicate a tremendous decrease in abundance that may be attributable to Hurricane Isaac inundating our study sites between the 2012 and 2013 breeding seasons (Stouffer et al. 2013). Analyses of the patterns of seaside sparrow and marsh rice rat abundance across oiled and unoiled study areas await the collection of additional data. Preliminary seaside sparrow nesting data combined from 2012 and 2013 indicate that nests on unoiled sites were significantly more likely to fledge than those on oiled sites (likelihood ratio, G(1) = 10.9, p = .001; figure 3). This work is also ongoing, and eventual long-term analyses will require a detailed consideration of potentially confounding variables across study sites (Parker et al. 2013).

Figure 3.

The percentage of seaside sparrow nests that survived to fledging on oiled (n = 3) and unoiled (n = 3) plots studied in 2012; a fourth oiled plot was added in 2013. The number of nests with a known outcome (fledge or fail) is included within each bar; the figure does not reflect nests with unknown fates (13% overall).

Assessing the ecological impacts of the Macondo blowout on resident terrestrial vertebrates

Widespread habitat alterations across Louisiana marshes following the Macondo spill have the potential to affect ecosystem structure and function with cascading adverse effects for higher trophic levels. Specifically, direct contact with oil from the Macondo spill caused almost complete mortality of the two dominant Louisiana salt-marsh plant species relative to unoiled reference locations (figure 1c; Lin and Mendelssohn 2012). In addition, heavy oiling of marshes weakens the underlying soil, creating a deep undercut of the upper marsh edge and thereby increasing shoreline erosion rates (McClenachan et al. 2013). These physical changes to salt-marsh ecosystem structure following the Macondo spill likely contributed to marked declines in abundance of invertebrate species, including insects, spiders, and crabs (McCall and Pennings 2012), along Louisiana marshes; it remains to be seen whether PAH exposure may have also contributed to their decline. On the basis of what is known about the diet of seaside sparrow and marsh rice rat, the declines in these prey resources may have consequences for the species’ overall ecology. For example, a reduction in primary consumer abundance can restrict the abundance of secondary consumers by driving these higher consumers to forage for alternative resources or to move to different locations in search of available prey (Rush et al. 2010a). Movements could affect the extent of contact with oiled marsh habitat but could also lead to increased intra- and interspecific competition. Moreover, marsh rice rats and seaside sparrows may compete for both prey and nesting sites (Post 1981). Increased competition between these two common species could have important demographic consequences for either or both species. Tracking changes in the movement patterns of terrestrial vertebrates coupled with prey community metrics (e.g., composition, diversity, abundance) is therefore essential for understanding the ecological impacts of oil on species’ demography.

Importantly, however, it is necessary to connect changes in prey populations with actual changes in a predator's foraging ecology. To address this, the diet of these species is being determined through gut-content analyses of adults and through nestling throat ligatures. Prey switching has not been documented in terrestrial salt-marsh vertebrates following an oil spill. However, using stable isotopes of carbon and nitrogen, clapper rails were observed to use different invertebrate prey species depending on the nesting habitat within estuaries (Rush et al. 2010b). We are using similar tools (e.g., stable isotopes of carbon and nitrogen and fatty acids) to determine whether the potentially reduced abundance of invertebrate prey species poses consequences to fitness or reproductive success in marsh rice rats and seaside sparrows. Specifically, dietary linkages among multiple trophic levels, from primary producers to secondary consumers, are being characterized, and through these trophic linkages, we anticipate a greater understanding of the energy flow and food web structure of salt-marsh communities and how they respond to disturbance. However, cascading ecological effects in the seaside sparrow or the marsh rice rat may be delayed if they are mediated through changes in inter­mediary populations (i.e., prey; Peterson et al. 2003), which warrants long-term monitoring.

Conclusions

Substantial advances have been made in recent years regarding the complex molecular and cellular mechanisms involved in physiological responses to pollutants. Still, determining the broad population- and community-level consequences is necessary to establish the ecological relevance of pollution (Hinton et al. 2005). In the aftermath of the Macondo blowout, the most comprehensive ecotoxicological studies must consider both broad and focused perspectives (Fodrie et al. 2014). For terrestrial vertebrates, our challenges are to understand the links of oil exposure (e.g., via SCAT and sediment PAH data) to cellular responses (e.g., CYP1A expression), to individual outcomes (e.g., hormone levels, nesting success), to population dynamics (e.g., density and distribution), and to community structure (e.g., trophic shifts). In the field, these challenges are compounded by changes in these relationships over the course of recovery and by the lack of baseline data and the irregularity of oil in the salt marsh. Furthermore, the effects of multiple interacting stressors in these ecosystems remain poorly understood. For example, the loss of marsh habitat in coastal Louisiana due to development, hurricanes, subsidence, salinization, or erosion (Craig 1979) could complicate our interpretations if the focal populations are found to be declining following the Macondo spill, particularly in the absence of corroborative data. Other naturally encountered stressors, such as pathogens, may also vary with exposure to oil, again complicating the interpretation of costs to species (Whitehead 2013). The ecological and physiological effects of oil outlined above can, themselves, interact (e.g., shifts in prey resources following a spill could result in a concurrent shift in PAH exposure when ingestion is a primary route of exposure), each with potential consequences to fitness. The work outlined here provides a framework for identifying the complex effects of oil on terrestrial vertebrates in salt marshes; long-term study and an integrative approach are clearly essential for success in this endeavor.

Financial support was provided by a grant from BP and the Gulf of Mexico Research Initiative to the Coastal Waters Consortium and by the Louisiana Department of Wildlife and Fisheries. We are grateful for field assistance provided by R. Eileen Butterfield, Tracy Burkhard, Richard Gibbons, Mark Herse, Ryan Leeson, Emilie Ospina, Laura Southcott, and Joseph Welklin. We thank three anonymous reviewers for insightful comments that improved the manuscript.

References cited

Akcha
F
Burgeot
T
Narbonne
JF
Garrigues
P
Douben
PE
Metabolic activation of PAHs: Role of DNA adduct formation in induced carcinogenesis
PAHs: An Ecotoxicological Perspective
2003
Wiley
(pg. 
65
-
80
)
Albers
PH.
Birds and polycyclic aromatic hydrocarbons
Avian and Poultry Biology Reviews
2006
, vol. 
17
 (pg. 
125
-
140
)
Alonso-Alvarez
C
Munilla
I
López-Alonso
M
Velando
A
Sublethal toxicity of the Prestige oil spill on yellow-legged gulls
Environment International
2007
, vol. 
33
 (pg. 
773
-
781
)
Bowyer
RT
Testa
WJ
Faro
JB
Habitat selection and home ranges of river otters in a marine environment: Effect of the Exxon Valdez oil spill
Journal of Mammalogy
1995
, vol. 
76
 (pg. 
1
-
11
)
Brooks
AC
Gaskell
PN
Maltby
LL
Importance of prey and predator feeding behaviors for trophic transfer and secondary poisoning
Environmental Science and Technology
2009
, vol. 
43
 (pg. 
7916
-
7923
)
Craig
NJ
Turner
RE
Day
JW
Jr
Land loss in coastal Louisiana (U.S.A.)
Environmental Management
1979
, vol. 
3
 (pg. 
133
-
144
)
Dundee
HA
Rossman
DA
The Amphibians and Reptiles of Louisiana
1996
Louisiana State University Press
Esler
D
, et al. 
Cytochrome P4501A biomarker indication of oil exposure in harlequin ducks up to 20 years after the Exxon Valdez oil spill
Environmental Toxicology and Chemistry
2010
, vol. 
29
 (pg. 
1138
-
1145
)
Espinosa-Reyes
G
Torres-Dosal
A
Ilizaliturri
C
Gonzalez-Mille
D
Diaz-Barriga
F
Mejia-Saavedra
J
Wild rodents (Dipodomys merriami) used as biomonitors in contaminated mining sites
Journal of Environmental Science and Health A
2010
, vol. 
45
 (pg. 
82
-
89
)
Eubanks
BW
Hellgren
EC
Nawrot
JR
Bluett
RD
Habitat associations of the marsh rice rat (Oryzomys palustris) in freshwater wetlands of southern Illinois
Journal of Mammalogy
2011
, vol. 
92
 (pg. 
552
-
560
)
Finch
BE
Wooten
KJ
Smith
PN
Embryotoxicity of weathered crude oil from the Gulf of Mexico in mallard ducks (Anas platyrhynchos)
Environmental Toxicology and Chemistry
2011
, vol. 
30
 (pg. 
1885
-
1891
)
Fodrie
FJ
Able
KW
Galvez
F
Heck
KL
Jr
Jensen
OP
Lopez-Duarte
PC
Martin
CW
Turner
RE
Whitehead
A
Integrating organismal and population responses of estuarine fishes in Macondo spill research
BioScience
2014
, vol. 
64
 (pg. 
778
-
788
[in this issue]
Fouchécourt
MO
Arnold
M
Berny
P
Videmann
B
Rether
B
Rivière
JL
Assessment of the bioavailability of PAHs in rats exposed to a polluted soil by natural routes: Induction of EROD activity and DNA adducts and PAH burden in both liver and lung
Environmental Research
1999
, vol. 
80
 (pg. 
330
-
339
)
Garshelis
DL.
Sea otter mortality estimated from carcasses collected after the Exxon Valdez oil spill
Conservation Biology
1997
, vol. 
11
 (pg. 
905
-
916
)
Golden
NH
Rattner
BA
Ranking terrestrial vertebrate species for utility in biomonitoring and vulnerability to environmental contaminants
Reviews of Environmental Contamination and Toxicology
2002
, vol. 
176
 (pg. 
67
-
136
)
Golet
GH
Seiser
PE
McGuire
AD
Roby
DD
Fischer
JB
Kuletz
KJ
Irons
DB
Dean
TA
Jewett
SC
Newman
SH
Longterm direct and indirect effects of the Exxon Valdez oil spill on pigeon guillemots in Prince William Sound, Alaska
Marine Ecological Progress Series
2002
, vol. 
241
 (pg. 
287
-
304
)
Grau
CR
Roudybush
T
Dobbs
J
Wathen
J
Altered yolk structure and reduced hatchability of eggs from birds fed single doses of petroleum oils
Science
1977
, vol. 
195
 (pg. 
779
-
781
)
Greenberg
R
Maldonado
JE
Droege
S
McDonald
MV
Tidal marshes: A global perspective on the evolution and conservation of their terrestrial vertebrates
BioScience
2006
, vol. 
56
 (pg. 
675
-
685
)
Hanson
JD
Indorf
JL
Swier
VJ
Bradley
RD
Molecular divergence within the Oryzomys palustris complex: Evidence for multiple species
Journal of Mammalogy
2010
, vol. 
91
 (pg. 
336
-
347
)
Head
JA
Kennedy
SW
Same-sample analysis of ethoxyresorufin-O-deethylase activity and cytochrome P4501A mRNA abundance in chicken embryo hepatocytes
Analytical Biochemistry
2007
, vol. 
360
 (pg. 
294
-
302
)
Hebert
CE
Arts
MT
Weseloh
DVC
Tracers can quantify food web structure and change
Environmental Science and Technology
2006
, vol. 
40
 (pg. 
5618
-
5623
)
Henkel
JR
Sigel
BJ
Taylor
CM
Large-scale impacts of the Deepwater Horizon oil spill: Can local disturbance affect distant ecosystems through migratory shorebirds?
Bioscience
2012
, vol. 
62
 (pg. 
676
-
685
)
Hinton
DE
Kullman
SW
Hardman
RC
Volz
DC
Chen
P-J
Carney
M
Bencic
DC
Resolving mechanisms of toxicity while pursuing ecotoxicological relevance?
Marine Pollution Bulletin
2005
, vol. 
51
 (pg. 
635
-
648
)
Holliday
DK
Roosenburg
WM
Elskus
AA
Spatial variation in polycyclic aromatic hydrocarbon concentrations in eggs of diamondback terrapins, Malaclemys terrapin, from the Patuxent River, Maryland
Bulletin of Environmental Contamination and Toxicology
2007
, vol. 
80
 (pg. 
119
-
122
)
Jenssen
BM.
Review article: Effects of oil pollution, chemically treated oil, and cleaning on thermal balance of birds
Environmental Pollution
1994
, vol. 
86
 (pg. 
207
-
215
)
Jones
DE
Magnin-Bissel
G
Holladay
SD
Detection of polycyclic aromatic hydrocarbons in the shed skins of corn snakes (Elaphe guttata)
Ecotoxicology and Environmental Safety
2009
, vol. 
72
 (pg. 
2033
-
2035
)
Khan
AA
Coppock
RW
Schuler
MM
Effects of multiple exposures of small doses of Pembina Cardium crude oil and diesel in rats
Archives of Environmental Contamination and Toxicology
2001
, vol. 
40
 (pg. 
418
-
424
)
Kruchek
BL.
Use of tidal marsh and upland habitats by the marsh rice rat (Oryzomys palustris)
Journal of Mammalogy
2004
, vol. 
85
 (pg. 
569
-
575
)
Lin
Q
Mendelssohn
IA
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
2012
, vol. 
46
 (pg. 
3737
-
3743
)
Lowry
GH
Pratt
HD
The Mammals of Louisiana and Its Adjacent Waters
1974
Louisiana State University Press
Luiselli
L
Akani
GC
Politano
E
Effects of habitat alteration caused by petrochemical activities and oil spills on the habitat use and interspecific relationships among four species of Afrotropical freshwater turtles
Biodiversity and Conservation
2006
, vol. 
15
 (pg. 
3751
-
3767
)
Malcolm
HM
Shore
RF
Douben
PE
Effects of PAHs on terrestrial and freshwater birds, mammals, and amphibians
PAHs: An Ecotoxicological Perspective
2003
Wiley
(pg. 
225
-
242
)
McCall
BD
Pennings
SC
Disturbance and recovery of salt marsh arthropod communities following BP Deepwater Horizon oil spill
PLOS ONE
2012
, vol. 
7
  
art. e32735
McClenachan
G
Turner
RE
Tweel
AW
Effects of oil on the rate and trajectory of Louisiana marsh shoreline erosion
Environmental Research Letters
2013
, vol. 
8
  
art. 044030
McDonald
MV.
Status Survey of Two Florida Seaside Sparrows and Taxonomic Review of the Seaside Sparrow Assemblage
1988
Florida Cooperative Fish and Wildlife Research Unit, School of Forestry and Conservation, University of Florida
 
Technical Report no. 32
Mendelssohn
IA
Andersen
GL
Baltz
DM
Caffey
RH
Carman
KR
Fleeger
JW
Joye
SB
Lin
Q
Maltby
E
Overton
EB
Oil impacts on coastal wetlands: implications for the Mississippi River Delta ecosystem after the Deepwater Horizon oil spill
Bioscience
2012
, vol. 
62
 (pg. 
562
-
574
)
Michel
J
, et al. 
Extent and degree of shoreline oiling: Deepwater Horizon oil spill, Gulf of Mexico, USA
PLOS ONE
2013
, vol. 
8
  
art. e65087
Neff
JM
Polycyclic Aromatic Hydrocarbons in the Aquatic Environment, Sources, Fates and Biological Effects
1979
Applied Science Publishers
Oris
JT
Roberts
AP
Wiens
JA
Cytochrome P450 1A (CYP1A) as a biomarker in oil spill assessments
Oil in the Environment: Legacies and Lessons of the Exxon Valdez Oil Spill
2013
Cambridge University Press
(pg. 
201
-
219
)
Parker
KR
Wiens
JA
Day
RH
Murphy
SM
Wiens
JA
Assessing effects and recovery from environmental accidents
Oil in the Environment: Legacies and Lessons of the Exxon Valdez Oil Spill
2013
Cambridge University Press
(pg. 
220
-
240
)
Paterson
G
Hebert
CE
Drouillard
KG
Haffner
GD
Congruent energy density trends of fish and birds reflect ecosystem change
Limnology and Oceanography
2014
, vol. 
59
 (pg. 
1171
-
1180
)
Peterson
CH
Rice
SD
Short
JW
Esler
D
Bodkin
JL
Ballachey
BE
Irons
DB
Long-term ecosystem response to the Exxon Valdez oil spill
Science
2003
, vol. 
302
 pg. 
2082
 
Post
W.
The influence of rice rats Oryzomys palustris on the habitat use of the seaside sparrow Ammospiza maritima
Behavioral Ecology and Sociobiology
1981
, vol. 
9
 (pg. 
35
-
40
)
Post
W
Greenlaw
JS
Poole
A
Seaside sparrow (Ammodramus maritimus)
Birds of North America
2009
American Ornithologists’ Union
 
doi:10.2173/bna.127
Reddy
CM
Eglinton
TI
Hounshell
A
White
HK
Xu
L
Gaines
RB
Frysinger
GS
The West Falmouth oil spill after thirty years: The persistence of petroleum hydrocarbons in marsh sediments
Environmental Science and Technology
2002
, vol. 
36
 (pg. 
4754
-
4760
)
Romero
LM
Wikelski
M
Severe effects of low-level oil contamination on wildlife predicted by the corticosterone-stress response: Preliminary data and a research agenda
Spill Science and Technology Bulletin
2002
, vol. 
7
 (pg. 
309
-
313
)
Rush
SA
Mordecai
R
Woodrey
MS
Cooper
RJ
Prey and habitat influences the movement of clapper rails in northern Gulf coast estuaries
Waterbirds
2010a
, vol. 
33
 (pg. 
389
-
396
)
Rush
SA
Olin
JA
Fisk
AT
Woodrey
MS
Cooper
RJ
Trophic relationships of a marsh bird differ between gulf coast estuaries
Estuaries and Coasts
2010b
, vol. 
33
 (pg. 
963
-
970
)
Short
JW
Irvine
GV
Mann
DH
Maselko
JM
Pella
JJ
Lindeberg
MR
Payne
JR
Driskell
WB
Rice
SD
Slightly weathered Exxon Valdez oil persists in Gulf of Alaska beach sediments after 16 years
Environmental Science and Technology
2007
, vol. 
41
 (pg. 
1245
-
1250
)
Shriver
W
Hodgman
TP
Hanson
AR
Poole
A
Nelson's sparrow (Ammodramus nelsoni)
The Birds of North America Online
2011
Cornell Lab of Ornithology
 
Skalski
JR.
Albers
PH
Heinz
GH
Ohlendorf
HM
Statistical design of wildlife toxicology studies
Environmental Contaminants and Terrestrial Vertebrates: Effects on Populations, Communities, and Ecosystems
2000
Society of Environmental Toxicology and Chemistry
(pg. 
95
-
107
)
Smith
PN
Cobb
GP
Harper
FM
Adair
BM
McMurry
ST
Comparison of white-footed mice and rice rats as biomonitors of polychlorinated biphenyl and metal contamination
Environmental Pollution
2002
, vol. 
119
 (pg. 
261
-
268
)
Smith
PN
Cobb
G
Godard-Codding
C
Hoff
D
McMurry
S
Rainwater
T
Reynolds
K
Contaminant exposure in terrestrial vertebrates
Environmental Pollution
2007
, vol. 
150
 (pg. 
41
-
64
)
Stouffer
PC
Taylor
SS
Woltmann
S
Bergeon Burns
CM
Shupe
TF
Bowen
MS
Staying alive on the edge of the Earth: Response of seaside sparrows (Ammodramus maritumus) to salt marsh inundation, with implications for storms, spills, and climate change
Proceedings of the 4th Louisiana Natural Resources Symposium
2013
(pg. 
82
-
93
)
Velando
A
Munilla
I
Leyenda
PM
Short-term indirect effects of the Prestige oil spill on European shags: Changes in availability of prey
Marine Ecology Progress Series
2005
, vol. 
302
 (pg. 
263
-
274
)
Warner
SE
Shriver
WG
Pepper
MA
Taylor
RJ
Mercury concentrations in tidal marsh sparrows and their use as bioindicators in Delaware Bay, USA
Environmental Monitoring and Assessment
2010
, vol. 
171
 (pg. 
671
-
679
)
Whitehead
A.
Interactions between oil-spill pollutants and natural stressors can compound ecotoxicological effects
Integrative and Comparative Biology
2013
, vol. 
53
 (pg. 
635
-
647
)
Wiens
JA
Day
RA
Murphy
SM
Parker
KR
Changing habitat and habitat use by birds after the Exxon Valdez oil spill, 1989–2001
Ecological Applications
2004
, vol. 
14
 (pg. 
1806
-
1825
)
Wikelski
M
Cooke
SJ
Conservation physiology
Trends in Ecology and Evolution
2006
, vol. 
21
 (pg. 
38
-
46
)