Thomas A. Dean

Exposure to hydrocarbons 10 years after the Exxon Valdez oil spill: evidence from cytochrome P4501A expression and biliary FACs in nearshore demersal fishes

Three biomarkers of hydrocarbon exposure, CYP1A in liver vascular endothelium, liver ethoxyresorufin O-deethylase (EROD), and biliary fluorescent aromatic compounds (FACs), were examined in the nearshore fishes, masked greenling (Hexagrammos octogrammus) and crescent gunnel (Pholis laeta), collected in Prince William Sound, Alaska, 7-10 years after the Exxon Valdez oil spill (EVOS). All biomarkers were elevated in fish collected from sites originally oiled, in comparison to fish from unoiled sites. In 1998, endothelial CYP1A in masked greenling from sites that were heavily oiled in 1989 was significantly higher than in fish collected outside the spill trajectory. In 1999, fishes collected from sites adjacent to intertidal mussel beds containing lingering Exxon Valdez oil had elevated endothelial CYP1A and EROD, and high concentrations of biliary FACs. Fishes from sites near unoiled mussel beds, but within the original spill trajectory, also showed evidence of hydrocarbon exposure, although there were no correlations between sediment petroleum hydrocarbon and any of the biomarkers. Our data show that 10 years after the spill, nearshore fishes within the original spill zone were still exposed to residual EVOS hydrocarbons.

The Distribution of Nearshore Fishes in Kelp and Eelgrass Communities in Prince William Sound, Alaska: Associations with Vegetation and Physical Habitat Characteristics

The nearshore (less than 20 m depth) demersal fish community in Prince William Sound, Alaska, is dominated by Pacific cod, Gadus macrocephalus, pricklebacks (mostly Arctic shanny Stichaeus punctatus), gunnels (mostly crescent gunnels Pholis laeta), a variety of greenlings (Hexagrammidae) and sculpins (Cottidae). During summer, the spatial distribution of fishes, over scales of 100s of m to 10s of km, varied by habitats characterized by different vegetation types. Juvenile Pacific cod and greenlings were numerically dominant in eelgrass, Zostera marina, beds. Pricklebacks and sculpins were dominant in areas with an understory of the kelps Agarum cribrosum and Laminaria saccharina. Greenlings and sculpins were the most abundant demersal fishes in more exposed sites with a canopy of Nereocystis luetkeana and an understory of L. bongardiana. Measured habitat variables, including vegetation type, slope, vegetation biomass, and substratum type, explained a significant proportion of the variation in the presence or absence of most fishes. The relative importance of different habitat characteristics varied between taxonomic groups of fishes. Vegetation type explained a significant proportion of variation for cod, rockfishes, and ronquils. Juvenile cod were closely associated with eelgrass, while rockfish and ronquils were associated with kelps. Pricklebacks and rockfishes were more frequently observed on steeply sloped shorelines, while ronquils were more often found at sites with higher biomass of vegetation. Within A. cribrosum habitats, more greenlings and sculpins were present at sites where algal biomass was higher. Also, sculpins were more abundant in deeper water and gunnels were more abundant in shallow water within this habitat. These associations may not have been causative. However, evidence suggests that some differences between fish communities in eelgrass and Agarum beds may have been causally related to vegetation characteristics. The possible roles of different vegetation types as refugia from predators or as sources of prey are discussed.

CORRELATES OF HARLEQUIN DUCK DENSITIES DURING WINTER IN PRINCE WILLIAM SOUND, ALASKA

Abstract We evaluated relationships of Harlequin Duck (Histrionicus histrionicus) densities to habitat attributes, history of habitat contamination by the 1989 Exxon Valdez oil spill, and prey biomass density and abundance during winters 1995–1997 in Prince William Sound, Alaska. Habitat features that explained variation in duck densities included distance to streams and reefs, degree of exposure to wind and wave action, and dominant substrate type. After accounting for these effects, densities were lower in oiled than unoiled areas, suggesting that population recovery from the oil spill was not complete, due either to lack of recovery from initial oil spill effects or continuing deleterious effects. Prey biomass density and abundance were not strongly related to duck densities after accounting for habitat and area effects. Traits of Harlequin Ducks that reflect their affiliation with naturally predictable winter habitats, such as strong site fidelity and intolerance of increased energy costs, may make their populations particularly vulnerable to chronic oil spill effects and slow to recover from population reductions, which may explain lower densities than expected on oiled areas nearly a decade following the oil spill.

Eelgrass (Zostera marina L.) in Prince William Sound, Alaska: Effects of the Exxon Valdez oil spill

Possible injury to, and recovery of, populations of eelgrass, Zostera marina L., in Prince William Sound were assessed following the Exxon Valdez oil spill by comparing populations at oiled vs reference sites between 1990 and 1995. Eelgrass beds in !heavily oiled bays were exposed to moderate concentrations of hydrocarbons. In 1990, a year after the spill, concentrations of total polycyclic aromatic hydrocarbons averaged nearly 4000 ng g-1 dry weight of sediment at oiled sites compared to less than 700 ng g-1 at reference sites. Injuries to eelgrass, if any, appeared to be slight and did not persist for more than a year after the spill. There were possible effects on the average density of shoots and flowering shoots, as these were 24 and 62% lower at oiled than at reference sites in 1990 (p < 0.10 for both). However, there were no differences between oiled and reference sites with respect to eelgrass biomass, seed density, seed germination or the incidence of normal mitosis in seedlings, and there were no signs of the elimiwation of eelgrass beds. Populations recovered from possible injuries by 1991, as there was a sharp decline in hydrocarbon concentrations and there were no differences in shoot or flowering shoot densities between oiled and reference sites in 1990 or subsequent years.

Effects of a large sewage spill on a kelp forest community: Catastrophe or disturbance?

Abstract San Diegos sewage outfall broke during winter 1992, spilling 7.1 × 10 8 litres/d of treated effluent in kelp forest depths for a two month period during an El Nino event. The ecological implications for the Point Loma kelp forest community were studied by comparing long term data with conditions during and after the spill. Surface ammonium concentrations within 1 km of the break were at potentially toxic levels, and light levels were reduced enough to have inhibited kelp germination and growth. However, because of El Nino conditions, it is unlikely that kelp would have germinated in the absence of a spill. Beyond 1 km, high ammonium concentrations benefitted the nutrient-depleted surface canopy of giant kelp ( Macrocystis pyrifera ). Measured sedimentation rates were significantly higher near the outfall during the spill and were strongly related to wave height; water motion, however, prevented sediment accumulation. Bioassays were conducted on a grid of stations surrounding the outfall. There were significant reductions in the density and growth of microscopic sporophytes of Macrocystis outplanted near the outfall during the spill, but this pattern disappeared in samples collected 11 d after the repair was completed and was not observed again. Sediments collected near the outfall during the spill significantly reduced Macrocystis germ tube elongation; a post repair assay showed no differences with respect to the outfall. No significant effects were observed in outplants of juvenile Macrocystis sporophytes, cup corals, and juvenile abalones. Video transects during the spill and subsequent diving observations provided no evidence of sediment accumulation or negative impacts on established animal populations. Kelp population dynamics at the permanent sites were predictable from existing population structure and El Nino conditions. Damage to kelps, apparently resulting from a combination of low light and nutrient conditions with mechanical damage from storms, construction activity, and barge anchor cables, was observed along the outfall immediately adjacent to the break point. Shortly after the outfall was repaired, upwelling improved conditions for kelp germination and growth, and the zone of maximum impact developed into a dense kelp forest. Suspension feeders, detritivores and sea urchins, whose natural history indicates they could have been affected by the spill, showed no unusual population changes. In the context of the continuum of disturbances observed in two decades of population studies at Point Loma, the spill was a modest disturbance similar to the natural vagaries of kelp recruitment. We emphasize that this spill was an intense but not chronic impact during an El Nino event that also stressed control areas. However, it is representative of massive spills in coastal regions, and the fact that a sewage spill of this magnitude had no lasting effects on a kelp forest community is of general interest.

HABITAT‐SPECIFIC RECOVERY OF SHALLOW SUBTIDAL COMMUNITIES FOLLOWING THE EXXON VALDEZ OIL SPILL

We contrasted impacts of the Exxon Valdez oil spill within kelp and eelgrass communities and examined recovery of these communities over a period of up to 10 yr after the spill. Impacts were greatest in sheltered bays that were subject to heavy oiling. The effects of oiling and associated cleanup activities on nearshore subtidal communities within these bays were more evident in eelgrass than in kelp habitats. In 1990, a year after the spill, concentrations of total polycyclic aromatic hydrocarbons (TPAHs) were higher, more groups of organisms were impacted, and more groups showed a negative response to oil (i.e., lower densities at oiled vs. reference sites) in eelgrass than in kelp beds. Recovery was slower in eelgrass than in kelp habitats. About 80% of the groups impacted in eelgrass beds failed to show clear signs of recovery by 1995, six years after the spill. In contrast, most impacted groups in kelp beds recovered within two years. The stronger, more persistent impacts in eelgrass beds were probab...

Densities of Barrow's goldeneyes during winter in Prince William Sound, Alaska in relation to habitat, food, and history of oil contamination

-We evaluated variation in densities of Barrows Goldeneyes (Bucephala islandica) during winter at 214 sites within oiled and unoiled study areas in Prince William Sound, Alaska in relation to physical habitat attributes, prey biomass, and history of habitat contamination by the 1989 Exxon Valdez oil spill. Using general linear model analyses, we found that goldeneye densities were positively associated with occurrence of a stream within 200 m, lack of exposure to wind and waves, and mixed (versus rocky) substrate. We speculate that these associations relate to habitat profitability via selection of beneficial attributes and avoidance of detrimental features. We also determined that biomass of blue mussels (Mytilus trossulus), the primary prey, was not related to Barrows Goldeneye densities; we suggest that mussel standing stock exceeds predation demands in our study areas and, thus, does not dictate goldeneye distribution. After accounting for habitat effects, we detected no effect of history of oil contamination on Barrows Goldeneye densities, suggesting that populations have recovered from the oil spill. Although other studies documented hydrocarbon exposure in Barrows Goldeneyes through at least 1997, either the level of exposure did not affect populations via reductions in survival, or effects of oil exposure were offset by immigration. Received 2 March 2000, accepted 6July 2000.

Seaweeds and the Exxon Valdez Oil Spill

ABSTRACT A three-year study, initiated in 1989, has evaluated the response of subtidal and intertidal seaweed communities to the Exxon Valdez oil spill and subsequent cleanup activities. The project was part of the coastal habitat injury assessment research sanctioned under the natural resource damage assessment program. A stratified random design was used to select oiled sites for the study. Paired control (unoiled) sites were then matched to the oiled sites. The most consistent effect found in subtidal populations in Prince William Sound was the higher relative abundance of small-size classes of kelps at the oiled sites, indicating the prior disappearance of larger plants. This disappearance was possibly caused by activities associated with the cleanup operations. Intertidal populations of algae were affected by the spill and cleanup in all three major areas studied: Prince William Sound, Cook Inlet-Kenai, and Kodiak-Alaskan Peninsula. The most obvious effect was a significant removal of the dominant in...

DAMAGE ASSESSMENT IN COASTAL HABITATS: LESSONS LEARNED FROM EXXON VALDEZ

ABSTRACT This paper examines alternative designs for the monitoring and assessment of damages of environmental impacts such as oil spills. The optimal design requires sampling at pairs of impacted (oiled) and control (unoiled) sites both before and after the event. However, this design proved impractical in evaluating impacts of the Exxon Valdez oil spill on nearshore subtidal communities, and may be impractical for future monitoring. An alternative design is discussed in which sampling is conducted at pairs of control and impact sites only after the impact.