Long-Term Fate and Persistence of Oil from the Exxon Valdez Oil Spill: Lessons Learned or History Repeated? Open Access

For over 20 years, scientists have studied the shorelines of Prince William Sound (PWS) to understand the distribution, fate, persistence, and bioavailability of Exxon Valdez oil residues that stranded on the shore in 1989 (Wells et al., 1995; Rice et al., 1996; Wiens, 2013). The grounding of the tanker resulted in the release of 258,000 barrels of Alaska North Slope crude oil into PWS. Shoreline surveys in 1989 found that approximately 783 km (16%) of the 4,800 km of the shoreline in PWS, Alaska, and another 1,300 km (13%) of the roughly 10,000 km of shoreline in the western Gulf of Alaska were oiled to varying degrees. Subsequent surveys were undertaken by the joint state, federal, and Exxon Shoreline Cleanup Assessment Team from 1990 to 1992 (Boehm et al., 1995; Neff et al., 1995; Page et al., 2008, 2013) and in more recent PWS studies by Exxon and the National Oceanic and Atmospheric Administration (NOAA) in 2001–2009 (Short et al., 2004; Page et al., 2008, 2013; Boehm et al., 2008; Michel et al., 2010; Nixon et al., 2013). Buried or subsurface oil (SSO) has been observed in the middle and upper tide zones of a small fraction of the shores where it was documented in 1991 (Short et al., 2004; Page et al., 2013). Currently, few locations remain with any significant SSO, but the mere presence of these SSO residues (SSORs) continues to drive hypotheses of continuing harm to wildlife (Short et al., 2007). This paper summarizes what is known about the persistence, bioavailability, and risk of SSOR and how our current understanding relates to observations from other well-studied oil spills.

Beginning in 1990, SSO deposits became an issue of concern and resulted in the following specific shoreline surveys (denoted by acronyms given by the joint federal, state, and Exxon teams):

• SSAT in 1990 documented intertidal SSO deposits, defined as oil >5 cm below surface boulder/cobble armor.

• MAYSAP (1991) and FINSAP (1992) surveyed sites where SSO was found by earlier surveys.

The number of shoreline subdivisions and segments surveyed decreased from year to year as subdivisions with no observed SSO in previous surveys were eliminated. These comprehensive shoreline surveys for SSO in 1989–1992 provided the basis for assessing oil distribution and persistence (Table 1; Neff, et al., 1995), allowing for a prediction of conditions prior to 2000. From 2001 through 2007, NOAA and Resource Planning, Incorporated (RPI, in 2007) conducted a series of shoreline surveys in PWS focusing on the presence of SSORs. The initial 2001 NOAA survey was based on results of the 1991 MAYSAP survey (Neff et al., 1995; Short et al., 2002) and served as much of the basis for the subsequent NOAA and RPI surveys. In parallel, Exxon and their scientific team performed shoreline surveys between 2001 and 2009 documenting the observed SSOR. These observations were classified as trace (TR), light oil residue (LOR), moderate oil residue (MOR), and heavy oil residue ([HOR) (Figure 1; Page et al., 2008), depending on the observations of oil in dug pits within the intertidal beaches. Along with these observations, the chemical compositional changes due to weathering, along with the depletion of the remaining shoreline SSO hydrocarbons, were quantitatively determined (Page et al., 2008; Boehm et al., 2008).

The results of the large and varied data set produced from the initial joint federal, state, and Exxon shoreline surveys, along with results from the independent shoreline surveys conducted by NOAA and Exxon, indicate that:

• Surface oil was removed rapidly; persistent residues on the surface weathered to inert asphalt and were generally only observed on boulder/cobble or bedrock shorelines

• SSO declined at rates from ~80%/year in 1989–1992 to ~4%/year after 2001, with those residues occurring in small isolated patches

• By 2009, most SSO on the shorelines was present as LOR, with minor amounts of scattered, sequestered HOR and MOR.

The amount and distribution of the remaining oil on the shoreline is a function of beach geomorphology and wave action exposure (Page et al., 2008, 2013). Currently, most remaining SSORs occur as sequestered deposits in low permeability sediment on boulder/cobble/gravel beaches that are protected by a surface boulder veneer. Why do boulder/cobble shorelines promote SSO persistence? An example cross section of a PWS boulder/cobble beach (Figure 2) shows how SSO deposits are protected by the boulder/cobble surface, and, in most cases, occurs as discontinuous lenses, often trapped by an underlying layer of peat. Results for the 2001 NOAA shoreline surveys show that only exposed boulder/cobble and sheltered boulder/cobble shoreline types contained SSORs designated as HOR or MOR (Figure 3).

As a result of extensive clean-up and natural processes in the dynamic PWS environment, the vast majority of shoreline oil was removed by 1992 (Neff et al., 1995). Even as early as 1992, the oil was very patchy and the size and distribution of these patches became smaller and more isolated over time. By 2001, any surface oiling was nearly completely absent and was represented by small patches (~1 m in length) of asphalt pavement on only some of the originally heavily-oiled shorelines. From 2001 until 2009, the remaining SSORs continued to decrease in amount and continued to weather. Those locations where SSOR was present in 1992, and again verified in 2001, were precisely those at which SSOR was found in the 2007–2009 surveys.

While the oil had been lightly to moderately weathered at the time of first landfall in 1989, a steady progression of weathering in the ensuing decades served to steadily remove most of the oil's more toxic components, including any and all volatile components, within the first 1–2 years, and the light 2- and 3-ringed polycyclic aromatic hydrocarbons (PAHs) thereafter. Comprehensive shoreline surveys conducted in 2007 by the authors indicated that the presence of samples with more than 30% of their PAHs remaining (less than 70% degraded) was a rare and very patchy occurrence (Boehm et al., 2008). The chemistry of the remaining SSOR indicated that it was generally very highly degraded (Figure 4) with nearly all but the most recalcitrant PAH compounds (i.e., the 4-, 5- and 6-ringed PAHs) completely depleted from the oil remnants. Most of the TR, LOR, and MOR samples were characterized as in the highly degraded range (Figure 5). This documented weathering sequence and extent is completely consistent with that observed in other oil spills (e.g. Atlas et al., 1981; Vandermeulen and Singh, 1994).

Because chemical weathering reduced monoaromatic and PAH concentrations, the toxicity of oil, as measured in standard amphipod bioassays performed on sediment samples collected in 1990–1993, diminished rapidly early in the spill history. The reduced toxicity of oil PAHs, the compounds of concern in petroleum, was demonstrated by an extensive set of bioassays conducted between 1989 and 1993. This decrease and the clear relationship between toxicity to sensitive amphipods and total PAH content in the shoreline substrate is clearly indicated in Figure 6, which summarizes the results of these amphipod toxicity tests (Page et al., 2002). From this testing, the threshold TPAH concentration was determined to be about 2,600 ppb (Page et al., 2002), with a LC10 of 4,100 ppb TPAH, similar to the Effects Range Low value reported by Long et al., 1998.

As seen with other spills, highly weathered, sequestered oil from the Exxon Valdez spill was still found in 2009 and most likely remains to the present in a few isolated localities. Patchy SSOR remains precisely because it is sequestered and largely, but not completely, isolated from most weathering processes. We know the mechanisms of why and where the oil is likely to remain (Page et al., 2008; Pope et al., 2010). Based on lessons from previous oil spills, this eventuality was well-known and predicted very early on. Boehm et al. (1995) stated that “Localized residues of weathered oil will no doubt exist beyond 1994 at certain locations, but their environmental significance will be negligible compared with other stresses ongoing in the sound.” The results of the NOAA 2001–2005 and the RPI 2007 shoreline studies show that of the nearly 10,000 shoreline pits excavated, over 95% had no visible oil (Figure 7). Of the pits with some SSOR observed, only 1.5% contained MOR or HOR, the SSOR suspected to have toxicological effects (Page et al., 2008). It is important to note that these surveys were specifically focused and targeted on those shoreline segments that were likely to have oil remaining (based on the earlier 1990–1992 surveys). Based on previous findings, if all the shoreline segments of PWS were to be surveyed, while a few more patches might occur, it is highly unlikely that any significant quantities of SSOR would have been found.

The overall context of hydrocarbons on the shorelines and in the subtidal sediment of PWS needs to be considered within the larger context of the spill. The sediments of PWS are not chemically pristine (Wooley, 2002). Sophisticated forensic analysis identified background levels of PAHs from hydrocarbon-rich natural deposits southeast of PWS and PAHs from commercial operations in sediments throughout PWS (Page et al., 1996). These levels represent the baseline conditions against which the fate of inputs from the oil spill can be measured. The background PAH signature from dated PWS sediment cores can be identified in ~150-year-old sediments (Page et al. 1996). The subtidal background PAH distribution contains a complex suite of petrogenic PAHs that are similar to those found in the Exxon Valdez crude oil cargo. In addition, past heavy fuel oil use at fish processing facilities and mines in PWS contribute to the baseline of bioavailable PAHs in PWS.

Direct measurements and observations confirm that the remaining SSORs exist as scattered sequestered deposits at specific well-known locations and are not readily bioavailable or bioaccessible. The SSORs are degrading naturally and pose no significant risk to the environment as seen by concentrations in mussels from PWS shorelines containing SSOR (Boehm et al., 2004). Examination of oil residues in other intertidal organisms from PWS reveals a similar trend as seen in Figure 8, which shows tissue PAH concentrations from a series of organisms, including mussels (Neff et al., 2006). In almost every case, the TPAH concentrations are below the upper level measured for mussels from reference locations in PWS from 1999–2002 (Figure 8).

In addition to findings of negligible bioavailability of SSOR to intertidal biota, extensive work has been performed examining the accessibility of SSOR to marine mammals (i.e., sea otters) and birds (e.g., harlequin ducks). These studies (Boehm et al., 2011; Neff et al., 2011) focused on the direct and data-driven examination of all plausible pathways for exposure of wildlife to SSOR. The locations of SSOR are restricted to locations on the shore and in substrate types where large clams do not occur and where sea otters do not dig foraging pits. A graphic illustration of the lack of a connection between locations of SSOR and where otters forage is shown in Figure 9. In addition, the feeding behavior of ducks does not involve digging into the substrate to the depth at which SSOR is located, and their prey are currently largely devoid of Exxon Valdez hydrocarbon residues. Therefore, based on direct measurements and multiple lines of data, it was concluded that it is not plausible that sea otters and harlequin ducks continue to be exposed to environmentally significant amounts of PAHs from the Exxon Valdez oil spill 20 years after the spill.

The findings of the extensive, published multidisciplinary studies confirm the lessons learned from all previous crude oil and heavy fuel oil spills:

1. 1)

   Oil on shorelines is rapidly and extensively weathered and removed by natural processes

2. 2)

   While buried oil may remain, this SSOR can be sequestered for decades in intertidal sediments at locations where the subsurface water flow required for erosion, dissolution, and biodegradation of the oil is low

3. 3)

   Sequestration limits the exposure of biota to the potentially harmful fractions of the SSO and does not readily allow accessibility of SSOR to foraging animals.

The fate and persistence of subsurface Exxon Valdez oil, as sequestered in isolated localities and being biologically unavailable, was predictable based on results of previous oil spill studies and was confirmed from field studies continuing from 1989 to 2009.