Something went terribly wrong with "science" in the aftermath of the Exxon Valdez oil spill as illustrated by the extreme differences in the findings of key studies between government and industry. These differences are symptomatic of underlying problems with the scientific and regulatory processes that were designed to produce quality science.

After the Exxon Valdez oil spill, as is the case of other large spills, the federal and state trustees for natural resources and the spiller, Exxon, designed and initiated studies to collect data to support their pre-conceived positions. For the governments, the position was that resources were injured and studies were designed to determine the nature of those damages. Exxon held the position that resources may have been injured, but recovery was rapid, and their studies were designed to determine what and how much was not injured, in order to minimize the Corporation's liability (see Appendix A).

Science is an open pursuit of truth with no preconceptions as to the outcome of results. However, the government and Exxon's studies were touted as "science" and were forced upon the scientific community to be treated as "science" for purposes of presentations and publications.

Part of the problem is that the scientific community is not prepared to handle studies that are premised on preconceptions, i.e. "spill science." Another part of the problem is that federal laws governing spill research simply do not work for catastrophic spills with minimal baseline data, because of the disproportionately large number of damage assessment studies needed for litigation and classified as confidential.

This section examines the roles of the scientific community, Congress, and the state and federal governments in improving the quality of science that becomes part of the public domain, particularly science undertaken in the aftermath of an oil spill.

The Role of the Scientific Community

Background: The scientific community relies on a peer review process for quality control of initial study design and final publication of research results. This process depends on scientists having unrestricted access to information necessary to determine if the study was properly designed to achieve its stated goals, if the raw data or summary statistics are sufficiently described to permit other investigators to verify the study, and if the results and conclusions are supported by the data.

In the case of the Exxon Valdez oil spill, confidentiality compromised the quality and success of the restoration, because scientists and the public were denied access to the data upon which the restoration plans were based (Cummings 1992 p. 366). The annual restoration plans, submitted by the Exxon Valdez oil spill (EVOS) Trustee Council, triggered outcries from public interest groups, who repeatedly objected to the lack of detailed information in the plans. Public interest groups were justifiably concerned that confidentiality destroyed the scientific peer review process and that public monies were being used to subsidize government studies. These concerns were verified by a 1993 General Accounting Office (GAO) audit of EVOS trustee council expenditures (GAO 1993, Phillips 1993a, Rosen 1993).

The upshot was, that, when the Prince William Sound ecosystem exhibited signs of serious and widespread disruption in both 1992 and 1993 (see section 4, 5-year status, this paper), the EVOS Trustee Council could not make definitive statements as to the cause, because the restoration program lacked a comprehensive strategy to address long-term oil-related damages. When frustrated fishermen demanded that comprehensive research be initiated immediately (by blockading oil tanker traffic in Prince William Sound for three days, see McCoy 1993), the Trustee Council agreed to set aside $5 million for such a study–four years after the spill. After four years of public outcry and millions of dollars spent on questionable studies, a comprehensive research plan has still not been put in place: clearly, this points to major shortcomings with research sequestered from public scrunity!

Another quality control problem arose in 1993 when government and Exxon scientists presented contradictory findings of Exxon Valdez oil spill studies at separate symposiums in different parts of the country with everyone maintaining that their studies would be peer-reviewed later, before publication!

The existing structure for peer review has no mechanism in place to pre-screen public presentations. Instead, the scientific community relies on the equivalent of an honor system: there is an entrenched assumption that peer pressure and potential liability to one’s reputation will produce quality presentations.

But peer pressure is not always an adequate substitute for peer review as was clearly demonstrated by the dichotomy in Exxon Valdez oil spill studies. These glaring discrepancies in results of supposedly high quality scientific papers demean both science and the peer review process.

Further, the discrepancies confuse and frustrate the public, who has a right to know the full environmental impacts of industrial activities that damage public resources. Public frustration and disgust can translate into reduced funding for science.

Recommendation: Oil spill response, damage, and restoration research conducted with public funds should be peer-reviewed during design and prior to publication, and should be competitively bid to ensure high quality, cost-effective research. Possible mechanisms to make this process work are discussed in the next section.

All quantitative data should be required to be reported with any paper submitted for both public presentations (see recommendation 2).

The Role of Congress

Two steps could be taken by Congress to improve public access to and public understanding of scientific research in the aftermath of an oil spill.

(1) Require cooperative research during a spill response.

Background. Immediate spill response, cleanup, and longer-term research are regulated under the Clean Water Act(2). Restoration of natural resources, and a means to pay for it, are regulated by the natural resource damage assessment (NRDA) process under the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA)(3). Only the NRDA studies that provide the government with a monetary calculation for damages to natural resources are designed for litigation and subject to confidentiality restrictions. However, not all studies initiated for restoration are NRDA studies, and the ones that are not NRDA studies, at least in theory, do not have to be confidential.

Cooperation among the state and federal governments and the spiller is encouraged by both laws, but required in neither. In fact, the NRDA regulations specifically emphasize coordination of research to avoid duplication and to improve cost effectiveness. (Cummings 1992 p. 365, 368-372). One provision of the NRDA regulations states: "The Assessment Plan shall contain procedures and schedules for sharing data, split samples, and results of analyses, when requested, with any identified potentially responsible parties [spiller] and other natural resource trustees" (emphasis added) (supra note 51 in Cummings 1992 p. 371).

Despite these regulations, government and Exxon scientists performed their studies separately, but duplicated each others’ efforts often to the same sampling site (Houghton et al. 1991, 1993a, b & Page et al. 1993; Sharr 1989, 1990, 1991 & Maki et al. 1993). While limited information sharing between government and industry scientists is not atypical of "spill science" (because of liability and litigation), duplicate studies, with no basis for comparison, can increase research and litigation costs and confuse public understanding of spill impacts.

In response to the public outcry after the Exxon Valdez oil spill, Congress established an Interagency Coordinating Committee (ICC) to develop a program for cooperative, cost-effective oil pollution research and prevention and response technology development with federal agencies, industry, universities, research institutions, state governments and other nations (OPA ’90 ). However, despite the intent, this would not prevent the situation that developed after the Exxon Valdez oil spill because it does not require cooperative research after the spill.

Recommendation: Congress should require cooperative research during spill response and restoration. The laws governing oil spill response should include, in regulation, provisions for

A mandatory cooperative agreement for comprehensive research should encompass three key elements. First, the general framework for the comprehensive research plan should include both short- and long-term studies of key individual species (and related key prey organisms and predators), key community assemblages and chemical analyses (water, sediment, and mussel tissue). The comprehensive plan should integrate all prior research in the area, including baseline and management research, and it should be peer-reviewed.

Second, a list of "cooperative" studies that all parties will conduct (either jointly or separately) must be developed, along with a quality control program for these studies and a protocol for data reporting, including submission of quantitative data. Cooperative studies must include the chemical analyses and other studies that are central to the comprehensive plan (and litigation).

Third, parties must agree that presentation, publication and use in court of cooperative studies is contingent upon adherence to the established protocol. This agreement is critical, because it gives the public a basis for understanding the effects of a spill by providing a basis for comparing key studies that underpin the comprehensive research plan.

Without a requirement in federal laws for cooperative research during spill response and restoration, more attention will continue to be focused on accounting for the differences among spill studies than is spent discussing the implications for recovery of the ecosystem and future spills. This is the legacy to date of the Exxon Valdez. We need to learn from, rather than repeat, our mistakes.

(2) Require and fund cooperative baseline research.

Background: The federal laws governing spill research simply do not work for catastrophic spills with minimal baseline data, because of the disproportionately large number of damage assessment studies needed for litigation and classified as confidential. In situations such as the Exxon Valdez, effort becomes focused on "legal questions of whether data should or will be released, not on the more important issue of restoration of the environment" (Cummings 1992, p. 407). Once the focus on restoration is lost, things rapidly progress from bad to worse, as can be seen by reviewing events in Alaska (Appendix C).

In response to public outcry after the Exxon Valdez oil spill for "open" spill response and restoration science, Congress created a fund for oil spill response and damage assessment research, available on an open, competitive bid basis for research leading to publication (OPA90 Title VII). The concept was to enable governments to pursue NRDA studies for damage assessment purposes, while allowing scientists to pursue open research, free of the tort process, and to encourage long-term environmental monitoring studies, while providing the public with timely information on spill impacts.

However, the fund created by Congress is subject to restrictions imposed by the Graham-Rudmann zero sum spending. Agencies that are part of OPA90, Title VII, are authorized to spend money from the fund for oil spill effects research, but the funds have yet to be appropriated for this purpose, because any agency that receives funds from OPA90 has to offset the amount received from within its own budget. In effect, this means that oil spill effects research has to compete with all the other programs and projects within each agency listed in OPA90 Title VII. One type of research seldom wins out in these internal funding battles. So while the fund exists on paper, it does not exist in practice. This is not good enough.

The public must insist that politicians recognize the true environmental, social and economic costs of our dependence on fossil fuel.

Congress must recognize that the price for oil dependency must be to commit the effort and funds for baseline research necessary to understand the risk to natural and human resources. Further, Congress must ensure that the responsible agencies actually conduct a constant level of oil spill effects research, which must include baseline monitoring prior to approval of oil and gas lease sales.

Recommendation: Federal and state agencies should work cooperatively with the public as required by OPA90 to develop comprehensive baseline monitoring programs that incorporate the latest fate and effects research and include subsistence harvest information, where appropriate. To fund this research, Congress should require each agency listed in OPA90 to commit a percentage of its budget for oil spill effects research as defined in the Act. A comprehensive baseline monitoring program should be completed prior to approval of oil and gas lease sales on federal lands.

The Role of Federal & State Governments

Two steps could be taken by federal and state to improve public awareness and understanding of oil spill effects.

(1) State Trustee for Natural Resources

BACKGROUND: To parallel the national effort, the State of Alaska must also recognize the environmental, social and economic costs of its dependency on revenues from fossil fuels. These oil revenues contribute about 85% of the State's operating budget. The price for this dependency must be to commit the effort and funds for baseline research necessary to understand the risks from oil development to Alaska's natural resources and to protect the interests of those Alaskans at risk from oil development.

A relatively small number of local Alaskan residents in spill-impacted communities are paying the full costs for the Exxon Valdez spill, for Alaska's dependency on oil revenues, and for the nation's dependency on oil. Although Alaska depends on oil revenues, the state does not take extra precautions to safeguard the public and the environment against the risks posed by oil development. In fact, Alaska is the only state without a Trustee for Natural Resources!

Recommendation: Alaska must have a Trustee for Natural Resources immediately. Given the magnitude of both oil development and the resources at risk, the Trustee should be a dedicated position within the Alaska Department of Fish & Game. The State should fund the Trustee position from the general operating budget as part of the baseline management research required by ADF&G. The State Trustee should be a member of the Regional Response Team and actively develop contingency plans for baseline studies and damage assessments around the State in cooperation with federal trustees and the public. A comprehensive baseline monitoring program should be completed prior to approval of oil and gas lease sales on state lands.

(2) Peer Review of Response, Damage, & Restoration Studies

Background: If Congress acts on the recommendation for requiring a mandatory comprehensive research plan, then studies by state and federal agencies to assess damage to and recovery of natural resources will already be of sufficient quality to publish–but, that’s after the next spill. What about the problems that exist in the aftermath of the Exxon Valdez?

The EVOS Trustee Council set up an internal peer review process that has turned out to be counter-productive to the restoration process. The EVOS Trustee Council hired a person to be in charge of reviewing restoration studies prior to their release to the public. However, restoration studies are highly politicized, and have become even more so with evidence of long-term effects. Evidence of long-term effects from oil carries substantial political liabilities for federal and state governments and the oil industry in terms of increased public demands for spill prevention and response measures, restricted areas for oil and gas development, and increased insurance coverage. These political liabilities are not in the best interests of a nation or state so dependent on oil.

The restoration studies and their peer-review must be isolated from the political processes that would attempt to minimize or diffuse the importance of these long-term effects studies and their results. It is essential that peer review of the restoration studies be conducted by the scientific community, rather than by persons hired by political appointees.

The overall quality of the peer review process would also be improved by opening the process to the scientific community, which has the expertise necessary to critique the diverse and complex issues covered in the restoration research. It is unrealistic to expect one or a small number of reviewers with specific fields of knowledge to critique the range of issues addressed in the restoration research.

The EVOS Trustee peer review process is also an unnecessary expense of public funds ($500,000 per year). Further, it is slowing the release of information to the public, because there are only a small number of reviewers for a large number of studies.

Recommendation: State and federal agencies should place a higher priority on publishing oil spill research directly into scientific journals for maximum public circulation.

The EVOS Trustee internal review process should be restructured so that senior authors send their papers out to the scientific community for internal peer review before submission either as a Natural Resource Damage Assessment (NRDA) final report or to a journal for publication.