| 1 | SPECIAL MEETING TO ASSESS SCIENTIFIC ISSUES IN LAKEWIDE MANAGEMENT PLANS FOR CRITICAL POLLUTANTS |
The 1987 Protocol amending the 1978 Great Lakes Water Quality Agreement contains a renegotiated Annex II concerning Remedial Action Plans and Lakewide Management Plans. The International Joint Commission was given the responsibility of reviewing and commenting on plans that were submitted to the Commission. In the 1997-1999 Revised Priorities of the Commission concerning the work to be undertaken by the boards under the Great Lakes Water Quality Agreement, the Commission requested advice on the adequacy of the plans. The Great Lakes Science Advisory Board organized a Meeting to Assess Scientific Issues in Relation to Lakewide Management Plans; it was held on February 25 and 26, 1999. It was organized following the format in Annex 2 of the Agreement as follows: Definition of the Threat to Human Health; Definition of the Threat to Aquatic Life; Evaluation of Sources, Load Reductions and Pathways; and Surveillance and Monitoring to Track Effectiveness.
| 1.1 | Definition of the Threat to Human Health |
Research findings indicate that critical subpopulations continue to be exposed to persistent toxic substances (PTSs), particularly from fish consumption as the major pathway of exposure. Increased fish consumption is associated with body burdens of persistent toxic substances that are two to four times greater than in the general population. Many men and women consume Great Lakes fish during most of their reproductive years resulting in exposures to persistent toxic substances from the Great Lakes that cause disturbances in reproductive parameters (Courval et al. 1999) and demonstrate neurobehavioural and developmental deficits in newborns (Jacobson et al. 1985; Jacobson and Jacobson 1993, 1996; Lonky et al. 1996) and cancer (Moysich et al. 1999). The more highly exposed subpopulations include anglers, particularly among the urban poor, commercial fishermen, charter boat captains, Native Americans, and particular ethnic groups such as Hmong and Vietnamese immigrants. The more susceptible ones include pregnant women, fetuses and nursing infants, children and the elderly (Johnston et al. 1998; De Rosa et al. 1999).
The participants acknowledged that the preparation of a Lakewide Management Plan is a difficult task, and particularly as it relates to defining the threat to human health from critical pollutants. First, it was acknowledged that there is little political will (Great Lakes Water Quality Board 1993) to prepare such coherent statements because of liability concerns and because the costs of clean up are substantial. The fact is, the general population does not like paying higher taxes or higher prices and it is ultimately society, in general, that is the source of political pressure. But at the technical level, it was recognized by participants that there is a more subtle difficulty. The medical profession is organized mainly on the basis of responding to individuals who present themselves at the doctor's office with a disease or disorder to be treated. In contrast, exposure to pollutants tends to have effects on populations and particularly on the normative functioning of individuals within the population (International Joint Commission 1994). The loss of five or six I.Q. points in the most highly exposed individuals in a critical subpopulation cannot be diagnosed. Thus, there is a need to expand the definition of human health beyond the disease model and to include consideration of functional capacity. For example, a downward shift in the mean I.Q. of an exposed subpopulation could represent a loss of intellectual capacity with significant economic and social consequences. This is due in part to the fact that a small shift in the mean of a distribution can have a profound effect on those individuals in the tails of the distribution (Weiss 1997; Great Lakes Science Advisory Board 1997).
Community Health and Endocrine Disruptors
Epidemiology is a rational approach to studying the incidences and causes of diseases. However, it may not be a particularly feasible method for investigating dysfunction within a community. Not only may the statistical requirements be particularly stringent resulting in surveys requiring large sample sizes, but also estimates of exposures to pollutants can be unreliable or unavailable. There are factors other than environmental exposures that are determinants of human health, including: genetics, lifestyle, psychological factors, socio-economic factors and how individuals interact with the health care system. The teratogenic effects of environmental pollutants are unlikely to arouse the apprehensions of clinicians because they are both subtle and non specific, even though they are significant for the exposed subpopulations (Colborn and Clements,1992; Colborn et al. 1998, 1999; Johnson et al. 1998; De Rosa et al. 1999).
Health authorities in both countries have acknowledged the need for an official response to the perception within communities that they have been exposed to chemicals and that health has been affected. Generally, this is not a priority issue with local or regional levels of government compared with traditional public health concerns. There is a perception at the local community level, among those involved in the preparation of the Remedial Action Plans and the Lakewide Management Plans, that there is no mechanism for gathering the information needed by the physicians and health authorities at the local level, to document what is occurring as a result of exposures to contaminants. Much of the evidence is anecdotal and concerns reports of effects on children and is therefore delivered in an emotional manner without objectivity, and with great fear and trepidation and by people who are suffering. One central question is whether clusters of health effects indicate unusual exposures.
In the United States, the Agency for Toxic Substances and Disease Registry, which is part of the Public Health Service, may be called in by the state authorities to undertake a health and exposure survey. In Canada, the responsibility for health protection is a provincial responsibility. Health Canada has integrated existing information on the risks to human health, and as part of the Great Lakes Health Effects Program, compiled a handbook detailing a community-based approach to undertaking exposure assessments. The challenge is how to utilize these approaches to community health in preparing Lakewide Management Plans and Remedial Action Plans. For example one of the issues brought up by several LaMP managers was the problem of scale, and the question of whether human health questions could be addressed at the local, rather than the regional and national levels. Are the health and exposure issues in a community likely to be different from another community 50 or 100 km away? Much of the information available is collected at the national or regional level and does not give the detail required to assess health effects at the local level. Similarly, there is a need for outreach, by federal, state and provincial health authorities, to the communities communicating the significance of the information at the regional and local levels for preparation of the Lakewide Management Plans and the Remedial Action Plans.
Fish Advisories
For the past 30 years, the response to the finding of dangerous levels of pollutants in fish has been to close commercial fisheries and to publish fish advisories concerning the quantity and frequency of consumption of different sport fish species that could be eaten from a particular location. In 1993, the Council of Great Lakes Governors appointed a panel to establish uniform fish advisories that were to be both consistent among jurisdictions and protective of human health. Similarly, the Province of Ontario publishes an annual Guide to Eating Ontario Sport Fish. All of the Great Lakes waters in the U.S. have some kind of advisory. In 1996, the most recent year for which there are national data, the number of U.S. bodies of water with advisories rose by 453 to a total of 2,193, representing a 26 percent increase over 1995. In Canada, 2,617 advisories were in effect in 1996, and all resulted from contamination with one or more of five pollutants: mercury, PCBs, dioxins/furans, toxaphene and Mirex. While the various advisories generally contain useful information, there is no information about the effects of these pollutants on human health if larger amounts of fish are consumed. Surveys have been undertaken of the effectiveness of fish advisories as a means of protecting human health. In one study, awareness of the fish advisories was especially low among women, suggesting the need for improved communication of the information to female consumers (Tilden et al. 1997).
Based on the weight of evidence (Colborn and Clements 1992; Colborn et al. 1998, 1999; Johnson et al. 1998; De Rosa et al. 1999), the Great Lakes Sciences Advisory Board concludes that :
The Great Lakes Science Advisory Board recommends the following.
| 1.2 | Definition of The Threat to Aquatic Life |
Contrasting Approaches
The session on the threat to aquatic life was dominated by the contrasting approaches of the fish and wildlife researchers. While there have been demands that those responsible for the preparation of the LaMPs incorporate existing science into the documents, several factors have prevented this being done in a competent and timely manner, including the availability of qualified staff familiar with the scientific literature and the institutional arrangements. The most important factor, however, has been the ambiguity about whether the Lakewide Management Plans shall be designed to reduce loadings of critical pollutants or whether they shall embody a systematic and comprehensive ecosystem approach in restoring and protecting beneficial uses.
A small number of scientists have been investigating the effects of critical pollutants on fish and wildlife populations in the Great Lakes basin, during the past 40 years. In the 1960s and 1970s, they reported the gross effects of persistent toxic substances on the reproduction and development of wildlife and in ranch mink dependent on Great Lakes fish. Wildlife populations that have been affected by exposures to persistent toxic substances are generally non-game species, and are not managed as a resource. Consequently, the number interested is relatively small and their influence is limited. Effective action has traditionally depended on preparing a scientifically defensible case relating the injury to wildlife populations to a specific substance or class of substances and subsequently persuading the regulatory officials to take the necessary steps to prohibit the activities involving the substance. This approach has resulted in the successful restoration of many populations of wildlife species extirpated by exposures to persistent toxic substances, but not for certain sensitive species in areas, such as Lake Ontario, that remain heavily contaminated.
In contrast, a large number of scientists are involved in research, monitoring and management of fisheries populations in the Great Lakes. There have been extensive changes in species composition that are generally expressed, since the 1970s, in ecological terms. Management objectives are formulated in terms of losses and gains in biodiversity, relating to extinctions and introductions respectively, and to changes in ecosystem structure and function, sustainability and balance, and are based on the four major categories of threats to Great Lakes fisheries: introduced flora and fauna; habitat destruction; overfishing; and proliferation of fish diseases (Hartman 1988). In contrast to non-game wildlife, Great Lakes fisheries management is of interest to many parties. Fisheries managers perceive that the greatest challenge to planning and management in the Great Lakes basin is maintaining a sense of balance among the various ecological, social and political systems involved. Fisheries managers believe that the greater the number of people with an interest who get involved in the process, the greater the likelihood of success. Fisheries management in the Great Lakes is a mature consultative process.
The contrast between the toxicological approach of the wildlife biologists and the ecological approach of most fisheries researchers is reflected in the advice given to managers involved in the preparation of the Lakewide Management Plans, and in the interpretation of the Agreement implementation. The toxicological approach is aimed at identifying the specific substance or class of substances that caused the observed injury to the wildlife, whereas the ecological approach is directed toward describing the many factors that may be stressing the populations.
The issue of causality lies at the heart of the definitions for the Lakewide Management Plans. There are extensive case studies that have been prepared relating the observations of reproductive failure and deformities in wildlife to exposures to critical pollutants such as DDT and metabolites, PCBs, and dioxins. These case studies have been prepared using the criteria incorporated by the United States Environmental Protection Agency (U.S. EPA 1992) into the guidelines for ecological risk assessment.
The ecological interpretation of the Agreement is based on the premise that changes in Great Lakes fish populations resulted from the cumulative impacts of many stressors including the introduction of exotic species, habitat changes, eutrophication, overfishing, proliferation of fish diseases and selective stocking of sport fish. Fisheries ecologists have not been able to ascertain whether all the changes in the Great Lakes fisheries were the results of many factors acting in concert at the local level or whether some indigenous species in certain locations were made extinct through a single factor, or through a primary factor acting in concert with other stressors.
A Challenge to the Multi-Causal Explanation
In the past decade there has been a challenge to this multi-causal explanation of the changes in the fisheries populations in the Great Lakes. With evidence of the presence of high levels of 2,3,7,8-tetrachlorodibenzo-p-dioxin in Lake Ontario, in the early 1980s, and the availability of an established methodology for risk assessment, researchers studied several lines of evidence to undertake a retrospective risk assessment of the effects of dioxin on Lake Ontario lake trout. The Lakewide Management Plan, prepared concerning Lake Ontario, concluded that the "contribution of persistent toxic substances to the loss of certain fisheries is unclear because fish populations were already severely degraded by the time that significant levels of contaminants began to be released to the environment." If Cook et al. (1997) have interpreted the evidence correctly, then this LaMP conclusion warrants reassessment.
The data on the harvesting of lake trout from Lake Ontario reveal that there was a decline in the status of the population that started about 1880. Through the stocking of hatchery-reared fry, there was a large population increase in the early 1900s, and a maximum harvest of nearly five hundred thousand kilograms in 1928. These data indicate several important factors in the exploitation of the lake trout stocks in Lake Ontario.
The probable concentrations of PCDDs, PCDFs and PCBs in the Lake Ontario lake trout eggs have been reconstructed retrospectively back to 1920. The results were used with the accumulation factors from sediment to biota adjusted for historical trends in contaminant distribution between water and sediment. There was a sharp rise in the probable concentration starting as early as the 1930s and continuing to a peak in the late 1960s. Toxicological evidence from experimental exposure of eggs of lake trout to 2,3,7,8-tetrachlorodibenzo-p-dioxin has, surprisingly, shown that the early life stages of this species are extremely sensitive to this compound. The decline in Lake Ontario lake trout populations after 1940 was inversely correlated with the increasing exposures to TCDD toxicity. The toxicity of TCDD and related chemicals in Lake Ontario was estimated to have completely eliminated the capability for lake trout reproduction and recruitment by 1950 (Cook et al. 1997).
The demise of the Lake Ontario lake trout poses the question whether other species were extirpated from Lake Ontario by this same cause. For example, the deep-water sculpin, which was the natural prey of the lake trout in the profundal area was apparently extirpated by the 1960s, and the populations of burbot, lake herring and lake whitefish practically disappeared. The information assembled on these contaminants provides a hypothesis to apply to historical data on other populations of other fish species in other Great Lakes that were extirpated during this century.
Undoubtedly, other stressors, such as the continuation of a commercial fishery, together with the depredations of the sea lamprey may have contributed to the declines in the Lake Ontario lake trout population. Similarly, other stressors such as alewife predation on fry, may contribute to the present difficulty in reestablishing naturally reproducing populations of Lake Ontario lake trout. The important corollary for research managers arising from the conclusion of Cook et al. (1997) is whether other stressors acting in concert at the local level could have brought about the extinction of the stocks of Lake Ontario lake trout. This toxicological hypothesis (Cook et al. 1997) strongly suggests that the TCDD toxicity alone would have caused the extinction of this Lake Ontario species, as was observed in the early 1960s. Similarly, this toxicological hypothesis, which closely conforms to the classes of evidence relating exposures to persistent toxic substances and effects in wildlife populations, challenges multi-causal ecological assumptions and statements, that are now being used as the basis for the preparation of the Lakewide Management Plans.
Differing Interpretations of the Great Lakes Water Quality Agreement
The purpose of the 1978 Agreement "to restore and maintain the chemical, physical, and biological integrity of the waters of the Great Lakes Basin Ecosystem" by eliminating or reducing the discharge of pollutants, was ascertained by Oakley in the Review of the Changes in the 1978 Great Lakes Water Quality Agreement (1979) to be an agreement on water quality, and not to be an agreement on ecosystem management. In 1985, the National Research Council of the United States and the Royal Society of Canada viewed the Agreement as An Evolving Instrument of Ecosystem Management (NRC and RSC 1985). Session participants also held differing views as to the intent and implementation of the Agreement. Some maintained that the Agreement is an instrument of ecosystem management and provides an appropriate and effective mechanism for accommodating a multitude of issues associated with the restoration of the integrity of the waters of the Great Lakes Basin Ecosystem. Others argued that its intent is focused more specifically on critical pollutants, and that a broader focus can limit progress by reducing resources available to address specific initiatives such as LaMPs. In this latter view the Agreement was never intended to be an instrument of ecosystem management; therefore it has inherent limitations, both in terms of its scope and its institutional capabilities, to achieve a broader purpose. Similar differences of opinion arose among the meeting participants as to whether the primary goal of a LaMP is to restore beneficial uses or to reduce loadings of critical pollutants.
The SAB recommends the following.
The SAB recognizes that there are certain resource management issues that relate directly to water quality, such as the restoration of fish populations extirpated by releases of PTSs (Cook et al. 1997), or the impact of agricultural practices on water quality (Great Lakes Commission 1996). At the same time, fisheries management activities take into account many factors that some consider to have a less direct relationship to the Agreement per se, but which might benefit from coordinated action or discussion, for example, fish stocking. Similarly, other Great Lakes institutions, such as the Great Lakes Commission, conduct a broad range of program and management reviews that could provide a basis for greater Commissioner, or staff, interaction with the IJC.
The SAB recommends the following.
| 1.3 | Evaluation of Sources, Loads Reductions and Pathways. |
Overview
The panel on sources, load reductions and pathways included experts on the utilization of existing data, on the generation of new data from surveys, and on modeling. Its comments focused on scientific challenges that they considered crucial to reducing loadings of critical pollutants in the Lakewide Management Plans. The topics discussed included: the role of zebra mussels on predictions of fish burdens; the availability of atmospheric monitoring data; access to industrial emissions data; the relative importance of contaminated riverine sediments; and the role that the LaMP process can play in improving the scientific understanding of toxic cycling in the lakes. Several important conclusions appeared to reflect consensus. Furthermore, many of the conclusions reached showed enthusiasm for the LaMP process and the panelists' optimism about the LaMPs' potential for success in achieving progress. The recommendations summarized below focus on specific actions that need to occur to ensure this success.
Scientific Issues that the LaMPs Should Consider
All the panelists recognized the effort involved in the development and implementation of the Lakewide Management Plans. While difficult, the identification of sources and decisions about actions for reducing loads are not without precedent. Their discussion included many examples of issues that had been previously considered or examined through scientific efforts.
Because most of the panelists are not involved directly in any of the LaMPs, the discussion focused on matters of science rather than planning issues. The panel was concerned that their comments not be construed as criticism, but rather as suggestions and scientific advice from experts knowledgeable about toxics behaviour in the lakes. Four issues are highlighted from the discussion as follows.
The SAB recommends the following.
| 1.4 | Surveillance and Monitoring to Track Effectiveness |
In the expert panel on Surveillance and Monitoring, eight experts presented their views on the relevant scientific issues. The presentations covered a wide range of topics including the design and intended uses of data from surveillance and monitoring programs. Several key themes emerged as follows.
Implications of Fiscal Restraint on Surveillance and Monitoring Programs
Many speakers made reference to the serious budget cuts that have occurred in Great Lakes programs over the past decade. For many agencies, these cuts have forced a fundamental reconfiguration of surveillance and monitoring programs. Sometimes, this reconfiguration has been reflected in reduced monitoring frequency or spatial coverage. In other programs, there has been a conscious decision to move away from large-scale lakewide programs toward local studies that can provide more detailed and accurate loading estimates, but which fail to provide a regional or lakewide perspective. Long-term programs focusing on key indicator species have received less funding and attention in recent years, to the detriment of our being able to track temporal and spatial trends.
Several speakers emphasized the point that the lakes themselves are dynamic systems, completely apart from actions that may change pollutant loadings to them. Dramatic food web changes have already been documented in some systems, and these changes are thought to be the result of complex phenomena, probably including, but not limited to pollutant loading changes. There is a need for ongoing adaptation of surveillance and monitoring programs so that they are responsive to the dynamic nature of the lakes.
Data Used for Many Purposes
It is apparent that under the prevailing climate of fiscal restraint, few programs have the resources to collect data for a single purpose. Rather, most agencies are collecting data that may be used for many purposes, some of them not intended when the program was designed. Most LaMPs must therefore patch together data from a variety of sources, and of varying ages and analytical methodologies, and try to interpret them in a lakewide context. This has been a challenging task, especially since LaMPs are required to track the effectiveness of remedial measures and the impacts on beneficial uses of critical pollutants. To complicate interpretation, much of the available data is not in a form that is readily understood by most LaMP participants. As a result, data may be misinterpreted or simply ignored as unusable. Many scientists who spoke at the session, both those on the panel and those in the audience, expressed concern that LaMPs may contain outdated or incomplete information, and where that occurs may convey an inaccurate or misleading message. They believe that the LaMPs are significantly weakened by the omission of current and valuable, if highly technical, data.
Those participants with LaMP planning responsibilities had a different perspective on this problem. They agreed that the LaMP analysis may be less complete and/or less accurate than desirable, if critical information is omitted. They stated, however, that their frustration arises in trying to make sense of scientific data without the close involvement of the scientific community. They spoke of their attempts to piece together a mosaic of data, little of it collected for the LaMP effort, and of trying to make sense of information that is contradictory, confusing or simply unintelligible.
Surveillance and Monitoring as Part of a Larger LaMP Process
Several speakers made reference to the complex interactions between a LaMP, the scientific community and regulatory agencies. Some of the dissatisfaction that was expressed about the LaMP process may in fact be a result of confusion about an appropriate division of responsibilities and an appropriate mechanism for consultation among the various players legitimately involved in LaMP-related activities.
Three key roles emerged from the workshop discussion.
The SAB recommends the following.