Stakeholders, Risk from Mercury, and the Savannah River Site: Iterative and Inclusive Solutions to Deal with Risk from Fish Consumption

  • Joanna BurgerEmail author


Many states issue consumption advisories to provide information, mainly to anglers, on the risk from eating fish from contaminated water bodies. The Savannah River passes between South Carolina and Georgia, yet, in 1999, the state-issued consumption advisories for self-caught fish were not in agreement. This chapter examines a stakeholder-driven process that involved state and federal regulators, wildlife biologists, Center for Disease Control, Department of Energy (DOE), fishers themselves, and others to reduce risk for people eating self-caught fish from the river adjacent to the Savannah River Site (a DOE facility). The process included problem formulation, stakeholder identification, identification of the scientific data needed to answer the key questions, development of studies to address these questions, refinement based on stakeholder collaboration, and then development of a mechanism to advise potentially affected persons of the risk. In sum, data on fishing behavior, consumption patterns and mercury levels in fish indicated that people who ate fish frequently were at risk from excess mercury exposure from eating some fish, and an information brochure embraced by the several regulatory agencies and jurisdictions was developed that specifically addressed these issues for people fishing in the Savannah River. This solution sidestepped competing jurisdictional issues between the two states and allowed all parties to create a Fish Fact Sheet brochure that could be distributed annually to those fishing along the Savannah River.


Environmental Protection Agency Fish Consumption Mercury Level Largemouth Bass Mercury Exposure 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

5.1 Introduction

The threat to humans and ecosystems from anthropogenic contaminants is a complex and persistent by-product of our industrialized and urbanized society. Among the several environmental exposures that remain from the Cold War Legacy are stockpiles of radionuclide and chemical wastes, contaminated buildings, and contamination of soils, sediment, and rivers. Mercury can remain in these media for decades and centuries. For mercury, the key feature is that inorganic mercury can be converted by anaerobic bacteria to methylmercury, can bioaccumulate in organisms, and can then be amplified up the food chain to top-level predators, such as sharks, eagles, wolves, lions, and people (Downs et al. 1998). Since mercury is a neurotoxin that can adversely affect eggs, chicks, and young animals, as well as developing fetuses and children, it is important to reduce mercury exposure in biota (including humans) and ecosystems. Excessive mercury exposure from consuming fish high in mercury can even adversely affect adults (Hightower and Moore 2003). In this chapter, all analytic results refer to total mercury, of which on average about 90% would be methylmercury (Jewett et al. 2003).

Contamination can be reduced in the environment by reducing or eliminating the source of the contamination, and by cleaning up sites that are already contaminated by released contaminants. However, sometimes it is not possible to remove the source, or to clean up the contamination, and immediate responses to address the risks are necessary. In some cases, the short-term solution is to manage the exposure through education or intervention or by physically blocking exposure pathways (such as erecting fences). In all cases, sufficient scientific information is necessary to document the extent of the potential exposure, locate exposure pathways, and thus define the potential risk to humans and the environment from the exposure. Obtaining such data is often difficult, time-consuming, and expensive, but the information is needed nonetheless.

Many sites, especially streams, rivers, and lakes, are contaminated with mercury (EPA 2009a) from both local and distant sources. While the method of eliminating the risks from mercury should be source reduction and mercury removal in ecosystems, interim early response risk reduction strategies for humans and the environment are essential tools needed by conscientious public policy makers, governmental agencies, health professionals, and the public at large. In many cases, however, the extent of mercury contamination on- and off-site is unknown: both the potential exposure of consumers (including humans) and the potential risk to consumers are unknown, making it difficult to devise a reasonable strategy for reducing risk to humans and the environment. Moving forward with a risk management strategy requires defining the problem, obtaining site-specific information on contaminants and exposure pathways, determining exposure information, examining risk, and devising a strategy for reducing risk to the public (in this case, from fish consumption). Optimally, the process should be iterative, interactive, inclusive, and collaborative with as many of the interested and affected people, organizations and agencies as possible.

5.1.1 Objectives for This Chapter

This chapter addresses the issues posed by mercury contamination in a riverine system adjacent to a U. S. Department of Energy (DOE) site. At the start of the research, both the extent of the contamination in fish and the extent of human consumption of fish were unknown. Thus the potential risk was unknown. And if there was a risk, both the source and possible risk-reduction solutions were unclear.

In this chapter, I describe the process of examining and managing the potential risk from mercury to people who eat fish from the Savannah River that flows adjacent to the DOE’s Savannah River Site (SRS), including (1) Defining the problem, (2) Identifying the stakeholders, (3) Deciding what information was necessary to examine the potential risk from fish consumption, (4) Obtaining data to determine the risk, (5) Evaluating the data, (6) Determining a path forward to reduce risk to people who eat fish, and (7) Evaluating whether the strategy was effective in informing the public and reducing potential risk (Fig. 5.1). In this chapter the popularly used, gender-neutral term “fishers” is used to refer to fishermen, fisherwomen, fisher children, and their families. Throughout this whole process, a range of stakeholders was involved, and the major focus of the chapter is examining the role of stakeholders in improving the science that led to a reasonable solution to reduce exposure from fish consumption.
Fig. 5.1

Overall process for developing a path forward to resolve discrepancies and conflicts over the potential risk from mercury in fish consumed from the Savannah River (flowing between South Carolina and Georgia)

The main environmental problem addressed in this chapter is thus how to identify and deal with mercury levels in fish that might pose a health threat to people, and the focus is on how to integrate and collaborate with a full range of stakeholders to obtain sufficient data to evaluate the risk, and to devise methods to reduce risk. Stakeholders were integral to all phases, and greatly improved both the science and the management of the issue. I use the term “stakeholders” broadly to include any individuals or organizations that are interested in mercury in fish in the Savannah River (see Table 5.1).
Table 5.1

Key stakeholders involved in examining risks to people from mercury in fish in the Savannah River

Agency or group

Role and interests

Fisherfolk and people who eat the fish they catch

People who fish along the Savannah River eat their own fish, distribute it to friends and neighbors, and give it to churches and others for fish fries. Their interests are in knowing what fish are safe to eat, which fish to avoid, and how to optimize fish consumption with risk reduction

General public

People who occasionally come to the Savannah River to fish or recreate, and people who are interested in biota and ecosystem health. Interested in understanding contaminant levels in fish

U.S. Department of Energy (DOE)

Landowner of the Savannah River Site (SRS) that runs adjacent to the Savannah River, and is responsible for some mercury contamination in Steel Creek (on site) and in the river. Interested in reducing risk to fish consumers

U.S. Environmental Protection Agency (EPA)

Federal agency responsible for human health and the environment, and that has some jurisdiction for waters jointly held by two states. Interested in risk reduction and in having uniform fish consumption advisories, as well as having oversight responsibility for CERCL cleanup on these lands

Citizen’s Advisory Board (for DOE) (CAB)

DOE-appointed body to provide advice on environmental issues and cleanup at the SRS. Interested in monitoring DOE’s activities and providing advice to them about issues such as mercury in fish

Savannah River Ecology Laboratory (University of Georgia, on SRS) (SREL)

Ecology laboratory of the University of Georgia with facilities adjacent to the site, and largely funded by DOE at the time. Interested in sound science and interactions with both DOE and the public about this issue

Centers for Disease Control (CDC)

U.S. federal body that monitors diseases and had conducted limited studies of mercury contamination on the river

South Carolina Department of Health and Environmental Control (SCDHEC)

State agency responsible for issuing consumption advisories (for South Carolina residents). Interested in mercury levels in fish from both an ecological and a human health risk perspective. Also interested in efficacy of fish consumption advisories and in materials for the public

Georgia Department of Natural Resources (GDNR)

State agency responsible for issuing consumption advisories (for Georgia residents). Interested in mercury levels in fish from both an ecological and a human health risk perspective. Also interested in efficacy of fish consumption advisories and in materials for the public

Consortium for Risk Evaluation with Stakeholder Participation (CRESP)

DOE-funded, independent research entity that conducts original research, reviews documents and reports, and involves stakeholders in research and management processes

The Savannah River flows between South Carolina and Georgia

Briefly, consumption information was needed from the stakeholders themselves to determine whether there was a risk to them, and it was necessary to build credibility with these same stakeholders (people possibly at risk) so that when the data and its interpretation were available and had been translated into a Fish Fact statement, the stakeholders (fishers) would be sufficiently trustful to follow the Fish Fact Sheet advise once they understood it. I will show that the desired risk reduction could likely not have been obtained without the stakeholder involvement in the data development process. Thus, the process of securing the information from them (and for them) was at the same time a preparatory step for building the kind of relationship with them and their consciousness that led to acceptance of the Fish Fact Sheet. When data were then turned into advice that depended on shaping their behavior, the groundwork for their willingness to do so had been laid. This case study shows that stakeholder inclusion in the data development process lays the foundation for successful risk management.

5.1.2 Mercury in the Environment

Mercury enters the environment from both natural and anthropogenic (human-generated) sources, and it is the latter that are of greatest concern because of both historic and recent changes in mercury levels. The increasing demand for electric power generation that releases mercury into the atmosphere from burning coal results in both local and regional pollution through atmospheric transport and deposition. In addition, mercury exposure can come from point-sources that release mercury into streams and rivers.

Mercury occurs naturally in seawater, and coastal waters receive mercury runoff from land, input from rivers, and airborne deposition. Biomethylation of mercury occurs in sediment, allowing for food chain biomagnifications (Montiero et al. 1996; Downs et al. 1998). Mercury in fish tissue may be six orders of magnitude higher than the mercury concentration in the water column (Scudder et al. 2009). Anthropogenic sources of mercury account for about 80% of the annual inputs of mercury to the environment. For many areas of the world, atmospheric deposition, both regional and global, is the primary source of mercury (Fitzgerald and Mason 1996; Driscoll et al. 2006). The global contribution of mercury to the atmosphere is unevenly distributed. The Asian countries contribute about 54% of the mercury to total atmospheric sources. China alone contributes 28% to the total emissions, followed by Africa (18%) and Europe (11%, Pacyna et al. 2006). These sources are primarily from electric power generation, although regionally mercury inputs can be due to other industrial processes, as well as from gold mining extraction (Mueezzinoglu 2003). Hospitals have traditionally contributed to local mercury sources (SFWMD 2007 and their earlier reports).

Mercury is a persistent toxicant that bioaccumulates in animals (including humans; Nichols 2001), making it critical to identify the sources of mercury, evaluate exposure to humans and the environment, and determine the risk it poses to humans and the environment. Armed with such knowledge, managers and regulators can lower mercury emissions to a level that reduces adverse effects on individuals and populations, and other risk managers (including health professionals) can develop strategies to reduce mercury exposure. For example, in the Everglades of south Florida, knowledge of the high levels of mercury in predatory fish led to enacting controls on emissions by local power plants and other industries, which ultimately led to a drastic reduction in the mercury levels in the fish eaten by birds and mammals, including people (Davis and Ogden 1994; Lange et al. 1994; SFWMD 2007). Even in the Everglades, however, it was necessary to issue fish consumption advisories to reduce human exposure to mercury.

5.2 Background on Fish Advisories

5.2.1 Fish Consumption Advisories

In the absence of an effective cleanup strategy, or way to rapidly reduce mercury in biota (such as fish) and in ecosystems to immediate risk issues, fish consumption advisories are often issued. To be sure, there is a general distrust of governmental agencies, even in other countries (Wentholt et al. 2009), and thus fishing advisories are often ignored (Burger and Gochfeld 2006). Still, although mitigation of the mercury contamination and its source is clearly preferable for many reasons, fish advisories and their dependence on getting affected parties to change their normal patterns of behavior are sometimes the only effective near-term risk management alternative. This approach to risk reduction has shifted the burden from pollution reduction to personal behavior (Jakus et al. 1997; Halkier 1999).

In general, states are responsible for issuing contaminant advisories to the public, should they become necessary, although the Environmental Protection Agency (EPA) and the Federal Food and Drug Administration (FDA) can issue them. The U.S. EPA posts a “National Listing of Fish Advisories” every 2 years (EPA 2009a), and its most recent edition notes that all 50 states, the District of Columbia, American Samoa and Guam (U.S. territories), and five Native American Tribes have fish consumption advisories for some local waters. Fish advisories are not regulations, but are recommendations to help protect human health. Fish advisories are developed for the general population, and for sensitive populations (such as pregnant women and children). Approximately 18 million lake acres (43% of U.S. lakes) and 1.4 million river miles (39% of U.S. total river miles) were under advisory in 2008 (EPA 2009a), so the problem is by no means local only to the Savannah River. States may be hindered in their issuance of fish advisories by lack of information on either consumption rates or contaminant levels in fish, organizational problems, or by jurisdictional issues when two or more states share the same waters (Chess et al. 2005).

All 50 states have advisories for mercury, and 80% of all fish consumption advisories were issued in part because of mercury. Mercury accounts for the greatest number of advisories for lakes (about 17,000 acres) and rivers (over 1.3 million miles, EPA 2009a). Mercury thus remains one of our most important environmental issues, both from point-source and from atmospheric sources.

As noted, compliance with fish consumption advisories is sometimes low, leading to questions about the efficacy of such advisories as a public health policy (Connelly and Knuth 1998; Burger 2000; Jardine 2003). However, Hispanic fishermen from Newark Bay (New Jersey) showed a willingness to change their consumption behavior when presented with clear risk information (Burger et al. 1999a, b; Pflugh et al. 1999), and others have shown a decline in fish consumption among pregnant women following a federal mercury advisory issued in January 2001 (Oken et al. 2003). People have to know about advisories to follow them, and often people are not aware of them (Burger 2005; Burger and Gochfeld 2006). And agencies need to know whether the advisories are effective and result in changes in behavior.

5.2.2 Why Fish Advisories are Necessary: The Role of Fish in Diets

Fishing provides a nutritious source of protein and is very popular in both urban and rural areas of the United States, and elsewhere in the World (Toth and Brown 1997; Burger et al. 1992, 1993, 2001b, c; Ramos and Crain 2001). Fishing not only provides fish and shellfish to eat, but it also provides a range of social benefits that include interactions with family and friends, allows people to get away from the stresses of life, and provides opportunities for people to commune with nature (Fleming et al. 1995; Toth and Brown 1997; Burger 2002; Burger and Gochfeld 2006). Increasing attention to health and nutrition in the media has increased the public’s consumption of fish, even among those who never fish themselves (NOAA 2004).

In the United States, there has been a general upward trend in seafood consumption since the 1960s, despite an increase in price and warnings about contaminants. The trend has waxed with nutrition advice and waned with hazard advisories (FOA 1998), but continues to increase, now exceeding 7.4 kg/capita per year (NOAA 2004). Increases in contaminant levels in fish have led to the necessity to issue consumption advisories to deal with the risk.

Fish are a healthy source of protein, provide omega-3 (n-3) fatty acids that are generally accepted to reduce cholesterol levels, and reduce the incidence of heart disease, stroke, and preterm delivery (Anderson and Wiener 1995; Patterson 2002; Albert et al. 2002). Iribarren et al. (2004) showed a positive relationship between consumption of fish with high n-3 fatty acids and a lower likelihood of high hostility in young adults.

However, contaminant levels in some fish are sufficiently high to potentially cause adverse human health effects, making it necessary to consider both the risks and benefits of fish consumption. (Gochfeld and Burger 2005) Adverse effects from contaminants in fish include counteracting the cardioprotective effects (Guallar et al. 2002), damaging unborn babies and young children (IOM 1991; Neuringer et al. 1994; ATSDR 1996; Iso and Rexrode 2001; Olsen and Secher 2002; Moya 2004), and adversely affecting adult behavior and physiology (Hightower and Moore 2003; Hites et al. 2004). There is a positive relationship between mercury and polychlorinated biphenyl (PCB) levels in fish, fish consumption by pregnant women, and deficits in neurobehavioral development in children (IOM 1991; Sparks and Shepherd 1994; Schantz 1996; NRC 2000; Schantz et al. 2003). There is a decline in fecundity in women who consume large quantities of contaminated fish from Lake Ontario (Buck et al. 2000). There is also a suggestion that mercury affects blood pressure (Vupputuri et al. 2005). Generally, there is a positive relationship between mercury levels in people and fish consumption (Knobeloch et al. 2005; Johnsson et al. 2005). These largely epidemiology studies have demonstrated a relationship between mercury levels in fish they eat and mercury levels in people, and between mercury levels in human tissue and adverse effects. And although these relationships are correlational, they are persuasive in identifying the need to issue consumption advisories.

The extensive discussion about what the “safe” level of exposure is may be partly political; it is surely controversial, and will continue to be so for some time (Stern 1993; NRC 2000; Stern et al. 2004). Even the role of occasional peak exposures vs. chronic lower level exposures to methylmercury requires closer attention, and, depending on how the science of that question evolves in the near term, it may also be essential to develop single-meal fish consumption advisories for fish species high in methylmercury (Ginsberg and Toal 2000).

The bottom line, however, is that fish are an important and nutritious form of protein, and people should not stop eating fish. However, they should be aware of the relative benefits (in terms of omega-3 levels) and risks (contaminant levels) and optimize their eating patterns to reduce the risks while increasing the benefits I have cited. Obviously, they cannot do so without understandable, credible, and persuasive information.

5.3 Mercury, Fish Consumption and Stakeholders from the Savannah River

The general problem faced in this example is how to deal with mercury in fish in the Savannah River, and the solution involved a multistep process (Fig. 5.1). Each of the major steps are described below, with further refinements of the process and how stakeholders were involved. That mercury in fish from the Savannah River might pose a human health risk had previously been identified by all the state and federal agencies with responsibility for human health.

The controversy surrounding the issue involved the DOE’s assertion that no one fished on the site itself (because of access restrictions and inability to get on site to fish in Steel Creek) and that no one fished adjacent to the SRS. Thus, the Department contended there was no risk to the fishing public. The state agencies and the federal Environmental Protection Agency were unsure whether people fished along the river at the edge of SRS, whether there was indeed any specific risk from the fish that resided in the waters along the SRS, and whether the resultant fishing/consumption patterns and mercury levels in fish from this stretch of the river posed a human health risk.

5.3.1 Background on Contamination and Conflicts Site Description

The Savannah River originates in the southern Appalachians of North Carolina, passes through South Carolina and Georgia, and flows into the Atlantic Ocean. It winds through several large reservoirs, past various industrial sites (including chemical facilities), nuclear power plants, and the SRS. The original mercury contamination of the Savannah River came from chemical plants upstream from SRS.

The SRS (33.1°N, 81.3°W) is a 780-km2 former nuclear weapons production and current research facility operated by the DOE since the early 1950s (Fig. 5.2). The DOE used the river as a source of cooling water for nuclear reactors, and water was discharged to artificial thermal cooling reservoirs. Streams from the SRS flow directly into the Savannah River, and fish can move freely between the on-site streams and the river (Workman and McLeod 1990). Although there is controversy about the upstream sources of mercury to the Savannah River, the on-site SRS activities themselves also resulted in contamination by heavy metals (including mercury) and radionuclides, and there is some discharge of mercury from the coal-burning power plant on DOE site (Kvartek et al. 1994; Sugg et al. 1995). Further, the use of water for cooling the reactors redistributed the mercury on SRS lands, and in the marshes and streams on site, and off-site. These sources could result in higher mercury levels in fish adjacent to the SRS. Atmospheric deposition of mercury is relatively high in this region (>10 μg/m2/yr, EPA 1980; Downs et al. 1998). If mercury exposure was coming from the SRS, then mercury levels should be lower above the site than along or below it (and they generally were not). Regardless of the source of mercury, DOE is mindful of its economic and social role in the region (Greenberg et al. 1998).
Fig. 5.2

Map of the Savannah River showing the location of the Department of Energy’s Savannah River site Consumption Advisories for the Savannah River

At the time (1999), the South Carolina Department of Health and Environmental Control (SCDHEC) had issued fish consumption advisories that included the Savannah River (SCDHEC 1999). While Georgia did not issue an advisory for its side of the Savannah River, it did issue “Guidelines for eating fish from Georgia waters” (GDNR 1999). Thus, important discrepancies existed in the advice given by their states to fishers and consumers. Since EPA, through its regional office in Atlanta, Georgia, provides CERCLA (Comprehensive Environmental Response Compensation and Liability Act) oversight for all federal facilities, it had some interest in the mercury contamination in fish issue. EPA wanted each state to maintain its jurisdictional primacy with respect to human health, while insisting that a tangible form of risk communication resulted. EPA was at the time particularly attentive to its mandate (Federal Order 12898) to address environmental justice issues (EPA 2002, 2009b; DOE 2009).

There was resistance to the need for fish advisories (by Georgia), confusion about the likely pattern of public response to the consumption advisories, and important differences of opinion among the several public authorities as to whether there was significant fishing in the Savannah River along the SRS. The Consortium for Risk Evaluation with Stakeholder Participation (CRESP) was encouraged by DOE (and then also volunteered) to help resolve the issues by developing and providing new data to address these conflicts, and by facilitating discussions and collaborations among all parties.

5.3.2 Identification of Stakeholders

One of the most important steps to formulating issues and solving environmental problems is the level and timing of involvement of stakeholders. A mantra among most public agencies is “early and often,” a mantra more often stated than enacted. While there were a number of obvious stakeholders (fisherfolk, state agencies responsible for issuing advisories) in situations such as this, there is a fuller range of stakeholders that should be included (Table 5.1). Additionally, although knowledge and expertise may vary among stakeholders, they all have something worthwhile to contribute to the discussion. In particular, often long-time residents, fisherfolk, and others familiar with the site will have information not available in the literature.

Sufficient time and attention needs to be devoted to identifying the relevant stakeholders, and for the SRS, this process involved talking to the relevant federal and state agencies, attending meetings (particularly of the Citizens’ Advisory Board (CAB)), and talking with the fishers themselves. That is, CRESP talked to the regulatory agencies (GDNR, SCDHEC, EPA), as well as DOE, and they each suggested others (such as CDC, CAB) that should also be included. Then the CDC and CAB were asked who else should be included. This resulted in a list of the key stakeholders that needed to be involved throughout the process. In the case of mercury in fish from the Savannah River, the key stakeholders involved federal and state agencies, advisory boards, a scientific research laboratory, and scientists (from the CRESP), as well as the people fishing and consuming the fish. Further, information was put on the CRESP web site, making it possible for the general public to access information and provide comments to CRESP (

5.3.3 Formulating the Problem

Formulating the problem and identification of stakeholders is an iterative process that involves initial problem formulation, followed by meetings with stakeholders, and then a reformulation of the problem in collaboration with the key stakeholders. Although the individuals (or agencies) who initiated the process may initially define the problem, once it is identified, it is critical to involve all stakeholders to refine the problem, the hypotheses to be tested, the temporal and spatial features of the issue, and the possible methodologies to employ. Problem formulation should involve both determination of assessment endpoints (what is important) and measurement end-points (what can be measured to reflect the important questions, Norton et al. 1992).

The general problem faced by DOE and the other stakeholders was whether mercury levels in fish actually did pose a risk to humans (and to other biota) who consumed them at the levels they were, in fact, consuming the fish. The state and federal agencies did not think there were very many people who fished above, along, or just below the SRS (Fig. 5.3). Thus, the problem identified was to determine whether any people were at risk from eating fish from the Savannah River, which involved examining both fishing and consumption patterns, and contaminants in fish.
Fig. 5.3

Photographs of fishermen along the Savannah River who were interviewed for the study. Another factor necessary to consider is that some fishermen give inferior fish (often with higher mercury levels) to others

While there was general agreement on the problem itself, there were priority and other nuanced differences among the diverse different stakeholders. In particular, the primary questions of concern to the several kinds of stakeholders differed (Table 5.2). The regulatory agencies were primarily interested in whether they needed to modify their consumption advisories, and if so, how. EPA, on the one hand, was interested in having uniformity between the states, but on the other hand respected the two states’ rights to develop the advisories. Additionally, EPA was also responsible for overseeing cleanup on SRS, and both EPA and DOE were interested in arriving at an agreeable solution to deal with the mercury contamination in fish that would accommodate the different interests and perspectives of the two adjoining states. Of course, one potential outcome was a requirement for DOE to conduct further remediation to remove mercury from on-site source (that could move off-site via streams and creeks).
Table 5.2

The problem as seen through the eyes of different stakeholders, including self-interest and management goals


Problem formulation and questions of interest

Fisherfolk and the General Public

What fish are safe to eat, and in what quantities?

Department of Energy

Are remediation actions necessary to prevent mercury contamination in the fish or other biota?

Citizens’ Advisory Board

What advice should we give to DOE regarding cleanup or risk from mercury in fish (because of human exposure through consumption)?

Environmental Protection Agency

Should we intervene in the issuance of fish consumption advisories? How can advisories be similar for the two states? What are the implications for cleanup at SRS?

Savannah River Ecology Laboratory

What levels of mercury are there in fish, and what is the import to functioning ecosystems?

Center for Disease Control

What is the extent of exposure that could prove harmful to human health?

South Carolina Department of Health and Environmental Control

Are there any regulatory controls or actions required? Are any fish at risk? Should there be any changes in fish consumption advisories?

Georgia department of natural resources

Are any fish at risk? Should there be any changes in fish consumption advisories?

Consortium for Risk Evaluation with Stakeholder Participation

What data need to be collected to address the major concerns of the different stakeholders? How can the data be used to formulate a solution for the fisherfolk?

Discussions were held among the above parties in open meetings with each group individually, and in meetings (and conference calls) among groups (e.g., GDNR, SCDHEC, EPA, DOE). The discussions among the range of stakeholders resulted in agreement that data were needed to address the basic questions, which included the following: (1) What are the levels of mercury in fish from the Savannah River? (2) Do the levels of mercury vary above, along, and below the Savannah Rive Site? (3) Is there fishing along this stretch of the Savannah River, and (4) What are the consumption rates (and do they vary by gender or ethnicity)? All stakeholders agreed that site-specific data were required to determine if there was a risk to people from consuming fish from the Savannah River. And all agreed that if there was a risk, they had to address what could or should be done about it. Some agencies were willing to issue advisories on existing information, others not until the site specific data became available.

As a result of the discussions, two studies were designed to assess (1) fishing behavior and consumption patterns in anglers along the Savannah River (Fig. 5.4) and (2) mercury levels in fish as a function of species, size, and location along the river. In the latter case, fish were collected from above SRS, along SRS, and below SRS (Fig. 5.1).

5.3.4 Fishing Behavior and Consumption Patterns

In response to stakeholder requests, these two separate studies were initiated by CRESP to provide the information needed to move forward on the advisory process: (1) fishing behavior and consumption patterns and (2) mercury levels in fish as a function of species, size of fish, trophic level, and location of collection. The major steps for examining fishing behavior and consumption patterns involved establishing the study site (above, along, and below the SRS), designing a survey instrument to ascertain fishing behavior and consumption patterns, conducting a pilot study to make sure that the survey instrument was appropriate, conducting the study, analyzing the data, and presenting the results to all the stakeholders in a series of meetings (Fig. 5.4, see Burger et al. 2001a for more details).
Fig. 5.4

Schematic of the process involved in conducting a stakeholder-driven survey of fishing behavior and consumption patterns of fishers along the Savannah River. EPA environmental protection agency; DOE department of energy; CAB citizens’ advisory board for SRS; SREL university of Georgia’s Savannah River ecology laboratory located adjacent to SRS; CDC centers for disease control; GDNR Georgia department of natural resources; SCDHEC South Carolina department of health and environmental control. Note: “Fisherman” includes all fishers of all ages and genders.CRESP was involved in all phases and conducted the study

The process of designing the survey instrument involved the full range of stakeholders, both in the initial design and in the redesign of the instrument. The stakeholders provided information not only on what questions to ask but on the overall protocol. While the scientists were responsible for the initial design of the survey form, the information gathered in the iterative process assured that information gathered would satisfy the regulatory agencies’ needs as well as inform the fishing public.

A pilot study of 40 fishers was essential to make sure that the study questions were relevant to people fishing along the Savannah River, and could be administered easily. People fishing along the river provided input during the pilot study. In subsequent meetings with all the stakeholders, considerable refinements were made to the survey instrument to make it relevant to the fishers and to the agencies involved. As a result of this process, we added questions about cooking practices, specific fish, and the age at which children first consumed fish. The pilot study also allowed CRESP researchers to determine the sample size necessary for appropriate statistical analysis. All other components of the study were similarly taken to the stakeholders for comments and suggestions in open meetings or in conference calls.

Two aspects of the stakeholder involvement process bear comment: (1) inclusion of local scientists, and (2) use of local interviewers familiar with the culture and locations along the river (and who spoke the local language). The involvement of local scientists from the Savannah River Ecology Laboratory (SREL) provided CRESP with advice on the form of questions, advice on how to approach local fishers, advice on the protocol, and logistical information. The training of local people (fisherfolk themselves) as interviewers ensured that the interviews were conducted in an appropriate manner (with sufficient time for conversing amiably). Both were essential to the process.

The results of the study of fishing behavior and consumption patterns indicated that (1) portion size increased with the number of times people ate self-caught fish, (2) a substantial number of people ate more fish than that used by state agencies to compute risk to recreational fishers (19 kg/year for SC), (3) some people consume more fish than the default assumption (50 kg/yr), and (4) blacks consumed more locally caught fish per year than did whites, putting them at greater risk from mercury exposure (Fig. 5.5; Burger et al. 1999b) Further, people with less education, and those with lower incomes, ate more fish than others. Thus, the data clearly indicated that there was a problem with people eating more fish than agencies thought (Fig. 5.5), and it was an environmental justice issue as well (Bullard and Wright 1987; Bullard 1990, 1994; EPA 1994, 2002, 2009b)
Fig. 5.5

Relative consumption patterns of black and white people interviewed who were fishing along the Savannah River. 19 kg/year is the fish consumption rate used to design the fish consumption advisories used at the time, and 50 kg/year was the maximum default assumption

5.3.5 Mercury in Fish and Risk to Fish Consumers

Determining mercury levels in fish requires designing a protocol, collecting the fish, and analyzing the fish according to the protocol. A key aspect is to select fish species that are of interest to the fishing public, regulators, managers, DOE, and other stakeholders. A smaller group of stakeholders was involved in the process, although all the stakeholders named above made comments about the initial study design, locations for collecting, and fish species to be examined (Fig. 5.6). Fish were collected by electroshocking the fish to force them to the surface; all fish were dissected at SREL and shipped to Rutgers University for analysis.

Not surprisingly, we found that there were significant differences in mercury levels among fish, with bowfin (Amia calva, a fish largely eaten by black people) and largemouth bass (Micropterus salmoides, a preferred fish) having the highest mercury concentrations (Burger et al. 2001b, c). There were few differences in mercury levels in fish collected above, along, and below the Savannah River. Where there were differences, they were small and contradictory. Mercury levels were highest in bowfin above the SRS, while mercury levels in largemouth bass were highest below the SRS (Burger et al. 2001b). The lack of a strong difference among the three study locations may relate to the migratory patterns of the fish, moving up and down river. Thus, the data did not, in fact, provide strong support for SRS being a major contributor to mercury levels in fish. On the other hand, those same migratory patterns also did not exonerate the site as being a contributing source to the mercury levels in the fish. Data from the study were presented to the full range of stakeholders, many of whom suggested other analyses or questions to address with the data. In all cases, these analyses were performed and returned to the appropriate stakeholders. In general, this improved the analyses by making them more relevant to the users (Figs. 5.4 and 5.6).
Fig. 5.6

Schematic showing the overall process involved in examining the levels of mercury found in fish from the Savannah River. The full ranges of stakeholders are those described in Fig. 5.4

The ultimate goal of the study was to determine if people who eat fish from the Savannah River are at risk. The consumption study indicated that many people were eating fish at higher levels than those used to develop the fish consumption advisories (Burger et al. 1999b), and the mercury study in fish indicated that levels varied by species and size of the fish (Burger et al. 2001b). Although several methods were used to examine risk from consuming fish from the Savannah River (Burger et al. 2001b), one of the easiest to understand is to use the action levels developed by various states or other agencies, such as the FDA. This action level is the mercury level used by the FDA to condemn a shipment of commercial fish between states or prevent importation. While these levels were initially designed to prevent commerce between states, they can serve as a guide for fish that should generally be avoided.

The U.S. FDA (FDA 2001, 2005) has an action level of 1.0 μg/g (=ppm), and has issued a series of consumption advisories based on methylmercury, advising that pregnant women and women of childbearing age who may become pregnant should limit their fish consumption, should avoid eating four types of marine fish (shark, swordfish, king mackerel, tilefish), and should also limit their consumption of all other fish to just 12 oz per week (FDA 2003).

The FDA Action Level is not a health-based level. Some states have developed action levels, and many states use 0.5 ppm or even 0.3 ppm. Since information on levels of mercury is reported in ppm, providing this information can aid in making informed decisions. Table 5.3 shows the mean mercury levels for the fish, along with the percent of samples greater than 0.5 and 1.0 ppm. This indicates that many of the fish caught in the river, both by recreationists and by subsistence fishermen, have mercury levels that are of concern. At a minimum, this analysis allows people to decide which fish they should generally avoid (bowfin, largemouth bass), and which are safe to eat in nearly unlimited quantities (sunfish).
Table 5.3

Mercury concentrations (arithmetic mean and standard error) (μg/g wet weight  =  ppm) and percent of each species exceeding 0.5 and 1.0 ppm (after Burger et al. 2001b, c)


Mean (μg/g)  ±  SE

% >0.5 ppm

% >1 ppm


Amia calva

0.94  ±  0.05



Largemouth bass

Micropterus salmoides

0.46  ±  0.04



Chain pickerel

Esox niger

0.36  ±  0.03



Yellow perch

Perca flavescens

0.28  ±  0.02



Spotted sucker

Minytrema melanops

0.27  ±  0.04



Black crappie

Pomoxis nigromaculatus

0.24  ±  0.02



Channel catfish

Ictalurus punctatus

0.20  ±  0.02



American eel

Anguilla rostrata

0.15  ±  0.03



Bluegill sunfish

Lepomis macrochirus

0.14  ±  0.02



Red-breast sunfish

Lepomis auritus

0.13  ±  0.02



5.3.6 Development of a Fish Fact Sheet

As noted, in many regions, including along the Savannah River, adherence to the advice given in [compliance with] state-issued fish consumption advisories is low (Burger et al. 1999b; Burger and Gochfeld 2006; Wentholt et al. 2009). Low compliance can be due to a number of factors, such as confusing messages, conflicting advisories (as was the case with the Savannah River), controversies concerning the health benefits and risks of fish consumption, personal preferences, and an unwillingness to comply because of personal beliefs or preferences. In short, people often ignore advisories because they distrust the government, don’t believe it applies to them, assume they can detect tainted fish, or just love fish.

After numerous meetings and conferences with the full complement of stakeholders, it was agreed that there were people (particularly blacks) who were consuming more fish than the advisories advocated, and that they ate significantly more than the amount used to devise the advisories. Information on consumption and mercury levels indicated that some people (not only pregnant women), are at risk from consuming fish from the Savannah River. This led to agreement among the several responsible public entities that advice in a form likely to be understood by the affected community (i.e., a brochure or other communication device) was needed that gave more information on species to avoid (and to eat), preferred cooking methods, and the adverse effects of mercury (particularly on fetuses and young children). The device (brochure) and the method by which it was delivered needed to be clear, simple, readable, attractive, and responsive to the information needs of the fishing public. Since advisories were generally being ignored, there needed to be a plan for bringing it to people’s attention. One recommendation was that the brochures should be distributed to people fishing on the Savannah River (not just left at clinics and fishing tackle shops, as is the case with the state-generated consumption advice). The concept emerged that the brochure should be distributed by local student interns during the summer, an approach which had the added advantage that it involved yet another group of stakeholders.

This process also involved other steps that included all the major stakeholders. It involved even more conference calls and meetings than the science-based studies because it involved direct actions, as well as an agreed-upon Fish Fact Sheet (Fig. 5.7). Given the diverse interests that we have identified among the public entities (see Table 5.1), not only did it take time for the several public entities to agree on the message, it required many meetings to agree upon the exact wording even though everyone agreed there was a need for a directed message to consumers of fish (particularly pregnant women) from this area of the Savannah River.
Fig. 5.7

Schematic of the iterative and interactive process of developing a fish fact sheet for consumers of fish from the Savannah River. EPA environmental protection agency; DOE department of energy; CAB citizens’ advisory board for SRS; SREL university of Georgia’s Savannah River ecology laboratory located adjacent to SRS; CDC centers for disease control; GDNR Georgia department of natural resources; SCDHEC South Carolina department of health and environmental control. CRESP was involved in all phases and conducted the study

Having data on consumption patterns (including what species of fish were preferred by local anglers) and on contaminant levels was a crucial aid to bringing all stakeholders together. Since this information could have resulted in disagreements between the states, and with the EPA about the issuance of consumption advisories, we sought instead to find common ground. By finding a communications mechanism that effectively removed the jurisdictional issues (between the states and with EPA), it was possible to work on a design and wording of the brochure that was acceptable to everyone, yet provided the public with much-needed information.

Before embarking on creating a brochure, the group agreed on several principles: (1) fish are a healthy source of protein, (2) there was a population of the fisher public at risk because they ate more fish with higher mercury levels than previously suspected, (3) information on demographics was useful in helping outreach and communication specialists, (4) pregnant women and children should be the focus, (5) complete site characterization and extensive knowledge of the pathways was not necessary to develop a brochure for fish consumption, and (6) it should be clear to everyone, including the fishers, that the brochure represented a consensus among the two state agencies and EPA (Burger et al. 2001d). This was accomplished by adding logos of each agency on the brochure, and giving contact phone numbers of individuals from each of the three relevant agencies.

In informal meetings it was agreed that EPA would provide the initial content, and SCDHEC would write the initial draft. This was followed by numerous communications concerning the intent and working of the draft, including conference calls where representatives of all agencies were involved. The draft was also reviewed by the CAB, SRS–CDC health effects committees, and others for content and presentation (Fig. 5.7). While the process took more time than a top–down approach by any agency, it resulted in a product that all parties were pleased with. This was the first such information brochure about fish consumption that involved two states and the EPA. Further, this agreement contributed to the willingness of DOE to fund an intern to distribute the brochure during each fishing season.

Taken as a whole, the processes described here are examples of how credible scientific data gathered with stakeholder involvement can in turn be used to implement important risk-reduction management and policies, and to provide a basis for consensus building on difficult risk communication issues. It further suggests that consensus building and risk communication are continuing processes that involve assimilation of new information on contaminants in the food chain, variations in mercury levels among fish species (and sizes), state and federal laws, public policy, and public responses.

5.3.7 Efficacy of the Fish Fact Sheet: Evaluating the Stakeholder Involvement Process Itself

Having reached consensus about a path forward, and developed a Fish Fact Sheet to distribute to fishers above, along, and below the SRS, we wanted to determine whether the brochure was clear and the meaning received. We designed a questionnaire to determine if we were reaching the target audience, using the same overall methods as described for the other stakeholder-driven process described above. Our objectives were to determine (1) whether people fishing had previously read the Fish Fact Sheet or had heard about consumption advisories, (2) what major message they obtained from the Sheet, (3) who they felt the Fact Sheet was aimed at, and who should get it, and (4) who should be concerned about health risks from fish consumption. We also gathered the same demographic information as obtained in the fish consumption study. The same interviewers (local fishers who had lived there all their lives) conducted the interviews by first giving the Fish Fact Sheet to them, asking them to read it, and then interviewing them. Nearly everyone approached participated, and carefully read the brochure before answering the questions.

Over half of those interviewed felt that everyone should get the Fish Fact Sheet, even if they did not fish along the river, and the messages they obtained were accurate (Burger and Waishwell 2001). The major message, that people should limit fish intake in some way (e.g., by eating fish lower in mercury), was recognized by 86% of the people, a very high comprehension rate for this kind of information. Further, many people asked for additional copies for family members and friends.

This last step in the stakeholder-driven process to address potential mercury exposure of people fishing on the Savannah River is one usually not taken for environmental actions. That is, remediation, restoration, and educational and behavioral intervention actions taken around contaminated sites are not evaluated to see if they are effective in reducing risk to people fishing and to other community members living around the site. Even large-scale remediation/restoration projects seldom conduct in-depth follow-up studies to determine efficacy and effectiveness. Merely continuing to monitor levels of contaminants in media (soil, sediment), the usual method for determining success, may not effectively determine whether contaminants have reached the food chain, and ultimately humans.

This step in the process was again time-consuming, since it involved developing a questionnaire specifically for this project, pilot testing it, conducting the interviews, analyzing the results, and having stakeholders involved in the process. However, the end results convinced the agencies that the Fish Fact Sheet was effective, and that the distribution was worth the effort. It further indicated that the primary stakeholders, the fishers and their family/friends who ate the fish, were interested in the message, wanted more brochures to distribute, and obtained the correct message. That the message was designed specifically for this target audience improved the willingness to read the brochure, distribute it to friends and family, and act on the advice. Further, the interviewers found that those interviewed were very interested in more information about the source of contamination, ecological pathways, exposure, risk levels, and why the river had not been cleaned up (Burger and Waishwell 2001).

5.4 The Role of Science in the Process

The role of science was to provide site-specific information that was credible and addressed the questions as identified in the problem formulation phase. It was the presence of the site-specific data on consumption patterns and behavior that convinced the agencies (SCDHEC, GDNR, EPA) that there was a problem, DOE and CDC that something needed to be done, and the whole group that it was necessary to stop worrying about jurisdiction issues and past differences in the issuing of advisories. By changing the game from disagreements over consumption advisories to consensus over the message that needed to reach stakeholders, the science could form the basis for moving forward. Without site-specific data dealing with the key aspects of the issue, the parties had not agreed on any solution. The data provided a consensus for moving forward.

The two main advantages of collaborative science are that there can be some disagreements among the scientists about the approaches and methods to be used in the studies (which result in better approaches), and the scientists have to be willing to make changes in the design and protocol as a result of collaborations. That is, scientists usually like to design their own studies individually, and conduct them as they wish. Collaborative science requires that scientists listen to different viewpoints, and fully intend to implement changes in protocols.

5.5 The Role of Stakeholders in the Process

Jardine, Burger, and others suggested that public participation in the establishment of fish consumption advisories would greatly improve not only the advisories themselves, but compliance (Jardine 2003; Jardine et al. 2003; Burger and Gochfeld 2006). Jardine established 14 guiding principles for public participation that included the inclusion of community needs and values into the advisory process, in a timely fashion, in a collegial fashion, and with transparency. This happened at Savannah River.

The full range of stakeholders were involved in every step of the process, from problem formulation, identification of the key questions, design of the studies (one on fishing behavior and consumption patterns, and one on mercury levels in fish), examining the results of the pilot study with the survey questionnaire, redesigning the survey form, examining the final analyses, suggesting additional analyses, participating in public meetings, deciding on a path forward, creating a brochure, and testing the efficacy of the brochure. While the interviews themselves and the collection of fish for analysis were conducted by scientists, this involved using local research assistants who were also fishers as interviewers, and scientists from both CRESP and SREL (refer to Table 5.4). The beginning-to-end stakeholder involvement process was not only good public policy, rather it was fundamental to actually improving the quality of the scientific results and laid the groundwork for successful risk reduction once the risk management strategy had been defined.
Table 5.4

Role of different stakeholders in refinement of the fishing behavior and consumption patterns study along the Savannah River










Formulating the problem









Identifying additional stakeholders









Designing of survey instrument









Modification of survey instrument









Conducting study





Analyzing results




Presenting and commenting on preliminary results









Suggesting additional analyses









Reanalyzing data after input




Meetings with stakeholders









Modifying analyses









Writing scientific papers






Discussing path forward









Involvement in developing fish fact sheet









Efficacy survey of fish fact sheet







Meetings with stakeholders









Writing scientific paper


Distributing fish fact sheet to fishing public along the river (each summer)







Anglers  =  fisherfolk; other abbreviations found in Table 5.1. An X indicates that the agency or group contributed to this aspect of the study. CRESP is not included because they participated in all phases. Fisherfolk and the General Public are combined for this table as anglers

The collaborative process of inclusion of regulators, other state and federal agencies, scientists, citizens’ groups, fishers, and others requires the recognition that there will be disagreements. And in this case, there were times when phone calls were aborted, meetings got heated or canceled until issues could be resolved among two or more parties, and everyone felt the process might fall apart. But the bottom line was that everyone wanted to address the issue because there was a clear need to determine if the fish had high mercury levels and if there were people at risk. The agencies all recognized that the fishers were confused about the safety of fish from this stretch of the Savannah River, and that a clear, united message had to be presented that had the blessing of the state (SCDHEC, GDNR) and federal agencies (EPA), and the cooperation and agreement of DOE. More importantly, the message (form or content) had to be on target for the people fishing along the Savannah River, as well as clear, concise, and accurate.

5.6 Lessons Learned and Paths Forward

Until we succeed in removing or at least containing environmental contaminants and preventing further contamination, we must deal with reducing risk to humans and the environment, often by education or by blocking the exposure pathway. Both education and consumption advisories are short-term measures to reduce exposure and to deal with high levels of mercury contamination until such time as source reduction has significantly reduced the risk. These are solutions until our nation’s waters are sufficiently clean so that all fish can be eaten as often as desired. The decline in PCB concentrations in the environment over the past 25 years provides a basis for optimism (EPA 2009a).

5.6.1 Conflict Resolution

During the process there were obviously contentious issues that revolved around (1) jurisdictional autonomy (state’s rights), (2) federal authority to obtain uniformity for advisories pertaining to the same waters, (3) different health objectives (long-term goals of CDC vs. immediate mandate of the states to issue advisories), and (4) source reduction, cleanup standards (for DOE) vs. behavioral modifications in the form of issuance of specific fish consumption advisories or information brochures for this section of the Savannah River. These were major and important concerns, and they were always present as undercurrents to discussions. However, these old and contentious issues were mitigated by shifting the discussion from consumption advisories to information brochures that everyone could agree were needed.

There were several times during the nearly 3-year process that one or more parties became annoyed with progress or the views of others, disgusted with delays, or impatient with jurisdictional issues. What held the process together was the desire for site-specific data gathered from fishers and their families on fishing and consumption, site-specific data on mercury levels in fish gathered to answer very specific questions, and a desire to move forward to provide the public with a unified message. The role of CRESP was to facilitate the process and to design the initial protocol (for others to collaborate on modifying), conduct the study with help from local interviewers who were long-time residents and fishers and from SREL scientists. Since all parties participated in both the design of the survey form and the design of the mercury in fish study, there was uniform buy-in of the resultant data.

5.6.2 Lessons Learned

All participants in the process learned the importance of inclusion, iteration, and involvement of the state and federal agencies, local scientists, citizen’s boards, and the fishers themselves. While the process took longer than some participants had expected or desired at the outset, it did result in resolution and in coming up with a communication plan (and a brochure) that all parties could agree met the needs of the fishers and their families. The inclusion of the wide range of stakeholders early, and at every point in the process, meant that everyone was aware of progress and had an opportunity to participate. While not everyone agreed with every decision about protocols, methods, approaches, or data analysis, everyone had a chance to comment, make suggestions, and to have their concerns addressed and questions clarified.

Some of the main lessons learned are described in Table 5.5. In the final analysis, the issue of risk from mercury that might have been increased by SRS had been festering among the stakeholders for some time without any resolution. It was only the stakeholder-driven process that moved it forward.
Table 5.5

The main lessons learned from considering the risks from mercury in fish in the Savannah River, that flows between Georgia and South Carolina

1.Define the problem initially

2.Identify the stakeholders early and be inclusive

3.Involve stakeholders early and often

4.Include all the relevant state and federal agencies

5.Include a range of scientists, including local agency or university scientists

6.Modify the survey instrument to fit local biological and social conditions

7.Use local fisherfolk to conduct the interviews (with scientific training)

8.Conduct a pilot study to determine appropriate sample sizes, and refine survey forms

9.Be prepared for the process to take longer than a top–down approach

10.Be prepared (the scientists) to change the objectives, protocols, methods, and data collection, and to reanalyze data at the request of a range of stakeholders

11.Recognition by all participants that iteration and inclusion of others is essential, and that all viewpoints are valid

12.Be prepared for disagreements, and find ways to change the definition of the immediate issue to one that everyone can work with

The stakeholder-driven and collaborative process involved a wide range of state and federal agencies, citizens groups, research scientists and laboratories, and the fishing public



I particularly thank Michael Gochfeld, Charles W. Powers, David S. Kosson, Bernard D. Goldstein, Lynne Waishwell, and Camilla Warren for valuable discussions about SRS, science, stakeholders, and fish consumption; Shane Boring, I.Lehr Brisbin Jr., Patricia A. Cunningham, Karen F. Gaines, J.Whitfield Gibbons, and Joel Snodgrass for advice and logistical help while on SRS; Caron Chess, James Clarke, MichaelGreenberg, and Lisa Bliss for helpful discussions about science over the years; Tracey Shelly and Robert Marino (SCDHEC), Randy Manning (GDNR), John Stockwell (USEPA), Thomas Johnson and Wade Whittaker (USDOE), Christian Jeitner, Taryn Pittfield, and Mark Donio (Rutgers University) for technical help. This research was funded mainly by the Consortium for Stakeholder Participation (CRESP) through a contract from the Department of Energy (AI # DE-FFG-26-00NT 40938 and DE-FC01-06EW07053) to Vanderbilt University and Rutgers University, as well as the Nuclear Regulatory Commission (NRC 38-07-502M02), NIEHS (P30ES005022), and the New Jersey Department of Environmental Protection. The conclusions and interpretations reported herein are the sole responsibility of the author, and should not in any way be interpreted as representing the views of the funding agencies.


  1. Albert CM, Campos H, et al (2002) Blood levels of long-chain n-3 fatty acids and the risk of sudden death. New Engl J Med 346:1113–1118CrossRefGoogle Scholar
  2. Anderson PD, Wiener JB (1995) Eating fish. In: Graham JD, Wiener JB (ed) Risk versus risk: tradeoffs in protecting health and the environment. Harvard University Press, CambridgeGoogle Scholar
  3. Agency for Toxic Substances and Disease Registry (ATSDR) (1996) Mercury awareness for the public. Hazard Subst Public Health 6:4–5Google Scholar
  4. Buck GM, Vena JE, Schisterman, EF, et al (2000) Parental consumption of contaminated sport fish from Lake Ontario and predicted fecundability. Epidem 11:388–383CrossRefGoogle Scholar
  5. Bullard RD (1990) Dumping in Dixie: race, class, and environmental quality. Westview Press, ColoradoGoogle Scholar
  6. Bullard RD (1994) Overcoming racism in environmental decision making. Environment 30:39–44Google Scholar
  7. Bullard RD, Wright BH (1987) Environmentalism and the politics of equity: emergent trends in the black community. Mid-Am Rev Sociol 12: 21–32Google Scholar
  8. Burger J (2000) Consumption advisories and compliance: The fishing public and the deamplification of risk. J Environ Plan Manage 43:471–488CrossRefGoogle Scholar
  9. Burger J (2002) Consumption patterns and why people fish. Environ Res 90:125–135CrossRefGoogle Scholar
  10. Burger J (2005) Fishing, fish consumption and knowledge about advisories in college students and others in central New Jersey. Environ Res 98:268–275CrossRefGoogle Scholar
  11. Burger J, Gochfeld M (2006) A framework and information needs for the management of the risks from consumption of self-caught fish. Environ Res 101:275–285CrossRefGoogle Scholar
  12. Burger J, Waishwell L (2001) Are we reaching the target audience? Evaluation of a Fish Fact Sheet. Sci Total Environ 277:77–86CrossRefGoogle Scholar
  13. Burger J, Gaines KF, Gochfeld M (2001a) Ethnic differences in risk from mercury among Savannah River fishermen. Risk Anal 21:533–544CrossRefGoogle Scholar
  14. Burger J, Cooper K, Gochfeld, M (1992) Exposure assessment for heavy metal ingestion from a sport fish in Puerto Rico: estimating risk for local fishermen. J Toxicol Environ Health 36:355–365CrossRefGoogle Scholar
  15. Burger J, Staine K, Gochfeld M (1993) Fishing in contaminated waters: knowledge and risk perception of hazards by fishermen in New York City. J Toxicol Environ Health 3:95–105CrossRefGoogle Scholar
  16. Burger J, Pflugh KK, Lurig L, von Hagen LA, von Hagen SA (1999a) Fishing in urban New Jersey: Ethnicity affects information sources, perception, and compliance. Risk Anal 19:217–229Google Scholar
  17. Burger JW, Stephens CS, Boring CS, Kuklinski M, Gibbons JW, Gochfeld M (1999b) Factors in exposure assessment: Ethnic and socioeconomic differences in Fishing and Consumption of fish caught along the Savannah River. Risk Anal 19:427–438Google Scholar
  18. Burger J, Gaines KF, Gochfeld M (2001b) Ethnic differences in risk from mercury among Savannah River fishermen. Risk Anal 21:533–544CrossRefGoogle Scholar
  19. Burger J, Gaines KF, Boring CS et al (2001c) Mercury and selenium in fish from the Savannah River: species, trophic level, and locational differences. Environ Res 87:108–118CrossRefGoogle Scholar
  20. Burger J, Gochfeld M, et al (2001d) Science, policy, stakeholders, and fish consumption advisories: developing a fish fact sheet for the Savannah River. Environ Manage 27:501–514CrossRefGoogle Scholar
  21. Chess C, Burger J, McDermott MH (2005) Speaking like a state: environmental justice and fish consumption advisories. Soc Nat Res 18:267–278CrossRefGoogle Scholar
  22. Connelly NA, Knuth BA (1998) Evaluating risk communication: examining target audience perceptions about four presentation formats for fish consumption health advisory information. Risk Anal 18:649–659CrossRefGoogle Scholar
  23. Davis S Ogden JC (1994) Everglades: the ecosystem and its restoration. CRC Press, FloridaGoogle Scholar
  24. Department of Energy (DOE) (2009) Environmental justice at the U.S. Department of Energy. Accessed 9 Feb 2010
  25. Downs SG, Macleod CL, Lester JN (1998) Mercury precipitation and its relation to bioaccumulation in fish: a literature review. Water Air Soil Pollut 108:149–187CrossRefGoogle Scholar
  26. Driscoll CT, Abbot M, et al (2006) Airsheds and watersheds. In: Harris R, Krabbenhoft DP, Mason R, Murray MW, Reash R, Saltman T, (ed) Ecosystem responses to mercury contamination. CRC Press, FloridaGoogle Scholar
  27. EPA (Environmental Protection Agency) (1980) Ambient water quality criteria for mercury. US Environmental Protection Agency Report 440/5-80-058. Natl Tech Inform Serv, 5285 Port Royal Road, Springfield, VAGoogle Scholar
  28. Environmental Protection Agency (EPA) (1994) Environmental justice initiatives 1993. (EPA 2000-R-93-001) US Environmental Protection Agency WashingtonGoogle Scholar
  29. Environmental Protection Agency (EPA) (2002) National Environmental Justice Advisory Council: fish consumption and environmental justice­publications/ej/fish_consump_report_1102.pdf. Accessed April 2009
  30. EPA (Environmental Protection Agency) (2009a) National listing of fish advisories: technical fact sheet: 2008 biennial national listing. Accessed 9 Nov 2009
  31. Environmental Protection Agency (EPA) (2009b) Environmental justice: compliance and enviorcement. Accesed 1 Feb 2010
  32. Food and Drug Administration (FDA) (2001) FDA consumer advisory. Accessed 1 Dec 2005
  33. Food and Drug Administration (FDA) (2003) FDA consumer advisory. Accessed 1 Jan 2008
  34. Food and Drug Administration (FDA) (2005) What you need to know about mercury in fish and shellfish – March 2004: 2004 EPA and FDA Advice For: Women Who Might Become Pregnant Women Who are Pregnant Nursing Mothers Young Children. Accessed 8 June 2009
  35. Fitzgerald WF, Mason RP (1996) The global mercury cycle: oceanic and atmospheric aspects. In: Baeyens W, Ebinghaus R, Vasiliev O (ed) Global and regional mercury cycles: sources, forces and mass balances. Kluwer Academic Publishers, DordrechtGoogle Scholar
  36. Fleming LE, Watkins S et al (1995) Mercury exposure in humans through food consumption from the Everglades of Florida. Water Air Soil Pollut 80:41–48CrossRefGoogle Scholar
  37. FOA (1998) Seafood safety – Economics of hazard analysis and critical control point (HACCP) programmes. United Nations Food and Agricultural Organization, Fisheries Technical Paper # 381, Rome, Italy. Accessed 10 March 2008
  38. GDNR (Georgia Department of Natural Resources) (1999) Guidelines for eating fish from Georgia waters. GDNR, AtlantaGoogle Scholar
  39. Ginsberg GL, Toal BF (2000) Development of a single-meal fish consumption advisory for methylmercury. Risk Anal 20:41–47CrossRefGoogle Scholar
  40. Gochfeld M, Burger J (2005) Good fish/bad fish: a composite dose and benefit-risk curve. NeuroToxicol 26:511–520CrossRefGoogle Scholar
  41. Greenberg M, Krueckeberg D et al (1998) Socioeconomic impacts of U.S. nuclear weapons facilities: a local scale analysis of Savannah River, 1950-1993. Appl Geograph 18:101–116CrossRefGoogle Scholar
  42. Guallar E, Sanz-Gallardo MI, et al (2002) Heavy metals and Myocardial Infarction Study Group: Mercury, fish oils, and the risk of myocardial infarction. New Eng J Med 347:1747–1754CrossRefGoogle Scholar
  43. Halkier B (1999) Consequences of the politicization of consumption: the example of environmentally friendly consumption practices. J Environ Policy Plan 1:25–41Google Scholar
  44. Hightower JM, Moore D (2003) Mercury levels in high-end consumers of fish. Environ Health Perspect 111:604–608CrossRefGoogle Scholar
  45. Hites RA, Foran JA et al (2004) Global assessment of organic contaminants in farmed salmon. Science 303:226–229CrossRefGoogle Scholar
  46. Institute of Medicine (IOM) (1991) Seafood Safety. National Academy Press. WashingtonGoogle Scholar
  47. Iribarren C, Markovitz JH et al (2004) Dietary intake of n-3, n-6 fatty acids and fish: relationship with hostility in young adults – the CARDIA study. Eur J Clin Nutr 58:24–31CrossRefGoogle Scholar
  48. Iso H, Rexrode KM (2001) Intake of fish and omega-3 fatty acids and risk of stroke in women. J Am Med Assoc 285:304–312CrossRefGoogle Scholar
  49. Jakus PM, Downing M, Bevelhimer MS, Fly JM (1997) Do sportfish consumption advisories affect reservoir anglers’ site choice? Agricult Res Econom Rev 26:196–204Google Scholar
  50. Jardine CG (2003) Development of a public participation and communication protocol for establishing fish consumption advisories. Risk Anal 23:461–471CrossRefGoogle Scholar
  51. Jardine CG, Hrudey SE et al (2003) Risk management frameworks for human health and environmental risks. J Toxicol Environ Health Part B 6:569–641Google Scholar
  52. Jewett SC, Zhang X et al (2003) Comparison of mercury and methylmercury in northern pike and Arctic grayling from western Alaskan rivers. Chemosphere 50:386–392CrossRefGoogle Scholar
  53. Johnsson C, Schutz A, Sallsten G (2005) Impact of consumption of freshwater fish on mercury levels in hair, blood, urine, and alveolar air. J Toxicol Environ Health (part A). 68:129–141CrossRefGoogle Scholar
  54. Knobeloch L, Anderson HA et al (2005) Fish consumption, advisory awareness, and hair mercury levels among women of childbearing age. Environ Res 97:220–227CrossRefGoogle Scholar
  55. Kvartek EJ, Carlton WH et al (1994) Assessment of mercury in the Savannah River Site environment. Westinghouse Savannah River Co (WSRC-TR-94-0218ET). AikenGoogle Scholar
  56. Lange TR, Royals HE, Connor LL (1994) Mercury accumulation in largemouth bass (Micropterus salmoides) in a Florida lake. Environ Contam Toxico l27:466–471Google Scholar
  57. Montiero LR, Costa V, Furness RW, Santos RS (1996) Mercury concentrations in prey fish indicate enhanced bioaccumulation in mesopelagic environments. Mar Ecol Prog Ser 141:21–25CrossRefGoogle Scholar
  58. Moya J (2004) Overview of fish consumption rates in the United States. Human Ecol Risk Assess 10:1195–1211CrossRefGoogle Scholar
  59. Mueezzinoglu, A (2003) A review of environmental consideration on gold mining and production. Critical Reviews in Environmental Science and Technology 33:45–71CrossRefGoogle Scholar
  60. National Research Council (NRC) (2000) Toxicological effects of methylmercury. National Academy Press. WashingtonGoogle Scholar
  61. Neuringer M, Reisbick S, Janowsky J (1994) The role of n-3 fatty acids in visual and cognitive development: Current evidence and methods of assessment. J Pediatr 125:39–47CrossRefGoogle Scholar
  62. Nichols JW (2001) Use of indicators in ecological risk assessment for persistent, bioaccumulative toxicants. Human Ecol Risk Assess 7:1043–1057CrossRefGoogle Scholar
  63. NOAA. (2004) Seafood consumption rose again in 2003. NOAA Magazine, U.S. Commerce Dept., National Oceanographic and Atmospheric Administration. Accessed 10 March 2005
  64. Norton SB, Rodier DR et al (1992) A framework for ecological risk assessment at the EPA. Environ Toxicol Chem 11:1663–1672CrossRefGoogle Scholar
  65. Oken E, Kleinman KP et al (2003) Decline in fish consumption among pregnant women after a national mercury advisory. Obstet Gynecol 102:346–351CrossRefGoogle Scholar
  66. Olsen SF, Secher NJ (2002) Low consumption of seafood in early pregnancy as a risk factor for preterm delivery: prospective cohort study. Brit Med J 324:447–450CrossRefGoogle Scholar
  67. Pacyna EG, Pacyna JM, Steenhuisen F, Wilson, S (2006) Global anthropogenic mercury emissions inventory for 2000. Atmosph Environ 40:4048–4063CrossRefGoogle Scholar
  68. Patterson J (2002) Introduction – comparative dietary risk: balance the risks and benefits of fish consumption. Comments Toxicol 8:337–344CrossRefGoogle Scholar
  69. Pflugh KK, Lurig L, vonHagen LA, vonHagen S, Burger J (1999) Urban angler’s perception of risk from contaminated fish. Sci Total Environ 228:203–218CrossRefGoogle Scholar
  70. Ramos AM, Crain EF (2001) Potential health risks of recreational fishing in New York City. Ambulat Pediatr 1:252–255CrossRefGoogle Scholar
  71. SCDHEC (South Carolina Department of Health and Environmental Control) (1999) Public health evaluation: Cesium-137 and strontium-90 in fish. Attachment to the fish consumption advisory for the Savannah River, ColumbiaGoogle Scholar
  72. Schantz SL (1996) Developmental neurotoxicity of PCBs in humans: What do we know and where do we go from here? Neurotox Teratol 18:217–227CrossRefGoogle Scholar
  73. Schantz SL, Widholm JJ, Rice DC (2003) Effects of PCB exposure on neuropsychological function in children. Environ Health Perspect 111:357–376CrossRefGoogle Scholar
  74. Scudder BC, Chaser LC et al (2009) Mercury in fish, bed sediment, and water from streams across the United States, 1998–2005. U.S. Dept of Interior, Report 2009–5109. Reston, Virginia. 74 ppGoogle Scholar
  75. SFWMD (South Florida Water Management District) (2007) South Florida Environmental Report: 2007 (G. Redfield, ed). West Palm Beach. SFWMDGoogle Scholar
  76. Sparks P, Shepherd R (1994) Public perceptions of the potential hazards associated with food production: an empirical study. Risk Anal 14:799–808CrossRefGoogle Scholar
  77. Stern AH (1993) Re-evaluation of the reference dose for methylmercury and assessment of current exposure levels. Risk Anal 13:355–364CrossRefGoogle Scholar
  78. Stern AH, Jacobson JL, Ryan L, Burke TA (2004) Do recent data from the Seychelles Islands alter the conclusions of the NRC Report on the toxicological effects of methylmercury? Environ Health 3:2Google Scholar
  79. Sugg DW, Chesser RK, Brooks JA, Grasman BT (1995) The association of DNA damage to concentrations of mercury and radiocesium in largemouth bass. Environ Toxicol Chem 14:661–668CrossRefGoogle Scholar
  80. Toth Jr JF, Brown RB (1997) Racial and gender meanings of why people participate in recreational fishing. Leisure Sci 19:129–146CrossRefGoogle Scholar
  81. Vupputuri S, Longnecker MP, Daniels JL, Xuguang G, Sandler DP (2005) Blood mercury level and blood pressure among US women: results from the national Health and Nutrition Examination Survey 1999–2000. Environ Res 97:195–200CrossRefGoogle Scholar
  82. Wentholt MTA, Rose G, Konig A, Marvin HJP, Frewer LJ (2009) The views of key stakeholders on an evolving food risk governance framework: results from a Delphi study. Food Policy 34:539–548CrossRefGoogle Scholar
  83. Workman SW, McLeod KW (1990) Vegetation of the Savannah River Site: major community types. Savannah River Ecology Laboratory (SRO-NERP-19). AikenGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC 2011

Authors and Affiliations

  1. 1.Division of Life SciencesEnvironmental and Occupational Health Sciences Institute (EOHSI), Consortium for Risk Evaluation with Stakeholder Participation (CRESP), and Rutgers UniversityPiscatawayUSA

Personalised recommendations