 |
||||
|
||||
|
|
 |
||
|
||||
      |
What are the Causes, Consequences and Correctives of Fish Contamination in the Detroit River AOC that Cause Health Consumption Advisories?
Primary Investigator:
Donna Kashian - NOAA/GLERL
Co-Investigators:
Jonathan Bulkley, Larissa Sano - University of Michigan*
Ken Drouillard, Doug Haffner - University of Windsor*
Ann Krause - University of Toledo*
Executive Summary of Rationale
The Detroit River remains under several fish consumption advisories, which are beneficial use impairments that both impact human health and affect economic revenue. In terms of the former, there are consumption restrictions for various fish species that apply to sensitive sub-groups, such as children and women of childbearing age. In terms of the latter, consumption advisories on highly prized fish species such as walleye and yellow perch likely impact economic revenue of the region (Johnson 2000).
Despite the critical importance of these advisories, little progress has been made in developing effective management strategies. For example, although sediment remediation efforts in the Detroit River have totaled more than $120 million dollars, there is little evidence of ecosystem improvement (Heidtke et al. 2003). Many uncertainties also remain regarding the primary drivers of these advisories, with key uncertainties including the relative contribution of sediment hot spots, the role of point versus non-point contaminant sources, and the appropriateness of tissue trigger-levels in identifying threshold action levels for consumption advisories. Because of these complexities, solutions for remedying consumption advisories will require novel approaches directed at both decreasing body burden levels in fish over the long-term and reducing human health risks in the short- and long-term.
Proposed Work
Current/Ongoing
Specific activities for the first year include: 1) Continue our workshop series designed to integrate research, monitoring and policy efforts among the United States and Canada, 2) begin compiling and synthesizing available data on sediment, water, fish movement and fish contaminants in the Detroit River, 3) using available data, begin developing a bioaccumulation model for walleye in the Detroit River, 4) begin developing a food web network model for the Detroit River, and 5) work with the Michigan Department of Community Health to assess trigger-levels used by state for fish consumption advisories.
Scientific Rationale
For both the Detroit River and the rest of the state, the majority of fish consumption advisories are due to either polychlorinated biphenyls (PCBs) or mercury (Hg). Fish consumption advisories in the Detroit River cover a range of species including carp, freshwater drum, northern pike, redhorse sucker, walleye, and yellow perch. Consistent with the Integrated Assessment (IA) framework, we will use available data from the Detroit River to develop models that both identify possible drivers of elevated fish contaminant body burdens and facilitate a more risk-based approach of tissue trigger-levels for consumption advisories. This integrated assessment approach is particularly useful for consumption advisories as it provides the ability to integrate and organize complex data in a manner that can help inform management decisions.
In addition, the IA framework explicitly fosters collaboration and participation of multiple interested groups. We will capitalize on this component of IAs, by seeking the active participation of different stakeholder groups in developing logic models that identify the goals of the scientific assessment and the connections between the science and management or policy outcomes. This science-policy connection will be the focus of workshops designed to evaluate frameworks identifying the goals and desired outcomes from the scientific assessment component of the IA. This information will be used to guide management initiatives.
Although this project will focus on organic contaminants in the Detroit River, the outcome of this effort should provide a framework for addressing advisories for other contaminants of concern, specifically mercury, in the Detroit River and at other areas of concern in Michigan and for identifying measures that can help with delisting criteria.
Objectives:
1) Synthesize and summarize the status and trends of fish contaminant levels and advisories in the Detroit River, through an assessment of the trends in the data as well as documentation of the history of the fish advisory in the river (Canadian and U.S.).
2) Describe the abiotic, biotic, and human health causes and consequences of fish consumption advisories, with an emphasis on model simulation and explanatory analyses. This effort will focus on the environmental conditions that contribute to fish consumption advisories in addition to other factors that may exacerbate human health risks.
3) Identify the key uncertainties regarding the drivers of consumption advisories for use in prioritizing future research and monitoring efforts and in helping guide management and policy directives.
4) Utilize information on the causes and consequences of consumption advisories for providing technical guidance in implementing policy and management options. This will include a focus on short-term measures that reduce direct threats to human health and longer-term objectives to reduce overall body burden of fish in the Detroit River relative to reference areas.
Although this project will initially focus on organic contaminants in the Detroit River, we envision this effort as a pilot project that can be expanded to other contaminants of concern in the Detroit River and to other impacted areas in the state of Michigan that are under consumption advisories.
Methodology:
A) Document the status and trends of environmental, social, and economic conditions related to fish consumption advisories in the Detroit River.
A critical element of this project will be to synthesize and summarize available information and data related to fish consumption advisories in the Detroit River. This will generally be accomplished by gleaning information from the scientific literature, reports, and available datasets. The final summary will be structured to provide a comprehensive document that structures the current state of knowledge about these advisories.
Environmental trends related to fish consumption advisories: Multiple available datasets from the Detroit River will be used for this component of the IA. Data sets are derived from both Canada and the U.S.
B) Describe the environmental, social, and economic causes and consequences of the fish consumption advisories
To help assess the causes of fish consumption advisories, available data on contaminants and fish body burdens in the Detroit River (detailed above) will be used to develop models to identify potential drivers of these advisories.
Model 1: Probabilistic bioaccumulation model for walleye
An important consideration in identifying the causes of consumption advisories in the Detroit River is the relative role of sediment contaminant (i.e., legacy contaminants and hot spots) compared to water concentrations (i.e., current input). Deciphering between these two contaminant sources is further compounded by the mobility of the sport fish populations most of concern. Previous studies using food web bioaccumulation models in the Detroit River have suggested a strong bias with respect to larger, mobile sport fish, which may be due to the movement of fish outside of the modeling region (Drouillard et al. 2003). To help identify the drivers of elevated body burdens in these larger sport fish, a probabilistic bioaccumulation model incorporating fish movements will be used to develop a risk-based assessment for identifying potential sediment hotspots that might drive consumption advisories for walleye, a representative sport fish. This food web bioaccumulation model will be used in conjunction with contaminant water concentrations and sediment concentrations to help predict steady-state sport fish contaminant residues for each of 20 modeling zones distributed within the Detroit River and the adjacent Detroit River – Lake St. Clair corridor.
Model 2: Network analysis model – trophic web and contaminants
The basic mechanism for PCB body burdens of fish species utilized by humans is understood to be bioaccumulation from the food the fish eat, what is not clearly understood is the relative role of the food web in creating these burdens. We seek to determine the subsystem structure of the spatially-explicit food web in the Detroit River by identifying groups of highly interacting populations of fish, benthic invertebrates, and plankton; and quantify the influence of the subsystem structure and function on PCB body burdens of those populations.
The spatially explicit network of the food web will be developed primarily from data collected specifically on the food web of the Detroit River (DRMMF 2003). This network will represent the feeding interactions as indicated by diet data as well as an understanding of the spatial movement patterns of the population in question. The subsystem structure will be analyzed using the statistical software package KliqueFinder (Frank 1996).
Model 3: Fish tissue trigger-levels for consumption advisories
We propose to re-evaluate the current trigger-values used in issuing fish consumption advisories. This will be accomplished through a review of relevant gray- and peer-reviewed literature aimed at documenting the history of contaminant advisories in the Detroit River, both for the U.S. and Canadian side, evaluating more recent data and information regarding trigger-values and contaminant effects, and review approaches of other states and jurisdictions in establishing consumption advisories. Trigger-values have not been re-evaluated by the state of Michigan since the late-1990s and should be re-assessed to determine the utility and application of the data given advances and more recent studies on PCB levels in fish. We will work on developing a more probabilistic risk-based approach for identifying trigger-value thresholds.
C) Provide technical guidance for implementing each option
Information about the causes and consequences of fish consumption advisories generated from this project will be used to develop technical guidance for the local and state officials addressing Detroit River consumption advisories. This technical guidance will be developed through two approaches: The first approach will be the application of social network methods to assess the organizational network involved with fish consumption. The second approach is through collaborative workshops to develop an explicit evaluation process. We will hold bi-annual workshops for all proposal participants and critical stakeholder groups. At these workshops, we will discuss the desired outcomes with respect to consumption advisories and ways in which this project can be used to inform management and policy directives in the Detroit River.
D) Provide an assessment of the levels of certainty associated with the information provided in each of the steps
Output from this effort will be evaluated to identify key parameters influencing our understanding of fish consumption advisories causes. Where appropriate, a quantitative assessment of uncertainty will be conducted primarily using Monte Carlo approaches (Asante-Duah 2002). This type of approach will be used primarily for assessing: 1) the key parameters in the bioaccumulation model with the highest level of uncertainty including water and sediment contaminant concentrations, fish movement patterns, and physiological parameters driving contaminant uptake and depuration; 2) the main parameters used by the State in issuing consumption advisories; and 3) the relative source contribution of contaminant input.
Governmental/Societal Relevance
The Great Lakes are continually impacted by natural and anthropogenic contaminates causing a wide range of ecosystem health and economic effects. NOAA has responsibility to further the understanding and predict changes in earth’s environment, including the Great Lakes, in order to meet economic, societal and environmental needs. The purpose of this work is to identify the causes, consequences and correctives of fish contamination in the Detroit River that result in health consumption advisories. This project has the potential to impact government policy regarding the appropriateness of current tissue trigger-levels and identifying threshold action levels for consumption advisories. Results from this study can aid in management decisions by providing technical guidance in implementing policy and management options aimed at identifying measures that reduce direct threats to human health and longer-term objectives to reduce overall body burden of fish in the Detroit River relative to reference areas.
Relevance to Ecosystem Forecasting
A goal of this project it to provide forecasts of conditions under various policy options. Models 1 and 2 provide a set of tools that can be used as a decision support framework for policy options. The policy options to be explored will come from the stakeholder process. Model 1 provides an opportunity to determine the relative importance of water and sediments in the bioaccumulation of PCBs in fish. Policy scenarios that involve either decreasing PCBs in the water or decreasing PCBs in the sediments can be explored with this model to determine which one may have a bigger impact on PCB bioaccumulation for fish species of interest and determine if existing water quality and sediment quality objectives for PCBs are protective against fish consumption advisories for different species of fish. Furthermore, we propose to evaluate interaction between fish movements and aerial sediment contamination to determine sediment quality objectives that would be protective against fish advisories and distinguish point sources (water quality) from legacy contaminant issues.
Model 2 can be adapted using an agent-based approach to explore what-if scenarios in policy options for stakeholders during their workshops. An agent-based approach has been applied in a diversity of social and natural sciences, including environmental management (Bousquet and Le Page 2004). In agent-based modeling, researchers and stakeholders can set up decision rules within a network and simulates outcomes based on those decision rules, where the decision rules can have stochastic properties. Although similar to the more widely applied individual-based modeling (IBM) in ecology, it is particularly suited for networks over IBMs and other traditional dynamic models employed in ecology because it is guided by an organizational view of complexity (Bousquet and Le Page 2004). The policy scenarios that would be particularly suited for such an exercise are those involving the reduction of PCBs in spatial areas of the river, but stakeholders may come up with others as well.
Stakeholders will also be involved in the process of identifying which management units (RAPs and LaMPs) need to be involved and which management actions (point source control), sediment remediation would be most effective on a per management unit basis. The results on trigger-levels from Model 3 would provide context for evaluating the effectiveness of the policy option at reaching the stakeholders’ goals and help identify biological factors which influence fish advisories.
Cited References
Bousquet, F. and C. Le Page. 2004. Multi-agent simulations and ecosystem management: a review. Ecological Modeling. 176:313-332.
DRMMF Interpretive Report. 2003. Report Submitted to the Detroit River Canadian Cleanup Committee, April 2003. Prepared by the Great Lakes Institute for Environmental Research, University of Windsor, Windsor, ON, Canada.
Drouillard KG. 2003. Use of bioaccumulation models to perform hazard assessments for sport fish consumption advisories in the Detroit River AOC. Workshop Presentation: Staying on Target: A Regional Workshop for Establishing Restoration Targets for Great Lakes Areas of Concern, June 6-7, 2003, Romulus, MI, U.S.A.
Frank, K.A. 1996. Mapping interactions within and between cohesive subgroups. Social Networks. 18:93-119.
Frank, K.A., Zhao, Y., and Borman, K. 2004. Social capital and the diffusion of innovations within organizations: application to the implementation of computer technology in schools. Sociology of Education. 77:148-171.
Hedtke, TM, J Hartig, B Yu. 2003. Evaluating ecosystem results of PCB control measures within the Detroit River-Western Lake Erie Basin. Great Lakes National Program Office, U.S. Environmental Protection Agency, Chicago, IL, EPA-905-R-03-001, pp. 108-114.
Hesse, J. 1997. Case study: Sport fish consumption advisories in the Great Lakes region. In Environmental Risk Harmonization: Federal and State Approaches to Environmental Hazards in the USA. Edited by M.A. Kamrin. Wiley, New York.
Johnson, TB, RC Haas. 2004. Fish and fisheries of the Detroit River. Detroit River State of the Strait Conference, Dec. 2, University of Windsor, Windsor, On.
Krause, A.E., K.A. Frank, D.M. Mason, R.E. Ulanowicz, and W.W. Taylor. 2003. Compartments revealed in food-web structure. Nature. 426:282-285.
Simon, H.A. 1962. The architecture of complexity. Proceedings of the American Philosophical Society. 106:467-482.
Products
New products developed for placement along fishing areas of the Detroit River
Michigan Family Fish Consumpion Guide (.pdf)*
*Link leads off GLERL's website
