The Role of Statistics in the Development of National Fish Monitoring Programs for Endocrine Disruption in Canada
Kelly R. Munkittrick, Tim J. Barrett
Canadian Rivers Institute, University of New Brunswick, Saint John, NB, Canada

During the late 1980s studies found reduced gonad sizes, delayed maturation and changes in secondary sex characteristics near effluent discharges from some Canadian pulp mills. The findings led to the development of a national, standardized, cyclical monitoring program for fish and benthic invertebrate communities all of the >130 Canadian mills in 1992. This program was expanded to include metal mines with liquid effluent discharges a decade later. The industry-funded cyclical program is tiered, and decisions on subsequent monitoring depend on the sizes of effects and consistency of patterns of responses. Statistics played a number of unique roles in the development of this national program, including developing methods for determining critical effects sizes, setting alpha and power levels, and developing and applying methods for calculating the magnitude of difference on an national basis, in an environment open to multi-stakeholder input and debate. In the fish program, more than 65 species have been used for monitoring across the country. Given the diversity of sampling strategies and discharge environments, statistics played a critical role in defining optimal sampling periods for all 65 species, and for conducting meta-analyses for comparison of results across species on a national basis. We will review some of the important steps and points of controversy.

Keywords: Endocrine disruptors; Fish populations; Monitoring; Critical effect size

Biography: Kelly Munkittrick is the Associate Director of the Canadian Rivers Institute and holds a Tier 1 Canada Research Chair in Ecosystem Health Assessment at the University of New Brunswick in Saint John, New Brunswick, Canada. He is also the Program Leader for the Watersheds and Ecosystems Theme of the Canadian Water Network, and prior to his appointment at UNB, he worked for 11 years for the Canadian federal government, in Environment Canada and Fisheries and Oceans. His research interests are related to assessing the environmental impacts of industrial and agricultural activities, and on developing methods for environmental effects monitoring and for the cumulative effects assessment of multiple stressors on aquatic environments. He currently have active projects on assessing environmental impacts in Sri Lanka, Bhutan, Chile, Uruguay, Brazil, the US and Canada, and has worked, taught or given invited lectures in more than 25 countries.