Development of an Ecosystem Sensitivity Model Regarding Mercury Levels in Fish Using a Preference Modeling Methodology: Application to the Canadian Boreal System

Abstract

The objective of this study is to use a preference modeling methodology as a predictive tool to roughly assess the sensitivity of the ecosystems regarding mercury (Hg) concentrations in fish. We apply a preference modeling methodology to rank lakes within the boreal forest from highest to lowest Hg concentrations in fish using simple environmental factors. Among the numerous variables influencing Hg fate in the environment, we only retain simple key indicators that are expected to influence Hg concentrations in fish tissue such as watershed characteristics of the lake (percentage of the catchment area of the lake, ratio of drainage area versus lake area, percentage of the drainage area of the lake as wetlands, land use, and clear-cutting), lake characteristics (chlorophyll, dissolved organic carbon, pH, and fishing intensity), and atmospheric Hg inputs. Preliminary results of modeling that we carried out using a set of Canadian lakes of boreal forest data are promising. With only a minimum set of criteria, we are able to reproduce the trends of Hg contamination in fish caught in six regions of the Canadian boreal forest and classify the sensitivity of the ecosystems to Hg loadings in three categories: high, medium, and low.