2005 Salt Lake City Annual Meeting (October 16–19, 2005)

Paper No. 8
Presentation Time: 10:00 AM

MERCURY IN FISH: A GEOLOGIST PERSPECTIVE


BEHLING, Stuart James, NA, NA, Superior National Forest, 8901 Grand Ave. Pl, Duluth, MN 55808, sbehling@fs.fed.us

Mercury contamination of fish has been a health concern in Northeastern Minnesota since testing began over 30 years ago. In spite of this concern, less than 15% of the lakes in this area have been tested. The data show that fish mercury contamination varies considerably between lakes but no pattern of distribution has been recognized. The data also show that fish mercury contamination varies between species and increases with length of fish. For example, walleyes are more contaminated than northern pike of the same size, from the same lake, and longer fish have more mercury than shorter ones. However, there is not enough data available from most lakes to plot these contamination trends. A model that could predict fish mercury contamination levels by species and length of fish would be a great asset in understanding the extent and distribution of fish mercury in this area.

In this study, Northeastern Minnesota was divided into groups of geologically related lakes termed lake provinces. Ten lake provinces were identified in two distinct geomorphic areas; seven are in an area of bedrock controlled terrain and three are in an area of glacial deposition. In the bedrock controlled terrain, the size shape and orientation of the lakes are directly related to structural or compositional weakness in the underlying bedrock. Weathering and glacial erosion exploited these weaknesses to produce distinct patterns of lakes that were used to identify the provinces. In the areas of glacial deposition, the lake provinces were associated with the deposits left by three distinct glacial advances. When fish mercury was plotted against length of fish by province, the correlation coefficients increased significantly. The R squared values for the provinces ranged from .58 to .81 as compared to an R squared value of .32 for all samples from the study area plotted collectively. A map showing the distribution of mercury contamination in walleyes and northern pike was produced using the results from this ongoing study.