Abstract
Controls on mercury bioaccumulation in lotic ecosystems are not well understood. During 2007-2009, we studied mercury and stable isotope spatial patterns of macroinvertebrates and fishes from two medium-sized (<80 km(2)) forested basins in contrasting settings. Samples were collected seasonally from multiple sites across the Fishing Brook basin (FB(NY)), in New York's Adirondack Mountains, and the McTier Creek basin (MC(SC)), in South Carolina's Coastal Plain. Mean methylmercury (MeHg) concentrations within macroinvertebrate feeding groups, and mean total mercury (THg) concentrations within most fish feeding groups were similar between the two regions. However, mean THg concentrations in game fish and forage fish, overall, were much lower in FB(NY) (1300 and 590 ng/g dw, respectively) than in MC(SC) (2300 and 780 ng/g dw, respectively), due to lower trophic positions of these groups from FB(NY) (means 3.3 and 2.7, respectively) than MC(SC) (means 3.7 and 3.3, respectively). Much larger spatial variation in topography and water chemistry across FB(NY) contributed to greater spatial variation in biotic Hg and positive correlations with dissolved MeHg and organic carbon in streamwater. Hydrologic transport distance (HTD) was negatively correlated with biotic Hg across FB(NY), and was a better predictor than wetland density. The small range of landscape conditions across MC(SC) resulted in no consistent spatial patterns, and no discernable correspondence with local-scale environmental factors. This study demonstrates the importance of local-scale environmental factors to mercury bioaccumulation in topographically heterogeneous landscapes, and provides evidence that food-chain length can be an important predictor of broad-scale differences in Hg bioaccumulation among streams.