0 u ) Mercury and Organic Carbon Relationships in Streams Draining Forested Upland / Peatland Watersheds

We determined the fluxes of total mercury (HgT), total organic carbon (TOC), and dissolved organic carbon (DOC) from five upland/ peatland watersheds at the watershed outlet. The difference between TOC and DOC was defined as particulate OC (POC). Concentrations of HgT showed moderate to strong relationships with POC (I?* = 0.77) when ah watersheds were grouped. Although POC only accounts for 10 to 20% of the OC transported, we estimate that it is associated with 52 to 80% of the HgT transported from the five watersheds. Total transport of HgT from the watersheds ranged from 0.70 to 2.82 pg rn-’ yr-‘. Watershed geometry and hydrology play important roles in determining the intluence of OC on HgT transport in forested watersheds. Watershed properties such as peatland area have considerable promise as predictors for estimating HgT transport in streams draining forested watersheds in the Great Lakes States. T HE fate and transport of total mercury (HgT) in terrestrial environments has been little studied and poorly understood. A large knowledge gap currently exists regarding the transport of HgT in and through the terrestrial environment, even though it is widely recognized that terrestrial transport is important in determining loading of atmospherically deposited HgT to the aquatic environment (Lindqvist, 1991). In most watersheds in northern latitudes, peatlands play an important role in cycling of HgT and organic carbon (OC) (Driscoll et al., 1994). In soil systems, HgT and soil organic matter are closely related (Grigal et al., 1994; Roulet and Lucotte, 1995) as are HgT and OC in soil solution (Aastrup et al., 1991). The link continues to the aquatic system. Total Hg is positively correlated with OC in stream/runoff waters (Hurley et al., 1995; Lee and Iverfeldt, 1991; Johansson et al., 1991; Johansson and Iverfeldt, 1994) and in lake waters (Meili et al., 1991; Lee and Iverfeldt, 1991; Driscoll et al., 1994; Driscoll et al., 1995; Sorensen et al., 1990). It is therefore apparent that OC plays an important role in the watershed cycling of HgT, the hydrological pathways that control the terrestrial transport of OC also influence the terrestrial transport and cycling of HgT (Kolka, 1996). If a close relationship between OC and HgT in solution exists and can be quantified, then R.K. Kolka, Dep. of Forestry, Univ. of Kentucky, Lexington, KY 40546; D.F. Grigal and E.A. Nater, Dep. of Soil, Water, and Climate, Univ. of Minnesota, St. Paul, MN 55108; and E.S. Verry, USDA Forest Service, North Central Forest Exp. Stn., Grand Rapids, MN 55744. Received 23 Apr. 1998. *Corresponding author (rkolk2@ pop.uky.edu). Published in J. Environ. Qual. 28:766-775 (1999). predictions of sites and processes associated with enhanced HgT transport will be possible. Watershed parameters have also shown promise in predicting HgT fluxes and concentrations. Hurley et al. (1995) found correlations with watershed land use (urban, wetland, forest, and agriculture) and HgT for 39 Wisconsin rivers. Driscoll et al. (1995) found that the percentage of near-shore wetlands in a watershed influences HgT concentration in lakes of the Adirondack region of New York. Commonly wetlands, especially peatlands, hold large reserves of OC. Considering wetlands are the source of flow to many surface waters, OC transported from peatlands likely influences HgT transport to lakes and streams. Separating the influence that wetlands and associated uplands have on total watershed cycling of HgT and OC would allow watershed parameters such as wetland area, wetland type, upland/ wetland area ratio, or other physically based watershed metrics to be used to predict HgT loading to surface waters. The objective of this study was to compare and contrast the relationships between HgT and OC concentrations and fluxes from stream waters draining five forested watersheds and assess if watershed parameters influence this relationship. MATERIALS AND METHODS