Streamwater Particulate Mercury and Suspended Sediment Dynamics in a Forested Headwater Catchment

Abstract

Due to the inherent differences in bioavailability and transport properties of particulate and dissolved mercury (HgP and HgD), it is important to understand the processes by which each is mobilized from soil to stream. Currently, there is a paucity of HgP data in the literature despite the fact that it can be the dominant fraction in some systems. We analyzed HgP in conjunction with volatile solids (VS, an estimate of organic content) and total suspended solids (TSS) and investigated the viability of using turbidity as a surrogate measure of HgP. Samples were collected for flow conditions ranging from 72 to 8,223 L s−1 during October 2009 through March 2010 in a 10.5-km2 forested headwater catchment. Total Hg concentrations ranged from 0.28 to 49.60 ng L−1, with the relative amount of HgP increasing with discharge from approximately 40% to 97%. Storm dynamics of HgP and HgD were not consistent, indicating unique controls on the export of each fraction. During high-flow events, HgP was consistently higher on the rising limb of the hydrograph compared with the receding limb for a range of discharge events, with this hysteresis contributing to a degraded relationship between HgP and streamflow. Overall, HgP was strongly positively correlated with VS (r 2 = 0.97), confirming the known association with organic carbon. Due to a consistent organic fraction of the suspended solids (34 ± 6%), HgP was also well correlated with TSS (r 2 = 0.95), with an average of 0.10 ng of HgP per milligram of TSS in this system. Stream turbidity measured with an in situ sonde also had a strong correlation with TSS (r 2 = 0.91), enabling commutative association with VS (r 2 = 0.86) and HgP (r 2 = 0.76). Turbidity can explain more than twice the temporal variance in HgP concentrations (n = 50) compared with discharge (r 2 = 0.76 versus 0.36), which leads to improved monitoring of HgP dynamics and quantification of mass fluxes.