Simulated watershed mercury and nitrate flux responses to multiple land cover conversion scenarios

Abstract

Water quality and toxic exposure science is transitioning towards analysis of multiple stressors rather than one particular environmental concern (e.g., mercury) or a group of similarly reacting chemicals (e.g., nutrients). However, two of the most important water quality constituents affecting both human and ecosystem health today, reactive nitrogen (N(r) ) and methylmercury (MeHg), are often assessed separately for their independent effects on water quality. With the continued pressure of landscape modifications on water quality, a challenge remains in understanding the concurrent watershed flux response of both N(r) and MeHg to such physical stressors, particularly at the spatial scale (regional watersheds) and within the mixed land cover type systems that most decision-making processes are conducted. We simulate the annual average and monthly flux responses of Hg (MeHg and total mercury [HgT]), NO(3) -N, and runoff to four land cover change scenarios in the Haw River Watershed (NC, USA), a headwater system in the Cape Fear River Basin. Fluxes are simulated using a process-based, spatially explicit watershed Grid-Based Mercury Model (GBMM) and a NO(3) -N watershed flux model we developed to link to GBMM. Results suggest that annual NO(3) -N and Hg fluxes increase and decrease concomitantly to land cover change; however, the magnitude of the changes in NO(3) -N, MeHg, HgT, and water fluxes vary considerably between different land cover conversion scenarios. Converting pasture land to a suburbanized landscape elicited the greatest increase in runoff and MeHg, HgT, and NO(3) -N fluxes among all four conversion scenarios. Our findings provide insight for multi-stressor ecological exposure research and management of coastal eutrophication resulting from elevated N(r) loadings and exposure risk due to elevated concentrations of MeHg in fish tissue.

Pre-anthropocene mercury residues in North American freshwater fish

“…As noted, residue levels are strongly influenced by watershed and water body attributes, which can vary significantly between regions and among individual lakes and rivers within a region (Bradley et al ; Chasar et al ; Golden and Knightes ; USEPA ). One of these attributes is the areal extent of wetlands within a watershed, because wetlands are generally understood to increase the conversion of ionic Hg (Hg[II]), loaded from surrounding uplands and atmospheric deposition, to MeHg (Bradley et al ; Brigham et al ; Lorey and Driscoll ; Selvendiran, Driscoll, Bushey, et al ; Selvendiran, Driscoll, Montesdeoca, et al ).…”

“…These results further suggest that reductions in anthropogenic emissions on a global scale will not lead, with absolute certainty, to residues less than current regulatory criteria in any specific waterbody. As noted, residue levels are strongly influenced by watershed and water body attributes, which can vary significantly between regions and among individual lakes and rivers within a region (Bradley et al 2013;Chasar et al 2009;Golden and Knightes 2011;USEPA 1997). One of these attributes is the areal extent of wetlands within a watershed, because wetlands are generally understood to increase the conversion of ionic Hg (Hg[II]), loaded from surrounding uplands and atmospheric deposition, to MeHg (Bradley et al 2013;Brigham et al 2009;Lorey and Driscoll 1999;.…”

Characterizing mercury concentrations and fluxes in a Coastal Plain watershed: Insights from dynamic modeling and data

“…Details of GBMM's equations and sensitivity analysis are previously published [ Golden and Knightes , 2011]. GBMM has been applied and validated in multiple settings in the southeastern United States, including the Piedmont Physiographic Province [ Golden and Knightes , 2011; Golden et al , 2010] and Coastal Plain Province [ Dai et al , 2005; Feaster et al , 2010]. Final input parameters for McTier Creek watershed are found elsewhere for the hydrology module [ Feaster et al , 2010].…”

Climate change and watershed mercury export: a multiple projection and model analysis

“…The mass balance of Hg at the watershed outlet is estimated using the equations where C s is the concentration of Hg in watershed soils (µg m −3 ); L is the Hg load (µg d −1 ); L p is the Hg atmospheric deposition load on pervious land (µg d −1 ); L f is the Hg atmospheric deposition load on forest land (µg d −1 ); L d is the litter decomposition Hg load on forestland (µg/d); K r is the reduction rate constant (d −1 ), where reduced Hg is assumed to immediately volatilize and is considered a loss from the watershed; K l is the leaching loss constant (d −1 ); K ro is the runoff loss constant (d −1 ); K e is the erosion loss constant (d −1 ); V s is the watershed soil volume (m 3 ); A c is the grid area (m 2 ); and z d is the watershed soil mixing depth (m). A full description of GBMM's equations and sensitivity analysis , final input parameters for McTier Creek Watershed for the hydrology module , and parameters for the sediment and Hg modules in McTier Creek are detailed elsewhere.…”