Nitrate Controls Methyl Mercury Production in a Streambed Bioreactor

Abstract

Organic carbon bioreactors provide low-cost, passive treatment of a variety of environmental contaminants but can have undesirable side effects in some cases. This study examines the production of methyl mercury (MeHg) in a streambed bioreactor consisting of 40 m³ of wood chips and designed to treat nitrate (NO₃) in an agricultural drainage ditch in southern Ontario (Avon site). The reactor provides 30 to 100% removal of NO₃-N concentrations of 0.6 to 4.4 mg L(-1), but sulfate (SO₄(2-)) reducing conditions develop when NO₃ removal is complete. Sulfate reducing conditions are known to stimulation the production of MeHg in natural wetlands. Over one seasonal cycle, effluent MeHg ranged from 0.01 to 0.76 ng L(-1) and total Hg ranged from 1.3 to 3.4 ng L(-1). During all sampling events when reducing conditions were only sufficient to promote NO₃(-) reduction (or denitrification) ( = 5, late fall 2009, winter 2010), MeHg concentrations decreased in the reactor and it was a net sink for MeHg (mean flux of -5.1 μg m(-2) yr(-1)). During all sampling events when SO₄(2-) reducing conditions were present ( = 6, early fall 2009, spring 2010), MeHg concentrations increased in the reactor and it was a strong source of MeHg to the stream (mean flux of 15.2 μg m(-2) yr(-1)). Total Hg was consistently removed in the reactor (10 of 11 sampling events) and was correlated to the total suspended sediment load ( r² = 0.69), which was removed in the reactor by physical filtration. This study shows that organic carbon bioreactors can be a strong source of MeHg production when SO₄(2-) reducing conditions develop; however, maintaining NO₃-N concentrations > 0.5 mg L suppresses the production of MeHg.

Watershed features and stream water quality: Gaining insight through path analysis in a Midwest urban landscape, U.S.A.

“…However, intermixing of urban land and crop land is typical for cities in the Midwest Cornbelt region, and this analysis provides evidence that streams in this landscape may receive greater amounts of pollutants from agricultural than from urban sources (Carpenter et al, 1998;Dinnes et al, 2002; , 1999;Herringshaw et al, 2011). Attention to the potential negative effects of elevated nitrogen levels are important and several strategies, as among them vegetated buffer strips (e.g., Osborne & Kovacic, 1993;Schultz, Isenhart, Simpkins, & Colletti, 2004) or nitrogen bioreactors (e.g., Christianson et al, 2012;Shih, Robertson, Schiff, & Rudolph, 2011) could be adapted for these settings.…”

Nitrate and phosphate removal from agricultural subsurface drainage using laboratory woodchip bioreactors and recycled steel byproduct filters

“…Researchers also suggest maintaining nitrate concentration above 0.5 mg N/L in the bioreactor to avoid extreme reducing conditions which could promote sulfate reduction and methyl mercury production (Shih et al, 2011;Cooke and Bell, 2014). However, these design considerations may lead to low nitrate removal efficiency and high nitrite accumulation.…”

Internal hydraulics of an agricultural drainage denitrification bioreactor

“…Even attempts to artificially increase temperatures within a bioreactor (e.g., passive solar heating, Cameron and Schipper, 2011), may not be sufficient to overcome low retention times inherent to spring flow conditions. Recently, concern has been raised about the potential for denitrification systems experiencing SO 4 2− reducing conditions to produce methylated mercury (Hudson and Cooke, 2011;Shih et al, 2011). Shih et al (2011) recommended maintaining at least 0.5 mg L −1 NO 3 − -N in the effluent to minimize this concern.…”

“…Recently, concern has been raised about the potential for denitrification systems experiencing SO 4 2− reducing conditions to produce methylated mercury (Hudson and Cooke, 2011;Shih et al, 2011). Shih et al (2011) recommended maintaining at least 0.5 mg L −1 NO 3 − -N in the effluent to minimize this concern. Here, the effluent NO 3 − -N concentration was reduced to below this level on the August 24 sample event with a corresponding decrease in SO 4 2− concentration across the bioreactor (Fig.…”