Soil mercury and its response to atmospheric mercury deposition across the northeastern United States

Abstract

Abstract. Terrestrial soil is a large reservoir of atmospherically deposited mercury (Hg). However, few studies have evaluated the accumulation of Hg in terrestrial ecosystems in the northeastern United States, a region which is sensitive to atmospheric Hg deposition. We characterized Hg and organic matter in soil profiles from 139 sampling sites for five subregions across the northeastern United States and estimated atmospheric Hg deposition to these sites by combining numerical modeling with experimental data from the literature. We did not observe any significant relationships between current net atmospheric Hg deposition and soil Hg concentrations or pools, even though soils are a net sink for Hg inputs. Soil Hg appears to be preserved relative to organic carbon (OC) and/or nitrogen (N) in the soil matrix, as a significant negative relationship was observed between the ratios of Hg/OC and OC/N (r ¼ 0.54, P , 0.0001) that shapes the horizonal distribution patterns. We estimated that atmospheric Hg deposition since 1850 (3.97 mg/m 2 ) accounts for 102% of the Hg pool in the organic horizons (3.88 mg/m 2 ) and 19% of the total soil Hg pool (21.32 mg/m 2 ), except for the southern New England (SNE) subregion. The mean residence time for soil Hg was estimated to be 1800 years, except SNE which was 800 years. These patterns suggest that in addition to atmospheric deposition, the accumulation of soil Hg is linked to the mineral diagenetic and soil development processes in the region.

Association of dissolved mercury with dissolved organic carbon in U.S. rivers and streams: The role of watershed soil organic carbon

“…Dry deposition was not included due to a lack of nationally available high‐resolution data. Dry deposition may exceed wet deposition in some regions [ Schroeder and Munthe , ; Sakata et al ., ; Graydon et al ., ; Risch et al ., ; Yu et al ., ] and could potentially be a significant factor affecting Hg availability in such settings. However, when modeled dry deposition was included for five of the sites, Riscassi and Scanlon [] found that the Hg D :DOC ratio actually declined as total deposition increased.…”

“…It does not appear that recent wet deposition controls the spatial variability of the stream Hg D :DOC ratio between sites, but it is important to note that MDN data does not reflect historical atmospheric Hg loading to the soil. Previous studies have reported a disconnect between recent atmospheric Hg deposition and Hg in streams [ Hintelmann et al ., ; Brigham et al ., ; Oswald et al ., ], lakes [ Harris et al ., ], and soils [ Obrist et al ., ; Yu et al ., ]. This suggests a time lag between the deposition of Hg to terrestrial watersheds and its transport to streams or a possible loss of Hg during transport (e.g., volatilization).…”

“…The emergent Hg D :DOC ratios found in streams are the product of complex interactions associated with the cycling of carbon and Hg in soils, including decomposition, volatilization, and translocation. Previous research has characterized these specific processes through direct sampling of watershed soils [ Hintelmann et al ., ; Demers et al ., ; Selvendiran et al ., ; Obrist et al ., ; Demers et al ., ; Oswald et al ., ; Yu et al ., ], but a generalized, process‐based theory that can be applied across a diverse range of watersheds remains a current research priority. While the present analysis may place some constraints on such a theory, process‐based models that explicitly consider the timescales associated with carbon and Hg cycling in watershed soils are ultimately desired.…”

“…Observational data sets of how pools of soil Hg scale with SOC are relatively scarce in the literature. A recent study characterized soils pools throughout the northeastern United States [ Yu et al ., ] from measurements of Hg and SOC concentrations in the O and B horizons. Based on data from 122 soil pits (raw data from X. Yu (personal communication, 2014)), soil Hg pools (ng m −2 ) scale with SOC pools (mg m −2 ) to the 0.57 (±0.16) power (r 2 = 0.33), which suggests nonlinearity consistent with our theoretical expectations.…”

Using foliar and forest floor mercury concentrations to assess spatial patterns of mercury deposition

“…Obrist et al (2011) hypothesized that this relationship is likely due to differences from centuries of Hg inputs known as "legacy" Hg. However, because the Adirondacks is a region with relatively uniform soil age, disturbance history, and Hg deposition history, it would appear that decomposition processes and nutrient status may be regulating this pattern, and Hg enrichment within the forest floor profile seems to be driven more by ecosystem processes than by historic deposition patterns (Yu et al, 2014).…”