Search USGS⌕ Search

Geology topics

S.J. Nelson

Publications and source records attributed to S.J. Nelson.

4 recordsLinked to original sources

Snow refugia: Managing temperate forest canopies to maintain winter conditions

Climate change is reducing snowpack across temperate regions with negative consequences for human and natural systems. Because forest canopies create microclimates that preserve snowpack, managing forests to support snow refugia—defined here as areas that remain relatively buffered from contemporary climate change over time that sustain snow quality, quantity, and/or timing appropriate to the landscape—could reduce climate change impacts on snow cover, sustaining the benefits of snow. We review the current understanding of how forest canopies affect snow, finding that while closed-conifer forests and snow interactions have been extensively studied in western North America, there are knowledge gaps for deciduous and mixed forests with dormant season leaf loss. We propose that there is an optimal, intermediate zone along a gradient of dormant season canopy cover (DSCC; the proportion of the ground area covered by the canopy during the dormant season), where peak snowpack depth and the potential for snow refugia will be greatest because the canopy-mediated effects of snowpack sheltering (which can preserve snowpack) outweigh those of snowfall interception (which can limit snowpack). As an initial test of our hypothesis, we leveraged snowpack measurements in the northeastern United States spanning the DSCC gradient (low, <25% DSCC; medium, 25%–50% DSCC; and high, >50% DSCC), including from 2 sites in Old Town, Maine; 12 sites in Acadia National Park, Maine; and 30 sites in the northern White Mountains of New Hampshire. Medium DSCC forests (typically mature mixed coniferous–deciduous forests) exhibited the deepest peak snowpacks, likely due to reduced snowfall interception compared to high DSCC forests and reduced snowpack loss compared to low DSCC forests. Many snow accumulation or snowpack studies focus on the contrast between coniferous and open sites, but our results indicate a need for enhanced focus on mixed canopy sites that could serve as snow refugia. Measurements of snowpack depth and timing across a wider range of forest canopies would advance understanding of canopy–snow interactions, expand the monitoring of changing winters, and support management of forests and snow-dependent species in the face of climate change.

Ecosphere↗

A comparison of winter mercury accumulation at forested and no-canopy sites measured with different snow sampling techniques

Atmospheric mercury (Hg) is delivered to ecosystems via rain, snow, cloud/fog, and dry deposition. The importance of snow, especially snow that has passed through the forest canopy (throughfall), in delivering Hg to terrestrial ecosystems has received little attention in the literature. The snowpack is a dynamic system that links atmospheric deposition and ecosystem cycling through deposition and emission of deposited Hg. To examine the magnitude of Hg delivery via snowfall, and to illuminate processes affecting Hg flux to catchments during winter (cold season), Hg in snow in no-canopy areas and under forest canopies measured with four collection methods were compared: (1) Hg in wet precipitation as measured by the Mercury Deposition Network (MDN) for the site in Acadia National Park, Maine, USA, (2) event throughfall (collected after snowfall cessation for accumulations of >8 cm), (3) season-long throughfall collected using the same apparatus for event sampling but deployed for the entire cold season, and (4) snowpack sampling. Estimates (mean ± SE) of Hg deposition using these methods during the 91-day cold season in 2004–2005 at conifer sites showed that season-long throughfall Hg flux (1.80 μg/m 2 ) < snowpack Hg (2.38 ± 0.68 μg/m 2 ) < event throughfall flux (5.63 ± 0.38 μg/m 2 ). Mercury deposition at the MDN site (0.91 μg/m 2 ) was similar to that measured at other no-canopy sites in the area using the other methods, but was 3.4 times less than was measured under conifer canopies using the event sampling regime. This indicates that snow accumulated under the forest canopy received Hg from the overstory or exhibited less re-emission of Hg deposited in snow relative to open areas. The soil surface of field-scale plots were sprayed with a natural rain water sample that contained an Hg tracer ( 202 Hg) just prior to the first snowfall to explore whether some snowpack Hg might be explained from soil emissions. The appearance of the 202 Hg tracer in the snowpack (0–64% of the total Hg mass in the snowpack) suggests that movement of Hg from the soil into the snowpack is possible. However, as with any tracer study the 202 Hg tracer may not precisely represent the reactivity and mobility of natural Hg in soils.

Applied Geochemistry↗

Landscape controls on mercury in streamwater at Acadia National Park, USA

Fall and spring streamwater samples were analyzed for total mercury (Hg) and major ions from 47 locations on Mount Desert Island in Maine. Samples were collected in zones that were burned in a major wildfire in 1947 and in zones that were not burned. We hypothesized that Hg concentrations in streamwater would be higher from unburned sites than burned watersheds, because fire would volatilize stored Hg. The Hg concentrations, based on burn history, were not statistically distinct. However, significant statistical associations were noted between Hg and the amount of wetlands in the drainage systems and with streamwater dissolved organic carbon (DOC). An unexpected result was that wetlands mobilized more Hg by generating more DOC in total, but upland DOC was more efficient at transporting Hg because it transports more Hg per unit DOC. Mercury concentrations were higher in samples collected at lower elevations. Mercury was positively correlated with relative discharge, although this effect was not distinguished from the DOC association. In this research, sample site elevation and the presence of upstream wetlands and their associated DOC affected Hg concentrations more strongly than burn history. ?? Springer Science + Business Media B.V. 2007.

Environmental Monitoring and Assessment↗

Mass balances of mercury and nitrogen in burned and unburned forested watersheds at Acadia National Park, Maine, USA

Precipitation and streamwater samples were collected from 16 November 1999 to 17 November 2000 in two watersheds at Acadia National Park, Maine, and analyzed for mercury (Hg) and dissolved inorganic nitrogen (DIN, nitrate plus ammonium). Cadillac Brook watershed burned in a 1947 fire that destroyed vegetation and soil organic matter. We hypothesized that Hg deposition would be higher at Hadlock Brook (the reference watershed, 10.2 ??g/m2/year) than Cadillac (9.4 ??g/m2/year) because of the greater scavenging efficiency of the softwood vegetation in Hadlock. We also hypothesized the Hg and DIN export from Cadillac Brook would be lower than Hadlock Brook because of elemental volatilization during the fire, along with subsequently lower rates of atmospheric deposition in a watershed with abundant bare soil and bedrock, and regenerating vegetation. Consistent with these hypotheses, Hg export was lower from Cadillac Brook watershed (0.4 ??g/m2/year) than from Hadlock Brook watershed (1.3 ??g/m2/year). DIN export from Cadillac Brook (11.5 eq/ ha/year) was lower than Hadlock Brook (92.5 eq/ha/year). These data show that ??50 years following a wildfire there was lower atmospheric deposition due to changes in forest species composition, lower soil pools, and greater ecosystem retention for both Hg and DIN. ?? Springer Science + Business Media B.V. 2006.

Environmental Monitoring and Assessment↗