Print PageOne of the principal predictions of climate change models is an increase in extreme weather events, including heavy precipitation and exacerbated drought severity. Because these events could lead to dramatic changes in snowpack and soil moisture in alpine, Arctic, and Antarctic ecosystems, it is crucial to understand how spatial variation in these parameters can affect other soil properties. The McMurdo Dry Valleys Long Term Ecological Research (LTER) site in Antarctica provides an ideal opportunity for this type of study, since extreme conditions constrain life to simple communities whose interactions with abiotic soil properties are more easily interpretable. We implemented a spatially-explicit sampling scheme that utilized both sides (exposed and subnivian) of an 8-year-old snowfence manipulation in the Taylor Valley of Antarctica. We sampled 89 points at pairwise distances of 25 cm to 16 meters for snow depth, soil moisture, temperature, nitrates, salinity, C:N ratios, pH, and soil respiration. Our research questions are as follows: 1) What is the degree of spatial heterogeneity in soil chemistry and respiration in subnivian compared to adjacent exposed soils? 2) How does fine-scale variation in snow depth influence chemistry and respiration of soils? 3) How does small-scale variation in soil chemistry influence soil respiration?
Results/Conclusions
We employed geostatistics to analyze spatial variability by fitting the data into semivariograms. For all of our analyses, the nuggets were 0, indicating negligible sampling error. Spatial variability (calculated as the sill) of pH and inorganic nitrogen were higher in the exposed site, but variability of total carbon was the same for both sites. The following soil properties were more variable in subnivian soils: snow depth, soil moisture, CO2 flux, total nitrogen, C:N, and conductivity. For the majority of the factors measured, the subnivian site appeared to confer greater spatial variability of soil properties than the exposed site. We found evidence of a positive linear correlation between snow depth and soil moisture. Additionally, average pH at exposed and subnivian sites were both high (9.483 and 9.598, respectively). Mean C:N ratios were also similar at both sites (exposed 9.113; subnivian 9.034). These results will help us better understand how future alteration in snowpack from climate change will impact soil biogeochemistry and respiration; furthermore, they provide a much-improved framework for the design of future studies, ensuring the most efficient sampling possible to maximize independent comparisons between points.
