B21C-0063:
Sensing winter soil respiration dynamics in near-real time

Tuesday, 16 December 2014
Alexandra Contosta1, Elizabeth A Burakowski2, Ruth K Varner1 and Serita D Frey3, (1)University of New Hampshire (UNH), Institute for the Study of Earth, Oceans, and Space (EOS), Durham, NH, United States, (2)National Center for Atmospheric Research, Boulder, CO, United States, (3)University of New Hampshire (UNH), Department of Natural Resources and the Environment, Durham, NH, United States
Abstract:
Some of the largest reductions in seasonal snow cover are projected to occur in temperate latitudes. Limited measurements from these ecosystems indicate that winter soil respiration releases as much as 30% of carbon fixed during the previous growing season. This respiration is possible with a snowpack that insulates soil from ambient fluctuations in climate. However, relationships among snowpack, soil temperature, soil moisture, and winter soil respiration in temperate regions are not well-understood. Most studies have infrequently sampled soil respiration and its drivers, and most measurements have been limited to the soil surface. We made near-real time, continuous measurements of temperature, moisture, and CO2 fluxes from the soil profile, through the snowpack, and into the atmosphere in a deciduous forest of New Hampshire, USA. We coupled these data with daily sampling of snow depth and snow water equivalent (SWE). Our objectives were to continuously measure soil CO2 production (Psoil) and CO2 flux through the snowpack (Fsnow) and to compare Fsnow and Psoil with environmental drivers. We found that Fsnow was more dynamic than Psoil, changing as much as 30% over several days with shifting environmental conditions. Multiple regression indicated that SWE, air temperature, surface soil temperature, surface soil CO2 concentrations, and soil moisture at 15 cm were significant predictors of Fsnow. The transition of surface temperature from below to above 0°C was particularly important as it represented a phase change from ice to liquid water. Only air temperature and soil moisture at 15 cm were significant drivers of Psoil, where higher moisture at 15 cm resulted in lower Psoil rates. Time series analysis showed that Fsnow lagged 40 days behind Psoil. This lag may be due to slow CO2 diffusion through soil to overlying snow under high moisture conditions. Our results suggest that surface soil CO2 losses are driven by rapid changes in snow cover, surface temperature, and surface moisture while winter soil CO2 production is driven by subsurface moisture conditions. Our finding that subsurface moisture regulated winter soil respiration demonstrates the transformative power of sensors; winter measurements of soil CO2 concentrations and moisture levels would not be possible with traditional methods.