Impacts of a reduced winter snowpack on soil and stem carbon dioxide fluxes in a temperate hardwood forest

Northern forests typically have a continuous snowpack for much of the winter, which insulates the underlying soil from sub-freezing air temperatures. However, winters with a late-developing or intermittent snowpack occasionally occur and may result in colder soil temperatures and a greater frequency and severity of soil frost. Such variability in winter climate can affect rates of soil respiration and may contribute to interannual variability in biosphere-atmosphere carbon (C) exchange. Climate models predict a reduction in snowpack depth and duration by the end of the 21st century in the northeastern U.S. The reduction in winter snowpack may have important implications for ecosystem C fluxes. While past research has contributed significantly to our understanding of the terrestrial C cycle, linkages between above- and belowground processes and their coupled response to changes in winter climate remain uncertain. We are conducting a snow removal experiment to quantify the impacts of a reduced winter snowpack on forest-atmosphere C exchange in mixed stands of red maple (Acer rubrum) and red oak (Quercus rubra) at Harvard Forest in Petersham, MA. Snow was removed for the first 5 weeks of winter and ecosystem C losses (i.e. from soil respiration and stem CO2 efflux) and uptake (i.e. canopy C exchange) were quantified in each plot. Here we report on ecosystem C losses. During the first year of this multi-year study snow removal increased the depth and duration of soil frost, increased the frequency of freeze-thaw cycles, and impeded soil warming in the spring. These changes in soil frost dynamics tended to increase ecosystem C losses from bulk soil respiration. Greater root + rhizosphere respiration, rather than changes in heterotrophic soil respiration, appears to be driving this increase in bulk soil respiration. Red oaks experiencing soil frost tended to have lower rates of stem CO2 efflux during the fall. In contrast, stem CO2 efflux from red maples experiencing soil frost tended to be lower during the growing season but these differences decreased during the fall. Our results suggest that soil frost has a stronger impact on root + rhizosphere respiration than heterotrophic soil respiration. Therefore, greater belowground C allocation might be driving the observed increase in CO2 losses from bulk soil respiration in response to soil frost. Furthermore, the response of stem CO2 efflux to soil frost appears to vary by tree species, highlighting the importance of understanding species specific responses to environmental perturbations.