Process-based isotope partitioning of winter soil respiration in a subalpine ecosystem reveals importance of rhizospheric respiration

Abstract

Deep snow in sub-alpine ecosystems may reduce or eliminate soil freezing, thus contributing to the potential for winter soil respiration to account for a significant fraction of annual CO2 efflux to the atmosphere. Quantification of carbon loss from soils requires separation of respiration produced by roots and rhizosphere organisms from that produced by heterotrophic, decomposer organisms because the former does not result in a net loss of stored carbon. Our objective was to quantify winter soil respiration rates in a sub-alpine forest and meadow, and to partition that flux into its rhizosphere and heterotrophic components. We were particularly interested in comparing early winter soil respiration to late winter/early spring soil respiration of each component because previous work has shown a consistent increase in soil respiration of subalpine systems from early winter to late winter/spring. Field data on the total soil CO2 flux and its carbon isotope composition were coupled with data from laboratory incubations using a novel process-based stable isotope mixing model implemented in a hierarchical Bayesian framework. We found that soil respiration generally increased from early to later winter and was greatest mid-summer. After correcting for the effect of wind on snowpack δ13C–CO2, the δ13C of soil-respired CO2 varied little over winter, and the contributions of rhizospheric (~35 %) and heterotrophic (~65 %) respiration were relatively constant. The significance of winter respiration from the rhizosphere and apparent coupling of increases in rhizospheric and heterotrophic respiration in late winter are likely to be important for predicting changes in soil carbon in sub-alpine ecosystems.