Soil respiration across a permafrost transition zone: spatial structure and environmental correlates

• Published in: Biogeosciences
• Main Authors: Stegen, James C., Anderson, Carolyn G., Bond-Lamberty, Ben, Crump, Alex R., Chen, Xingyuan, Hess, Nancy
• Format: Article in Journal/Newspaper
• Language: English
• Published: Copernicus Publications 2017
• Subjects: article; Verlagsveröffentlichung; permafrost
• Online Access: https://doi.org/10.5194/bg-14-4341-2017; https://noa.gwlb.de/receive/cop_mods_00008505; https://noa.gwlb.de/servlets/MCRFileNodeServlet/cop_derivate_00008462/bg-14-4341-2017.pdf; https://bg.copernicus.org/articles/14/4341/2017/bg-14-4341-2017.pdf

Soil respiration is a key ecosystem function whereby shifts in respiration rates can shift systems from carbon sinks to sources. Soil respiration in permafrost-associated systems is particularly important given climate change driven permafrost thaw that leads to significant uncertainty in resulting ecosystem carbon dynamics. Here we characterize the spatial structure and environmental drivers of soil respiration across a permafrost transition zone. We find that soil respiration is characterized by a non-linear threshold that occurs at active-layer depths greater than 140 cm. We also find that within each season, tree basal area is a dominant driver of soil respiration regardless of spatial scale, but only in spatial domains with significant spatial variability in basal area. Our analyses further show that spatial variation (the coefficient of variation) and mean-variance power-law scaling of soil respiration in our boreal system are consistent with previous work in other ecosystems (e.g., tropical forests) and in population ecology, respectively. Comparing our results to those in other ecosystems suggests that temporally stable features such as tree-stand structure are often primary drivers of spatial variation in soil respiration. If so, this provides an opportunity to better estimate the magnitude and spatial variation in soil respiration through remote sensing. Combining such an approach with broader knowledge of thresholding behavior – here related to active layer depth – would provide empirical constraints on models aimed at predicting ecosystem responses to ongoing permafrost thaw.