Temperature and substrate controls on intra‐annual variation in ecosystem respiration in two subarctic vegetation types

Abstract Arctic ecosystems are important in the context of climate change because they are expected to undergo the most rapid temperature increases, and could provide a globally significant release of CO 2 to the atmosphere from their extensive bulk soil organic carbon reserves. Understanding the relative contributions of bulk soil organic matter and plant‐associated carbon pools to ecosystem respiration is critical to predicting the response of arctic ecosystem net carbon balance to climate change. In this study, we determined the variation in ecosystem respiration rates from birch forest understory and heath tundra vegetation types in northern Sweden through a full annual cycle. We used a plant biomass removal treatment to differentiate bulk soil organic matter respiration from total ecosystem respiration in each vegetation type. Plant‐associated and bulk soil organic matter carbon pools each contributed significantly to ecosystem respiration during most phases of winter and summer in the two vegetation types. Ecosystem respiration rates through the year did not differ significantly between vegetation types despite substantial differences in biomass pools, soil depth and temperature regime. Most (76–92%) of the intra‐annual variation in ecosystem respiration rates from these two common mesic subarctic ecosystems was explained using a first‐order exponential equation relating respiration to substrate chemical quality and soil temperature. Removal of plants and their current year's litter significantly reduced the sensitivity of ecosystem respiration to intra‐annual variations in soil temperature for both vegetation types, indicating that respiration derived from recent plant carbon fixation was more temperature sensitive than respiration from bulk soil organic matter carbon stores. Accurate assessment of the potential for positive feedbacks from high‐latitude ecosystems to CO 2 ‐induced climate change will require the development of ecosystem‐level physiological models of net carbon exchange that differentiate the responses of major C pools, that account for effects of vegetation type, and that integrate over summer and winter seasons.