Decade of permafrost thaw in a subarctic palsa mire alters carbon fluxes without affecting net carbon balance

Snow depth increases observed in some artic regions and its insulations effects have led to a winter-warming of permafrost-containing peatlands. Permafrost thaw and the temperature-dependent decomposition of previously frozen carbon (C) is currently considered as one of the most important feedbacks between the artic and the global climate system. However, the magnitude of this feedback remains uncertain because winter effects are rarely integrated and predicted from mechanisms active in both surface (young) and thawing deep (old) peat layers.

Laboratory incubation studies of permafrost soils, in situ carbon flux measurements in ecosystem-scale permafrost thaw experiments, or measurements made across naturally degrading permafrost gradients have been used to improve our knowledge about the net effects of winter-warming in permafrost C storage. The results from these studies, however, are biased by imprecision in long-term (decadal to millennial) effects due to the short time scale of the experiments. Gradient studies may show longer-term responses but suffer from uncertainties because measurements are usually taken during the summer, thus ignoring the long cold season. The need for robust estimates of the long-term effect of permafrost thaw on the net C balance, which integrates year-round C fluxes sets the basis of this study.

Here, we quantified the effects of long-term in situ permafrost thaw in the net C balance of a permafrost-containing peatland subjected to a 10-years snow manipulation experiment. In short, we used a peat age modelling approach to quantify the effect of winter-warming on net ecosystem production as well as on the underlying changes in surface C inputs and losses along the whole peat continuum. Contrary to our hypothesis, winter-warming did not affect the net ecosystem production regardless of the increased old C losses. This minimum overall effect is due to the strong reduction on the young C losses from the upper active layer associated to the new water saturated conditions and the decline in bryophytes. Our findings highlight the need to incorporate long-term year-round responses in C fluxes when estimating the net effect of winter-warming on permafrost C storage. We also demonstrate that thaw-induced changes in moisture conditions and plant communities are key factors to predicting future climate change feedbacks between the artic soil C pool and the global climate system.