Warming-induced contrasts in snow depth drive the future trajectory of soil carbon loss across the Arctic-Boreal region

Abstract

The Arctic-Boreal region is projected to experience spatially divergent trends in snow depth following climate change. However, the impact of these spatial trends has remained largely unexplored, despite potentially large consequences for the carbon cycle. To address this knowledge gap, we forced a customised arctic version of the dynamic vegetation model LPJ-GUESS with daily CMIP6 outputs from a global climate model (MRI-ESM2-0) under three climate scenarios. We find that snow depths increased the most in the coldest, northernmost regions, insulating the soil, which led to increased heterotrophic respiration and reduced carbon residence times. We emphasise the need for improved projections of future snow depth - in particular diverging trends across landscapes - to more accurately simulate the strength of Arctic-Boreal carbon feedbacks and their impact on global climate. In a warming climate, snow depth increases fastest in the coldest Arctic regions, which leads to enhanced respiration and reduced carbon residence times compared to less cold areas, suggest simulations of climate-change scenarios with an Arctic-specific vegetation model.