The Torneträsk System - A basis for predicting future subarctic ecosystems

Arctic and subarctic areas have experienced a rapid warming and substantial increases in precipitation in recent decades. The frequency and intensity of some extreme events, such as fires, winter warming events, extreme rainfall, and droughts, has also increased. These climatic changes and other anthropogenic factors have caused profound changes in arctic and subarctic ecosystems with important implications for the local residents and for the global population, which are likely to exacerbate under the predicted climate change scenarios. Thus, a better understanding of potential future ecosystem changes is paramount for defining climate change mitigation goals and adaptation strategies.Dynamic ecosystem models are powerful tools to study the influences of climatic and other drivers on ecosystem processes. Nevertheless, predictions of ecosystem changes still hold large uncertainties, arising mostly from insufficient observational data, lack of process understanding, difficulties in quantifying the effects of different ecosystem processes and their interactions, and/or model limitations in representing these interacting processes. The Torneträsk area, in the Swedish subarctic, has an unrivalled history of environmental observations spanning over 100 years, and is one of the most studied sites in the Arctic. The area has undergone substantial climatic and ecosystem changes. By studying its rapidly-transforming ecosystems we can obtain critically important information needed to improve our understanding and predictions of future ecosystem changes at a larger scale. This thesis summarized and ranked the direct and indirect drivers of ecosystem change in the Torneträsk area, and proposed research priorities to improve predictions of ecosystem change. Winter warming events (WWEs) were the top-ranked research priority. Hence, this thesis further examined the impacts of WWEs on subarctic ecosystems using monitoring data, manipulation experiments and modelling. The monitoring and manipulation data suggest an increasingly strong warming effect of WWEs on permafrost, especially rain on snow events occurrying in the presence of thick snowpacks. The modeling experiments in LPJ-GUESS indicated a strong cooling effect of WWEs on ground temperature, driven mostly by changes in snow insulation, which resulted in profound changes in the biogeochemical fluxes of magnitudes comparable to long-term climatic changes. We identified several modeling gaps that may explain the mismatch between the model- and the observational-based impacts of WWEs on ground temperatures, including 1) the lack of surface energy balance in LPJ-GUESS, the model’s daily timestep that neglects sub-daily freeze-thaw cycles within the snowpack, and 3) the model’s simplistic water retention scheme that minimizes the amount of water retained in the snowpack and hence the amount of latent heat release upon freezing. Addressing these issues is paramount for accurately estimating future ecosystem changes and their inplications on the arctic’s carbon balance. Climate change, including long term changes and short-lasting events such as WWEs, affected lowland permafrost sites in the Torneträsk area differently, depending on the site-specific climatíc and environmental conditions. This resulted in permafrost thaw rates decreasing Eastwards. This thesis revealed, through metagenomic sequencing and greenhouse gas measurements in three peatlands across this thaw gradient, that different rates and stages of permafrost degradation influence greenhouse gas exchange through an altered taxonomic structure and function of the microbial communities. This highlights the need for expanding the monitoring of peatland fluxes and microbial dynamics that is currently based on very few sites