Modelling wildfire and post-fire carbon budgets of a boreal forest under a changing climate

Understanding the dynamics of post-fire carbon recovery is critical for managing boreal forests under climate change. This study evaluates the LPJ-GUESS-BLAZE models’ performance in simulating vegetation structure, fire emissions, and post-fire carbon dynamics at the stand scale in a boreal forest in Sweden. Combining current climate and future scenarios (SSP1–2.6, SSP5–8.5), we investigated the impacts of post-fire management strategies, including no management and different reforestation options, on the forest carbon balance. Our results showed that the model accurately simulated unburnt vegetation structure. However, it overestimated fire carbon emissions compared to in-situ estimates. Post-fire carbon flux simulations revealed that avoiding salvage-logging of fire-surviving trees could result in the highest net carbon uptake in the near future (2019–2060). Reforestation strategies with a combination of conifer and broad-leaved trees consistently resulted in higher net C uptake than monoculture or mixed conifer options across all scenarios. Warming scenarios further accelerated carbon recovery, reducing the years needed to reach the carbon compensation point (CCP), i.e. the time when cumulative carbon uptake by reforestation balances the total carbon lost due to fire and post-fire emissions. Nevertheless, our study also revealed that the modelled rate of forest recovery was more rapid than observed. Our findings suggest that addressing specific model issues (e.g. fuel load and early stand development) can increase the robustness of modelled fire carbon emissions and CCP estimates