Impacts of fire management practices on forested peatland carbon pools and fluxes in boreal Canada

The boreal region of western Canada is predicted to have larger periods of summer drought with anthropogenic climate change, and as a result, an increase in the extent, frequency, and intensity of forest fires. Further, over 50% of the peatlands in the region are forested, where not only are the trees be susceptible to burning, but also the smoldering of organic rich peat. Peat smoldering can cause large carbon losses to the atmosphere and is a challenge to extinguish. Therefore, fuel management treatments to prevent and limit fire in these ecosystems may be increasingly needed, especially in regions where forested peatlands are close to human infrastructure. Field studies have shown that although treatments such as mulching and/or thinning trees can reduce the amount peat that is burned, the burning of vegetation and surface compression can lead to greater carbon losses compared to the original site conditions, suggesting that there are trade-offs between supressing smoldering combustion and altering surface fuels. Further, field studies to date have only been able to report carbon losses post-fire, and the changes in carbon uptake and release through recovery as well as the change in the probability that a fire will occur have yet to be fully explored or are part of ongoing monitoring.

My project aims to test how the timing and combination of treatments impacts peatland carbon losses and recovery from fire using the Canadian Model for Peatlands (CaMP), which tracks carbon pools and fluxes pre- and post-disturbances. This work will support continued field studies and explore recovery trajectories on timescales of 50-100 years. My work will also help to include more fire scenarios within CaMP that can be used to predict carbon losses from fire at a national scale. My progress on the project to date has been to parameterize the model for fuel management treatments and fire using a relationship between the water content of peat and depth of burn. Four treatments are being tested: removing all the trees, removing some trees, mulching the trees in place, and compressing the soils. My initial parameterization of the treatments has shown that dead moss cover with tree removal can result in higher potential depth of burn than what is originally expected at the site. The effect of dead moss cover diminishes as live moss colonizes the surface and can potentially be mitigated by adding mulch or compressing the peat post-treatment. Therefore, the timing between fuel management treatments and fire will likely be critical for minimizing soil carbon loss. Future work will involve running model scenarios for each treatment under different fire frequencies and drought conditions.