https://doi.org/10.7939/r3-19fc-k379

Wildfire and permafrost thaw have been common disturbances in the boreal zone for millennia and are now intensified by warming due to human-made climate change. The Taiga Plains ecozone in northwestern Canada is warming at a faster rate than other regions. In this ecozone, permafrost is found at relatively low latitudes due to the vast abundance of peatlands, whose large, frozen carbon (C) and nitrogen (N) stocks are vulnerable to warming. Ecosystems and atmosphere exchange C and N as greenhouse gas (GHG) fluxes such as carbon dioxide (CO2), methane (CH4), and nitrous oxide (N2O), which are altered, directly by warming and indirectly by wildfire and permafrost thaw. This may lead to further warming by an increase in net radiative GHG forcing, resulting in positive climate feedback. I examined the effects of wildfire and permafrost thaw on the GHG balance in three field studies covering three peatland complexes at the southern limit of permafrost across the Taiga Plains at two spatial and temporal scales: a) by monthly static chamber flux measurements from plots affected by wildfire or permafrost thaw (chapter 2), b) by continuous eddy covariance measurements at the landscape scale including different permafrost extents (chapter 3) and different years of wildfire (chapter 4). Previously often neglected, the GHG N2O was the focus of chapter 2, where I compared the GHG exchange from soil plots affected by different stages of thermokarst (=permafrost thaw) and wildfire disturbance to the ones of intact permafrost plateaus. I found uptake of N2O by peat plateaus which decreases post-fire but increased post-thaw, the latter driven by N2O soil gas concentrations below-ambient, warmer soil temperatures and higher soil moisture contents. From a net radiative GHG forcing point of view, changes in N2O were minor compared to alterations in soil respiration and CH4, determined by the monthly chamber-based flux measurements over one growing season. To quantify differences in CO2 and CH4 with a higher temporal resolution, I then investigated multi-year and high-frequency eddy covariance datasets. In chapter 3, I contrasted two peatland complexes with different extents of permafrost, sporadic (<50% permafrost) and discontinuous (>50% permafrost), and their fluxes of CO2 and CH4 as well as resulting balances of C and net radiative GHG forcing. The net uptake of CO2 and the release of CH4 was higher at the sporadic permafrost site due to the higher abundance of permafrost-free wetlands in the landscape. Consequently, thaw-induced shrinking of permafrost extent will thus lead to increases in peatland C uptake driven by CO2 along with increases in net radiative GHG forcing driven by CH4. In chapter 4, I investigated the effects of wildfire on the CO2 exchange post-fire. CO2 release was large from a recently burned permafrost peatland, accumulating up to half of the C combustion losses. However, the massive CO2 losses were not sustained into the decade after the wildfire due to the recovery of the peatland vegetation. However, old C may be reintroduced to the atmosphere, as the active layer rapidly deepens towards the end of the growing season in the years after fire compared to unburned sites. My doctoral research adds knowledge to our limited understanding of the magnitude and direction of change of GHG fluxes of permafrost peatlands in a warmer world. Here, I quantify future climate feedback resulting from thawing permafrost and wildfires increasing in severity and frequency. My findings suggest that the two disturbances are not only similar in their aerial extent across the Taiga Plains, but also in their effects on increased net radiative GHG forcing, leading to further warming by both, CO2 release following wildfire and CH4 following permafrost thaw.

https://doi.org/10.7939/r3-19fc-k379