Search for: All records

Abstract Wildfires are a major disturbance to forest carbon (C) balance through both immediate combustion emissions and post-fire ecosystem dynamics. Here we used a process-based biogeochemistry model, the Terrestrial Ecosystem Model (TEM), to simulate C budget in Alaska and Canada during 1986–2016, as impacted by fire disturbances. We extracted the data of difference Normalized Burn Ratio (dNBR) for fires from Landsat TM/ETM imagery and estimated the proportion of vegetation and soil C combustion. We observed that the region was a C source of 2.74 Pg C during the 31-year period. The observed C loss, 57.1 Tg C year −1 , was attributed to fire emissions, overwhelming the net ecosystem production (1.9 Tg C year −1 ) in the region. Our simulated direct emissions for Alaska and Canada are within the range of field measurements and other model estimates. As burn severity increased, combustion emission tended to switch from vegetation origin towards soil origin. When dNBR is below 300, fires increase soil temperature and decrease soil moisture and thus, enhance soil respiration. However, the post-fire soil respiration decreases for moderate or high burn severity. The proportion of post-fire soil emission in total emissions increased with burn severity. Net nitrogen mineralization gradually recovered after fire, enhancing net primary production. Net ecosystem production recovered fast under higher burn severities. The impact of fire disturbance on the C balance of northern ecosystems and the associated uncertainties can be better characterized with long-term, prior-, during- and post-disturbance data across the geospatial spectrum. Our findings suggest that the regional source of carbon to the atmosphere will persist if the observed forest wildfire occurrence and severity continues into the future. 

ABSTRACT Climate change has increased the size and frequency of wildfires across the boreal biome. Severe wildfires in boreal forests have been found to trigger shifts from evergreen to deciduous canopies, which has cascading effects on carbon and nitrogen cycling. Ecosystem productivity and carbon uptake in boreal forests are strongly linked with nitrogen, and Earth system models increasingly depend on our understanding of the nitrogen balance to predict post‐fire carbon uptake. To investigate the post‐fire boreal nitrogen balance, we combined a mass balance approach and literature synthesis to estimate rates of nitrogen accumulation and nitrogen inputs across a network of 18 boreal wildfire chronosequences that varied in both wildfire regime and post‐fire canopy type, comprising 527 forest stands. We found that deciduous‐ or mixed‐dominance boreal forests establishing after severe, stand‐replacing fires had the highest nitrogen accumulation rates (15.7 ± 3.8 kg ha⁻¹ year⁻¹), while evergreen‐dominated forests establishing after surface‐ or mixed‐severity fires had the lowest nitrogen accumulation rates (1.4 ± 1.1 kg ha⁻¹ year⁻¹). Annual known inputs from nitrogen deposition and biological nitrogen fixation combined, estimated from published data, largely failed to explain the rate of nitrogen accumulation, particularly in deciduous or mixed‐dominance forests establishing after stand‐replacing fires, suggesting that the origins of most nitrogen in these forest types remain poorly understood. As the frequency of severe wildfires increases across the boreal biome and shifts toward deciduous canopies become more common, our study reveals a large knowledge gap in the resulting nitrogen balance that needs to be resolved in order to improve predictions of forest carbon uptake.