Forests store a significant portion of terrestrial carbon, but this storage capacity is vulnerable to disturbance, such as wildfire. Accurate accounting of carbon emissions from wildfires is critical for the assessment of greenhouse gas reduction mandates. There is a misconception about the amount of carbon released in a wildfire: while fires can be destructive to ecosystems and built environments, less than 5% of tree biomass actually combusts. This has caused overestimation of carbon emissions through modeling of recent fires and projected future fires.
Wildfires across the western US have been increasing in frequency and intensity, with resulting tree mortality leading to potential increased carbon release and decreased carbon uptake. Forest recovery following a wildfire depends on pre and post- fire climatic conditions, and burn extent and severity. Forest management techniques have been promoted to increase resilience to increasingly severe fires, but the impact of these techniques is unknown. Recovery of forested systems is difficult to predict, compounding the difficulty of using management tools effectively. We are using forest inventory and management datasets combined with ecosystem models to understand how wildfires and forest management have shaped current carbon storage and to predict impacts of future fire and management strategies on ecosystem carbon trajectories
Results/Conclusions
Using the biogeochemical model DayCent, we show that emissions estimates from widely-used combustion percentages are on average 145 to 182% greater than those based on observations. We found that most models misrepresent forest structural changes following fire, affecting both combustion during fire and subsequent decay, leading to faster release of CO2. These overestimates undervalue persistent forest carbon storage after disturbance and are being used as the basis of forest management plans to increase harvest to prevent future CO2 emissions from fire. Using these improvements, we estimate that over the last 15 years western US forests have released 800 Tg CO2e through fire, approximately 5% of fossil fuel emissions. We show that combustion estimates from commonly used ecosystem models can be wildly overestimated, and using these modeled emissions estimates in climate policy with greenhouse gas reduction mandates can lead to under-effective climate-mitigation policy. We will continue using ecosystem models to better understand how modeled combustion can be improved to predict better estimates of fire impacts on carbon storage throughout the western US. In addition, we will continue to analyze publicly available forest inventory and management datasets to elucidate management effects on carbon storage.