Fuels are the only component of the fire triangle that forest and fire managers can alter to change fire behavior. There have been numerous studies examining how fuel reduction treatments and salvage logging alter fire behavior, severity, and its’ ecological impacts. However, less attention has been paid to how different forest management objectives may influence fire severity in multi-owner landscapes, despite costly and politically contentious suppression of wildfires that do not acknowledge ownership boundaries. In 2013, the Douglas Complex burned over 19,000 ha of Oregon & California Railroad (O&C) lands in Southwestern Oregon, USA. This checkerboard landscape of private industrial and federal forestland provides a unique opportunity to quantify the effects of forest management practices on wildfire severity. We bring together geospatial data on fire progression, fire weather, topography, pre-fire forest conditions, and past management activities to represent the different factors that influence fire behavior. Using ensemble machine learning and spatial generalized linear modelling techniques, we disentangled the relative importance of these factors on fire severity (relative differenced normalized burn ratio, RdNBR) derived from Landsat imagery.
Results/Conclusions
Daily fire weather (burning index) was the most important predictor of fire severity across the Douglas Complex. Ownership and stand age were also important driver of fire severity, with younger and structurally homogeneous stands on private industrial forests displaying higher fire severity compared to older and more structural complex forests on federal lands. After accounting for fire weather, topography, biomass, and stand age, private industrial forestlands burned at higher severity (increased RdNBR = 75.6, SE = 21.7). Counter to the dominant narrative where decades of fire exclusion in older forests has resulted in greater wildfire risk and severity, these findings indicate that spatial homogenization of fuels, rather than absolute fuel loading, is a significant driver of wildfire severity.