Natural wildfires in Sierra Nevada wilderness areas

In chapter one I use satellite-based estimates of fire severity to identify factors contributing to observed spatial patterns of fire severity in two large natural fires. Relative humidity and dominant vegetation type were important predictor variables explaining patterns of fire severity. The lowest fire severity corresponded with increased relative humidity. In both fires, forest stands that were dominated by lodgepole pine (Pinus contorta) burned at highest severity, while red fir (Abies magnifica) stands corresponded with the lowest fire severities.; In chapter two I evaluate the effects of wildland fire use (WFU) programs in two Sierra Nevada wilderness areas. I reconstructed historical fire occurrence and tree recruitment to make inferences on the effects of WFU programs on forest structure. Historically fires burned every 6 to 9 years, which moderated tree recruitment. Fire suppression policies established in the early 1900's successfully excluded fire and allowed for unprecedented tree recruitment. Despite the substantial changes in forest structure and composition, the frequency and extent of fires during the current WFU period approaches that of historical levels.; In chapter 3 I analyze interactions between naturally-ignited fires, and assess to what extent the environments in which they burn influence these interactions. Using mapped fire perimeters for a 150 km2 region of California's Sierra Nevada and satellite-based estimates of burn severity, I demonstrate that when allowed to burn relatively freely over three decades, fires can be a self-limiting process. When the amount of time between successive fires is under nine years, and when fire weather is not extreme, probabilities of reburning are extremely low.; In chapter 4 I use information from multiple, naturally-occurring fires in two Sierra Nevada wilderness areas to identify stand and landscape factors influencing fire scarring in trees, as well as provide direct comparisons of reconstructed fire extent and frequency to actual fire extent and frequency based on fire atlases. When intervals between successive fires are short, probabilities of scarring were low. In all WFU fires but one, reconstructed extent of fires was substantially smaller than actual fire extent. As a result, fire-scar reconstructed estimates of fire rotation were much longer than actual fire rotation.