Climate, Fuel, and Land Use Shaped the Spatial Pattern of Wildfire in California's Sierra Nevada. Issue 2 (25th February 2021)

• Record Type: Journal Article
• Authors: Chen, Bin; Jin, Yufang; Scaduto, Erica; Moritz, Max A.; Goulden, Michael L.; Randerson, James T.
• Is Part Of: Journal of geophysical research. Volume 126:Issue 2(2021)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-02-25
• Subjects: drought -- fire -- fire management -- maximum entropy model -- wildland‐urban interface (WUI); Geobiology -- Periodicals; Biogeochemistry -- Periodicals; Biotic communities -- Periodicals; Geophysics -- Periodicals; 577.14
• DOI: 10.1029/2020JG005786
• Languages: English
• ISSNs: 2169-8953

Abstract

Abstract: California's Sierra Nevada has experienced a large increase in wildfire activities over recent decades. This intensifying fire regime has coincided with a warming climate and increasing human activity, but the relative importance of the biophysical and anthropogenic drivers of wildfire remains unclear across this diverse landscape, especially at a finer spatial scale. We used multisource geospatial data sets of fire occurrence, and human, climatic, and biophysical variables to examine the spatial pattern and controls on Sierra Nevada wildfires averaged from 1984 to 2017. The maximum entropy model driven by both biophysical and anthropogenic variables predicted the spatial distribution of fire probability well, with an area under the curve (AUC) score of 0.81. Model diagnostics revealed that aspects of the climate, including vapor pressure deficit (VPD), temperature, and burning index (difficulty of control), dominated the spatial patterns of fire probability across the whole Sierra Nevada region. The VPD was the leading control, with a relative contribution of 32.1%. Population density and fuel amount were also significant drivers, each accounting for 15.8%–12.4% of relative contribution. VPD and burning index were the most important factors for fire probability in higher elevation forest, while population density was comparatively more important in the lower elevation forest regions of the Sierra Nevada. Our findings improved our understanding of the relative importance of various factors in shaping the spatial patterns of historical fire probability in the Sierra Nevada and across various subecoregions, providing insights for targeting spatially varying forest management strategies to limit potential future increases in wildfires. Key Points: Maximum entropy models driven by biophysical and anthropogenic variables predicted the spatial distribution of wildfires accurately Vapor pressure deficit dominated the spatial patterns of fire probability, and population density and fuel amount were significant drivers The relative importance of biophysical versus human controls varied with subecoregions