Optimized Hot Spot and Directional Distribution Analyses Characterize the Spatiotemporal Variation of Large Wildfires in Washington, USA, 1970−2020

Abstract

Spatiotemporal analysis of fire activity is vital for determining why wildfires occur where they do, assessing wildfire risks, and developing locally relevant wildfire risk reduction strategies. Using various spatial statistical methods, we determined hot spots of large wildfires (> 100 acres) in Washington, the United States, and mapped spatiotemporal variations in large wildfire activity from 1970 to 2020. Our results found that all hot spots are located east of the crest of the Cascade Range. Our spatiotemporal analysis found that the geographic area wherein most of the state’s acres burned has shrunk considerably since 1970 and has become concentrated over the north-central portion of the state over time. This concentration of large wildfire activity in north-central Washington was previously unquantified and may provide important information for hazard mitigation efforts in that area. Our results highlight the advantages of using spatial statistical methods that could aid the development of natural hazard mitigation plans and risk reduction strategies by characterizing previous hazard occurrences spatially and spatiotemporally.