During periods of heightened wildland fire activity in the United States, multiagency coordinating groups must prioritize among multiple on-going fires to allocate scarce suppression resources. While many studies have explored factors that influence wildfire suppression expenditures and personnel allocation, understanding the specific factors that affect daily wildfire prioritization has remained unexplored. In this study, we first examine wildfire reporting and ranking processes across different regions of the United States to provide insight into criteria used for fire ranking. We then focus on examining the 12 criteria used for ranking fires daily by California's multiagency coordination group. We developed a computer program to replicate the California prioritization process and found that fire rankings generated by this program align well with the historical rankings, indicating close adherence of California's fire managers to their ranking rules. A correlation analysis revealed weak correlations among the 12 criteria, suggesting that no criterion should serve as a proxy for another during fire priority evaluations. We further applied a Random Forest machine learning model, which identified threats and damage to structures, fire size, and evacuations as the most impactful criteria in determining fire priority. Our findings can benefit wildfire decision makers by providing clear insights into the existing wildfire priority assessment process, so that adjustments to the process can be made for better management outcomes. Policymakers can also leverage these insights to develop evidence-based fire management policies, regulations, and practices that promote more efficient responses to fire risks while fostering greater public trust in fire management efforts.
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号