Post-wildfire housing recovery is a complex process for which systematically collected data remains scarce. Consequently, our ability to anticipate obstacles and plan for housing recovery from future events is limited. This study leverages housing permit datasets collected in Santa Rosa and Unincorporated Sonoma County, impacted by the 2017 Tubbs Fire, and Paradise, impacted by the 2018 Camp Fire. Permit and tax assessor data are combined to gain insights into the recovery processes for these communities. Although the percentage of rebuilt destroyed homes varies significantly between regions, the peak construction demand occurs around 1.5 years after each wildfire, with a substantial decline in the reconstruction rate after 2.5 years. Moreover, the pace of transition from permit application to reconstruction completion is similar across all three regions. Using this finding, we propose a methodology to forecast the number of parcels rebuilt per unit of time based on observations from prior events. A proof-of-concept application of the proposed methodology indicates that it estimates long-term housing recovery patterns based on permit application data collected within one year of the event. These findings indicate that a longitudinal housing recovery data database would help forecast housing recovery from future disasters by providing a source for early empirical validation of predictive models.
Read full abstract