Hazard Mitigation & Government Response in The Bay Area: Assessing the Intersection and Causation of Wildfires, and Floods through the Application of Remote Sensing Technologies

Floods and wildfires both present significant hazards to humans and infrastructure. Often less acknowledged is the coupling between them, which we propose to analyze using remote sensing data. Increased drought in California has led to ubiquitous wildfires, setting up conditions that enhance the risk of flash flooding due to mudflows, soil, and topographic changes. Even light rain can lead to devastating flash flooding and mudflows if wildfire has removed most of the vegetation cover. Additionally, wildfires can create infrastructural deformation that heightens overall risk of flooding. Furthermore, current government programs designed to reward flood readiness do not recognize both disasters in how they compound each other while also adhering to equitable measures for disadvantaged communities. Here we ask, how do we accurately assess these intersecting occurrences to assess overall hazard? And more specifically, can we do this using readily available remote sensing data to be able to take more local risk factors into account?

Here we describe our approach to analyze, predict, and track water buildup caused by wildfire scarring in the California Bay Area through the combination of remote sensing InSAR data, computational methods, and algorithmic models. By using remote sensing data, we seek to analyze and manage risk in areas where local support infrastructure is lacking. Many remote sensing studies have been developed to sense near-real time flood extent. Nonetheless, the lack of detailed elevation data caused by fires can prevent accurate risk assessments, especially in poorly-mapped communities. We combine newly available worldwide elevation data from the Copernicus digital elevation model with low latency spaceborne SAR images in order to predict fire damage and flood depths on a neighborhood basis. We will use a combination of radar intensity based methods and InSAR methods and topographical mapping to obtain the aerial extent of scarred grounds, to estimate potential water depths to identify the highest risk areas. Our goal is to thus create better model predictions of risk during flooding and help better create and deploy response strategies.