Rapid hazard prediction and assessment of post-fire debris flows using UAV lidar: Eaton Fire, California

Although post-wildfire debris flows pose significant hazards to property and life, robust prediction systems for the magnitude and timing of debris flows remain limited. This knowledge gap exists in part due to sparse data on historical debris-flow events. In response to the Eaton Fire, part of the broader 2025 Los Angeles Wildfire event, we deployed unpiloted aerial vehicle (UAV) lidar systems to map topographic change before and after significant rainfall events. We perform two types of analyses. First, through volumetric measurements of post-rainstorm sediment accumulation in 14 debris basins, we quantify the volumes of debris flows mobilized during a major post-fire rainstorm. Second, using UAV lidar surveys acquired before the first major rainfall, we measure the volume of sediment mobilized as dry ravel during the wildfire. We compare the pre- and post-rainstorm surveys to evaluate whether the debris flows were sourced predominantly from dry-ravel cones filling the channel network. We show that measuring the dry-ravel volumes provides a data-driven approach to predicting hazards. By integrating the UAV lidar with field observations, time-lapse cameras, rainfall monitoring, and airborne lidar, we present a framework for capturing the spatial and temporal dynamics of debris flows. Finally, we discuss logistical challenges related to airspace access and weather constraints, and offer recommendations for integrating UAV lidar into routine postfire hazard mitigation workflows. This study underscores the transformative potential of UAV lidar monitoring for improving postfire debris-flow hazard prediction, emergency response, and long-term planning.