COMPARING PRE- AND POST-FIRE DEBRIS-FLOW HAZARD ASSESSMENTS IN THE HERMITS PEAK / CALF CANYON BURN AREA, NEW MEXICO

Recently burned watersheds are susceptible to flash flooding and debris flow from intense rainfall. The USGS conducts rapid assessments of debris-flow hazard following wildfire to assist partners with post-fire planning and emergency response. In the southwestern U.S., fire season often coincides with the start of the monsoon, leaving little time to fully develop those plans. Multiple methods have been employed to estimate potential post-fire hazards for advanced planning, but these methods have not been tested extensively using data from real fires. Here, we evaluate a pre-fire modeling approach using data from the Hermits Peak / Calf Canyon (HPCC) burn area in New Mexico. This pre-fire assessment uses a historical distribution of burn severity for a variety of vegetation types to construct a synthetic distribution of the differenced normalized burn ratio (DNBR) and soil burn severity. The simulated DNBR and soil burn severity are then used as inputs to the same models used to estimate post-fire debris flow hazards for real wildfires. Two wildfire scenarios were constructed using this method. The first scenario represents the 50^th percentile of historical distributions. The second scenario represents a more severe wildfire corresponding to the 84^th percentile of historical distributions. We compare the pre- and post-fire assessments of the HPCC burn area to evaluate the pre-fire model performance. First, we compare simulated burn severity to the field-validated soil burn severity. Second, we compare predicted debris-flow likelihood, volume, and rainfall thresholds using both observed and simulated burn severity distributions. We find the simulated burn severity for the two scenarios create a more uniform distribution compared to field-validated soil burn severity maps. However, the 84^th percentile scenario is in reasonable agreement with the observed burn severity on steep slopes, which are the likely zones of debris-flow initiation. This correlation translates into an adequate agreement in debris-flow predictions between those made with the field-validated soil burn severity and the 84^th percentile wildfire scenario. Although more testing is needed, these results indicate that the pre-fire modeling approach is a useful tool for advance post-fire planning for wildfires