THE USGS LANDSLIDE HAZARDS PROGRAM’S RESPONSE TO THE 2022 WILDFIRES IN NEW MEXICO

In the spring and early summer of 2022, New Mexico experienced its most severe wildfire season on record, including two of the largest wildfires in state history. A substantial fraction of the area burned was steep terrain, which is susceptible to flash flooding and debris flows after wildfire. To assist local, state, and federal agencies in post-fire planning and emergency response, the USGS Landslide Hazards Program conducted rapid debris-flow hazard assessments for seven burn areas covering an area equivalent to the size of the state of Rhode Island (3,200 km^2). These assessments estimated (1) the likelihood of debris flow across the burn area, (2) the rainfall thresholds that must be exceeded to trigger flows, and (3) the potential volume of these flows. The assessment for the largest fire, the Hermits Peak / Calf Canyon fire, indicates it has the highest susceptibility to debris flow of the seven fires evaluated. Estimated 15-minute rainfall thresholds for the northern and southern halves of the fire were 23 and 30 mm/hr, respectively. For comparison, rainstorms with one-year recurrence intervals in the area have a 15-minute rainfall intensity ranging between 50-70 mm/hr, indicating that it is highly likely that 2022 monsoonal rainfall will be substantially above the thresholds for debris-flow initiation.

To evaluate the accuracy of the hazard assessment and improve the models for future applications, the USGS selected the Gallinas Creek watershed within the Hermits Peak/Calf Canyon burn area for intensive monitoring. This watershed has a dense network of rain and stream gages established by the USGS New Mexico Water Science Center shortly after the fire. We complemented this network with an additional station in a 1.2 km^2 sub watershed to record flow characteristics and timing with geophones and a video camera. Over the next two years, systematic field observations of flow activity will be made by the USGS and partners following rainstorms. These data will be used to test how well the hazard models match observations and to refine rainfall thresholds used for warning.