Comparing ground-based lightning detection networks near wildfire points-of-origin

Abstract

AbstractLightning detection and attribution to wildfire ignitions is a critical component of fire management worldwide to both reduce hazards of wildfire to values-at-risk and to enhance the potential for wildland fire to provide resource benefits in fire-adapted ecosystems. We compared two operational ground-based lightning detection networks used by fire managers to identify cloud-to-ground strokes within operationally-relevant distances (1.6 km) of the origins of 4408 western United States lightning-ignited wildfires spanning May–September 2020. Applying two sets of constraints–varying holdover time and applying a quality control measure–we found strokes were co-detected near 55–65% of fires, increasing to 65–79% for detection by at least one network, with neither network detecting lightning near 1024–1666 fires. Because each network detected strokes near 136–376 unique fires, the use of both networks is suggested to increase the probability of identifying potential fire starts. Given the number of fires with network-unique detections and no detections by either network, improvements in lightning detection networks are recommended given increasing fire hazard.