Abstract
Wildfires are increasingly affecting electric power systems through two coupled pathways: ignitions initiated by grid assets (e.g., faults, conductor–vegetation contact, arcing) and damage or operational disruption when fires reach overhead transmission corridors, substations, and supporting structures. Utilities, therefore, need quantitative, spatially explicit methods that turn fire-behavior uncertainty into actionable indicators at the asset level.This conference paper presents an operational workflow for building wildfire ensembles and deriving burn probability and time-to-impact metrics for power transmission infrastructure. The workflow builds on open-data landscape preparation, combining public forest maps and inventories with LiDAR-derived canopy structure to generate high-resolution fuels and canopy layers. Fire spread is simulated using FlamMap together with a calibrated Cellular Automata (CA) model. Ensemble modelling propagates uncertainty in ignition location, wind regimes, and fuel moisture, so outputs can be expressed as probability and arrival-time ranges rather than single deterministic estimates. The outputs are post-processed into gridded burn-probability and fire-arrival-time maps, and then aggregated within corridor buffers to produce segment-level exposure indices for planning and operations. In this contribution, we focus on the methodology and its data requirements; a full case-study validation and quantitative performance assessment are the subject of ongoing work.
Operational Wildfire Ensembles for Electric Grid Assets: Mapping Burn Probability and Exposure
