Global Power Outage Detection at the Kilometer Scale from Satellite Nighttime Lights

Julian Rice , Ning Sun, Karthik Balaguru, Tse-Chun Chen, Brent Daniel, Drew Suarez

Satellite nighttime light observations show great promise for remote sensing of power outages, yet have only been utilized in limited case studies to date. Here, we present generalized detection of power outages across the globe at ~1-kilometer resolution. Our method, the Outage frameWork for Luminosity Interruptions (OWL-I), reveals the power outages can be reliably detected from VIIRS/Black Marble nighttime light observations with no knowledge of the driver event or infrastructural records. Confounding factors that have complicated past nighttime light outage detection efforts---including lunar illumination, satellite viewing geometry, snow albedo, and temporal factors---are, for the first time, jointly corrected in a unified framework. While conventional outage detection methods from the literature are shown to be prone to false positive detections, our method exhibits dramatically improved detection skill and remains well calibrated and skillful even for extreme events it was never trained on. When applied to the complete satellite nighttime light observational record 2012--2024, the structure, extent, and time-evolution of power outages are revealed with unprecedented fidelity. These results enable spatially detailed analyses of the drivers and impacts of power disruptions, supporting power grid resilience planning, hazard impact assessment, and emergency management research.