Using Nighttime Light Satellite Imagery to Identify Power Outages and Affected Communities After the 2021 Western Kentucky Tornado Disaster

Keywords: Social Vulnerability, Power Outages, Nighttime Light (NTL), Tornado Damages, Disaster Response, Midwestern U.S.
Abstract Type: Poster Abstract

Abstract

Accurately estimating power outage duration and extent from natural disasters is essential for prioritizing resources to protect socially and medically vulnerable communities. Tornadoes, increasing in frequency and intensity across the Midwestern U.S., often leave rural populations especially exposed to prolonged power disruptions. While nighttime light (NTL) satellite imagery is a promising, cost-effective tool for assessing disaster impacts, its application in rapid response has been limited by the need for clear, post-disaster images, which are often delayed. The short duration and localized nature of tornadoes, however, allow for faster acquisition of clear NTL imagery after the event, opening new avenues for rapid damage assessment.

This study investigates the feasibility of using NTL data to delineate the affected areas and monitor power restoration following the 2021 Western Kentucky Tornado. By analyzing changes in NTL brightness before and after the tornado, we developed an approach to determine the spatial extent, duration, and intensity of power outages caused by the disaster. To evaluate the accuracy and reliability of this approach, NTL-derived outage data were compared with reports from utility companies. In collaboration with Kentucky Emergency Management, our findings are expected to enhance the ability of emergency responders and public health officials to target aid and resources more effectively, particularly to protect medically vulnerable populations dependent on continuous power. This project not only demonstrates the value of NTL imagery in tornado disaster response but also contributes a cost-effective and scalable tool for future recovery efforts in similar events.