Mapping Disasters & Tracking Recovery in Conflict Zones Using Nighttime Lights

The capability to map disasters and track recovery thereafter has seen substantial enhancement from the development of connected communication and imaging technology. In this work, we demonstrate a series of techniques for assessing the damage and the recovery process from the humanitarian crisis in Sana'a, Yemen. In particular, we make use of the Visible Infrared Imaging Radiometer (VIIRS) Day/Night Band (DNB), which produces nightly data on illumination levels globally at 750m resolution. While nearly all work that has previously been performed using the VIIRS DNB dataset uses monthly or annual composite data, we develop novel methods that leverage the nightly availability of this data source, particularly for mapping the scale and extents of illumination changes, accurately identifying the day when bombing occurred and illumination conditions changed, characterizing electricity outage patterns, and tracing the recovery of nighttime lighting both spatially and temporally. For each of these methods, we compare with publicly-available datasets to verify the accuracy of our observations. In total, these data, available only 1 day after an event, and the resulting analyses can be especially valuable for disaster response as well as targeting and tracking investments in recovery.