In response to the COVID-19 pandemic, governments around the world have enacted
widespread physical distancing measures to prevent and control virus transmission. Quantitative, spatially-disaggregated information about the population-scale shifts in activity that have resulted from these measures is extremely scarce, particularly for regions outside of Europe and the US. Public health institutions often must make decisions about control measures with limited region-specific data about how they will affect societal behavior, patterns of exposure, and infection outcomes. The Visible Infrared Imaging Radiometer Suite Day/Night Band (VIIRS DNB), a new-generation space-borne low-light imager, has the potential to track short-term changes in human activity, but the capability has not yet been applied to a cross-country analysis of COVID-19 responses. Here, we show initial evidence of VIIRS-DNB-detected changes to societal behavior patterns from COVID-19 control measures. We examine multiyear (2015–2020) daily time-series data derived from NASA’s Black Marble VIIRS nighttime lights product (VNP46A2) covering 584 urban areas, in 17 countries in the Middle East. Our analysis specifically focuses on the first stage of the pandemic and the Ramadan period, to understand how communities have adhered to COVID-19 measures when they were contrary to long-standing socio-cultural practices. Nighttime lights capture the onset of national curfews and lockdowns well, but also expose the inconsistent response to control measures both across and within countries. In conflict-afflicted countries, lockdowns and curfews resulted in less change in activity, highlighting the compound health and security threats that fragile states face. Our findings show how satellite measurements can aid in assessing the public response to COVID-19 physical distancing policies and the socio-cultural factors that shape their success, especially in data-sparse regions.