remote sensing Article Cross-Sensor Nighttime Lights Image Calibration for DMSP/OLS and SNPP/VIIRS with Residual U-Net Dmitry Nechaev 1 , Mikhail Zhizhin 2,3, * , Alexey Poyda 1,4 , Tilottama Ghosh 2 , Feng-Chi Hsu 2 and Christopher Elvidge 2   Citation: Nechaev, D.; Zhizhin, M.; Poyda, A.; Ghosh, T.; Hsu, F.-C.; Elvidge, C. Cross-Sensor Nighttime Lights Image Calibration for DMSP/OLS and SNPP/VIIRS with Residual U-Net. Remote Sens. 2021, 13, 5026. https://doi.org/10.3390/ rs13245026 Academic Editors: Gang Chen and Yasushi Yamaguchi Received: 7 September 2021 Accepted: 6 December 2021 Published: 10 December 2021 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affil- iations. Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). 1 Moscow Institute of Physics and Technology, 117303 Moscow, Russia; nechaev.dv@phystech.edu (D.N.); poyda_aa@nrcki.ru (A.P.) 2 Earth Observation Group, Payne Institute, Colorado School of Mines, Golden, CO 80401, USA; tghosh@mines.edu (T.G.); fengchihsu@mines.edu (F.-C.H.); celvidge@mines.edu (C.E.) 3 Space Research Institute, Russian Academy of Sciences, 117997 Moscow, Russia 4 NRC “Kurchatov Institute”, 123182 Moscow, Russia * Correspondence: mzhizhin@mines.edu Abstract: Remote sensing of nighttime lights (NTL) is widely used in socio-economic studies of economic growth, urbanization, stability of power grid, environmental light pollution, pandemics and military conflicts. Currently, NTL data are collected with two sensors: (1) Operational Line-scan System (OLS) onboard the satellites from the Defense Meteorology Satellite Program (DMSP) and (2) Visible Infrared Imaging Radiometer Suite (VIIRS) onboard the Suomi NPP (SNPP) and NOAA- 20 satellites from the Joint Polar Satellite System (JPSS). However, the nighttime images acquired by these two sensors are incompatible in spatial resolution and dynamic range. To address this problem, we propose a method for the cross-sensor calibration with residual U-net convolutional neural network (CNN). The CNN produces DMSP-like NTL composites from the VIIRS annual NTL composites. The pixel radiances predicted from VIIRS are highly correlated with NTL observed with OLS (0.96 < R2 < 0.99). The method can be used to extend long-term series of annual NTL after the end of DMSP mission or to cross-calibrate same year NTL from different satellites to study diurnal variations. Keywords: nighttime lights; cross-sensor calibration; DMSP/OLS; SNPP/VIIRS; convolutional neural network 1. Introduction Defense Meteorology Satellite Program (DMSP) was launched in 1962 and since then its satellites with Operational Line-scan System (OLS) serve as a source of valuable night- time light (NTL) data of the Earth surface. Starting from 1992 the DMSP satellites broadcast digital images, which were post-processed by the NOAA (currently Colorado School of Mines) Earth Observation Group (EOG) into global annual average and background removed NTL maps. With annual data stretch from 1992 to 2013, it makes DMSP Nighttime Lights the longest data series available for nocturnal remote sensing on human activities [1]. In 2011 Suomi NPP (SNPP) satellite with Visible Infrared Imaging Radiometer Suite (VIIRS) was launched. VIIRS instrument is also capable of detecting dim light sources at night. Annual VIIRS maps of nighttime lights are published from 2013 to 2019 [2]. NTL maps made with DMSP (DNL) or VIIRS (VNL) are widely utilized in research of human activity, economy and ecology [3]. Despite the value of the long-term archives from DMSP and VIIRS, their potential is not fully utilized. The main reason is that there is a significant difference between the OLS and VIIRS sensors (Table 1). VIIRS is significantly superior to OLS in terms of spatial, temporal, and radiometric resolution [4]. Due to the lack of calibration, DMSP data have no radiometric measurement units whilst VIIRS output is measured in nW/cm 2 /sr. Remote Sens. 2021, 13, 5026. https://doi.org/10.3390/rs13245026 https://www.mdpi.com/journal/remotesensing