A feasible framework to downscale NPP-VIIRS nighttime light imagery using multi-source spatial variables and geographically weighted regression

• DOI: 10.1016/j.jag.2021.102513
• Publication Date: 2021-08-26T09:52:23Z
• Authors: Yang Ye; Linyan Huang; Qiming Zheng; Chenxin Liang; Baiyu Dong; Jinsong Deng; Xiuzhen Han
• Keywords: Environmental sciences; Impervious surface detection; Physical geography; 550; Nighttime light (NTL); Downscaling; 0401 agriculture, forestry, and fisheries; GE1-350; 04 agricultural and veterinary sciences; Geographically weighted regression (GWR); GB3-5030

The cloud-free monthly composite of global nighttime light (NTL) data the Suomi National Polar-orbiting Partnership with Visible Infrared Imaging Radiometer Suite (NPP-VIIRS) day/night band (DNB) provides indispensable indications human activities and settlements. However, coarse spatial resolution (15 arc sec) NTL imagery greatly restricts its application potential. This study proposes a feasible framework to downscale NPP-VIIRS using muti-source variables geographically weighted regression (GWR) method. High-resolution auxiliary were acquired from Landsat 8 OLI/ TIRS social media platforms. GWR-based downscaling procedures consequently implemented obtain at 100-m resolution. downscaled validated against Loujia1-01 based on coefficient determination (R2) root-mean-square error (RMSE). results suggest that quality was suitably improved after downscaling, yielding higher R2 (0.604 vs. 0.568) lower RMSE (8.828 9.870 nW/cm2/sr) values than those original data. Finally, extendedly applied detect impervious surfaces, had accuracy NTL. Therefore, this facilitates improvement by thus enabling more accurate applications.