Evaluation of impervious surface extraction based on Qimingxing-1 nighttime light and point of interest data

Abstract

Impervious surface extraction is essential for environmental management and urban studies. Although nighttime light data have been widely used for this purpose, few studies have utilized the high-resolution (21 m) Qimingxing-1 data released in 2022, which offer significantly higher spatial resolution compared to the widely used NPP-VIIRS product (750 m). In this study, the effectiveness of Qimingxing-1 nighttime light data for impervious surface extraction was investigated for the first time using the highly urbanized city of Dongguan as an example. We combined nighttime light data with POI data to create a ‘Nighttime-POI’ (NP) index and extracted impervious surfaces from four datasets (Qimingxing-1, NPP-VIIRS, QM_NP, and NPP_NP) using the thresholding method. Comparisons with reference data from Landsat 8 and Sentinel-2 revealed that the Qimingxing-1 data improved extraction accuracy by 2.7% over NPP-VIIRS. The Qimingxing-1 data could more accurately reflect the spatial distribution of impervious surfaces and imperviousness degree. Additionally, combining nighttime light data with POI data further enhanced accuracy by 2–3%, resulting in more accurate boundaries and detailed information on impervious surfaces. In conclusion, the Qimingxing-1 data are highly advantageous for impervious surface extraction. The high-precision impervious surface results can assist the government in promptly comprehending the dynamics of urban development.

Keywords

• Nighttime light
• Qimingxing-1
• VIIRS
• impervious surfaces
• POI
• built-up area