Advancing SAR monitoring of urban impervious surface with a new polarimetric scattering mixture analysis approach

Abstract

Accurate monitoring of urban impervious surfaces (UIS) changes is crucial in understanding the anthropogenic activities of urbanization and its environmental, economic, and social sustainability. Over the past two decades, synthetic aperture radar (SAR) has posed huge potential for urban monitoring with its all-weather and all-day imaging capability over rainy and cloudy regions around the world. Unfortunately, some critical challenges associated with UIS identification from data have been identified due to the high diversity of UIS and the different SAR imaging mechanisms from optical remote sensing. To address the challenges, this study proposed a new method of polarimetric scattering mixture analysis (PSMA) by modeling the diverse scattering mechanisms of various UIS categories. C- and L-band full-polarization SAR data in four typical urban areas from central, eastern, and southern China, were employed to evaluate the performance of the method. Experimental results demonstrate that the proposed method achieves an overall accuracy (OA) exceeding 96 % across all study cases. Compared with traditional methods, PSMA significantly reduces land cover confusion and enhances the producer's (PA) and user's accuracy (UA) for UIS, particularly for “road”, by up to 13.4 % and 9.5 %, respectively. Non-impervious surfaces such as “soil” and “vegetation” witness a notable increase in PA and UA by up to 10 % and 20.5 %, respectively. These promising results underscore the effectiveness of the proposed method for accurate urban monitoring and understanding land cover scattering confusion over complex urban environments.