Mapping impervious surfaces in the Xiangjiang River basin based on remote sensing spectral indices: a case study in Chang-Zhu-Tan region

摘要

Increased impervious surfaces pose significant threats to the hydrologic cycle of the Xiangjiang River basin as a consequence of urbanization. Quantifying the percentage of imperviousness within the Xiangjiang River basin is important to pollution control and watershed management. Per-pixel and sub-pixel methods have been widely used for analyzing impervious surface changes, but these methods are considered as complicated, computationally intensive, and sometimes subjective, especially when applied to a large geographic area. In this paper, normalized difference built-up index (NDBI), normalized difference impervious surface index (NDISI), normalized difference vegetation index (NDVI) and enhanced built-up and bareness index (EBBI) were respectively used to estimate impervious surfaces in Chang-Zhu-Tan region (CZT) of the Xiangjiang River basin, and a comparative analyses was conducted. Then the optimum spectral index was chosen to map the percentage of impervious surfaces for the study area. The results show that the spectral index of NDBI has the optimum estimation of large-scale impervious surfaces, and the percentage of imperviousness in CZT was 13.87%. The water quality in CZT was characterized as "protected", indicating that water quality protection in the plain areas of CZT is imperative.