Coupling Modified Linear Spectral Mixture Analysis and Soil Conservation Service Curve Number (SCS-CN) Models to Simulate Surface Runoff: Application to the Main Urban Area of Guangzhou, China

Abstract

Land surface characteristics, including soil type, terrain slope, and antecedent soil moisture, have significant impacts on surface runoff during heavy precipitation in highly urbanized areas. In this study, a Linear Spectral Mixture Analysis (LSMA) method is modified to extract high-precision impervious surface, vegetation, and soil fractions. In the modified LSMA method, the representative endmembers are first selected by combining a high-resolution image from Google Earth; the unmixing results of the LSMA are then post-processed to reduce errors of misclassification with Normalized Difference Built-up Index (NDBI) and Normalized Difference Vegetation Index (NDVI). The modified LSMA is applied to the Landsat 8 Operational Land Imager (OLI) image from 18 October 2015 of the main urban area of Guangzhou city. The experimental result indicates that the modified LSMA shows improved extraction performance compared with the conventional LSMA, as it can significantly reduce the bias and root-mean-square error (RMSE). The improved impervious surface, vegetation, and soil fractions are used to calculate the composite curve number (CN) for each pixel according to the Soil Conservation Service curve number (SCS-CN) model. The composite CN is then adjusted with regional data of the terrain slope and total 5-day antecedent precipitation. Finally, the surface runoff is simulated with the SCS-CN model by combining the adjusted CN and real precipitation data at 1 p.m., 4 May 2015.

Coupling the Modified Linear Spectral Mixture Analysis and Pixel-Swapping Methods for Improving Subpixel Water Mapping: Application to the Pearl River Delta, China

“…To extract high-precision endmember class fractions, a modified LSMA method was developed by integrating spectral indices, including the normalized difference built-up index, the normalized difference bare-soil index, and the normalized difference vegetation index [2,40]. Highprecision surface water fractions are required to map the spatial distribution of water at a subpixel scale, which determines the precision of subpixel water mapping.…”

“…However, large errors exist in the LSMA results, partly because some endmembers are misclassified during the process of endmember selection. Following the work of Xu et al [2], a two-step method is applied to improve the accuracy of endmember selection by combining the Maximum Noise Fraction (MNF) and a high-resolution image from Google Earth. However, the numerical problems of LSMA and the imprecise selection of pure endmembers directly result in large errors in the LSMA results; for instance, some water endmember fractions in the pure water pixels may be misclassified as other endmembers.…”

“…In general, the water fraction of the pure water pixels is set to 100%. Some studies have shown that an error of approximately 10% still exists in the LSMA results [2,53]. Therefore, to reduce uncertainties introduced by LSMA, water fractions less than 10% in the mixed water-land pixels are set to 0, and these mixed water-land pixels are not used in subpixel water mapping.…”

“…The areas of forest, farmland, wetland, and river floodplain can act as a sponge for storing water and regulating runoff, which are important for reduction, mitigation, and prevention of rainstorm waterlogging and flood disasters. The Pearl River Delta (PRD) is frequently struck by heavy rainstorms, which often results in urban waterlogging [2]. Therefore, accurate land surface water mapping at higher resolutions is very relevant.…”

“…However, the greatest spectral similarity of endmembers within each pixel constrains the prediction of inter-class boundaries among pixels, which still leads to certain difficulty in the accurate extraction of water fractions with subpixel precision [29,39]. To extract high-precision endmember class fractions, a modified LSMA method was developed by integrating spectral indices, including the normalized difference built-up index, the normalized difference bare-soil index, and the normalized difference vegetation index [2,40]. High-precision surface water fractions are required to map the spatial distribution of water at a subpixel scale, which determines the precision of subpixel water mapping.…”