Estimating urban impervious surface percentage with multi-source remote sensing data

Impervious surface is a significant factor in monitoring urban development and environmental quality. However, accurate and cost-effective extraction of impervious surface is still a challenge. In light of the increasing availability of multisource remote sensing data from different imaging sensors, this study developed a method to map large-area impervious surface percentage at the sub-pixel level using multi-source remote sensing data. A case study in Shenzhen was conducted for this purpose based on a classification and regression tree (CART) algorithm to SPOT, Landsat ETM+ images. Experiment results indicate that both of the data are capable of mapping urban impervious surface percentage (ISP) with a reasonable accuracy, but the SPOT image has a better performance of impervious surface percent (ISP) estimation accuracy owing to its higher spatial resolution compared with Landsat ETM+.