Remote sensing-based retrieval of ground impervious surfaces

Worldwide land use change and urban spatial expansion have replaced the vegetation-dominated natural landscape with various impervious surfaces. This replacement has brought about significant negative impacts on the global ecological environment and has raised public awareness of the emergence of this key ecological environment indicator. Impervious surfaces have become an important consideration in many environmental-or socioeconomic-related studies. Quickly gathering information regarding the magnitude, location, geometry, and spatial pattern of impervious surfaces and accurately quantifying the dynamic information on impervious surfaces have become urgent issues to be addressed. Today's remote sensing technology can provide a promising solution to this problem owing to its rapid, repetitive, synoptic, and multi-scale Earth observation. The remote sensing of impervious surfaces has made considerable progress after its development in 2000, and various innovative techniques for the retrieval of impervious surface information have been proposed in the last decade. Therefore, we examined these innovative approaches and focused on their advantages and disadvantages through a literature review. Chinese research and achievements regarding the remote sensing of impervious surfaces were also summarized. The current remote sensing of impervious surfaces has made great progress, and many of the techniques for the information extraction and classification of impervious surfaces achieve an accuracy of over 85%. Nevertheless, the mapping of impervious surfaces remains a challenge. The main problem is the confusion between impervious surface information and bare soil/shadow information, which affects the accurate retrieval of impervious surface information. Most impervious surface materials are made of or directly from rock, sand, or clayish soil. Thus, impervious surfaces exhibit similar spectral characteristics. Existing multispectral remote sensors lack sufficient spectral resolution to distinguish impervious surface materials from bare soil. Thus, using the techniques on the basis of spectral characteristics alone hampers the improvement of the accuracy of impervious surface inversion. Other secondary data, such as LiDAR data, are expected to help solve this bottleneck in future research on the remote sensing-based retrieval of impervious surfaces. © 2016, Science Press. All right reserved.