Long-term monitoring of the urban impervious surface mapping using time series Landsat imagery: A 23-year case study of the city of Wuhan in China

摘要

Impervious surface is a significant indicator of urban environments. In order to better understand the pattern of urban expansion, this study adopted random forest algorithm for a long-term urban impervious surface mapping of Wuhan in China. We produced 23-year (1993 to 2015) urban impervious surface mapping data via a dense Landsat time series and the results indicated that the impervious surface area in Wuhan increased rapidly from 4.33% to 15.34% of the total land area from 1993 to 2015. Moreover these map showed that the loss of agricultural land plays a major role in the development of urbanization and impervious surface spread from center area to rural areas, and NDVI showed a higher influence than MNDWI and BCI. Specifically, it was a more important finding that obvious decreasing tendency within the old city zones when they are faced the transformation of old urban renewal. With respect to the formulation of the new urban area, the impervious surface percent can be regarded as an indication of land use because the number and distribution of urban impervious areas have a close and inter-annually correlation with land use/cover types. Finally, it was clearly that RF has a good performance to deal with massive data.