Land use and land cover change detection using satellite remote sensing techniques in the mountainous Three Gorges Area, China

The Three Gorges Area (TGA), along the Yangtze River, China has been experiencing drastic land use and land cover (LU/LC) changes since the commencement of the construction of the Three Gorges Dam in 1994. These changes may have environmental impacts. However, information about the changes is limited and difficult to obtain. In this paper, optical satellite imagery is used to detect LU/LC changes during the study period 1987-2006 for the complicated, high-relief environment. A case study was conducted in Zigui County of Hubei Province within the TGA. A site-specific procedure using a decision rule-based classification method and a post classification change detection technique is developed to combine spectral and spatial knowledge in the classification of multi-temporal images. Using the decision rule-based classification method, overall accuracies of LU/LC maps of seven classes between 73.4% and 89.5% were obtained, increased by 4-5% over that using the traditional method. The results show that the main trend in LU/LC change in the study area throughout the monitoring period was a steady reduction in natural vegetation areas. About 32% of the total area of natural vegetation, including forest, shrub and grass, was lost to built up areas, crop fields and orchards.