Monitoring Urban Land Cover Change

The rapid population has led to increase for space in urban areas as a result urban expansion; this might pose serious threat to those managers of the cities. Up to date information are required by planners and policy makers. This work examines the rapid expansion of the Las Vegas metropolitan area, one of the fastest growing regions in the United States from 1984 to 2001 using Landsat TM and ETM+ images. In this work, a post classification comparison approach was employed and the PCA-based was also implemented in order to compare its results. Land cover maps were produced using the Maximum Likelihood algorithm Supervised Classification. It was found that the built-up land had expanded by 11% at the expense of the barren land cover class. The overall accuracy assessment for the two techniques revealed that the PCA based produced an accuracy of 88.8% which was slightly higher than the independently classified images and proved to be an efficient technique in change detection in a semi-arid region. The results of the work can be improved upon by using advanced methods such as spectral mixture analysis and or with higher resolution images