研究期間：10108~10207;Rapid urbanization is creating societal impacts on the environment attributed to the increasing population. Understanding spatio-temporal dimensions of land-use changes and environmental risks that shape the urbanization is thus important to the process of urban planning. The goal of this project is to investigate the urban growth and flood risks due to rapid urbanization through remotely-sensed data and GIS technique. Following out the objective, the project is designed for a three-year period from 2012 to 2014. In the first year (2012), mapping the urban growth and landscape changes using Landsat and Rapideye images will be processed using two methods: linear mixture model (LMM) and artificial neural networks (ANNs). The purpose of using these two methods is to verify their effectiveness for urban growth mapping in the study area. The classification results will be validated using ground reference data collected from the field surveys. In the second year (2013), we carry out monitoring floods and their spatiotemporal dynamics using Landsat, Rapideye and MODIS images. The MODIS will be used to generate time-series maps showing spatiotemporal changes of flood inundation. However, flood maps will be also generated from Landsat and Rapideye data for comparison with MODIS-derived flood maps and for flood risk analysis. In the third year (2014), we will try to assess flood risks due to urbanization. Spatiotemporal variation of flooding inundation with respect to landuse change in the detected urban area will be evaluated, involving topographic and geographical data, for disaster risk accessment. Potential flood risk will be evaluated using elevation-based spatial analysis and remotely-sensed data. It is believed that the results could be useful for policymakers to devise a better urban planning and flood risk management strategy in the future.