Noppon Choosri. Cellular Automata for predicting mangrove forest area changes : a case study of Ban Laem district, Phetchaburi province. Master's Degree(Information Management on Environments and Resources). Mahidol University. : Mahidol University, 2009-04-22.
Cellular Automata for predicting mangrove forest area changes : a case study of Ban Laem district, Phetchaburi province
Abstract:
Mangroves play a crucial role supporting life along the shoreline. However, mangrove area is decreasing continuously due to urbanization and human activities. This research aims to understand and predict by an effect of urban expansion on mangrove. A cellular automata model, a local effect model, was integrated with remote sensing to keep track of land use change of Ban Laem district, Phetchaburi province between 1994 and 2004. The model focuses on three categories of the land use i.e., existing mangrove, human community and activities, degraded or regenerated mangrove area. A principle rule of cellular automata is that all the states can be switched to another state by considering their surrounding environments, transition rules of Moors neighborhood are applied to each cell (pixel) in each year. The results of the prediction are compared to a maximum likelihood classification to test the model. Results of the study showed that the CA model produces an overall accuracy of 91.07%. However, the category mangrove area only showed 42.98 % prediction accuracy