摘要
The prediction of shoreline erosion is vital for coastal management. This study aims to utilize geo-informatics technology to increase accuracy of a shoreline prediction model along two study sites in Samutprakarn province and in Prachuabkirikhan province. Predicting coastline change using remote sensing together with GIS （geographic information system） is a spat^o-temporal technology, which can continuously provide perspectives of coastal areas. Due to a long term of operational period of LANDSAT satellite, it is useful to enhance accuracy of prediction model. LANDSAT-5 TM images acquired during 1999-2009 were used to produce historical shoreline vectors. Physical data were modified to be input data of digital shoreline analysis system. The model was validated. Linear regressions were applied in order to derive equations of erosion magnitude. The result presents that averaged erosion and accretion rate along Samutprakarn province was 22.30 meters/year and 2.94 meters/year, respectively. On the other hand, the average rate of coastal erosion along Prachuabkirikhan province was much lower, being 2.48 meters/year while the accretion rate was approximately 4.11 meters/year. The predicted shoreline change at Samutprakarn province in 2019 is about -132.69 ~ 0.758 meters while at Prachuabkirikhan is 40.58 ~ 0.0012 meters. In conclusion, this prediction model focused the changing of shoreline in long term and accuracy of the model could be improved by increasing number of shorelines vectors, transect intervals and resolution of satellite images. Clearly, the model is flexible and can be applied in other particular areas for coastal zone management in Thailand.
