Chudech LOSIRI, Srinakharinwirot University, Thailand
A future land use and land cover change (LULC) is important phenomena which several organizations related to the spatial planning have been interested in. Samut Sakon is considering as agricultural and residential areas for supporting an increasing in population and housing of Bangkok Metropolis. This province plays attention to respond a demand from dwellers to locate suitable infrastructure.
To address above, this study divided the analysis procedure into two steps. The first is to utilize the multi-dates of Landsat images in 2007, 2012 and 2017 to determine the LULC into five classes like urban, agriculture, water, forest, and miscellaneous. The accuracy assessment was prepared to confirm the consistency between the classified maps and the reference data. Then, the change detection was investigated to quantify the predominant LULC change. The second is the future LULC prediction. Land change modeler (LCM) was used to predic LULC in 2032. Driving factors such as distance from built-up, water, road, and others were prepared to identify the influence of the LULC conversions. The Cramer's V was used to determine the level of the correlation between each factor and LULC change. A multilayer perceptron was used to create the transition potential maps for each conversion of LULC. A Markov chain calculated a transition probability matrix in the time step. The transition potential maps, and transition probability matrix were used to predict the LULC in 2017 to evaluate the efficiency of the model.
The result of the classification could see that the agriculture was the major LULC of Samut Sakon; however, its number had decreased dramatically. The LCM yielded the high accuracy result in the LULC simulation in 2017 which implied that it could be used for the prediction of the LULC in 2032 reasonably. The predicted result shows that the urban area in this province risen into about 32 percent which came from the conversion of agriculture and miscellaneous respectively.
Mots clés : Land use|Land cover |Samut Sakon|Geoinformatics |Land Change Modeling
A103777CL