Application of Geo-Informatics in Predicting Forest Land Use at the Doi Inthanon National Park, Chiang Mai Province

Abstract

                The objectives of this study were to apply Geo-informatics in change analysis and prediction of land use and land cover at the Doi Inthanon national park, Chiang Mai province. The visual interpretation classification technique was introduced to classify the land use during 2009, 2014 and 2019, using satellite images derived from LANDSAT-5 and LANDSAT-8. Five types of land cover were detected which are agricultural land, forest land, miscellaneous land, urban and build-up land, and water body.

                The land use changes during 2009-2014 indicated that areas under agricultural, miscellaneous, and urban and build-up increased by 43.50%, 4.45% and 2.05%, respectively and forest cover decreased by 50.00%, while the area under water body remained unaltered. Land use changes during 2014-2019 indicated that the area under urban construction and miscellaneous increased by 35.18% and 14.82%, respectively, while the forest and agricultural areas decreased by 40.39% and 9.61%, respectively and water body remained unaltered. An accuracy assessment using the confusion matrix indicated that the overall accuracy was 94.85%. We predicted the changes in land use during 2019-2024, using Land Change Modeler (LCM) and taking the land use during 2014-2019 as the base data. The predictions indicate that urban and build-up and miscellaneous land use during 2019-2024 would increase by 45.72% and 4.27%, respectively. Forest cover and agricultural area would decrease by 49.61% and 0.39% respectively, while the water body would remain unaltered. The predictions indicate to a change of land use from forest to urban and build-up area. We suggest a proactive forest management and surveillance to have a proper land use in the near future.