Multi-Scenario Simulations of Future Forest Cover Changes Influenced by Socio-Economic Development: A Case Study in the Chiang Mai-Lamphun Basin

Abstract: Changes in land cover in the Chiang Mai-Lamphun basin have been influenced by pressures of rapid socio-economic developments. The Markov-cellular automata and a multi-layer perceptron technique (Markov-CA-MLP) were employed to simulate three scenarios in 2021. Then, the future land cover maps in 2030 and 2050 were built based on the transition probability metric from 2021. The different scenarios were based on socio-economic schemes, which include the business-as-usual (BaU), the ecological protection scenario (EPS), and the baseline development scenario (BDS). A result of model validations using 534 ground survey points in 2021 showed that the BaU model in 2021 generated the highest overall accuracy (82.77%) with a Kappa value of 0.7846, quantity disagreement value of 0.0693, and allocation disagreement value of 0.1030. The projected BaU in 2050 revealed a decrease in forest land (6.70%). At the same time, the built-up and agricultural areas gained 5.57% and 0.88%, respectively. In BaU, class transformation between 2021 and 2050, including forests to agricultural areas (6.04%), agricultural areas to built-up areas (4.62%), and forests to built-up areas (1.59%). The BDS depicted the lowest accuracy level compared to BaU and EPS. Following this procedure, this study can provide scientific trends for possible land use management in the Chiang Mai-Lamphun basin based on the described socio-economic settings.