Exploring deep learning methods for analyzing land use change

The process of land use change (LUC) results from human interactions with the natural environment to meet the needs from societal development. Growing population leads to the depletion of the land resource which entails environmental consequences from local to global scales. Advanced analytical methods can help with the understanding of the complexity of LUC process. They can further benefit sustainable land development. The main objective of this thesis research is to evaluate the deep learning (DL) methods such as convolutional neural networks (CNN) and recurrent neural networks (RNN) for classifying and forecasting LUC. The results demonstrated that the CNN-based LU classification models achieved the model accuracy of above 95%, while the RNN-based models for short-term LUC forecasting had 86% forecasting accuracy. This thesis contributes to advancing the methods for LUC analysis and improving the understanding of LUC process for sustainable land management.