Drought is a complex natural disaster with cyclical occurrence, slow onset, and wide - ranging, long - lasting impacts that cause severe damage to socio - economic systems, the environment, and ecosystems. Effective monitoring and forecasting are crucial for early detection, severity assessment, and drought risk management. This study presents a novel approach to drought monitoring by integrating multi - source remote sensing data with artificial intelligence (AI) techniques. Landsat and Sentinel - 2 imagery were used to derive ten widely applied drought indices (NDVI, SAVI, VCI, NDWI, LSWI, MSI, NDDI, TCI, TVDI, VHI). Machine learning algorithms , including Random Forest (RF), Support Vector Machine (SVM), and Gradient Tree Boosting (GTB) , were employed to classify and predict drought risk based on a training dataset constructed from the TVDI index. The entire workflow was implemented on the Google Earth Engine (GEE) platform, enabling large - scale data processing and automation. The proposed method enhances accuracy, efficiency, and automation in drought monitoring, thereby supporting early warning systems and sustainable water resource management under the increasingly complex challenges of climate change.
Tran et al. (Fri,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: