The main objective of this visit is to leverage advanced deep learning-based spatiotemporal fusion techniques for integrating multi-source satellite imagery, such as Sentinel-2, Landsat, and MODIS. The goal is to accurately estimate the water balance components, with a specific focus on the snow component in Moroccan Atlas Mountains. The generated high-resolution time series will serve as crucial input for modeling hydrological parameters at the basin scale, enhancing our understanding and management of water resources.

This internship focused on the spatio-temporal fusion of multi-source satellite images for snow cover mapping in the Moroccan Atlas. The project aimed to address the limitations of current satellite sensors by combining data from Landsat-8 and Sentinel-2 using advanced fusion techniques, such as the Enhanced Spatial and Temporal Adaptive Reflectance Fusion Model (ESTARFM) and the Flexible Spatio-Temporal Data Fusion Model (FSDAF). These methods enabled the generation of high-resolution snow cover time series, essential for accurate snow monitoring and water resource management in semi-arid mountainous regions. The project successfully improved the temporal and spatial resolution of snow cover data to support more effective hydrological modeling and decision-making.