Mapping Crop Types Using Sentinel-2 Data Machine Learning and Monitoring Crop Phenology with Sentinel-1 Backscatter Time Series in Pays de Brest, Brittany, France

Crop supply and management is a global issue, particularly in the context of climate change rising urbanization. Accurate mapping monitoring specific crop types are crucial for studies. In this study, we proposed: (1) methodology to map two main winter crops (winter wheat barley) northern region Finistère with high-resolution Sentinel-2 data. Different classification approaches (the hierarchical classical direct extraction), methods (pixel-based (PBC) object-based (OBC)) were performed evaluated. Subsequently, (2) further study that involved phenology was carried out, based on previous results. The aim understand temporal behavior from sowing harvesting, identifying three important phenological statuses (germination, heading, ripening, including harvesting). Due high frequency precipitation our area, using Sentinel-1 C-band SAR backscatter time series data Google Earth Engine (GEE) platform. results showed achieved better accuracy when it compared extraction, an overall 0.932 kappa coefficient 0.888. Moreover, process, OBC reached cropland mapping, PBC proven more suitable extraction. Additionally, wheat, germination ripening (harvesting) phases can be identified at VV VH/VV polarizations, heading both VH polarizations. Secondly, able detect phase barley VH, polarizations VH/VV, finally,