Rice Crop Growth Stage Monitoring Using Harmonized Sentinel-2 and SAR Data in Google Earth Engine

Effective monitoring of rice crop growth stages is critical for optimizing agricultural management and yield prediction, especially in regions with extensive rice cultivation, such as Pavia, Italy. This study introduces a robust method for classifying rice crop growth stages by integrating Sentinel-l Synthetic Aperture Radar (SAR) and Sentinel-2 optical data within the Google Earth Engine (GEE) platform. Utilizing NDVI derived from Sentinel-2 and VH polarization data from Sentinel-I, we generated monthly composites spanning two growing seasons (2019–2020), creating a fused time-series dataset that captures the temporal dynamics of rice growth. A k-means clustering algorithm was applied to classify rice into early, middle, and late growth stages, with results validating the effectiveness of this integration for tracking crop phenology, particularly in cloud-prone conditions where optical data alone may be limited. The study demonstrates that combining optical and SAR data offers a scalable and accurate solution for monitoring rice phenology, enhancing data reliability and stage-specific classification. This approach is adaptable for large-scale agricultural monitoring and can be extended to different crop types and regions, thereby supporting precision agriculture, food security, and sustainable land management initiatives.