Accurate and timely estimates of crop area and yield are crucial for agricultural planning, food security, and market stability. This chapter presents a comprehensive approach for crop acreage and yield estimation using remote sensing, combining satellite imagery analysis with agronomic modeling. We first outline a methodology for mapping crop acreage using multi-spectral Sentinel-2 imagery, including advanced cloud masking (Cloud Score + algorithm), topographic correction, and Random Forest classification implemented on the Google Earth Engine platform. We then describe a semi-physical crop yield modeling approach that integrates satellite-derived indices with crop physiology, notably a light-use efficiency model driven by the fraction of absorbed photosynthetically active radiation (fAPAR), meteorological inputs (solar radiation, temperature), water stress indices, and crop-specific parameters like Radiation Use Efficiency (RUE) and Harvest Index (HI). Case studies from India, sub-Saharan Africa, and the United States demonstrate the accuracy and scalability of these methods—for example, remote sensing-based estimates of wheat area and production in Gujarat closely matched official statistics (within ~ 5–10%), and a national rice forecasting program achieved < 10% error at state scales. We discuss the challenges (cloud cover, small field sizes, model generalizability) and highlight future directions, including multi-source data fusion, machine learning (deep learning and data assimilation), and crowd-sourced ground truth. The results underscore that combining advanced satellite analytics with agronomic modeling can deliver near-real-time, spatially detailed crop insights, supporting decision-makers in achieving resilient and sustainable food systems.