Overall, with an encouragingly high R2 (0.65-0.80) and modest root mean square errors (12-20%), the results demonstrate that accurate estimations of agricultural productivity are attainable utilizing the suggested modeling framework. Though our results also show the potential of Sentinel-1, which has the advantage of not being influenced by clouds, Sentinel-2 was the most significant data source. The data were consistent over the years, which varied greatly in terms of crop yield and rainfall, as well as between the sites, which are distinguished by different crop compositions. Cloud-based data processing and open access remote sensing data sources are viable substitutes for enhancing generalization and long-term monitoring. However, the availability of high-quality reference data that accounts for variations in crop conditions and environment is a prerequisite for the successful operation of such a system. Creating crowdsourcing programs where local farmers are educated to gather and provide crop surveys across regions is one approach to accomplish this.