Integration of Remote Sensing and Crop Modeling for Yield Prediction: A Review

Abstract

Predicting crop yields accurately is very important because it helps ensure food security and allows farmers to manage their farms more effectively and make better decisions. Traditional methods of data collection, such as sending people to survey fields and using statistics, are often time-consuming and do not work well for large areas of land. With the rapid advancement of digital agriculture technologies, we can now use satellites, airplanes, and drones to collect data on crop performance, which is a significant advantage. Crop simulation models are computer programs that simulate how crops grow, where they are planted, and how they are affected by factors such as weather and soil conditions. Some commonly used models include DSSAT, APSIM, WOFOST, and AquaCrop. These models help us understand how crops will respond to different conditions, such as changes in weather or management practices. When we combine data from sensors with crop simulation models, we can make much more accurate predictions of crop yields. This is because we can incorporate a variety of data, such as plant health, soil moisture, and other relevant variables. Feeding this information into the computer models improves their accuracy and predictive capability. New techniques, such as integrating multiple types of data using machine learning and hybrid modeling, can further enhance these models. Therefore, the use of sensing technologies combined with crop simulation models is a powerful approach for monitoring crops and predicting yields, which supports sustainable agriculture, precision farming, and global food security.