Field Crop 2025, Vol.8, No.6, 293-300 http://cropscipublisher.com/index.php/fc 293 Feature Review Open Access Integrating Remote Sensing and Crop Modeling for Real-Time Yield Prediction inWheat Delong Wang, Pingping Yang, Jiong Fu Hainan Provincial Key Laboratory of Crop Molecular Breeding, Sanya, 572025, Hainan, China Corresponding email: jiong.fu@hitar.org Field Crop, 2025, Vol.8, No.6 doi: 10.5376/fc.2025.08.0030 Received: 13 Nov., 2025 Accepted: 24 Dec., 2025 Published: 13 Dec., 2025 Copyright © 2025 Wang et al., This is an open access article published under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Preferred citation for this article: Wang D.L., Yang P.P., and Fu J., 2025, Integrating remote sensing and crop modeling for real-time yield prediction in wheat, Field Crop, 8(6): 293-300 (doi: 10.5376/fc.2025.08.0029) Abstract Wheat is one of the most important cereal crops globally, and accurate yield prediction is critical for ensuring food security and supporting precision agriculture. Traditional estimation methods relying on field surveys are often time-consuming, labor-intensive, and limited in spatial coverage. This study integrates remote sensing technologies with crop modeling approaches to establish a real-time and scalable framework for wheat yield prediction. Specifically, we utilized satellite and UAV-based remote sensing data-including NDVI, LAI, and chlorophyll indices-combined with process-based crop models such as WOFOST and APSIM to simulate crop growth dynamics. The integration was achieved through data assimilation techniques that continuously feed remote observations into crop models, enabling dynamic calibration and validation across growth stages. A case study in the Indo-Gangetic Plains demonstrated that assimilating Sentinel-2 data with the WOFOST model significantly improved yield prediction accuracy and provided timely forecasts beneficial for regional decision-making. This integrated approach enhances both spatial and temporal resolution in yield monitoring, offering a more reliable foundation for precision management, early warning systems, and policy development. Future research should focus on incorporating artificial intelligence and machine learning algorithms to refine model performance and expanding open-access platforms for wider application in climate-resilient wheat production. Keywords Remote sensing; Crop modeling; Wheat yield prediction; Data assimilation; Precision agriculture 1 Introduction Wheat is almost everywhere on people's dining tables. It is not only a staple food in many countries but also an important source of calories and protein globally (Ma et al., 2024). When it comes to food security, wheat is often an unavoidable topic. Especially in the context of frequent climate change and continuous population growth, whether the yield of wheat can be accurately predicted is no longer just a scientific research issue, but is related to government decision-making, market fluctuations and even farmers' income (Cheng et al., 2022). However, the prediction methods commonly used in the past, such as field investigations and manual sampling, although seemingly intuitive, were labor-intensive, time-consuming, had small data volumes, and were also easily affected by regional differences. When encountering extreme weather or areas with complex terrain, these methods become even more inadequate. In addition, they have difficulty revealing the intricate relationships among genotypes, environmental conditions and agricultural management, and the prediction accuracy is naturally limited (Gawdiya et al., 2024). So, researchers began to try a new combination idea: using remote sensing technology and crop models "together". Satellite images can cover large agricultural areas in a very short time, providing dynamic information on crop growth, while models and machine learning algorithms are responsible for integrating these images, meteorological data, soil characteristics, etc., to form more comprehensive yield estimates (Bian et al., 2022). Some of the latest studies show that this approach of combining multi-source data not only makes the prediction results more timely and accurate, but also significantly helps local agricultural management and policy-making (Ashfaq et al., 2024).
RkJQdWJsaXNoZXIy MjQ4ODYzNA==