抄録
With the widespread application of big data technology in agriculture, the development and optimization of wheat growth models are entering a new stage of data-driven and intelligent approaches. This paper systematically reviews the progress in the construction and optimization of wheat growth models under the big data environment. First, it analyzes the key role of big data in model development, covering the diversification of data sources, methods of data preprocessing and integration, as well as the advantages of big data-driven models in expanding spatiotemporal scales and improving prediction accuracy. Second, it reviews the structural characteristics of current mainstream wheat growth models, parameterization and calibration methods, model validation and uncertainty analysis techniques, and summarizes strategies and practical approaches for model optimization. Then, it discusses the application performance of wheat growth simulation under different ecological and management conditions, emphasizing its practical value in precision agriculture, crop regulation, and agricultural decision support, while also pointing out challenges such as insufficient model generalization capacity, uneven data quality, and the complexity of technological integration. Finally, it looks ahead to future research trends, including multi-model coupling and integration, real-time dynamic simulation based on artificial intelligence, and the construction of interdisciplinary, open, and shared data platforms. This paper aims to provide systematic references for wheat growth modeling and application research, promoting the deep integration and innovative development of big data technology in agricultural system modeling.
