Released by the Institute of Agricultural Resources and Regional Planning under the Chinese Academy of Agricultural Sciences on World Soil Day, the model represents a shift in China’s farmland management approach. It aims to move from a focus on preserving land quantity to improving land quality, a task that traditional monitoring methods can no longer manage effectively.
Developed under the leadership of Tang Huajun, an academician of the Chinese Academy of Engineering, the model consists of a central foundation system supported by specialised vertical models. These sub-systems handle tasks such as field segmentation, crop classification and engineering quality inspections.
Wu Wenbin, director general of the CAAS Institute, said the AI model does more than provide observations. It can evaluate soil health, forecast changes and automatically create management plans, allowing full-life-cycle supervision for high-standard farmland.
China currently has more than 66.7 million hectares of such farmland, where soil fertility is maintained through modern technology and science-based agricultural practices.
In a pilot project conducted in Kunshan in Jiangsu Province, the AI system demonstrated its capability by dynamically adjusting fertiliser application based on weather patterns and crop growth stages, creating a complete closed-loop management system.
Industry experts say the introduction of this AI model will support the sustainable management of agricultural resources, enhance the quality of arable land and advance the development of high-standard farmland across the country.
