Water Resources Research Act Program

Details for Project ID ND_2023_Lin

Developing Machine Learning and Deep Learning Soil Moisture Models for Precision Agricultural Applications

Institute: North Dakota
Year Established: 2023 Start Date: 2023-09-01 End Date: 2024-05-15
Total Federal Funds: $2,500 Total Non-Federal Funds: $2,782

Principal Investigators: Zhulu Lin, Xin (Rex) Sun

Project Summary: Though the measurement of soil moisture at the point scale using in situ sensors and the estimation of soil moisture at larger scales using remote sensing technology has greatly improved in the past decades, there is a gap in the intermediate scales (10-100 m2) for soil moisture prediction, which is normally required when developing variable rate irrigation description maps for precision agricultural applications. This gap may be filled by taking advantage of the recent advancement in the field of artificial intelligence by developing machine learning and deep learning models to predict soil moisture. We plan to use soil properties and the meteorological data collected at the North Dakota Agricultural Weather Network stations located in the Red River of the North Basin to develop several machine learning models for soil moisture predictions for topsoil and subsoils to help farmers manage their field more efficiently and remain competitive in the industry.