Prediction of maize crop coefficient from UAV multisensor remote sensing using machine learning methods

Authors: SHAO, GUOMIN - Northwest A&f University; HAN, WENTING - Northwest A&f University; Zhang, Huihui; ZHANG, LIYUAN - Northwest A&f University; WANG, YI - Northwest A&f University; ZHANG, YU - Northwest A&f University

Submitted to: Agricultural Water Management
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/22/2022
Publication Date: 11/29/2022
Citation: Shao, G., Han, W., Zhang, H., Zhang, L., Wang, Y., Zhang, Y. 2022. Prediction of maize crop coefficient from UAV multisensor remote sensing using machine learning methods. Agricultural Water Management. 276. Article e108064. https://doi.org/10.1016/j.agwat.2022.108064.
DOI: https://doi.org/10.1016/j.agwat.2022.108064

Interpretive Summary: There is a need for prompt availability of crop water use estimation in precision irrigation. Farmers and irrigation schedulers often use crop coefficient (Kc) and reference evapotranspiration method in FAO56 manual. However, the Kc values need to be adjusted for crop types, growth stages, and other field conditions. It is necessary to develop a method or tool for users to determine Kc in a timely manner. In this study, we used an unmanned aerial vehicle (UAV)-based multi-types of sensor data to estimate Kc for irrigated maize in a semi-arid region in Northwest China. We found that UAV multispectral and thermal data showed greater contributions in the estimation of Kc and crop water use. The integration of UAV remote sensing and advanced data processing methods provides a promising tool to help farmers make decisions using timely mapped crop water needs, especially under water shortages or drought conditions.

Technical Abstract: In the upcoming irrigation management in agricultural production, accurate mapping of crop water consumption with a high spatial and temporal resolution at a farm scale is needed. In this study, we developed models for crop coefficient (Kc) estimation using unmanned aerial vehicle (UAV) remote sensing and machine learning techniques for irrigated maize in a semi-arid region in Northwest China. Kc values were calculated using a procedure given in FAO-56 using field measurements. Multispectral vegetation indices (VIs), vegetation fraction (VF), thermal-based VIs, and texture information (TI) were derived from UAV-based multispectral, RGB, and thermal infrared imagery, respectively. These remotely sensed variables and their combinations were used to develop prediction models using six machine learning algorithms (linear regression-LR, polynomial regression-PR, exponential regression-ER, random forest regression-RFR, support vector regression-SVR, and deep neural network-DNN). Among these models, the RFR method with the highest accuracy (R2 = 0.69; RMSE = 0.1019) was recommended to estimate maize Kc. The multispectral and thermal-based VIs and texture of the near-infrared band had greater contributions than RGB-based VF and TI in the Kc-RFR model under different irrigation treatments. Furthermore, the maize Kc-RFR prediction model had high accuracy in estimating cumulative evapotranspiration (R2 = 0.89, RMSE = 15.0 mm/stage) during different growth stages and daily soil water content (R2 = 0.85, RMSE = 0.0057) in the root zone. These results show that the integration of UAV remote sensing and ML provides a promising tool to help farmers make decisions using timely mapped crop water consumption, especially under water shortages or drought conditions.