Estimation of transpiration coefficient and aboveground biomass in maize using time-series UAV multispectral imagery

作     者：Guomin Shao Wenting Han Huihui Zhang Yi Wang Liyuan Zhang Yaxiao Niu Yu Zhang Pei Cao Guomin Shao;Wenting Han;Huihui Zhang;Yi Wang;Liyuan Zhang;Yaxiao Niu;Yu Zhang;Pei Cao

作者机构：College of Mechanical and Electronic EngineeringNorthwest A&F UniversityYangling 712100ShaanxiChina Key Laboratory of Agricultural Internet of ThingsMinistry of AgricultureYangling 712100ShaanxiChina Institute of Water-Saving Agriculture in Arid Areas of ChinaNorthwest A&F UniversityYangling 712100ShaanxiChina Water Management and Systems Research UnitUSDA-ARS2150 Centre AvenueBldg.D.Fort CollinsCO 80526USA College of InformationXi’an University of Finance and EconomicsXi’an 710100ShaanxiChina Institute of Soil and Water ConservationNorthwest A&F UniversityYangling 712100ShaanxiChina University of Chinese Academy of SciencesBeijing 100049China 

出 版 物：《The Crop Journal》 (作物学报（英文版）)

年 卷 期：2022年第10卷第5期

页      面：1376-1385页

核心收录：

学科分类：0710[理学-生物学] 082804[工学-农业电气化与自动化] 08[工学] 0828[工学-农业工程] 09[农学] 0901[农学-作物学] 

基　　金：funded by the National Natural Science Foundation of China (51979233) the Natural Science Basic Research Plan in Shaanxi Province of China (2022JQ-363)。 

主　　题：Crop transpiration Normalized difference red-edge index Unmanned aerial vehicles Random forest regression Biomass 

摘      要：Estimating spatial variation in crop transpiration coefficients(CTc) and aboveground biomass(AGB)rapidly and accurately by remote sensing can facilitate precision irrigation management in semiarid regions. This study developed and assessed a novel machine learning(ML) method for estimating CTc and AGB using time-series unmanned aerial vehicle(UAV)-based multispectral vegetation indices(VIs)of maize under several irrigation treatments at the field scale. Four ML regression methods: multiple linear regression(MLR), support vector regression(SVR), random forest regression(RFR), and adaptive boosting regression(ABR), were used to address the complex relationship between CTcand VIs. AGB was then estimated using exponential, logistic, sigmoid, and linear equations because of their clear mathematical formulations based on the optimal CTcestimation model. The UAV VIs-derived CTcusing the RFR estimation model yielded the highest accuracy(R^(2)= 0.91, RMSE = 0.0526, and n RMSE = 9.07%). The normalized difference red-edge index, transformed chlorophyll absorption in reflectance index, and simple ratio contributed significantly to the RFR-based CTcmodel. The accuracy of AGB estimation using nonlinear methods was higher than that using the linear method. The exponential method yielded the highest accuracy(R^(2)= 0.76, RMSE = 282.8 g m, and n RMSE = 39.24%) in both the 2018 and 2019 growing seasons. The study confirms that AGB estimation models based on cumulative CTcperformed well under several irrigation treatments using high-resolution time-series UAV multispectral VIs and can support irrigation management with high spatial precision at a field scale.