Skip to main content
ARS Home » Plains Area » Fort Collins, Colorado » Center for Agricultural Resources Research » Water Management and Systems Research » Research » Publications at this Location » Publication #378291
Research Project: Improving the Sustainability of Irrigated Farming Systems in Semi-Arid Regions

Location: Water Management and Systems Research

Title: Mapping maize crop coefficient Kc using random forest algorithm based on leaf area index and UAV-based multispectral vegetation indices

Author
item SHAO, GUOMIN - Northwest A&f University
item HAN, WENTING - Northwest A&f University
item Zhang, Huihui
item LIU, SHOUYANG - Institut National De La Recherche Agronomique (INRA)
item WANG, YI - Northwest A&f University
item ZHANG, LIYUAN - Northwest A&f University
item CUI, XIN - Northwest A&f University

Submitted to: Agricultural Water Management
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 4/1/2021
Publication Date: 4/28/2021
Citation: Shao, G., Han, W., Zhang, H., Liu, S., Wang, Y., Zhang, L., Cui, X. 2021. Mapping maize crop coefficient Kc using random forest algorithm based on leaf area index and UAV-based multispectral vegetation indices. Agricultural Water Management. 252. Article e106906. https://doi.org/10.1016/j.agwat.2021.106906.
DOI: https://doi.org/10.1016/j.agwat.2021.106906

Interpretive Summary: This research aimed to explore the potential of combining leaf area index (LAI) and unmanned aerial vehicle (UAV)-based multi-spectral vegetation indices (VIs) for estimating maize crop coefficient (Kc) at the field scale and to obtain Kc maps with high spatial-temporal resolution. The performance of the estimation model of the daily maize Kc derived by two algorithms (random forest and multiple linear regression) based on ground measured LAI and six multi-spectral VIs were evaluated under various irrigation conditions during the entire growing season. The crop evapotranspiration (ET) estimated by VIs-LAI was compared to soil water balance and FAO-56 estimated ET. Results showed that random forest algorithm could well estimate the maize Kc based on LAI and UAV-based VIs. The VIs based on Rededge-Red and Green-Red spectral bands were suitable predictors in Kc prediction model. It was also feasible to obtain the temporal and spatial distribution of the maize Kc based on normalized difference vegetation index derived LAI values and VIs. The study further confirmed that UAV-based multi-spectral remote sensing technology has a great potential in monitoring the distribution of water use and precision irrigation at field scales.

Technical Abstract: The lack of rapid access to spatial and temporal distribution of crop coefficient (Kc) impedes the application of Kc in precision irrigation. This research aimed to explore the potential of leaf area index (LAI) and unmanned aerial vehicle (UAV)-based multi-spectral vegetation indices (VIs) for estimating maize Kc at the field scale and to obtain Kc maps with high spatial-temporal resolution. The performance of the estimation model of the daily maize Kc derived by two algorithms (random forest regression-RFR and multiple linear regression-MLR) based on ground measured LAI and six multi-spectral VIs were evaluated under various irrigation conditions during the entire growing season. The crop evapotranspiration (ET) estimated by VIs-LAI was compared to soil water balance and FAO-56 estimated ET. Results showed that the RFR algorithm could well (R2 = 0.65) estimate the maize Kc based on LAI and UAV-based VIs. The VIs based on Rededge-Red and Green-Red spectral bands were suitable predictors in Kc prediction model. It was also feasible to obtain the temporal and spatial distribution of the maize Kc based on normalized difference vegetation index derived LAI values and VIs. The study further confirmed that UAV-based multi-spectral remote sensing technology has a great potential in monitoring the distribution of water use and precision irrigation application at field scales.