A multi-sensor analysis of selected reflectance-based crop coefficient models for daily maize evapotranspiration estimation

Authors: COSTA-FILHO, EDSON - Colorado State University; CHAVEZ, JOSE - Colorado State University; Zhang, Huihui

Submitted to: Journal of Agricultural Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 9/10/2023
Publication Date: 11/15/2023
Citation: Costa-Filho, E., Chavez, J.L., Zhang, H., Andales, A.A., Brown, A. 2023. A multi-scale analysis of reflectance-based crop coefficient models for daily maize evapotranspiration estimation. Journal of Agricultural Science. 15(12). https://doi.org/10.5539/jas.v15n12p1.
DOI: https://doi.org/10.5539/jas.v15n12p1

Interpretive Summary: In this advanced study, we focused on a critical aspect of agriculture: how to efficiently manage water for maize crops in a semi-arid climate. We used cutting-edge technology, such as various remote sensing devices mounted on satellites, drones, and handheld sensors, to collect data about the crops. We tested three different mathematical models known as reflectance-based crop coefficient models (RBCC). These models used different vegetation indices as input variables to predict daily maize evapotranspiration (ETa). The performance of these models using the data from two research sites in Colorado, where maize was being grown using different irrigation systems. Our findings revealed that no single remote sensing platform provided the most accurate results across all the models they tested. However, the handheld MSR sensor, operating at a distance of one meter from the plants, yielded the best data for one of the RBCC models. The Sentinel-2 satellite platform proved to be most effective for the other two RBCC models. These results highlight the importance of choosing the right combination of remote sensing tools and mathematical models to manage water effectively for maize crops in such dry regions. By doing so, farmers can make better irrigation decisions and contribute to sustainable water management in agriculture.

Technical Abstract: This study evaluated the performance of three reflectance-based crop coefficient models (RBCC) for daily maize evapotranspiration (ETa) using different spaceborne, airborne, and proximal multispectral data in a semi-arid climate region to identify the optimal multispectral sensor that gives the best ETa estimates for irrigation water management. The different remote sensing (RS) multispectral sensors were Landsat-8 (30 m), Sentinel-2 (10 m), Planet CubeSat (3 m), multispectral radiometer or MSR (1 m), and unmanned aerial vehicle or UAS (0.03 m). The algorithms of ETa evaluated in this study were three RBCC models using different vegetation indices as input variables. One RBCC had the normalized difference vegetation index (NDVI) as the RS input variable. Another RBCC model had soil-adjusted vegetation index (SAVI) variable as input, and the third RBCC model had fractional green vegetation cover (fc) as the RS predictor for daily maize ETa. The data for this study were from two maize research sites in Greeley and Fort Collins, Colorado, USA, in 2020 and 2021. The Greeley site had a subsurface drip system, while the Fort Collins site had surface irrigation (furrow). Daily maize ETa predictions were compared with observed daily maize ETa data from an Eddy Covariance system installed at each research site. Results indicated that no unique remote sensing platform provides more accurate results across all the RS of ETa algorithms evaluated. The MSR handheld sensor (1 m) provided the best remote sensing data for the SAVI-based RBCC model for daily maize ETa. Sentinel-2 was the best platform for the remaining two RBCC daily maize ETa algorithms. These results indicate the need to develop methods to improve remote sensing data from sub-optimal platforms to advance sustainable irrigation water management further.