An optimized surface aerodynamic temperature approach to estimate maize sensible heat flux and evapotranspiration

Authors: COSTA-FILHO, EDSON - COLORADO STATE UNIVERSITY; CHÁVEZ, JOSÉ - COLORADO STATE UNIVERSITY; Zhang, Huihui; ANDALES, ALLEN - COLORADO STATE UNIVERSITY

Submitted to: Agricultural and Forest Meteorology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 10/12/2021
Publication Date: 12/15/2021
Citation: Costa-Filho, E., Chávez, J.L., Zhang, H., Andales, A.A. 2021. An optimized surface aerodynamic temperature approach to estimate maize sensible heat flux and evapotranspiration. Agricultural and Forest Meteorology. 311. Article e108683. https://doi.org/10.1016/j.agrformet.2021.108683.
DOI: https://doi.org/10.1016/j.agrformet.2021.108683

Interpretive Summary: Accurately modeling aerodynamic surface temperature (To) has become relevant to improve crop evapotranspiration estimates from the surface energy balance approach. This study describes the development and validation of an optimized linear model of To, which serves as an input to model sensible heat in maize fields. The best-optimized linear model includes four variables: air temperature, nadir surface temperature, fractional vegetation cover, and a new variable that accounts for wind speed and the relative angle of attack between wind direction and crop row orientation. Data were collected from 2017 to 2019 on two maize fields at the Limited Irrigation Research Farm near Greeley, Colorado, USA. Micrometeorological data were measured on site. Stationary infrared thermometers measured nadir surface temperature. Multispectral surface reflectance data were collected on-site using a radiometer weekly. Measured net radiation and soil heat flux were acquired at ¼ and ½ of the field’s length. Measured sensible heat for calibrating the optimized To model was collected in 2017 and 2018 using two large-aperture scintillometers (LAS). Results indicated that the optimized To model improved the estimation of sensible heat fluxes by 31% and latent heat fluxes by 9% compared to its non-optimized counterpart linear model. The sensitivity analysis indicated that air temperature measurements and surface temperature added the most variability to estimates using the optimized model.

Technical Abstract: Typical errors for single-source sensible heat models range from 15-40% for vegetated surfaces, with 10 to 20% introduced by stability functions and remaining errors due to input data and assumptions regarding the aerodynamic surface temperature (To). Thus, accurately modeling To has become relevant to improve crop evapotranspiration estimates from the surface energy balance approach. This paper describes the development and validation of an optimized linear model of To, which serves as an input to model sensible heat in maize fields. The best-optimized linear model includes four variables: air temperature, nadir surface temperature, fractional vegetation cover, and a new variable that accounts for wind speed and the relative angle of attack between wind direction and crop row orientation. Data were collected from 2017 to 2019 on two maize fields at a Research Farm near Greeley, Colorado, USA. Micrometeorological data were measured in-situ at 3.30 m above the ground surface. Stationary infrared thermometers measured nadir surface temperature. Multispectral surface reflectance data were collected on-site using a radiometer weekly. Measured net radiation and soil heat flux were acquired at ¼ and ½ of the field’s length. Measured sensible heat for calibrating the optimized To model was collected in 2017 and 2018 using two large-aperture scintillometers (LAS). A one-way Analysis of Variance (ANOVA) test was performed to justify modeling To for different maize leaf area index values. A local sensitivity analysis study was performed to identify the most influential input variables in the optimized model. Results indicated that the optimized To model improved the estimation of sensible heat fluxes by 31% and latent heat fluxes by 9% compared to its non-optimized counterpart linear model. The sensitivity analysis indicated that air temperature measurements and surface temperature added the most variability to estimates using the optimized model.