Skip to main content
ARS Home » Plains Area » El Reno, Oklahoma » Oklahoma and Central Plains Agricultural Research Center » Livestock, Forage and Pasture Management Research Unit » Research » Publications at this Location » Publication #365122
Research Project: Integrated Agroecosystem Research to Enhance Forage and Food Production in the Southern Great Plains

Location: Livestock, Forage and Pasture Management Research Unit

Title: Modeling evapotranspiration of winter wheat using contextual and pixel-based surface energy balance models

Author
item KHAND, KUL - Oklahoma State University
item BHATTARAI, NISHAN - University Of Michigan
item TAGHVAEIAN, SALEH - Oklahoma State University
item Wagle, Pradeep
item Gowda, Prasanna
item ALDERMAN, PHILLIP - Oklahoma State University

Submitted to: Transactions of the ASABE
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 12/6/2020
Publication Date: 3/24/2021
Publication URL: https://handle.nal.usda.gov/10113/7329720
Citation: Khand, K., Bhattarai, N., Taghvaeian, S., Wagle, P., Gowda, P.H., Alderman, P. 2021. Modeling evapotranspiration of winter wheat using contextual and pixel-based surface energy balance models. Transactions of the ASABE. 64(2):507-519. https://doi.org/10.13031/trans.14087.
DOI: https://doi.org/10.13031/trans.14087

Interpretive Summary: Major limitations of the field-based evapotranspiration (ET) measurement methods are inability to capture ET from large heterogeneous areas and extensive maintenance requirements. Thus, remote sensing-based models are considered as viable options for mapping ET across various landscapes. However, it is necessary to compare performance of these models for estimating ET under different management practices at the same site to identify and develop robust models. Thus, we compared the performance of five remote sensing-based models: mapping evapotranspiration at high resolution with internalized calibration (METRIC), surface energy balance algorithm for land (SEBAL), triangular vegetation temperature (TVT), surface energy balance system (SEBS), and two-source energy balance (TSEB) for estimating daily ET for winter wheat fields under three different grazing (grain-only, graze-grain, and graze-out) and two tillage (no-till and conventional till) practices. The modeled ET was compared against eddy covariance measurements of ET. The SEBAL performed the best followed by TVT. The TSEB performed the worst among the five models and showed a strong tendency to overestimate ET under low vegetative conditions.

Technical Abstract: Surface energy balance (SEB) models based on remote sensing data are widely used to map evapotranspiration (ET) across various landscapes. However, their ability to capture ET from rainfed winter wheat (Triticum aestivum L.) is still not well understood. Understanding winter wheat ET dynamics is important in agricultural regions such as the Southern Great Plains (SGP) of the United States (US), where this crop is managed under different grazing and tillage practices. The goal of this study was to evaluate the performance of five widely used SEB models driven by remotely sensed data for winter wheat daily ET estimation in Oklahoma, US under variable grazing and tillage managements. These models include mapping evapotranspiration at high resolution with internalized calibration (METRIC), surface energy balance algorithm for land (SEBAL), triangular vegetation temperature (TVT), surface energy balance system (SEBS) and two-source energy balance (TSEB). The results were compared against observed daily ET data from six eddy covariance towers in rainfed winter wheat fields under three different grazing (grain-only, graze-grain, and graze-out) and two different tillage (conventional till and no-till) practices. Large variations were observed among the estimates of the SEB models. SEBAL and TSEB were found to be the best and worst performing models, respectively. SEBAL results showed that grain-only wheat had the highest mean daily ET, followed by graze-grain and graze-out wheat. Among the tillage treatments, conventional till had larger ET than no-till. However, none of these differences were statistically significant (p>0.05).