Submitted to: Irrigation Associations Exposition and Technical Conference Proceedings
Publication Type: Proceedings
Publication Acceptance Date: November 15, 2007
Publication Date: December 9, 2007
Citation: Chavez Eguez, J.L., Gowda, P., Howell, T.A., Copeland, K.S. 2007. Evaluating three evapotranspiration mapping algorithms with lysimetric data in the semi-arid Texas High Plains. In: Proceedings of the 28th Annual International Irrigation Show, December 9-11, 2007, San Diego, California. p. 268-283. 2007 CDROM. Interpretive Summary: Crop water use needs to be quantified in order to manage irrigation systems more efficiently. In this study, three remote sensing energy balance based methods (Two Source, METRIC, and Aerodynamic Temperature) were evaluated for their accuracy in estimating distributed evapotranspiration (ET). Data from five weighing lysimeters located at a USDA-ARS research facility in Bushland, Texas, and a Landsat 5 image acquired on July 23, 2006 were used for this purpose. Results showed that all three methods performed satisfactorily for mapping ET in the Texas High Plains. However, the Aerodynamic Temperature based ET model performed better than the other two models with a mean estimation error of -3.2 %. ET mean estimation error for the Two Source model was 7.4 % and 9.0 % for METRIC. More evaluation is needed for all major crops under different climatic conditions.
Technical Abstract: Ground water levels are declining at unsustainable rates in the Texas High Plains. Accurate evapotranspiration (ET) maps would provide valuable information on regional crop water use and hydrology. This study evaluated three remote sensing based algorithms for estimating ET rates for the Texas High Plains. Data from four large-scale weighing lysimeters (two each irrigated and dryland crops) at the Conservation Production Research Laboratory, USDA-ARS at Bushland, Texas, were used to evaluate the remote sensing methods. ET algorithms evaluated include Mapping Evapotranspiration at High Resolution using Internalized Calibration model (METRIC), Two-Source Energy Balance model (TSM), and an Aerodynamic Temperature based Energy Balance model (ATEB). A Landsat 5 TM image acquired on July 23, 2006 was used for estimating ET. Predicted ET values were compared with lysimetric data to determine how well the different ET models worked. A discussion of each model's strength and weaknesses, under the climatic conditions encountered in the Texas High Plains, is provided.