Use of eddy-covariance methods to “calibrate” simple estimators of evapotranspiration

Abstract. Direct measurement of actual evapotranspiration (ET) provides quantification of this large component of the hydrologic budget, but typically requires long periods of record and large instrumentation and labor costs. Simple surrogate methods of estimating ET, if “calibrated” to direct measurements of ET, provide a reliable means to quantify ET. Eddy-covariance measurements of ET were made for 12 years (2004-2015) at an unimproved bahiagrass (Paspalum notatum) pasture in Florida. These measurements were compared to annual rainfall derived from rain gage data and monthly potential ET (PET) obtained from a long-term (since 1995) U.S. Geological Survey (USGS) statewide, 2-kilometer, daily PET product. The annual proportion of ET to rainfall indicates a strong correlation (r²=0.86) to annual rainfall; the ratio increases linearly with decreasing rainfall. Monthly ET rates correlated closely (r²=0.84) to the USGS PET product. The results indicate that simple surrogate methods of estimating actual ET show positive potential in the humid Florida climate given the ready availability of historical rainfall and PET.