Analysis of Evaporative Flux Data for Various Climates

Abstract

Estimation of evapotranspiration is a key requirement of hydrologic balance studies and climate analysis. The study reported involved collection of precise weighing lysimeter and meteorological data from three sites representing distinct climates. The combined data set for daily amounts of evapotranspiration and meteorological variables covers a total of 19 years on either an annual or growing season basis. The pan evaporation, Priestley‐Taylor, original Penman, and Penman‐Monteith evapotranspiration estimating methods are compared with lysimeter measurements using a moving average of 1–30 days. The results indicate the applicability of the various methods as a function of climate regime and the reduction in standard error of the estimate and increase in the coefficient of determination as a function of length of the moving average period. The results can be used both to determine which methods are most applicable for different climates and the expected magnitude of the error as a function of the estimating interval. This study indicates that a 5–10‐day moving average can reduce the standard error of the estimate and increase the coefficient of determination significantly between estimated and measured reference evapotranspiration for several estimating methods for various climates.