Linear models for estimating annual and growing season reference evapotranspiration using averages of weather variables

摘要

We develop linear regression equations to estimate location-specific average annual reference evapotranspiration (ET_o) using one or more of annual averages of: incoming solar radiation (R_s), air temperature (T), relative humidity (RH), and wind speed (U). We also provide two sets of equations to estimate growing season ET_o, either using one or more of annual averages of Rs, T, RH, and U, or using growing season averages of the same variables. The equations are developed using the FAO-56 Penman-Monteith model ET_o estimates as a reference. Supporting weather data to develop the regression equations were from 102 locations (494 station-years) across the contiguous United States. The models were tested with additional data from 32 stations (114 station-years). To illustrate potential applications of the regression models, we estimate spatial patterns of annual ET_o and growing season ET_o across the contiguous United States using existing spatial datasets of annual averages of R_s, T, RH, and U. Other applications of the models provided may include rapid assessments of historical annual and growing season ET_o, evaluation of past ET_o trends, or evaluation of ET_o projected trends based on output from global climate models.