Some studies that investigate the climate change and hydrologic balance relationships utilize reference (potential) evapotranspiration (ETref) to either calculate the changes in trends and magnitude of actual ET or to determine changes in atmospheric demand. In such cases, it is important to acquire robust ETref estimates to correctly assess the impact of changes in meteorological variables on atmospheric evaporative demand, hydrologic balances, response of vegetation to climate, and their interactions. Despite its crucial importance, unfortunately, ETref is sometimes poorly addressed in climate change studies as some studies utilize temperature or radiation-based empirical equations due to various reasons (unavailability of climate data to solve combination-based energy balance equations, etc.). Since many climate variables that affect ETref rates have been changing and are expected to change in the future, single-variable equations for estimating the trend in ETref should be avoided due to the inherent nature of the trend passed to ETref from the variable. Here, we showed an integrated approach of practical and robust procedures that are already exist to estimate necessary climate variables [incoming shortwave radiation (R-s), net radiation (R-n), wind speed at 2-m (u(2)), relative humidity (RH), and vapor pressure deficit (VPD)] only from observed maximum and minimum air temperatures (T-max and T-min) and precipitation (P) data to be used in Penman-Monteith-type combination-based energy balance equations to predict grass-and alfalfa-reference evapotranspiration (ETo and ETr respectively). We analyzed the trends and magnitudes of change in meteorological variables for a 116-yr period from 1893 to 2008 in the agro-ecosystem-dominated Platte River Basin in central Nebraska, USA. Although we found a significant (P < 0.05) increase in T-min and T-avg at a rate 0.038 degrees C yr(-1) and 0.0187 degrees C yr(-1), respectively, and insignificant increase in u(2) and VPD, we observed a significant (P < 0.05) decline in ETref (-0.3596 mm yr(-1) for ETo and -0.3586 mm yr(-1) for ETr). We present data, analyses, and interpretation that the decrease in ETref is most likely due to significant (P < 0.05) increase in precipitation (0.87 mm yr(-1)) that results in significant reduction in R-s (-0.0223 MJ m(-2) yr(-1)) and, in turn in R-n (-0.0032 MJ m(-2) yr(-1)), which resulted in reduction in ETref because increase in P decreases available energy, which is primary driver of ETref. There was approximately 100 mm of increase in precipitation from 1893 to 2008 in the study location at a rate of about 0.90 mm yr(-1). Also, there was a significant increase in maximum daily precipitation, especially in the very high events (i.e., >70 mm d(-1)). We present detailed analyses of relationships between ETref and all meteorological variables. On an annual time step ETref significantly (P < 0.05) and inversely correlated to precipitation and RH, and significantly and positively correlated to T-max, T-avg, VPD, R-s, and R-n. We observed a higher degree of responsiveness of ETo to changes in meteorological variables than ETr, which may indicae that ETo may be more apposite to better detect the impact of changes in meteorological variables on ETref in climate change studies. (C) 2011 Elsevier B.V. All rights reserved.
