The Variability of ASCE Standardized Reference Evapotranspiration: A Rigorous, CONUS-Wide Decomposition and Attribution

Abstract. To fully attribute the variability of reference evapotranspiration to its drivers, a mean-value, second-moment uncertainty analysis is applied to a 30-year, CONUS-wide reanalysis of daily and annual tall-crop reference evapotranspiration as estimated by the ASCE Standardized Reference Evapotranspiration Equation driven by four variables from the North American Land Data Assimilation System phase 2 (NLDAS-2) reanalysis: temperature, specific humidity, wind speed, and downward shortwave radiation. The attribution methodology accounts for both the sensitivity of reference evapotranspiration to its drivers and their observed variabilities, and it permits the decomposition of reference evapotranspiration variability across CONUS at various timescales into contributions from each driver. An analytically derived expression of the sensitivity of daily ASCE Standardized Reference ET to each of the drivers is provided and mapped. Contrary to the assumption of much hydrologic practice, temperature is neither always nor everywhere the most significant driver of temporal variability in reference evapotranspiration. Instead, depending on regional hydroclimatology, season, and timescale, different drivers dominate; in fact, in many regions, temperature-based parameterizations should be avoided at all timescales.