Szilagyi, JzsefHobbins, MichaelJozsa, Janos2015-12-102015-12-101084-0699http://hdl.handle.net/1885/60317The original and modified versions of the advection-aridity (AA) model of regional evapotranspiration are tested with data from the Solar and Meteorological Surface Observation Network (SAMSON). The resulting long-term mean annual evapotranspiration estimates are validated against water balances of 25 watersheds that are minimally affected by human activity and contain at least one SAMSON station, as well as with similar closures of SAMSON-station/gridded precipitation and runoff. In general, model performance is very similar among the two versions, explaining at least 80% of the spatial variance in the long-term means, simultaneously remaining well within 10% of the water balance-based values in their station-averaged long-term mean annual evapotranspiration estimates. The modified AA model, however, can be used in humid as well as in arid regions with the same set of calibrated parameters, whereas the original AA model may require a recalibration.Keywords: Advection-Aridity model; Advection-aridity models; Human activities; Model performance; Precipitation and runoff; Recalibration; Regional evapotranspiration; Spatial variance; Surface observation; Water balance; Advection; Arid regions; Evaporation; Evapo Evaporation; Evapotranspiration; Surface water; Water balance200910.1061/(ASCE)HE.1943-5584.00000262016-02-24