The performance of evapotranspiration（ET） temporal scaling methods is affected by a number of factors.This paper compares the statistical characteristics of actual ET data measured by Lysimeter with ET data estimated on the basis of temporal scaling models.The study shows that on an "hourly-to-daily" scaling basis

the measured daily ET data show a closer agreement with the ET data estimated on the basis of surface resistance using morning measurements than with ET data estimated on the basis of evaporative fraction and crop coefficient.Additionally

the observed daily ET data show a closer agreement with ET data estimated on the basis of crop coefficient by using afternoon measurements than with ET data estimated using evaporative fraction and surface resistance.On a "daily-to-seasonal" scaling basis

the measured ET data show a higher correlation with ET data estimated on the basis of crop coefficient than that with data estimated on the basis of evaporative fraction and surface resistance.From the view of these results

we recommend considering the particular weather region

crop growing conditions

soil moisture content

temporal scale and observation facilities available before selecting an ET temporal scaling method for a given site.