Evapotranspiration quantification is a necessary task for water resources management in a watershed, since it is an important variable of the hydrological cycle. However, the estimation of the real evapotranspiration is a difficult task and researchers commonly use more than one method in order to validate the results. This paper presents a method for real evapotranspiration determination based on precipitation, runoff and groundwater level variation measured in the watershed. Using this method the real evapotranspiration in a pilot watershed was estimated to 900 mm per year. This result has been compared to potential evapotranspiration calculated by empirical and semi-empirical methods, based on climatologic data. The Thornthwaite method, with a potential evapotranspiration of 936 mm per year, presented the closest value to the water balance method. The water balance method based on hydrogeologic measurements showed to be suitable for real evapotranspiration estimation in the watershed. However it is necessary to observe the hypothesis and restrictions described in this paper to apply this method to another watershed.
water resources management; water balance; groundwater
