Evapotranspiration is an important indicator for the management and planning of water resources. The estimation of evapotranspiration is usually done trough bio-physical modeling, which requires the observation of multiple variables as well as the definition of the corresponding equations. In this paper, we evaluate the use of supervised machine learning as a strategy to get estimates of evapotranspiration from data observed in multiple meteorological stations in the Pampean region of Argentina. Particularly, we evaluate and compare regression methods for estimating the evapotranspiration from data collected during an extensive period of time, more than 40 years, from 24 stations placed in the region under study. The results obtained thus far are promising as they show the feasibility of applying a machine learning approach for obtaining accurate evapotranspiration estimates. The practical implications of these findings are relevant for the design of more efficient water monitoring systems in the country.