versión On-line ISSN 1850-468X
Hydrological modeling is a potent tool that can be useful to, among other things, determine the potential future availability of fresh water on Earth. In particular, the La Plata Basin supplies fresh water to millions of people across five countries. In a context of possible future changes on climate, river streamflows could be affected and in this sense it is of crucial importance to infer these potential changes for the upcoming decades. The objective of this paper is to determine the ability of three climate models (CNRM-CM3, ECHAM5-MPI/OM and GFDL-CM2.0) to simulate the observed discharges of the main La Plata Basin rivers. To force the distributed hydrology model, climate models output data included daily minimum and maximum temperature and precipitation during the last decade of last century (1990-1999). Results show that the three models display serious deficiencies to adequately capture the hydrological cycle of the basin and that these deficiencies are mainly due to the important errors in the temperature and precipitation fields over southern South America. This fact poses a strong limitation in the potential use of these models to perform inferences about future changes across the basin.
Palabras llave : Hydrological cycle; Climate models; La Plata Basin.