Abstract

El Niño-Southern Oscillation (ENSO) phenomenon strongly affects the interannual variability of hydroclimatic conditions. This variability is not considered in the traditional calibration and validation of hydrological models used for water resources planning and management. This paper presents an analysis, on a time-frequency basis, of ENSO effects on evaporation model errors. The wavelets transform was applied on hydroclimatic data series: monthly pan evaporation and sea surface temperature for the El Niño 3.4 region. An evaporation model was locally calibrated against measured pan evaporation data in Chillán, Chile. The difference between actual data and simulated values was correlated with the sea surface temperature, finding a high correlation in a 4-to 6-year period, associated with El Nino events. The classical approach used for model calibration and data processing did not allow the analysis of temporal patterns that are inherent to the time series and that can be captured using wavelets.