Daily streamflow forecasting using a wavelet transform and artificial neural network hybrid models

New wavelet and artificial neural network (WA) hybrid models are proposed for daily streamflow forecasting at 1, 3, 5 and 7 days ahead, based on the low-frequency components of the original signal (approximations). The results show that the proposed hybrid models give significantly better results than the classical artificial neural network (ANN) model for all tested situations. For short-term (1-day ahead) forecasts, information on higher-frequency signal components was essential to ensure good model performance. However, for forecasting more days ahead, lower-frequency components are needed as input to the proposed hybrid models. The WA models also proved to be effective for eliminating the lags often seen in daily streamflow forecasts obtained by classical ANN models.Editor D. Koutsoyiannis; Associate editor L. SeeCitation Santos, C.A.G. and Silva, G.B.L., 2013. Daily streamflow forecasting using a wavelet transform and artificial neural network hybrid models. Hydrological Sciences Journal, 59 (2), 312-324. ResumeNous proposons de nouveaux modeles hybrides d'ondelettes et de reseaux de neurones artificiels (WA) pour la prevision des debits journaliers aux horizons de 1, 3, 5 et 7 jours, bases sur les composantes basse frequence du signal d'origine (approximations). Les resultats montrent que les modeles hybrides proposes donnent des resultats significativement meilleurs que les reseaux de neurones artificiels classiques (ANN) pour toutes les situations etudiees. Pour les previsions a court terme (1 jour), des informations sur les composantes haute frequence du signal sont essentielles pour assurer une bonne performance du modele. Pour la prevision a plus long terme, les composants basse frequence sont necessaires comme entree des modeles hybrides proposes. Les modeles WA se sont egalement averes efficaces pour eliminer le delai souvent observe dans les previsions de debits journaliers obtenus par des modeles ANN classiques.