Real-time daily flow forecasting using black-box models, diffusion processes, and neural networks

The purpose of this study is to compare three modeling approaches used for the prediction of daily natural flows 1-7 days ahead. Linear black-box mode ls, which have been commonly used for modeling flows, constitute the first approach. The second approach, a linear type in the context of our applicat ion, is less known in the water resources field and is identified by the te rm diffusion process. The third approach uses models called neural networks , which have gained interest in many fields. All these approaches were test ed on 15 watersheds from the Saguenay - Lac-Saint-Jean hydrographic system, located in the province of Quebec, Canada. Because the watersheds possess different physical characteristics, the models were tested under several ru noff conditions. In this article, the focus is on results; all approaches a long with their conditions of use have been detailed elsewhere in the liter ature. The results obtained showed that neural networks constitute, for alm ost all the watersheds studied, the best approach to forecast daily natural flows. The more flexible structure of neural networks allows a best reprod uction of complex runoff conditions. However, neural networks are more sens itive to outliers present in observed natural flow series, which are used a s inputs in the three models tested. In practice, to model flows at specifi c periods of the year, it seems preferable to establish seasonal models. If a neural network has an inadequate structure for the period under consider ation, then it may produce less convincing results than the other two model ing approaches tested in this study.