Development of a coupled wavelet transform and neural network method for flow forecasting of non-perennial rivers in semi-arid watersheds

In this study, a method based on coupling discrete wavelet transforms (WA) and artificial neural networks (ANN) for flow forecasting applications in non-perennial rivers in semi-arid watersheds is proposed. The discrete à trous wavelet transform is used to decompose flow time series data into wavelet coefficients. The wavelet coefficients are then used as inputs into Levenberg Marquardt artificial neural network models to forecast flow. The relative performance of the coupled wavelet-neural network models (WA–ANN) was compared to regular artificial neural network (ANN) models for flow forecasting at lead times of 1 and 3 days for two different rivers in Cyprus (Kargotis at Evrychou and Xeros at Lazarides). In both cases, the coupled wavelet-neural network models were found to provide more accurate flow forecasts than the artificial neural network models. The results indicate that coupled wavelet-neural network models are a promising new method of short-term flow forecasting in non-perennial rivers in semi-arid watersheds such as those found in Cyprus.