monthly simulation of groundwater fluctuations using wavelet and dynamic neural network

simulation of groundwater fluctuations plays a crucial role in management of watersheds and water demand balancing. recently, wavelet analysis has been used widely in time series decomposition and coupling with neural networks for hydrological modeling. in this paper, the ability of the wavelet-dynamic artificial neural networks (w-ann) model was applied in forecasting one-month-ahead of groundwater level and compared to regular artificial neural networks (ann) and multi linear regression (mlr) models. the only variable used to develop the models was monthly groundwater level data recorded for ten years at two piezometers in the qom plain, iran. the results show that the mlr model overestimate the observed data and the performance of ann model hasn't enough accuracy, whereas the w-ann model with meyer mother wavelet and two decomposition levels, could predict one-month-ahead with nash-sutcliffe coefficient equal to 0.993 and 0.974 for piezometers 1 and 2 respectively.