Wavelet-based de-noising in groundwater quality and quantity prediction by an artificial neural network

Abstract The present study uses a wavelet-based clustering technique to identify spatially homogeneous clusters of groundwater quantity and quality data select the most effective input for feed-forward neural network (FFNN) model predict level (GL), pH HCO3? in groundwater. In second stage this methodology, first, GL, time series different piezometers were de-noised using threshold-based wavelet method impact noisy compared temporal modeling by artificial (ANN). results suggest that proposed decreases dimensionality layer consequently complexity FFNN with acceptable efficiency spatiotemporal simulation GL parameters. Also, application de-noising parameters ANN increases accuracy predictions, respectively, up 11.53, 11.94 38.85% on average.