Due to the need to simulate rainfall time series at different time scales for engineering purposes on one hand and lack of recordings for these parameters in small scales caused by the administrative and financial problems, on the other hand, disaggregation of rainfall time series to the desired scale is an essential topic in water resources engineering. In this study, to disaggregate Tabriz and Sahand rain gauges time series, the wavelet-artificial neural network (WANN) hybrid model is proposed according to nonlinear characteristics of the time scales. For this purpose, ten years of daily data from four rain gauges and monthly data from six rain gauges in Urmia Lake Basin were decomposed with wavelet transform. Then using mutual information and correlation coefficient criteria, the subseries were ranked and dominant subseries were used as input to ANN model for disaggregating the monthly rainfall time series into daily time series. Results obtained by the WANN disaggregation model were also compared with the results of ANN and conventional multiple linear regression models. The efficiency of WANN model at validation stage for Tabriz rain gauge showed an increase of up to 8. 5% and 33% with regards to ANN and multiple linear regression models. For Sahand rain gauge a respectively increase of up to 13. 7% and 26% were remarked. It was concluded that WANN hybrid model can be considered as an accurate model for disaggregation of the hydro-climatological time series.