High-precision precipitation data at the time and place scale, which is a main component of hydrological models, will greatly help to manage the water resources. Hence, in this research, evaluating precipitation data accuracy as well as their origin correction in Quantile method (RQUANT) was used to improve the performance of satellite data in the Yamchi Dam Basin in Ardabil province. The satellite data used included PERSIANN-CCS, PDIR-Now and GPM precipitation data on an hourly, daily and monthly time scale for the region's rainy months over 11 years that were selected by moderate precipitation, standard deviation and coefficient of variation. Comparison of the results obtained from GPM satellite with precipitation data showed that this product has better performance than PERSIANN-CCS and PDIR-Now models on a monthly scale. GPM data on a monthly scale had a MAE index and RMSE 4.66 and 9.70 mm per day, respectively and a correlation coefficient of 0.74; while these values for PERSIANN-CCS and PDIR-Now models were equal to 24.24, 62.03 and 0.36 and 07.09, 13.52, 0.27, respectively. Thus, the GPM satellite precipitation product was provided better performance on a daily scale. Since hourly data was required in flood analysis, then in this paper, precipitation data on an hourly scale was evaluated; the statistical values for GPM satellite were 0.91, 2.66 and 0.06, respectively. In general, among the study precipitation products, GPM precipitation gives better results, although on a daily and hourly scale, the desired results were not obtained compared to the measured data.