The vigorous development of satellite precipitation observations has gained widespread attention in the field of hydrology. Many studies have explored the reliability of hydrological models driven by satellite precipitation data (SPD) to reproduce streamflow. However, the reliability of SPD in the simulation of other hydrological processes remains unclear, even though a full understanding of the dynamic variations in hydrological processes is essential for the scientific management of water resources. In this study, we investigated the effect of SPD uncertainty on the distributed hydrological model in reproducing hydrological simulations. The investigation was conducted in the upper Huaihe River basin, which has a high-density ground precipitation observation network. The uncertainty inherent in 8 types of SPD in characterizing the temporal and spatial variations in precipitation was comprehensively examined. The Variable Infiltration Capacity hydrological model (VIC) was then applied to further explore the reliability of the simulated hydrological fluxes and states (HFS) driven by the selected SPD over multiple temporal-spatial scales. The results showed that: (1) the selected SPD tend to have greater uncertainty in recognizing spatial variations in precipitation than in identifying temporal variations; the VIC models driven by the selected SPD generally performed better in characterizing temporal variations in HFS than in reproducing their spatial variations. In addition, the corrected SPD (CSPD) are more suitable for HFS simulations than the uncorrected SPD (USPD); (2) the effective reproduction of the spatio-temporal variations in HFS has more stringent requirements on the quality of SPD than the satisfactory reproduction of streamflow. For the gridded scale simulations, the selected SPD-driven models exhibited large uncertainties in reproducing spatio-temporal variations in daily HFS. In contrast, although these models significantly improved the simulation performance of monthly HFS, they were still relatively deficient in reproducing their spatial variations. In terms of the basin-scale simulations, the CSPD-driven models tend to reliably reproduce monthly HFS; and (3) independent calibration can allow the selected SPD-driven models to reproduce streamflow more effectively, but it may not result in the best reproduction of HFS. This study provides guidance for strengthening the application of SPD in hydrological modeling and provides feedback for further improving the quality of SPD.