Abstract: Small watersheds are ideal objects for studying the evolution of small and microscale hydrology and water resources systems. A small watershed is the smallest unit for calculating river water and sediment production and is the best regional scale for hydrology and soil erosion research and management. Through RS technology, a climate model obtains precipitation estimation data and drives a distributed hydrological model to simulate and predict hydrological processes, which clarifies the inevitable trends of basin hydrology and water resources research. Using NOAA- CPC- US precipitation data as a reference, an analysis of the PERSIANN, PERSIANN- CDR, TRMM- 3B42V7, GPM- IMERG, StageIV, and ERA5 precipitation data products were compared for nine small watersheds in different regions of the United States. The accuracy of theseseven precipitation products allowed them to drive the CREST distributed hydrological model, which evaluated the hydrological simulation effects of the precipitation products. The study shows that the NOAA- CPC- US precipitation data product is the highest, followed in decreasing order by StageIV, PERSIANN-CDR,GPM-IMERG, PERSIANN,ERA5, and TRMM- 3B42V7. The precipitation estimation accuracy of each precipitation product in the small watersheds in the high latitudes and western mountains of the northern United States is lower; however, there is better precipitation accuracy in small watersheds in the central, southern, and eastern parts of the United States. In the hydrological simulation utility evaluation, the CREST model parameters were determined using seven kinds of precipitation products, respectively. After obtaining the set of parameters, the daily runoff process of the basin was simulated for the same verification period. The comparison results show that NOAA-CPC-US and StageIV have better effects on the hydrological simulation of small watersheds. However, caution should be exercised in hydrological simulations in the northern and western parts of the United States using PERSIANN, PERSIANN- CDR, GPM- IMERG, and ERA5 precipitation data, and the TRMM-3B42V7 simulation effect is not ideal.

Key words: Small Watersheds, CREST, Multi-source precipitation, Hydrological Simulation