With increasing temperatures, changing weather patterns and ongoing development, it is becoming increasingly important to clarify the evolution mechanism of future regional streamflow processes and their controlling factors. In this study, an integrated framework for watershed streamflow prediction based on a Global Climate Model (GCM), the Patch-generating Land Use Simulation model (PLUS), and the Soil and Water Assessment Tool (SWAT) was proposed in the middle Yellow River. The results indicate that, compared with the baseline period (1989–2018), levels of precipitation and maximum and minimum temperatures are expected to increase in the next 30 years, resulting in a warmer and wetter regional climate. Under various climate scenarios, the annual streamflow is projected to increase by 49.2–115.1%. The acreage of various land types may have tended to be saturated, and the main land types such as cropland, forest and grassland have little change (−6.6%-0.6%), so the impact on streamflow will be correspondingly reduced. Under various land use scenarios, the annual streamflow is projected to increase by 5.0%–7.3%. The annual average streamflow trends under the combined climate and land use scenarios are consistent with the climate change scenarios, while the mean values corresponding to the combined scenarios are higher than those of the single scenario. Findings show that climate change is the main driver influencing streamflow, with a contribution of 86.3%–95.1%. This study deepens understanding of the change pattern and influence mechanism of the streamflow process, which can provide a scientific basis for the development and refinement of regional ecological construction plans.