Objective Against the backdrop of climate change, the spatial and temporal distribution of precipitation within river basins has become increasingly uneven, with growing concerns over extreme drought and flood risks. This has exacerbated the imbalance between water supply and demand. Consequently, forecasting future precipitation and runoff trends within river basins has become a pressing research requirement in hydrogeological and water resources investigations.

Methods This study employs six climate models from the Sixth Coupled Model Intercomparison Project (CMIP6). Utilizing the Mann−Kendall (MK) trend test and linear trend estimation, it forecasts the spatiotemporal variations in precipitation within the Ganjiang River Basin from 2015 to 2100 under the SSP2−4.5 and SSP5−8.5 scenarios. The Variable Infiltration Capacity (VIC) model was employed to simulate future annual runoff and conduct trend analysis for the basin. The MK abrupt change test and sliding T−test were jointly applied to identify abrupt changes in annual runoff and assess their potential implications for extreme hydrological event management.

Results Future multi−year mean precipitation and runoff in the Ganjiang Basin under all six climate models showed varying degrees of increase compared to the reference period. Projected future annual precipitation and annual mean runoff exhibited an overall upward trend. During the flood season, precipitation and runoff in the Ganjiang Basin are projected to increase, while the dry season exhibits a decreasing trend. The magnitude of change under the SSP5−8.5 scenario is greater than that under SSP2−4.5, and the intra-annual distribution becomes more uneven. Annual runoff abrupt−change tests identified eight abrupt years, including 2076, with differences across models and scenarios indicating uncertainty in future hydrological−regime shifts.

Conclusions The integration of CMIP6 models with the VIC model, constructed using observed precipitation data, effectively captures the basin's hydrological response characteristics. This approach provides valuable support for future water resource planning and risk−informed flood/drought management in complex mountainous river basins.