Under climate change, drought events have a great impact on different regions of the world. Quantitative evaluation of regional drought characteristics is of great significance for regional drought risk aversion. This study simulates the changes in the historical and future hydrological runoff based on historical observation data, the Community Land Model-Distributed Time Variant Gain Model (CLM-DTVGM), and CMIP5 data. Subsequently, it reveals the historical and future hydrological drought characteristics in the Aksu River Basin under a changing environment. The conclusions are as follows: although both the asynchronous mutated precipitation and runoff in the upper reaches of Aksu River Basin increased, with a maximum increase of 3.3mm and 1.2mm/s, the synchronous mutation of precipitation and runoff in the downstream indicated that human activities only affected the runoff in Alaer station; CLM-DTGVM could be well applied in historical and future runoff simulation of the Aksu River basin with complex transformation of different land use under the changing environment; and the seasons of drought development, maximum intensity, and drought termination mainly occurred in winter, followed by spring, summer, and autumn, while human activities had an obvious impact on the drought characteristics in Aksu River Basin. These findings are of great significance for improving water-saving awareness and warning against drought risk.