Based on daily precipitation data of 15 rainfall stations in Zhangshui River Basin in the upper reaches of Ganjiang from 1957 to 2016, 12 extreme precipitation indexes were selected to analyze the spatial-temporal variation characteristics of this area using linear trend test, Modified Mann-Kendall non-parametric test, cumulative anomaly method, Pettitt test, Morlet analysis, Hurst index analysis and Kriging interpolation method. The results showed that over the last 60 years，the volume of rainstorm and its duration days, the heavy precipitation and its contribution rate, and the maximum number of wet days all showed an increasing trend, whereas the extremely strong precipitation and its contribution rate, the maximum number of dry days, the maximum annual rainfall and the maximum rainfall of five consecutive days showed a decreasing trend, which would be further strengthened in the future. Nevertheless heavy rainfall and its duration days showed an upward trend, with an inclination of weakening in the future. It is concluded that all the past trends of extreme precipitation indexes are not significant with no sign of abrupt changes, and the main cycle of 33–34 a with 1988 as the intermediate axis generally exists. The high value centers of heavy rainfall and its duration days, rainstorm and its duration days, heavy precipitation, maximum number of wet days, annual maximum rainfall and maximum rainfall of five consecutive days are all located near the Neiliang station in the southwest region of the basin, and show an upward trend. The low value areas are mainly distributed in the eastern region of the basin and present a downward trend. Their overall spatial distributions are roughly the same, both decreasing from southwest to northeast; However, the contribution rate of heavy precipitation and extremely strong precipitation decreases from southeast to northwest, and the maximum number of dry days decreases from east to west. In the future, the change trend of extreme precipitation in Zhangshui River Basin may deteriorate, and strengthening the prediction of extreme precipitation is of great significance to the prevention of flood and drought disasters in this river basin.