Daily precipitation data covering 1961-2005 from 67 meteorological stations across the upper Yangtze River basin are used to analyze spatial-temporal variability characteristics of precipitation structure in the context of climate change and human activities. The autocorrelation component in precipitation time series is eliminated by the Trend Free Pre-Whitening method, while the variability and trend in the precipitation structure are detected by the nonparametric Mann-Kendall method. Results show that over the upper Yangtze River basin and its sub-basins, the incidence rate of precipitation events for different durations decreases exponentially with the increase of precipitation durations; while, the corresponding contribution rate is found to be first increasing and then decreasing, with predominantly by short-duration precipitation events. It is detected that the year of 1976 was marked by an abrupt change for the contribution rate of short-duration (1 d and 2 d) precipitation events; while, the timing of abrupt changes for incidence rate and contribution rate of the precipitation events with long-durations of 6 and 10 days was in 1984 and in 1999 respectively. An upward trend is detected in the number of consecutive occurrences of short-duration precipitation events, with higher precipitation intensity and contributing to larger portion of total precipitation amount; while, the opposite is true for long-duration precipitation events. Tests are all significant for both above characteristics of precipitation structure amongst the Min River basin and the Dadu River basin, which are the sub-basins of the upper Yangtze River basin, as well as its main course areas.