We analyzed the statistical properties of maximum mean 7-day streamflow in Lake Poyang basin using eleven probability distribution functions. Estimate of parameters was performed using the L-moment technique. Goodness-of-fit was done based on Kolmogorov-Smimov statistic D. Mann-Kendall trend test was used to detect the trends within the streamflow extreme series. The research results indicate that; (1) Wakeby distribution is the good choice for description of the statistical behaviors of streamflow extremes in the study region; (2) Changing properties are varying from one station to another, implying different fluvial geomorphological features, climate changes and human activities; (3) Rainfall and hydraulic engineering is an important factor in effect the changes of extreme streamflow in the Lake Poyang basin. Higher frequency of hydrological and meteorological extremes may cause higher probability of floods in the study region. While hydraulic engineering, especially the construction of large-scale water conservancy facilities in the river, reduce the flood discharge. The impact of forest on flood flows is not obvious.