Abstract: To improve the calculation accuracy of design flood for long return period, this paper studies
application of high-order L-moments to flood frequency analysis, focusing on Monte-Carlo simulations for
statistical properties of the L-moments method and its influences on the return period flood quantile.
Parameters of GEV distribution and high-order L-moments were estimated for five annual maximum flow
series observed in Shaanxi province, and the fitting effect and calculation error of flood quantile evaluated.
The results show that this method has good statistical performance in flood analysis of long return period and it
gives a representation of the data series better than the lower-order methods.