资源环境科技发展态势分析平台

Changes in extreme precipitation are associated with changes in their probability distributions and the characteristics of quantiles derived from fitted distributions. In this study, the linear quantile regression method is employed to analyse spatio-temporal trends of extreme precipitation and to study the impact of North Atlantic tropical cyclones in the distribution of extreme precipitation for the southeastern United States. Daily annual maximum precipitation over the period of 64years (1950-2013) for 107 sites was used for the analysis. Our results show that changes in upper quantiles of the distributions of the extreme precipitation have occurred in the southeastern United States. Analysis of the potential changes in the distribution of the extreme precipitation by separating the historical record into two periods, that is, before and after 1981, reveals that upper-quantile trends have increasing magnitude in most of the sites for the latest time period. Analysis of the impact of tropical cyclones in the extreme precipitation distribution shows that overall the heavy rainfall events in the recent decades may have been caused by tropical cyclones. Such results are particularly useful for water managers who are more concerned with extreme values rather than the averaged one. Hence, our study has significant implication in environmental and infrastructural assessment as well as disaster risk management.