摘要: 世界上众多江河洪水序列形成的环境背景“一致性”已不复存在，传统极值流量分析的“极值理论”需要修正以适应这些现象。本文采用Pearson-Ⅲ型分布函数，以L-矩法估参。应用30a逐年滑动平均法分解变化环境下洪水分布参数演变规律，并对其成因及影响作了有益探讨。结果表明：1) 东江年最大日流量呈显著下降趋势，产生非一致性。除了降雨外，大中型水库调蓄影响是主要原因。2) 变化环境下，非一致性洪水序列引起分布参数改变，已导致高水尾端特征由“陡峭”变为“平缓”。3) 随分布参数的改变，指定标准P下(百年一遇)，其相应设计流量值也时刻发生改变，量级表现为由大减小。4) 在变化环境情况下，洪水重现期往往不是描述一场洪水的一个固定不变的属性。
Abstract:
The environment consistence of flood series of many rivers is no longer existed. And the normal method that used to analyze stream flow need to be adjusted to adapt these phenomena. This paper analyzes flood frequency by Pearson-Ⅲ distribution and estimate the parameters with the method of L-moments. The Changing properties, causes and impacts of parameters of the generalized extreme value distribution is analyzed by use of the 30-year moving average method. The result indicate that: 1) The inconsistent annual maximum daily flow series of stations in Dongjiang Basin shows a descend trend that caused by rainfall and construction of water conservancy projects, the latter is the major reason; 2) The flood series maintain inconsistency before and after environment change, the impacts on fitting curve of flood series showed an overall performance as upper tail from “gentle” to “steep”; 3) Because of the parameters is changing, so will the design flood magnitude, and its trend is descend; 4) The flood return period is no a constant property for describing a flood under the changing environment.