降雨时空分布量化及其在洪水过程分类中的应用

洪水过程是水利工程运行管理、防洪调度决策的重要依据，降雨的空间分布和降雨雨型是影响洪水过程形成的主要驱动因素。本文旨在采用描述降雨时空分布的量化指标，分析降雨时空分布特征与洪水过程的联系，探索量化指标在洪水过程分类中的应用。基于大清河流域北支紫荆关和南支阜平流域68场降雨洪水资料，计算了降雨一、二阶空间矩和一、二阶时间矩，分析了降雨时空分布特征对洪水过程线形状的影响，依据降雨时空特征信息采用K-medoids聚类分析法对洪水分类。结果表明：采用的降雨时空分布量化指标能够准确刻画两个流域的降雨中心空间分布和降雨雨峰时间分布。基于降雨时空特征指标的洪水聚类分析得到的不同洪水类型特征显著，两个流域的洪水主要有3种代表性类型，即尖瘦型、多洪峰长历时型和矮胖型洪水过程。研究结果验证了仅将降雨时空分布特征作为洪水分类依据的可行性，展示了其应用于流域洪水过程分类的潜力，降低了传统洪水过程分类方法对洪水数据的依赖性。本研究可为分析流域洪水形成过程的驱动因素、预测无资料流域洪水类型及揭示流域产汇流机理等提供科学依据。

Quantification of rainfall spatial and temporal distribution characteristics on the flood hydro- graph and its application in flood type classification

Flood hydrograph is an important basis for operation and management of hydro-project. The spatio-temporal characteristic of rainfall are generally the main driving factors affecting the flood hydrograph. In this paper,a set of rainfall spatio-temporal characteristic indexes (RSI) are used to analyze the 68 rainfall events in two study watersheds. Several combination scenarios are designed to study the influence of different rainfall spatio-temporal characteristics on hydrograph. K-medoids algorithm is used to classify the flood types with RSI as the input. The results show that the employed RSI could capture the spatio-temporal rainfall characteristics for study watersheds,facilitating a feasible quantitative method for rainfall characteristics analysis. The characteristics of three typical flood types from two study watersheds identified by K-medoids are representative. The research exhibits the potential of applying the RSI to the classification of flood types and provides a reference for analyzing the driving factors of watershed flood process.