基于Voronoi图的强降水过程空间非均匀性定量表征方法

摘要

To describe the characteristics quantitatively of the spatial-heterogeneities of regional persistent heavy precipitation,a new approach is promoted,which can generate quantitatively the spatialheterogeneities of regional persistent heavy precipitation based on the dynamic operation and computational geometry of the VoronoiDelaunay model.The daily precipitation dataset in China from 1959 to 2002 is used to build the time series of regional 5-day (10-day) persistent extreme precipitation events to test the application effectiveness,and this method is applied to analyze the spatial-heterogeneities,variations,and climatic trends of annual-extreme regional 5-day and (10-day) persistent precipitation.The results indicate that comparing with the traditional quantitative statistic method and Kriging interpolation method,this method can easily handle the spatial-heterogeneities of regional persistent heavy precipitation and is applicable in the climate change,climate features,and impact researches,and there is a notable advance trend for the occurring date of annual-extreme regional 5-day persistent precipitation from 1959 to 2002.%利用VoronoiDelaunay图模型影响范围和局部动态的特性,提出一种基于动态边界定量提取强降水过程空间非均匀分布特征的新方法,根据1959-2002年中国740站逐日降水资料,构建5日、10日两种时间尺度区域持续性年极端强降水过程序列,用以检验该方法的应用效果,并应用该方法分析了5日(10日)区域持续性年极端强降水过程空间非均匀分布特征及其演变规律.结果表明:与传统的数量统计方法和区域插值方法相比,该方法在强降水空间分布中心、过程内不同强度降水覆盖区域、降水集中区等的定量度量方面具有更高的分辨力和较好实际应用价值；气候趋势分析中在1959-2002年间,5日区域持续性年极端强降水过程出现日期有明显提前趋势.