IAHR Document Library

« Back to Library Homepage « Proceedings of the 35th IAHR World Congress (Chengdu, 2013)

Annual Maximum Rainfall Occurrence Dates Frequency Analysis Based on Mixed Von Mises Distribution

Author(s): Xiaozhong Sun; Jijian Lian; Kui Xu; Chao Ma

Linked Author(s): Jijian Lian, Chao Ma, Chao Ma

Keywords: No Keywords

Abstract: Von Mises distribution is one of the most commonly used distributions to describe the cyclical or seasonal variables, which is currently widely used in directional statistics. Mixed von Mises distributionis obtained by adding multiple von Mises distributions under a certain proportion, which could effectively fit the bimodality or multimodality characteristics of variables. In this paper, we attempt to utilize the mixed von Mises distribution to describe the occurrence dates of annual maximum rainfall by comparing with the single von Mises distribution. Taking Fuzhou area as an example, the calculation results demonstrate that the mixed von Mises distribution is more precisely than the single von Mises distribution and the annual maximum rainfall occurrence dates of Fuzhou region exhibit the bimodality characteristic and the two peak points of July 1 and September 5are obtained with a risk probability of 0. 39% and 1. 18% respectively. The probability of annual maximum rainfall occurs from June 13 to October 10was 73. 94% , in which 46. 25% occurs during August 10to September 27. Therefore, we achieve the conclusion that the annual maximum rainfall in Fuzhou area mainly occurs from June to October, especially for middle-Augusttolast-September. The average deviation (Bias) and the mean square error (RMSE) of the fitting resultsare0. 02 and 0. 09 by inspection, which further indicate that the von Mises distribution could accurately describe the bimodality characteristics of annual maximum rainfall occurrence dates in Fuzhou area. Furthermore, it could mine more rainfall characteristics and provide a reliable basis for the analysis of urban flood control by usingmixed von Mises distribution.


Year: 2013

Copyright © 2022 International Association for Hydro-Environment Engineering and Research. All rights reserved. | Terms and Conditions