琼州海峡海口站近岸风暴增水概率风险分析
Probability risk analysis of the nearshore storm surge at Haikou Station in Qiongzhou Strait
陈玲舫 1韩树宗 2车志伟 3罗耀 4谢强1
作者信息
- 1. 中国科学院深海科学与工程研究所,海南三亚572000
- 2. 中国海洋大学海洋与大气学院,山东青岛266100
- 3. 国家海洋局海口海洋环境监测中心站,海南海口570100
- 4. 中国科学院南海海洋研究所,广东广州510301
- 折叠
摘要
基于海口站1976～1997年逐时潮位和逐日最大风速资料,利用阿基米德Copula函数构建海口年最大增水与相应日期最大风速的联合概率分布模型.结果表明:1)广义极值分布可作为海口站年最大增水和相应日期最大风速的边缘分布.两个序列之间存在强正相关关系,G-H Copula函数更适用于作为海口站年最大增水和相应日期的最大风速联合概率分布的连接函数.2)两变量联合作用的同频率增水高度设计值与增水的单变量边缘分布设计值之间的相对差值约为7.5％.3)条件概率1(P(Y≥y｜X≥x))中同频率的年最大增水和相应风速的遭遇概率介于78.2％～80.9％,条件概率2(P(Y≥y｜X≤x))中同频率的年最大增水和相应风速两者的遭遇概率小于4.8％.
Abstract
Based on the hourly tidal level and daily maximum wind speed data at Haikou Station during 1976 ～ 1997,the joint probability distribution of yearly maximum surge height and corresponding wind speed has been analyzed with the Archimedean Copula functions.The main conclusions are as follows:1) Generalized extreme value distribution can be used as the optimal margin distribution for yearly maximum surge height and corresponding wind speed at Haikou station.There is a strong positive correlation between the two variables.Among the three Archimedean Copula functions,G-H Copula is a better contiguous function for joint distribution of the two variables.2) By comparing the joint distribution,the relative differences of the design values of the marginal distribution of surge heights are about 7.5％.3) Conditional probability 1 (P(Y≥y ｜ X≥x)) indicated that the encounter probabilities of the same frequency of yearly maximum surge heights and corresponding wind speeds fall in between 78.2％ ～ 80.9％,and the conditional probability 2(P(Y≥y ｜ X≤x)) indicated that the encounter probabilities of the same frequency of yearly maximum surge heights and corresponding wind speeds are less than 4.8％.
关键词
琼州海峡/Copula函数/联合分布/风暴增水/遭遇概率Key words
Qiongzhou Strait/Copula function/joint distribution/storm surge/encounter probability引用本文复制引用
基金项目
三亚市院地科技合作资助项目(2015YD01)
出版年
2017
NETL
NSTL
维普
万方数据