| 非平稳条件下北京市最大月降水量频率特征分析 |
| |
| 引用本文： | 韩 丽,黄俊雄,周 娜,李 超.非平稳条件下北京市最大月降水量频率特征分析[J].水文,2021,41(2):32-37,108. |
| |
| 作者姓名： | 韩 丽 黄俊雄 周 娜 李 超 |
| |
| 作者单位： | 北京市水科学技术研究院;中国矿业大学（北京）化学与环境工程学院 |
| |
| 基金项目： | 国家自然科学基金重点基金项目(41730749);中国北方再生水修复河道的水文变化及其生态效益项目(4173000223)。 |
| |
| 摘 要： | 为探究气候变化下极端降水的频率变化特征,基于北京市22个雨量站实测月降水量数据,以时间为协变量构建平稳和非平稳GEV模型,对北京市最大月降水量序列(极值降水序列)进行模拟和频率分析,并采用Bootstrap方法对频率分析结果的不确定性进行评价。结果表明:所有极值降水序列的最优概率分布模型均为非平稳GEV模型,该模型能够抓住序列随时间呈显著下降趋势的变化特征;由非平稳GEV模型估算得到的极值降水重现水平随时间呈减少趋势,这意味着未来极值降水导致洪涝灾害的风险在降低,但导致干旱的风险将增加;随着重现期的增加,极值降水重现水平估计值的不确定性也随之增大。
|
| 关 键 词： | 非平稳模型 GEV 最大月降水量 BOOTSTRAP 不确定性 |
Nonstationary Frequency Analysis of Annual Maximum Monthly Precipitation in Beijing |
| |
| HAN Li,HUANG Junxiong,ZHOU Na,LI Chao.Nonstationary Frequency Analysis of Annual Maximum Monthly Precipitation in Beijing[J].Hydrology,2021,41(2):32-37,108. |
| |
| Authors: | HAN Li HUANG Junxiong ZHOU Na LI Chao |
| |
| Institution: | (Beijing Water Science and Technology Institute,Beijing 100048,China;School of Chemical and Environmental Engineering,China University of Mining and Technology,Beijing 100083,China) |
| |
| Abstract: | To explore the variations of frequency in precipitation extremes under the changing climate,frequency analysis for the maximum monthly precipitation datasets(AM1 datasets)was carried out based on the observed monthly data at 22 stations in Beijing by using the stationary and nonstationary GEV models with time as the covariate.Furthermore,the uncertainties of the estimated return levels were also investigated by using the non-parametric Bootstrap method.Results show that significant downward trend are found for all AM1 datasets with an average decrease of 1.8 mm/year during 1956-2016.The optimal distribution models for all AM1 datasets are nonstationary GEV models(GEV1 or GEV11),rather than the stationary one,according to the VICc values,which is consistent with the results of trend test.The estimated return levels from the optimal models decrease with time,indicating that the risk of flood disasters caused by the maximum monthly precipitation in the future is decreasing,while the risk of drought is increasing.The uncertainties of the estimated return levels from both the stationary and the nonstationary models are relatively close.With the increase of the return period,the uncertainty of the return level estimation increases. |
| |
| Keywords: | nonstationary model GEV maximum monthly precipitation Bootstrap uncertainty |
| 本文献已被 维普 等数据库收录！ |
| 点击此处可从《水文》浏览原始摘要信息 |
| 点击此处可从《水文》下载免费的PDF全文 |
|
