Design flood estimation for global river networks based on machine learning models

Design flood estimation for global river networks based on machine learning models
复制标题

基于机器学习模型设计全球河网洪水估算

DOI:
10.5194/hess-25-5981-2021
复制
发表时间:
2021-11
影响因子:
6.3
通讯作者:
Bo Pang
Bo Pang
中科院分区:
地球科学2区
文献类型:
--
作者:
Gang Zhao;Paul Bates;Jeffrey Neal;Bo Pang

文献摘要

参考文献

相似文献

抽象。设计洪水估算是水文学的一项基本工作。在这项研究中,我们提出了一种基于机器学习的方法来估计全球设计洪水。这种方法包括三个阶段:(i)估计现场洪水频率曲线的全球测量站使用安德森-达令测试和贝叶斯马尔可夫链蒙特卡罗(MCMC)方法;(ii)聚类这些。站到子组使用K-均值模型的基础上,12个全球可用的集水区描述符;和(iii)开发一个回归模型。在每个子组的区域设计洪水估计使用相同的。全球共有11793个站点被选择用于模型开发,并对三种广泛使用的回归模型进行了设计洪水估算。研究结果表明:(1)当使用全部12个描述符进行聚类时,所提出的方法获得了最高的设计洪水估算精度,并且在训练和验证过程中考虑更多的描述符可以提高回归的性能;(2)在所有测试的回归模型中,支持向量机回归提供了最高的预测性能,100年一遇洪水估算的标准化误差为0.708;(3)热带、干旱、温带、寒冷和极地气候下的100年一遇设计洪水。
Abstract. Design flood estimation is a fundamental task in.hydrology. In this research, we propose a machine-learning-based approach to.estimate design floods globally. This approach involves three stages: (i)estimating at-site flood frequency curves for global gauging stations using the.Anderson–Darling test and a Bayesian Markov chain Monte Carlo (MCMC) method; (ii)clustering these.stations into subgroups using a K-means model based on 12 globally.available catchment descriptors; and (iii)developing a regression model in.each subgroup for regional design flood estimation using the same.descriptors. A total of 11 793 stations globally were selected for model.development, and three widely used regression models were compared for design.flood estimation. The results showed that (1)the proposed approach.achieved the highest accuracy for design flood estimation when using all.12 descriptors for clustering; and the performance of the regression was.improved by considering more descriptors during training and validation; (2)a support vector machine regression provided the highest prediction.performance amongst all regression models tested, with a root mean square.normalised error of 0.708 for 100-year return period flood estimation; (3)100-year design floods in tropical, arid, temperate, cold and polar climate.zones could be reliably estimated (i.e.
DOI: 10.1016/j.jhydrol.2015.05.023
发表时间: 2015-08
影响因子: 6.4
作者:
Jun Wang;Zhongmin Liang;Yiming Hu;Dong Wang
通讯作者: Jun Wang;Zhongmin Liang;Yiming Hu;Dong Wang
DOI: 10.1016/j.jhydrol.2007.10.050
发表时间: 2008-01
影响因子: 6.4
作者:
C. Shu;T. Ouarda
通讯作者: C. Shu;T. Ouarda
DOI: 10.1080/23754931.2015.1014272
发表时间: 2015-07
影响因子: --
作者:
E. Doxsey-Whitfield;K. Macmanus;Susana B. Adamo;L. Pistolesi;J. Squires;O. Borkovska;S. Baptista
通讯作者: E. Doxsey-Whitfield;K. Macmanus;Susana B. Adamo;L. Pistolesi;J. Squires;O. Borkovska;S. Baptista
DOI: 10.1038/nclimate2893
发表时间: 2016-04-01
影响因子: 30.7
作者:
Winsemius, Hessel C.;Aerts, Jeroen C. J. H.;Ward, Philip J.
通讯作者: Ward, Philip J.
DOI: 10.1088/1748-9326/11/9/094014
发表时间: 2016-09
影响因子: 6.7
作者:
M. Trigg;C. Birch;J. Neal;P. Bates;Andrew Smith;C. Sampson;Dai Yamazaki;Y. Hirabayashi;F. Pappenberger;E. Dutra;P. Ward;H. Winsemius;P. Salamon;F. Dottori;R. Rudari;M. Kappes;A. Simpson;G. Hadzilacos;T. Fewtrell
通讯作者: M. Trigg;C. Birch;J. Neal;P. Bates;Andrew Smith;C. Sampson;Dai Yamazaki;Y. Hirabayashi;F. Pappenberger;E. Dutra;P. Ward;H. Winsemius;P. Salamon;F. Dottori;R. Rudari;M. Kappes;A. Simpson;G. Hadzilacos;T. Fewtrell