A regional GEV scale-invariant framework for Intensity-Duration-Frequency analysis

被引:65
作者
Blanchet, J. [1 ]
Ceresetti, D. [1 ]
Molinie, G. [1 ]
Creutin, J. -D. [1 ]
机构
[1] Univ Grenoble Alpes, CNRS, LTHE, UMR 5564, F-38041 Grenoble, France
关键词
Extreme rainfall; Scale invariance; Extreme value statistics; Measurement frequency; Mediterranean region; CATASTROPHIC PRECIPITATING EVENTS; SOUTHERN FRANCE; RAINFALL; MODEL; EXTREMES; PREDICTABILITY; CURVES;
D O I
10.1016/j.jhydrol.2016.06.007
中图分类号
TU [建筑科学];
学科分类号
0813 ;
摘要
We propose in this paper a regional formulation of Intensity-Duration-Frequency curves of point-rainfall maxima in a scale-invariant Generalized Extreme Value (GEV) framework. The two assumptions we make is that extreme daily rainfall is GEV-distributed - which is justified by Extreme Value Theory (EVT) - and that extremes of aggregated daily rainfall follow simple-scaling relationships. Following these assumptions, we develop in a unified way a GEV simple-scaling model for extremes of aggregated daily rainfall over the range of durations where scaling applies. Then we propose a way of correcting this model for measurement frequency, giving a new GEV-scaling model for extremes of aggregated hourly rainfall. This model deviates from the simple-scaling assumption. This framework is applied to the Mediterranean region of Cevennes-Vivarais, France. A network of about 300 daily raingage stations covering the last 50 years and accumulated to span the range 1 day-1 week is used to fit the scale invariant GEV-model locally. By means of spatial interpolation of the model parameters, and correction for measurement frequency, we are able to build a regional model with good performances down to 1 h duration, even though only one hourly station is used to build the model. Finally we produce mean and return level maps within the region in the range 1 h-1 week and comment on the potential rain storms leading to these maps. (C) 2016 Elsevier B.V. All rights reserved.
引用
收藏
页码:82 / 95
页数:14
相关论文
共 47 条
[1]  
[Anonymous], 2001, INTRO STAT MODELING
[2]   Numerical simulation of orographic rainbands [J].
Anquetin, S ;
Minsicloux, F ;
Creutin, JD ;
Cosma, S .
JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES, 2003, 108 (D8)
[3]   Regional rainfall depth-duration-frequency equations for an alpine region [J].
Borga, M ;
Vezzani, C ;
Dalla Fontana, G .
NATURAL HAZARDS, 2005, 36 (1-2) :221-235
[4]   Scaling model of a rainfall intensity-duration-frequency relationship [J].
Bougadis, John ;
Adamowski, Kaz .
HYDROLOGICAL PROCESSES, 2006, 20 (17) :3747-3757
[5]   Scaling and multiscaling models of depth-duration-frequency curves for storm precipitation [J].
Burlando, P ;
Rosso, R .
JOURNAL OF HYDROLOGY, 1996, 187 (1-2) :45-64
[6]   Scaling properties of heavy rainfall at short duration: A regional analysis [J].
Ceresetti, D. ;
Molinie, G. ;
Creutin, J. -D. .
WATER RESOURCES RESEARCH, 2010, 46
[7]   A fully probabilistic approach to extreme rainfall modeling [J].
Coles, S ;
Pericchi, LR ;
Sisson, S .
JOURNAL OF HYDROLOGY, 2003, 273 (1-4) :35-50
[8]  
Cooley D, 2012, REVSTAT-STAT J, V10, P135
[9]   Statistical Modeling of Spatial Extremes [J].
Davison, A. C. ;
Padoan, S. A. ;
Ribatet, M. .
STATISTICAL SCIENCE, 2012, 27 (02) :161-186
[10]  
Davison A.C., 2003, Statistical Models