Home About HyMeX
Motivations
Science questions
Observation strategy
Modelling strategy
Target areas
Key documents
Organisation
International coordination
Working groups
Task teams
National contributions
Endorsements
Resources
Database
Data policy
Publications
Education and summer schools
Drifting balloons (BAMED)
SOP web page
Google maps data visualisation
Workshops Projects
LIAISE
ASICS-MED
MOBICLIMEX
MUSIC
IODA-MED
REMEMBER
FLOODSCALE
EXAEDRE
PICS
Offers Links Contacts
Science & Task teams
Science teams
Task teams
Implementation plan
Coordination
International Scientific Steering Committee (ISSC)
Executive Committee for Implementation and Science Coordination (EC-ISC)
Executive Committee - France (EC-Fr)
HyMeX France
HyMeX Italy
HyMeX Spain
Archive
by Mélèse, V., Blanchet, J. and Molinié, G.
Abstract:
We propose in this article a regional study of uncertainties in IDF curves derived from point-rainfall maxima. We develop two generalized extreme value models based on the simple scaling assumption, first in the frequentist framework and second in the Bayesian framework. Within the frequentist framework, uncertainties are obtained i) from the Gaussian density stemming from the asymptotic normality theorem of the maximum likelihood and ii) with a bootstrap procedure. Within the Bayesian framework, uncertainties are obtained from the posterior densities. We confront these two frameworks on the same database covering a large region of 100,000km2 in southern France with contrasted rainfall regime, in order to be able to draw conclusion that are not specific to the data. The two frameworks are applied to 405 hourly stations with data back to the 1980’s, accumulated in the range 3h–120h. We show that i) the Bayesian framework is more robust than the frequentist one to the starting point of the estimation procedure, ii) the posterior and the bootstrap densities are able to better adjust uncertainty estimation to the data than the Gaussian density, and iii) the bootstrap density give unreasonable confidence intervals, in particular for return levels associated to large return period. Therefore our recommendation goes towards the use of the Bayesian framework to compute uncertainty.
Reference:
Mélèse, V., Blanchet, J. and Molinié, G., 2018: Uncertainty estimation of Intensity–Duration–Frequency relationships: A regional analysisJournal of Hydrology, 558, 579-591.
Bibtex Entry:
@Article{Melese2018,
  author        = {Mélèse, V. and Blanchet, J. and Molinié, G.},
  title         = {Uncertainty estimation of Intensity–Duration–Frequency relationships: A regional analysis},
  journal       = {Journal of Hydrology},
  year          = {2018},
  volume        = {558},
  pages         = {579-591},
  issn          = {0022-1694},
  abstract      = {We propose in this article a regional study of uncertainties in IDF curves derived from point-rainfall maxima. We develop two generalized extreme value models based on the simple scaling assumption, first in the frequentist framework and second in the Bayesian framework. Within the frequentist framework, uncertainties are obtained i) from the Gaussian density stemming from the asymptotic normality theorem of the maximum likelihood and ii) with a bootstrap procedure. Within the Bayesian framework, uncertainties are obtained from the posterior densities. We confront these two frameworks on the same database covering a large region of 100,000km2 in southern France with contrasted rainfall regime, in order to be able to draw conclusion that are not specific to the data. The two frameworks are applied to 405 hourly stations with data back to the 1980’s, accumulated in the range 3h–120h. We show that i) the Bayesian framework is more robust than the frequentist one to the starting point of the estimation procedure, ii) the posterior and the bootstrap densities are able to better adjust uncertainty estimation to the data than the Gaussian density, and iii) the bootstrap density give unreasonable confidence intervals, in particular for return levels associated to large return period. Therefore our recommendation goes towards the use of the Bayesian framework to compute uncertainty.},
  copublication = {3: 3 Fr},
  doi           = {https://doi.org/10.1016/j.jhydrol.2017.07.054},
  owner         = {hymexw},
  timestamp     = {2019-01-25},
  url           = {http://www.sciencedirect.com/science/article/pii/S002216941730519X},
}