Skip to Main content Skip to Navigation
Theses

Quantification d'incertitudes et assimilation de données pour la modélisation hydrodynamique bidimensionnelle : application au modèle de prévision des hautes eaux de l’estuaire de la Gironde

Abstract : The accuracy of predictions is a major stake of flood forecast. This thesis deals with the development of data assimilation techniques in bidimensional numerical codes based on shallow water equations and, more specifically, the assessment of possible improvements in high water forecasting in the Gironde estuary through the development and use of these methods. The main characteristics of this work concern the unsteadiness of hydrodynamics, linked to both tidal propagation and the hydrometeorological event, and the correlation between the observations of the different available observation stations.The work carried out during the first part of the thesis focused on the proposal of a methodology for the treatment of uncertainties by global sensitivity analysis and variance decomposition for a code fed by time-dependent variables. This methodology made it possible to classify the sources of uncertainty based on a sample of the achievements of the Telemac2D numerical model for forecasting high water levels in the Gironde estuary, in order to prioritize the influence of the model's forcings and parameters.In a second step, in connection with the results of the previous study, a methodology was proposed to develop a simulation chain integrating the numerical code and an ensemblist data assimilation method, based on an Ensemble Kalman Filter, in order to reduce the uncertainties of the model related to its time-dependent forcings and parameterization. First, this methodology was applied to the Telemac2D model of the Gironde estuary with a twin experiment protocol whose objective is, on the one hand, to validate it and, on the other hand, to optimally determine the control vector, the measurement network and the parameters related to the proposed methodology. The gains for the forecast of water levels in the Gironde estuary are then evaluated using the observations available for several real events
Document type :
Theses
Complete list of metadata

https://pastel.archives-ouvertes.fr/tel-03020223
Contributor : Abes Star :  Contact
Submitted on : Monday, November 23, 2020 - 5:43:09 PM
Last modification on : Friday, December 11, 2020 - 11:29:24 AM
Long-term archiving on: : Thursday, February 25, 2021 - 1:37:50 PM

File

TH2020PESC1006.pdf
Version validated by the jury (STAR)

Identifiers

  • HAL Id : tel-03020223, version 1

Collections

Citation

Vanessya Laborie. Quantification d'incertitudes et assimilation de données pour la modélisation hydrodynamique bidimensionnelle : application au modèle de prévision des hautes eaux de l’estuaire de la Gironde. Ingénierie de l'environnement. Université Paris-Est, 2020. Français. ⟨NNT : 2020PESC1006⟩. ⟨tel-03020223⟩

Share

Metrics

Record views

149

Files downloads

85