Projects per year
Abstract
We propose to use a decomposition of largescale incremental four
dimensional (4DVar) data assimilation problems in order to make their
numerical solution more efficient. This decomposition is based on exploiting an adaptive hierarchy of the observations. Starting with a lowcardinality set and the solution of its corresponding optimization problem, observations are adaptively added based on a posteriori error estimates. The particular structure of the sequence of associated linear systems allows the use of a variant of the conjugate gradient algorithm which effectively exploits the fact that the number of observations is smaller than the size of the vector state in the 4DVar model. The method proposed is justified by deriving the relevant error estimates at different levels of the hierarchy and a practical computational technique is then derived. The new algorithm is tested on a 1Dwave equation and on the Lorenz96 system, the latter one being of special interest because of its similarity with Numerical Weather Prediction (NWP) systems.
dimensional (4DVar) data assimilation problems in order to make their
numerical solution more efficient. This decomposition is based on exploiting an adaptive hierarchy of the observations. Starting with a lowcardinality set and the solution of its corresponding optimization problem, observations are adaptively added based on a posteriori error estimates. The particular structure of the sequence of associated linear systems allows the use of a variant of the conjugate gradient algorithm which effectively exploits the fact that the number of observations is smaller than the size of the vector state in the 4DVar model. The method proposed is justified by deriving the relevant error estimates at different levels of the hierarchy and a practical computational technique is then derived. The new algorithm is tested on a 1Dwave equation and on the Lorenz96 system, the latter one being of special interest because of its similarity with Numerical Weather Prediction (NWP) systems.
Original language  English 

Place of Publication  Namur 
Publisher  Namur center for complex systems 
Number of pages  18 
Volume  NTR062013 
Publication status  Published  2013 
Keywords
 multilevel optimization
 adaptive algorithms
 data assimilation
Fingerprint Dive into the research topics of 'Adaptive Observations And Multilevel Optimization In Data Assimilation'. Together they form a unique fingerprint.
Projects
 2 Active

Recent developments in optimization methods for data assimilation in oceanography
SARTENAER, A., LALOYAUX, P., TOINT, P., Tshimanga Ilunga, J. & Gürol, S.
1/09/07 → …
Project: PHD

Activities

Institut National Polytechnique de Toulouse
Philippe Toint (Visiting researcher)
2017 → 2019Activity: Visiting an external institution types › Research/Teaching in a external institution

Data Assimilation for Weather Forecasting: Reducing the Curse of Dimensionality
Philippe Toint (Invited speaker)
1 Dec 2015Activity: Talk or presentation types › Oral presentation

Rapporteur de la thèse de E. Bergou
Philippe Toint (Examiner)
11 Dec 2014Activity: Examination types › External Thesis