Accelerating an ensemble of variational data assimilations with randomized preconditioning
Abstract
Ensembles of variational data assimilations (EDA) require solving systems of linear equations with iterative methods. The solution process can be accelerated using a limited memory preconditioner constructed with approximations of the leading eigenpairs of the Hessian matrix. Randomized methods for low-rank matrix approximations provide a feasible approach for computing these eigenpairs. These methods use a random sketching matrix to obtain a low-rank representation of the Hessian matrix, which is then used for computing the eigendecomposition. The sketching matrix highly influences the quality of the approximation. In this paper, we show how the structure of the EDA can be exploited to construct a suitable sketching matrix, i.e., using the differences of the right-hand sides of the linear systems of equations. Idealised numerical experiments with the Lorenz-96 model show that the resulting preconditioner is able to accelerate the EDA solution process for all ensemble members, even if constructed from the control member only.
Related
- A unified variational framework for augmented Lagrangian preconditioning of saddle point systems: applications to mixed-dimensional and fictitious domain methods
- Preconditioning proximal Galerkin: mesh and degree robust solvers for variational inequalities
- Accounting for model error in ensemble variational data assimilation through learned generative dynamics
- Comparative Performance Analysis of Optimization Solvers for Linear Support Vector Machines: Impact of Preconditioning and Dimensionality An Empirical Study Using the Benchopt Framework
- A parameterized block LU preconditioning for double saddle point problems
BibTeX
@misc{dauzickaitė2026accelerating,
title = {{Accelerating an ensemble of variational data assimilations with randomized preconditioning}},
author = {I. Daužickaitė and S. Gürol and M. Destouches and L. Berre and A. T. Weaver},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2605.23571},
eprint = {2605.23571},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2605.23571},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Accelerating an ensemble of variational data assimilations with randomized preconditioning
AU - I. Daužickaitė
AU - S. Gürol
AU - M. Destouches
AU - L. Berre
AU - A. T. Weaver
PY - 2026
DA - 2026/10/15
PB - arXiv
DO - 10.48550/arxiv.2605.23571
UR - https://arxiv.org/abs/2605.23571
AB - Ensembles of variational data assimilations (EDA) require solving systems of linear equations with iterative methods. The solution process can be accelerated using a limited memory preconditioner constructed with approximations of the leading eigenpairs of the Hessian matrix. Randomized methods for low-rank matrix approximations provide a feasible approach for computing these eigenpairs. These methods use a random sketching matrix to obtain a low-rank representation of the Hessian matrix, which is then used for computing the eigendecomposition. The sketching matrix highly influences the quality of the approximation. In this paper, we show how the structure of the EDA can be exploited to construct a suitable sketching matrix, i.e., using the differences of the right-hand sides of the linear systems of equations. Idealised numerical experiments with the Lorenz-96 model show that the resulting preconditioner is able to accelerate the EDA solution process for all ensemble members, even if constructed from the control member only.
ER -