English

4D-Var using Hessian approximation and backpropagation applied to automatically-differentiable numerical and machine learning models

Machine Learning 2024-08-07 v1 Dynamical Systems Geophysics

Abstract

Constraining a numerical weather prediction (NWP) model with observations via 4D variational (4D-Var) data assimilation is often difficult to implement in practice due to the need to develop and maintain a software-based tangent linear model and adjoint model. One of the most common 4D-Var algorithms uses an incremental update procedure, which has been shown to be an approximation of the Gauss-Newton method. Here we demonstrate that when using a forecast model that supports automatic differentiation, an efficient and in some cases more accurate alternative approximation of the Gauss-Newton method can be applied by combining backpropagation of errors with Hessian approximation. This approach can be used with either a conventional numerical model implemented within a software framework that supports automatic differentiation, or a machine learning (ML) based surrogate model. We test the new approach on a variety of Lorenz-96 and quasi-geostrophic models. The results indicate potential for a deeper integration of modeling, data assimilation, and new technologies in a next-generation of operational forecast systems that leverage weather models designed to support automatic differentiation.

Keywords

Cite

@article{arxiv.2408.02767,
  title  = {4D-Var using Hessian approximation and backpropagation applied to automatically-differentiable numerical and machine learning models},
  author = {Kylen Solvik and Stephen G. Penny and Stephan Hoyer},
  journal= {arXiv preprint arXiv:2408.02767},
  year   = {2024}
}

Comments

24 pages, 7 figures

R2 v1 2026-06-28T18:04:42.484Z