Mesh-Free Interpolant Observables for Continuous Data Assimilation

Abstract

This paper considers a nudging-based scheme for data assimilation for the two-dimensional (2D) Navier-Stokes equations (NSE) with periodic boundary conditions and studies the synchronization of the signal produced by this algorithm with the true signal, to which the observations correspond, in all higher-order Sobolev topologies. This work complements previous results in the literature where conditions were identified under which synchronization is guaranteed either with respect to only the $H^1$--topology, in the case of general observables, or to the analytic Gevrey topology, in the case of spectral observables. To accommodate the property of synchronization in the stronger topologies, the framework of general interpolant observable operators, originally introduced by Azouani, Olson, and Titi, is expanded to a far richer class of operators. A significant effort is dedicated to the development of this more expanded framework, specifically, their basic approximation properties, the identification of subclasses of such operators relevant to obtaining synchronization, as well as the detailed relation between the structure of these operators and the system regarding the syncrhonization property. One of the main features of this framework is its "mesh-free" aspect, which allows the observational data itself to dictate the subdivision of the domain. Lastly, estimates for the radius of the absorbing ball of the 2D NSE in all higher-order Sobolev norms are obtained, thus properly generalizing previously known bounds; such estimates are required for establishing the synchronization property of the algorithm in the higher-order topologies.

arXiv comment36 pages, some definitions revised, results adjusted accordingly, title changed

References in corpus (7)

• Discrete Data Assimilation in the Lorenz and 2D Navier--Stokes Equations
• Abridged continuous data assimilation for the 2D Navier-Stokes equations utilizing measurements of only one component of the velocity field
• Accuracy and Stability of The Continuous-Time 3DVAR Filter for The Navier-Stokes Equation
• Reconstruction of turbulent data with deep generative models for semantic inpainting from TURB-Rot database
• Global in time stability and accuracy of IMEX-FEM data assimilation schemes for the Navier-Stokes equations
• A discrete data assimilation scheme for the solutions of the 2D Navier-Stokes equations and their statistics
• Effect of hyperviscosity on the Navier-Stokes turbulence