Nonlinear stability and ergodicity of ensemble based Kalman filters

Xin T. Tong, Andrew J. Majda, David Kelly

Research output: Contribution to journalArticle

Abstract

The ensemble Kalman filter (EnKF) and ensemble square root filter (ESRF) are data assimilation methods used to combine high dimensional, nonlinear dynamical models with observed data. Despite their widespread usage in climate science and oil reservoir simulation, very little is known about the long-time behavior of these methods and why they are effective when applied with modest ensemble sizes in large dimensional turbulent dynamical systems. By following the basic principles of energy dissipation and controllability of filters, this paper establishes a simple, systematic and rigorous framework for the nonlinear analysis of EnKF and ESRF with arbitrary ensemble size, focusing on the dynamical properties of boundedness and geometric ergodicity. The time uniform boundedness guarantees that the filter estimate will not diverge to machine infinity in finite time, which is a potential threat for EnKF and ESQF known as the catastrophic filter divergence. Geometric ergodicity ensures in addition that the filter has a unique invariant measure and that initialization errors will dissipate exponentially in time. We establish these results by introducing a natural notion of observable energy dissipation. The time uniform bound is achieved through a simple Lyapunov function argument, this result applies to systems with complete observations and strong kinetic energy dissipation, but also to concrete examples with incomplete observations. With the Lyapunov function argument established, the geometric ergodicity is obtained by verifying the controllability of the filter processes; in particular, such analysis for ESQF relies on a careful multivariate perturbation analysis of the covariance eigen-structure.

Original languageEnglish (US)
Article number657
JournalNonlinearity
Volume29
Issue number2
DOIs
StatePublished - Jan 25 2016

Fingerprint

Nonlinear Stability
Kalman filters
Ergodicity
Kalman Filter
Energy dissipation
Ensemble
Lyapunov functions
Filter
Controllability
filters
Geometric Ergodicity
Ensemble Kalman Filter
Energy Dissipation
Liapunov functions
Nonlinear analysis
energy dissipation
Kinetic energy
controllability
Square root
Dynamical systems

Keywords

  • catastrophic filter divergence
  • ensemble Kalman filter
  • filter stability
  • geometric ergodicity
  • Lyapunov function

ASJC Scopus subject areas

  • Statistical and Nonlinear Physics
  • Mathematical Physics
  • Physics and Astronomy(all)
  • Applied Mathematics

Cite this

Nonlinear stability and ergodicity of ensemble based Kalman filters. / Tong, Xin T.; Majda, Andrew J.; Kelly, David.

In: Nonlinearity, Vol. 29, No. 2, 657, 25.01.2016.

Research output: Contribution to journalArticle

@article{07a42e247a6c47fa8285d263f07f0066,
title = "Nonlinear stability and ergodicity of ensemble based Kalman filters",
abstract = "The ensemble Kalman filter (EnKF) and ensemble square root filter (ESRF) are data assimilation methods used to combine high dimensional, nonlinear dynamical models with observed data. Despite their widespread usage in climate science and oil reservoir simulation, very little is known about the long-time behavior of these methods and why they are effective when applied with modest ensemble sizes in large dimensional turbulent dynamical systems. By following the basic principles of energy dissipation and controllability of filters, this paper establishes a simple, systematic and rigorous framework for the nonlinear analysis of EnKF and ESRF with arbitrary ensemble size, focusing on the dynamical properties of boundedness and geometric ergodicity. The time uniform boundedness guarantees that the filter estimate will not diverge to machine infinity in finite time, which is a potential threat for EnKF and ESQF known as the catastrophic filter divergence. Geometric ergodicity ensures in addition that the filter has a unique invariant measure and that initialization errors will dissipate exponentially in time. We establish these results by introducing a natural notion of observable energy dissipation. The time uniform bound is achieved through a simple Lyapunov function argument, this result applies to systems with complete observations and strong kinetic energy dissipation, but also to concrete examples with incomplete observations. With the Lyapunov function argument established, the geometric ergodicity is obtained by verifying the controllability of the filter processes; in particular, such analysis for ESQF relies on a careful multivariate perturbation analysis of the covariance eigen-structure.",
keywords = "catastrophic filter divergence, ensemble Kalman filter, filter stability, geometric ergodicity, Lyapunov function",
author = "Tong, {Xin T.} and Majda, {Andrew J.} and David Kelly",
year = "2016",
month = "1",
day = "25",
doi = "10.1088/0951-7715/29/2/657",
language = "English (US)",
volume = "29",
journal = "Nonlinearity",
issn = "0951-7715",
publisher = "IOP Publishing Ltd.",
number = "2",

}

TY - JOUR

T1 - Nonlinear stability and ergodicity of ensemble based Kalman filters

AU - Tong, Xin T.

AU - Majda, Andrew J.

AU - Kelly, David

PY - 2016/1/25

Y1 - 2016/1/25

N2 - The ensemble Kalman filter (EnKF) and ensemble square root filter (ESRF) are data assimilation methods used to combine high dimensional, nonlinear dynamical models with observed data. Despite their widespread usage in climate science and oil reservoir simulation, very little is known about the long-time behavior of these methods and why they are effective when applied with modest ensemble sizes in large dimensional turbulent dynamical systems. By following the basic principles of energy dissipation and controllability of filters, this paper establishes a simple, systematic and rigorous framework for the nonlinear analysis of EnKF and ESRF with arbitrary ensemble size, focusing on the dynamical properties of boundedness and geometric ergodicity. The time uniform boundedness guarantees that the filter estimate will not diverge to machine infinity in finite time, which is a potential threat for EnKF and ESQF known as the catastrophic filter divergence. Geometric ergodicity ensures in addition that the filter has a unique invariant measure and that initialization errors will dissipate exponentially in time. We establish these results by introducing a natural notion of observable energy dissipation. The time uniform bound is achieved through a simple Lyapunov function argument, this result applies to systems with complete observations and strong kinetic energy dissipation, but also to concrete examples with incomplete observations. With the Lyapunov function argument established, the geometric ergodicity is obtained by verifying the controllability of the filter processes; in particular, such analysis for ESQF relies on a careful multivariate perturbation analysis of the covariance eigen-structure.

AB - The ensemble Kalman filter (EnKF) and ensemble square root filter (ESRF) are data assimilation methods used to combine high dimensional, nonlinear dynamical models with observed data. Despite their widespread usage in climate science and oil reservoir simulation, very little is known about the long-time behavior of these methods and why they are effective when applied with modest ensemble sizes in large dimensional turbulent dynamical systems. By following the basic principles of energy dissipation and controllability of filters, this paper establishes a simple, systematic and rigorous framework for the nonlinear analysis of EnKF and ESRF with arbitrary ensemble size, focusing on the dynamical properties of boundedness and geometric ergodicity. The time uniform boundedness guarantees that the filter estimate will not diverge to machine infinity in finite time, which is a potential threat for EnKF and ESQF known as the catastrophic filter divergence. Geometric ergodicity ensures in addition that the filter has a unique invariant measure and that initialization errors will dissipate exponentially in time. We establish these results by introducing a natural notion of observable energy dissipation. The time uniform bound is achieved through a simple Lyapunov function argument, this result applies to systems with complete observations and strong kinetic energy dissipation, but also to concrete examples with incomplete observations. With the Lyapunov function argument established, the geometric ergodicity is obtained by verifying the controllability of the filter processes; in particular, such analysis for ESQF relies on a careful multivariate perturbation analysis of the covariance eigen-structure.

KW - catastrophic filter divergence

KW - ensemble Kalman filter

KW - filter stability

KW - geometric ergodicity

KW - Lyapunov function

UR - http://www.scopus.com/inward/record.url?scp=85000624325&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85000624325&partnerID=8YFLogxK

U2 - 10.1088/0951-7715/29/2/657

DO - 10.1088/0951-7715/29/2/657

M3 - Article

VL - 29

JO - Nonlinearity

JF - Nonlinearity

SN - 0951-7715

IS - 2

M1 - 657

ER -