Pith. sign in

REVIEW 3 major objections 4 minor 85 references

Long-time accuracy of ensemble Kalman filters for chaotic and machine-learned dynamical systems

T0 review · 3 major / 4 minor · reviewed 2026-08-11 · deepseek-v4-flash

Pith's one-line read The paper proves that a square-root ensemble Kalman filter with partial observations and inflation stays within a constant multiple of the noise level forever, and that machine-learned surrogate dynamics add only their…

desk verdict New and relevant results, but the proof of the central ensemble-to-mean-field lemma has a real gap that needs fixing. read the letter →

arxiv 2412.14318 v1 pith:BASO67NN submitted 2024-12-18 math.DS cs.NAmath.NAstat.ML

classification math.DScs.NAmath.NAstat.ML MSC 62F1568Q2560G3562M05
keywords ensembleKalmanfilterlong-timeaccuracydissipativechaoticdynamicalsystemssurrogatemodelsNavier-StokesequationsLorenzcovarianceinflationmean-field
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper asks whether ensemble Kalman filters, the standard tool for high-dimensional data assimilation, can be trusted over arbitrarily long time horizons when only a few coordinates are observed and the dynamics are chaotic. It answers yes: if the dynamics are dissipative and the unobserved degrees of freedom contract toward the observed ones fast enough, then with enough particles and enough covariance inflation, the filter's mean stays within a constant multiple of the observation noise level for all future times. The same conclusion holds when the forecast step is run with a machine-learned surrogate model, provided the surrogate's error in the unobserved part is small. The proof establishes accuracy first for an idealized mean-field filter, then shows that a finite ensemble stays close to that ideal filter. The conditions are verified for Lorenz-63, Lorenz-96, and the two-dimensional Navier-Stokes equations, so the result covers standard testbeds and application targets of data assimilation.

What carries the argument

The engine of the argument is the squeezing property, a detectability condition stating that the unobserved part of the difference of two trajectories contracts by a factor $\alpha<1$ after one forecast step. It is measured in the norm $V(u) = (\|u\|^2 + \beta\|Pu\|^2)^{1/2}$, which combines the full state norm with the observed-component norm. The proof uses a Lyapunov-style trace recurrence that forces the analysis covariance down to the noise level, and a small-ball lower bound on the empirical observation covariance $H\hat\Sigma_j H^*$ that lets the analysis gain be controlled with $N \ge 6k$ particles. The inflation parameter $a$ plays a dual role: it keeps the filter from trusting noisy observations too much in observed directions while keeping the empirical covariance invertible in the ensemble comparison step.

What would settle it

For the Lorenz-63 system with $H=(1,0,0)$, evaluate the supremum over the absorbing ball of $V^2((I-P)(\Psi(u)-\Psi(v)))/V^2(u-v)$ at the actual assimilation interval $\Delta t$; if the supremum reaches or exceeds 1, the squeezing inequality required by Assumption 2.1 fails and the theorem's conclusion is not guaranteed.

Watch

Extended reading notes

Core claim

The central discovery is Theorem 2.2 and its surrogate analogue Theorem 2.8: under Assumption 2.1, which combines an absorbing ball, local Lipschitz continuity, and the squeezing inequality $V^2((I-P)(\Psi(u)-\Psi(v))) \le \alpha V^2(u-v)$ with $\alpha<1$, the square-root ensemble Kalman filter with $N \ge 6k$ particles and covariance inflation $Q=aP$ satisfies $\limsup_{j\to\infty} \mathbb{E}\|\hat m_j - u_j\| \le C\varepsilon$. If the dynamics map is replaced by a surrogate $\Psi_s$ satisfying Assumption 2.7, the filter satisfies $\limsup_{j\to\infty} \mathbb{E}\|\hat m_j^s - u_j\| \le C_s(\varepsilon+\delta)$, where $\varepsilon$ is the observation noise level and $\delta$ is the surrogate's error in the unobserved components. The long-run error floor is set by the noise and the surrogate error, not by the chaotic attractor. The proof route is a mean-field Gaussian filter whose analysis covariance trace contracts geometrically, followed by a comparison showing the ensemble mean tracks the mean-field mean.

Load-bearing premise

The load-bearing premise is that the unobserved part of the difference between any two nearby states shrinks by a fixed factor less than one after one forecast step; the paper verifies this only for sufficiently frequent observations in the Lorenz and Navier-Stokes examples.

Editorial extensions

If this is right

  • For Lorenz-63, Lorenz-96, and the 2D Navier-Stokes equations with informative partial observations, the long-run filter error is bounded by $O(\varepsilon)$, so reducing observation noise directly improves state estimation over infinite time horizons.
  • An ensemble size $N \ge 6k$, independent of the state dimension, suffices for the accuracy guarantee, supporting the practical use of modest-sized ensembles in high-dimensional geophysical settings.
  • A machine-learned surrogate that is accurate only over a single assimilation cycle in the unobserved components can replace the true forecast model without losing the long-time accuracy guarantee; its error adds a $\delta$ term to the noise floor.
  • Sufficiently large covariance inflation is a required ingredient of the proof: inflation suppresses the observed-direction gain and prevents the empirical covariance from collapsing below the threshold needed for the ensemble comparison.
  • The results validate the common practice of cycling data assimilation with learned forecast models over long horizons even when those surrogates cannot forecast the attractor accurately over long timescales.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The theory suggests that training a surrogate to minimize error specifically in the unobserved components would directly lower the long-run filter error bound, whereas training on full-state or observed-coordinate losses alone may leave the bound uncontrolled.
  • Because the squeezing property is verified only for sufficiently small observation time steps, the practical reading is that frequent assimilation is needed; at long assimilation intervals the theorem gives no guarantee, and one should check the squeezing ratio numerically.
  • The $N \ge 6k$ requirement comes from a covariance lower-tail bound, so localization or deterministic covariance inflation may reduce the needed ensemble size, a testable extension the paper itself flags as an open direction.
  • The comparison strategy of ideal mean-field filter plus ensemble tracking might extend to nonlinear observations or non-Gaussian noise, but those settings would require additional conditions beyond the fixed linear observation model treated here.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 4 minor

Summary. The paper proves long-time accuracy bounds for square-root ensemble Kalman filters (EnKFs) with variance inflation for partially observed dissipative chaotic systems, including Lorenz-63, Lorenz-96, and the 2D Navier-Stokes equations. The main results, Theorem 2.2 and Theorem 2.8, state that under a squeezing/detectability condition (Assumption 2.1), an ensemble size N ≥ 6k, and sufficiently large inflation a, the analysis mean tracks the true state up to the observation noise level ε (and, for surrogate dynamics, up to ε plus the unobserved surrogate error δ). The proofs proceed by comparing the ensemble filter to an idealized mean-field Gaussian filter (Algorithm 3.1) and then bounding the ensemble-to-mean-field gap. Numerical experiments on Lorenz-96 with machine-learned surrogates illustrate the theoretical predictions.

Significance. If the main theorems are correct, this is a significant advance for the theory of ensemble Kalman filtering: it provides the first discrete-time, partially-observed accuracy guarantee for EnKFs without localization, in a setting that includes infinite-dimensional dynamics, and it validates the use of machine-learned surrogate models in data assimilation under an explicit accuracy condition on the unobserved components. The paper is well-structured, the assumptions are natural and are verified for several benchmark systems, and the mean-field comparison strategy is elegant. The surrogate-model result is practically relevant and the numerical experiments support the claims. However, the proof contains a load-bearing lemma (Lemma 3.4) that is false as stated, so the central claims are not yet established by the manuscript.

major comments (3)
  1. [Section 3, Lemma 3.4] The statement of Lemma 3.4 is false as written. The left-hand side E[max{1, λ_min(H bΣ H*)^(-q)}]^{1/q} is at least 1 for every a > 0, while the claimed upper bound 2C'/a is smaller than 1 whenever a > 2C'. The tail bound (3.19) in the proof cannot imply this statement; it would imply the corrected bound E[λ_min(H bΣ H*)^(-q)]^{1/q} ≤ C''/a without the max, via the layer-cake representation. The proof's assertion that the desired conclusion follows 'exactly as in [65]' is therefore incorrect. Since Theorem 3.3 relies on Lemma 3.4 with q=1 and q=2 to bound E[1/λ_min] and E[1/λ_min^2]^{1/2}, the proof of Theorem 2.2 currently rests on a false statement.
  2. [Section 3, Theorem 3.3; Section 2, Theorem 2.2] The proof of Theorem 3.3 uses Lemma 3.4 with q=2, but Lemma 3.4 only covers 1 ≤ q ≤ N/12. This requires N ≥ 24, whereas Theorem 2.2 and Theorem 3.3 assume only N ≥ 6k. For k ≤ 3, the condition N ≥ 6k does not imply N ≥ 24, so the invocation of Lemma 3.4 with q=2 is not justified under the stated assumptions. The ensemble size condition must be strengthened (e.g., to N ≥ max{6k, 24}) or an alternative argument must be supplied that yields the needed bound for N ≥ 6k.
  3. [Section 4, Theorem 4.2] The constants in the proof of Theorem 4.2 are not shown to be independent of ε and δ as claimed. In particular, c5 in (4.13) is defined with a factor (ε+δ), and c9 in (4.17) contains ε; these enter the final constant C3. The resulting bound contains quadratic terms in ε+δ, so the stated independence of C3 from ε and δ is not established. This is likely fixable by explicitly restricting ε+δ (e.g., to be bounded by 1) and absorbing the quadratic terms into the linear term, but the proof should state such a restriction and adjust the constants accordingly.
minor comments (4)
  1. [Section 3, Lemma 3.4] The lemma states the condition 'N ≥ min{6k, 12}', which is almost certainly a typo for 'N ≥ max{6k, 12}'. The proof uses N ≥ 6k for the tail bound and q ≤ N/12, so the lemma's own condition should be consistent with the subsequent use.
  2. [Section 3, Theorem 3.2] The proof says 'We assume without loss of generality that u0 ∈ B'. This is not entirely without loss for the mean-field filter because the analysis mean m_j is not projected into B; please clarify how the argument handles initial conditions outside B.
  3. [Section 5, numerical experiments] For the experiment illustrating Theorem 2.2, the inflation parameter is a = 1, while the theory requires a sufficiently large; the text would benefit from stating whether the chosen a satisfies the theoretical sufficient condition for the noise levels used.
  4. [General] The quantifier order for the inflation parameter a is ambiguous in Theorems 2.2 and 2.8: the proofs require a to be chosen large enough relative to ε (e.g., a ≥ 10NLcε²/k in the proof of Theorem 3.3). The statements should clarify whether a is allowed to depend on ε or whether a single a must work for a range of ε.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: theorems are derived from explicit assumptions, and the only overlapping-author citation ([75]) is an independent published proof, not an imported conclusion.

full rationale

This is a proof-based paper. Theorems 2.2 and 2.8 are derived from explicit Assumptions 2.1 and 2.7; no parameter is fitted to data and then renamed as a prediction. The ε-dependence in Theorem 2.2 arises from the analysis covariance bound in Lemma 3.1 and the Gronwall contraction α* < 1 in Theorem 3.2, not from any fitted constant. In Theorem 2.8, δ is an assumption (Assumption 2.7(3)) about surrogate error in the unobserved components, and the theorem states a bound linear in (ε+δ); this is an implication, not a definitional equivalence in which the output is built into the input. The only overlapping-author citation, [75], supplies proofs of the squeezing property for the Lorenz and Navier-Stokes examples; it is a published, parameter-free theorem with stated assumptions that do not include the present result, so it qualifies as independent support rather than a self-imported uniqueness or ansatz. Lemma 3.4 is a generalization of the independent result in [65] and is proved in the paper; whether the stated inverse-moment bound holds for large inflation a is a mathematical correctness concern, not a circularity concern. No derivation step in the paper reduces to its own conclusion by construction.

Assumptions & free parameters 1 free parameters · 5 assumptions · 0 invented entities

No new physical or mathematical entities are postulated. The central claims rest on stated assumptions about the dynamics and observations, on the inflation parameter a, and on external results from Mourtada and the authors' earlier work [75]. The examples in Section 2.1.3 are taken from [75] without re-derivation.

free parameters (1)
  • Variance inflation parameter a
    The algorithm adds Gaussian noise with covariance Q=aP to each particle. The theorems require a to be 'sufficiently large' with implicit lower bounds depending on problem constants, but no quantitative threshold is provided; a is a free algorithmic parameter, not fitted to data.
assumptions (5)
  • domain assumption Assumption 2.1: absorbing ball, local Lipschitz, squeezing property for the true dynamics Ψ with observation map H.
    Invoked in Lemma 3.1, Theorem 3.2 and throughout; verified for Lorenz-63, Lorenz-96, 2D Navier-Stokes in Section 2.1.3 via previous work [75].
  • domain assumption Assumption 2.7: surrogate model Ψs has bounded error κ, local Lipschitz continuity, and unobserved-part error at most δ.
    Needed for Theorem 2.8, Lemma 4.1, Theorems 4.2 and 4.3; this is the definition of surrogate fidelity.
  • domain assumption Observation operator satisfies HH* = I_k and P=H*H is an orthogonal projection.
    Used throughout to identify observed coordinates and bound Tr(P C P*) = Tr(H C H*); the Navier-Stokes observation operator is normalized to satisfy this.
  • standard math Mourtada's Theorem 4 and Lemma 7 on the lower tail of sample covariance matrices (Mourtada 2022, [65]).
    Used in Lemma 3.4 to obtain the λ_min(H bΣ H*) bound; the paper generalizes it to independent non-identically distributed samples and sketches the proof.
  • standard math Discrete Gronwall inequality, Jensen, Cauchy-Schwarz, Young's inequality.
    Recurrence arguments in Theorems 3.2, 3.3, 4.2, 4.3 rely on these standard tools.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Long-time accuracy of ensemble Kalman filters for chaotic and machine-learned dynamical systems." pith.science (2026). https://pith.science/paper/BASO67NN

@misc{pith2026241214318,
  author       = {Pith},
  title        = {Pith review of: Long-time accuracy of ensemble Kalman filters for chaotic and machine-learned dynamical systems},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BASO67NN}},
  note         = {Machine review of arXiv:2412.14318}
}
read the original abstract

Filtering is concerned with online estimation of the state of a dynamical system from partial and noisy observations. In applications where the state is high dimensional, ensemble Kalman filters are often the method of choice. This paper establishes long-time accuracy of ensemble Kalman filters. We introduce conditions on the dynamics and the observations under which the estimation error remains small in the long-time horizon. Our theory covers a wide class of partially-observed chaotic dynamical systems, which includes the Navier-Stokes equations and Lorenz models. In addition, we prove long-time accuracy of ensemble Kalman filters with surrogate dynamics, thus validating the use of machine-learned forecast models in ensemble data assimilation.

Figures

Figures reproduced from arXiv: 2412.14318 by the authors.

Figure 1
Figure 1. Average filter error over 50 Monte Carlo trials for decreasing observation noise levels, [PITH_FULL_IMAGE:figures/full_fig_p033_1.png] view at source ↗
Figure 2
Figure 2. The structure of Ψs . The output channels of CNN1 are divided into three groups of equal length, CNN(1) 1 , CNN(2) 1 , and CNN(3) 1 . The input channels to CNN2 are a concatenation of CNN(1) 1 and (CNN(1) 1 × CNN(2) 1 ), where the multiplication is point-wise. by each of the filters appear to track the true signal quite well. The error plots reveal that, as predicted by our theory, filtering with the low fidelity su… view at source ↗
Figure 3
Figure 3. Visualization of the true state trajectory over time, the ensemble means, the observed [PITH_FULL_IMAGE:figures/full_fig_p035_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: The left plot displays the average filter error over 50 Monte Carlo trials plotted with [PITH_FULL_IMAGE:figures/full_fig_p036_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

85 extracted references · 65 canonical work pages

  1. [65]

    Mourtada, Exact minimax risk for linear least squares, and the lower tail of sample covariance matrices, The Annals of Statistics, 50 (2022), pp

    J. Mourtada, Exact minimax risk for linear least squares, and the lower tail of sample covariance matrices, The Annals of Statistics, 50 (2022), pp. 2157–2178

  2. [1]

    Data Assimilation with Machine Learning Surrogate Models: A Case Study with FourCastNet

    M. Adrian, D. Sanz-Alonso, and R. Willett , Data assimilation with machine learning surrogate models: A case study with FourCastNet, arXiv preprint arXiv:2405.13180, (2024)

  3. [2]

    Al-Ghattas, J

    O. Al-Ghattas, J. Bao, and D. Sanz-Alonso , Ensemble Kalman filters with resampling, SIAM/ASA Journal on Uncertainty Quantification, 12 (2024), pp. 411–441

  4. [3]

    Covariance Operator Estimation: Sparsity, Lengthscale, and Ensemble Kalman Filters

    O. Al-Ghattas, J. Chen, D. Sanz-Alonso, and N. W aniorek , Covariance operator estimation: sparsity, lengthscale, and ensemble Kalman filters, arXiv preprint arXiv:2310.16933, (2023)

  5. [4]

    Al-Ghattas, J

    O. Al-Ghattas, J. Chen, D. Sanz-Alonso, and N. W aniorek , Optimal estimation of structured covariance operators, arXiv preprint arXiv:2408.02109, (2024)

  6. [5]

    Al-Ghattas and D

    O. Al-Ghattas and D. Sanz-Alonso , Non-asymptotic analysis of ensemble Kalman updates: effective dimension and localization, Information and Inference: A Journal of the IMA, 13 (2024), p. iaad043

  7. [6]

    J. L. Anderson, An ensemble adjustment Kalman filter for data assimilation, Monthly Weather Review, 129 (2001), pp. 2884–2903

  8. [7]

    Arcucci, J

    R. Arcucci, J. Zhu, S. Hu, and Y.-K. Guo , Deep data assimilation: integrating deep learning with data assimilation, Applied Sciences, 11 (2021), p. 1114

Show all 85 references
  1. [8]

    Azouani, E

    A. Azouani, E. Olson, and E. S. Titi , Continuous data assimilation using general interpolant ob- servables, Journal of Nonlinear Science, 24 (2014), pp. 277–304

  2. [9]

    E. Bach, R. Baptista, D. Sanz-Alonso, and A. Stuart , Inverse Problems and Data Assimilation: A Machine Learning Approach, arXiv preprint arXiv:2410.10523, (2024)

  3. [10]

    Bessaih, E

    H. Bessaih, E. Olson, and E. S. Titi , Continuous data assimilation with stochastically noisy data, Nonlinearity, 28 (2015), p. 729

  4. [11]

    K. Bi, L. Xie, H. Zhang, X. Chen, X. Gu, and Q. Tian , Accurate medium-range global weather LONG-TIME ACCURACY OF THE ENSEMBLE KALMAN FILTER 37 forecasting with 3d neural networks, Nature, 619 (2023), pp. 533–538

  5. [12]

    C. H. Bishop, B. J. Etherton, and S. J. Majumdar , Adaptive sampling with the ensemble transform Kalman filter. Part I: Theoretical aspects, Monthly Weather Review, 129 (2001), pp. 420–436

  6. [13]

    Bisw as and M

    A. Bisw as and M. Branicki , A unified framework for the analysis of accuracy and stability of a class of approximate Gaussian filters for the Navier-Stokes Equations, arXiv preprint arXiv:2402.14078, (2024)

  7. [14]

    Bocquet , Surrogate modeling for the climate sciences dynamics with machine learning and data assimilation, Frontiers in Applied Mathematics and Statistics, 9 (2023), p

    M. Bocquet , Surrogate modeling for the climate sciences dynamics with machine learning and data assimilation, Frontiers in Applied Mathematics and Statistics, 9 (2023), p. 1133226

  8. [15]

    Bocquet, J

    M. Bocquet, J. Brajard, A. Carrassi, and L. Bertino , Bayesian inference of chaotic dynamics by merging data assimilation, machine learning and expectation-maximization, Foundations of Data Science, 2 (2020), pp. 55–80

  9. [16]

    Brajard, A

    J. Brajard, A. Carrassi, M. Bocquet, and L. Bertino , Combining data assimilation and machine learning to emulate a dynamical model from sparse and noisy observations: A case study with the Lorenz 96 model, Journal of Computational Science, 44 (2020), p. 101171

  10. [17]

    Branicki, A

    M. Branicki, A. J. Majda, and K. J. La w , Accuracy of some approximate Gaussian filters for the Navier–Stokes equation in the presence of model error, Multiscale Modeling & Simulation, 16 (2018), pp. 1756–1794

  11. [18]

    C. E. Brett, K. F. Lam, K. La w, D. McCormick, M. R. Scott, and A. Stuart , Accuracy and stability of filters for dissipative PDEs, Physica D: Nonlinear Phenomena, 245 (2013), pp. 34–45

  12. [19]

    Cal vello, P

    E. Cal vello, P. Monmarché, A. M. Stuart, and U. V aes , Accuracy of the Ensemble Kalman Filter in the near-linear setting, arXiv preprint arXiv:2409.09800, (2024)

  13. [20]

    Cal vello, S

    E. Cal vello, S. Reich, and A. M. Stuart , Ensemble Kalman methods: a mean field perspective, arXiv preprint arXiv:2209.11371, (2022)

  14. [21]

    Carlson, A

    E. Carlson, A. F arhat, V. R. Martinez, and C. Victor , Determining modes, synchronization, and intertwinement, arXiv preprint arXiv:2408.01064, (2024)

  15. [22]

    Carlson, A

    E. Carlson, A. F arhat, V. R. Martinez, and C. Victor , On the infinite-nudging limit of the nudging filter for continuous data assimilation, arXiv preprint arXiv:2408.02646, (2024)

  16. [23]

    Carrillo, F

    J. Carrillo, F. Hoffmann, A. Stuart, and U. V aes , The mean-field ensemble Kalman filter: near- Gaussian setting, SIAM Journal on Numerical Analysis, 62 (2024), pp. 2549–2587

  17. [24]

    Chattopadhyay, M

    A. Chattopadhyay, M. Mustaf a, P. Hassanzadeh, E. Bach, and K. Kashinath , Towards physics- inspired data-driven weather forecasting: integrating data assimilation with a deep spatial-transformer- based U-NET in a case study with ERA5, Geoscientific Model Development, 15 (2022...

  18. [25]

    Chattopadhyay, E

    A. Chattopadhyay, E. Nabizadeh, E. Bach, and P. Hassanzadeh , Deep learning-enhanced ensemble-based data assimilation for high-dimensional nonlinear dynamical systems, Journal of Com- putational Physics, 477 (2023), p. 111918

  19. [26]

    K. Chen, T. Han, J. Gong, L. Bai, F. Ling, J.-J. Luo, X. Chen, L. Ma, T. Zhang, R. Su, et al., Fengwu: Pushing the skillful global medium-range weather forecast beyond 10 days lead, arXiv preprint arXiv:2304.02948, (2023)

  20. [27]

    Chen , Stochastic Methods for Modeling and Predicting Complex Dynamical Systems: Uncertainty Quantification, State Estimation, and Reduced-Order Models, Springer Nature, 2023

    N. Chen , Stochastic Methods for Modeling and Predicting Complex Dynamical Systems: Uncertainty Quantification, State Estimation, and Reduced-Order Models, Springer Nature, 2023

  21. [28]

    Chen and Y

    N. Chen and Y. Li , BAMCAFE: A Bayesian machine learning advanced forecast ensemble method for complex turbulent systems with partial observations, Chaos: An Interdisciplinary Journal of Nonlinear Science, 31 (2021)

  22. [29]

    Y. Chen, D. Sanz-Alonso, and R. Willett , Autodifferentiable ensemble Kalman filters, SIAM Jour- nal on Mathematics of Data Science, 4 (2022), pp. 801–833

  23. [30]

    Cheng, C

    S. Cheng, C. Quilodrán-Casas, S. Ouala, A. F archi, C. Liu, P. Tandeo, R. F ablet, D. Lu- cor, B. Iooss, J. Brajard, et al. , Machine learning with data assimilation and uncertainty quantification for dynamical systems: a review, IEEE/CAA Journal of Automatica Sinica, 10 (2023...

  24. [31]

    Ding and Q

    Z. Ding and Q. Li , Ensemble Kalman inversion: Mean-field limit and convergence analysis, Statistics and Computing, 31 (2021), pp. 1–21

  25. [32]

    Ding and Q

    Z. Ding and Q. Li , Ensemble Kalman sampler: Mean-field limit and convergence analysis, SIAM Journal on Mathematical Analysis, 53 (2021), pp. 1546–1578

  26. [33]

    O. G. Ernst, B. Sprungk, and H.-J. Starkloff , Analysis of the ensemble and polynomial chaos 38 D. SANZ-ALONSO AND N. WANIOREK Kalman filters in Bayesian inverse problems, SIAM/ASA Journal on Uncertainty Quantification, 3 (2015), pp. 823–851

  27. [34]

    G. Evensen, Sequential data assimilation with a nonlinear quasi-geostrophic model using Monte Carlo methods to forecast error statistics, Journal of Geophysical Research: Oceans, 99 (1994), pp. 10143– 10162

  28. [35]

    Evensen, F

    G. Evensen, F. C. Vossepoel, and P. J. V an Leeuwen , Data Assimilation Fundamentals: A Unified Formulation of the State and Parameter Estimation Problem, Springer Nature, 2022

  29. [36]

    F archi, P

    A. F archi, P. Laloyaux, M. Bona vita, and M. Bocquet , Using machine learning to correct model error in data assimilation and forecast applications, Quarterly Journal of the Royal Meteorological Society, 147 (2021), pp. 3067–3084

  30. [37]

    Foias, C

    C. Foias, C. F. Mondaini, and E. S. Titi , A discrete data assimilation scheme for the solutions of the two-dimensional Navier–Stokes equations and their statistics, SIAM Journal on Applied Dynamical Systems, 15 (2016), pp. 2109–2142

  31. [38]

    Furrer and T

    R. Furrer and T. Bengtsson , Estimation of high-dimensional prior and posterior covariance matrices in Kalman filter variants, Journal of Multivariate Analysis, 98 (2007), pp. 227–255

  32. [39]

    González-Tokman and B

    C. González-Tokman and B. R. Hunt , Ensemble data assimilation for hyperbolic systems, Physica D: Nonlinear Phenomena, 243 (2013), pp. 128–142

  33. [40]

    G. A. Gottw ald and S. Reich , Combining machine learning and data assimilation to forecast dy- namical systems from noisy partial observations, Chaos: An Interdisciplinary Journal of Nonlinear Science, 31 (2021)

  34. [41]

    G. A. Gottw ald and S. Reich , Supervised learning from noisy observations: Combining machine- learning techniques with data assimilation, Physica D: Nonlinear Phenomena, 423 (2021), p. 132911

  35. [42]

    Hamilton, T

    F. Hamilton, T. Berry, and T. Sauer , Ensemble Kalman filtering without a model, Physical Review X, 6 (2016), p. 011021

  36. [43]

    Hatfield, M

    S. Hatfield, M. Chantry, P. Dueben, P. Lopez, A. Geer, and T. Palmer , Building tangent- linear and adjoint models for data assimilation with neural networks, Journal of Advances in Modeling Earth Systems, 13 (2021), p. e2021MS002521

  37. [44]

    Hayden, E

    K. Hayden, E. Olson, and E. S. Titi , Discrete data assimilation in the Lorenz and 2D Navier–Stokes equations, Physica D: Nonlinear Phenomena, 240 (2011), pp. 1416–1425

  38. [45]

    Herty and G

    M. Herty and G. Visconti , Kinetic methods for inverse problems, Kinetic & Related Models, 12 (2019)

  39. [46]

    P. L. Houtekamer and H. L. Mitchell , Data assimilation using an ensemble Kalman filter technique, Monthly Weather Review, 126 (1998), pp. 796–811

  40. [47]

    Jiang, P

    R. Jiang, P. Y. Lu, E. Orlov a, and R. Willett , Training neural operators to preserve invariant measures of chaotic attractors, Advances in Neural Information Processing Systems, 36 (2024)

  41. [48]

    Kalnay, Atmospheric Modeling, Data Assimilation and Predictability, vol

    E. Kalnay, Atmospheric Modeling, Data Assimilation and Predictability, vol. 341, Cambridge University Press, 2003

  42. [49]

    Katzfuss, J

    M. Katzfuss, J. R. Stroud, and C. K. Wikle , Understanding the ensemble Kalman filter, The American Statistician, 70 (2016), pp. 350–357

  43. [50]

    D. T. Kelly, K. J. La w, and A. M. Stuart , Well-posedness and accuracy of the ensemble Kalman filter in discrete and continuous time, Nonlinearity, 27 (2014), p. 2579

  44. [51]

    D. P. Kingma , Adam: A method for stochastic optimization, arXiv preprint arXiv:1412.6980, (2014)

  45. [52]

    Kotsuki, K

    S. Kotsuki, K. Shiraishi, and A. Okazaki ,Integrating Ensemble Kalman Filter with AI-based Weather Prediction Model ClimaX, arXiv preprint arXiv:2407.17781, (2024)

  46. [53]

    Krasnopolsky and E

    V. Krasnopolsky and E. M. Center , Using machine learning for data assimilation, model physics, and post-processing model outputs, (2023)

  47. [54]

    Kwiatkowski and J

    E. Kwiatkowski and J. Mandel , Convergence of the square root ensemble Kalman filter in the large ensemble limit, SIAM/ASA Journal on Uncertainty Quantification, 3 (2015), pp. 1–17

  48. [55]

    R. Lam, A. Sanchez-Gonzalez, M. Willson, P. Wirnsberger, M. Fortunato, F. Alet, S. Ra vuri, T. Ew alds, Z. Eaton-Rosen, W. Hu, et al. , Graphcast: Learning skillful medium- range global weather forecasting, arXiv preprint arXiv:2212.12794, (2022)

  49. [56]

    K. La w, A. Shukla, and A. Stuart , Analysis of the 3dvar filter for thepartially observed lorenz’63 model, Discrete and Continuous Dynamical Systems, 34 (2013), pp. 1061–1078

  50. [57]

    K. La w, A. Stuart, and K. Zygalakis , Data Assimilation, vol. 214, Springer, 2015. LONG-TIME ACCURACY OF THE ENSEMBLE KALMAN FILTER 39

  51. [58]

    K. J. La w, D. Sanz-Alonso, A. Shukla, and A. M. Stuart , Filter accuracy for the Lorenz 96 model: Fixed versus adaptive observation operators, Physica D: Nonlinear Phenomena, 325 (2016), pp. 1–13

  52. [59]

    K. J. La w and A. M. Stuart , Evaluating data assimilation algorithms, Monthly Weather Review, 140 (2012), pp. 3757–3782

  53. [60]

    Le Gland, V

    F. Le Gland, V. Monbet, and V.-D. Tran , Large sample asymptotics for the ensemble Kalman filter, PhD thesis, INRIA, 2009

  54. [61]

    Mandel, L

    J. Mandel, L. Cobb, and J. D. Beezley , On the convergence of the ensemble Kalman filter, Appli- cations of Mathematics, 56 (2011), pp. 533–541

  55. [62]

    Maulik, V

    R. Maulik, V. Rao, J. W ang, G. Mengaldo, E. Constantinescu, B. Lusch, P. Balaprakash, I. Foster, and R. Kotamarthi , Efficient high-dimensional variational data assimilation with machine-learned reduced-order models, Geoscientific Model Development, 15 (2022), pp. 3433–3445

  56. [63]

    A. J. Moodey, A. S. La wless, R. W. Potthast, and P. J. V an Leeuwen ,Nonlinear error dynamics for cycled data assimilation methods, Inverse Problems, 29 (2013), p. 025002

  57. [64]

    C. Mou, L. M. Smith, and N. Chen , Combining stochastic parameterized reduced-order models with machine learning for data assimilation and uncertainty quantification with partial observations, Jour- nal of Advances in Modeling Earth Systems, 15 (2023), p. e2022MS003597

  58. [66]

    Nguyen, S

    D. Nguyen, S. Ouala, L. Drumetz, and R. F ablet , Em-like learning chaotic dynamics from noisy and partial observations, arXiv preprint arXiv:1903.10335, (2019)

  59. [67]

    Nguyen, J

    T. Nguyen, J. Brandstetter, A. Kapoor, J. K. Gupta, and A. Grover , Climax: A foundation model for weather and climate, arXiv preprint arXiv:2301.10343, (2023)

  60. [68]

    D. S. Oliver, A. C. Reynolds, and N. Liu , Inverse theory for petroleum reservoir characterization and history matching, 2008

  61. [69]

    J. Park, N. Yang, and N. Chandramoorthy , When are dynamical systems learned from time series data statistically accurate?, arXiv preprint arXiv:2411.06311, (2024)

  62. [70]

    Pathak, S

    J. Pathak, S. Subramanian, P. Harrington, S. Raja, A. Chattopadhyay, M. Mardani, T. Kurth, D. Hall, Z. Li, K. Azizzadenesheli, et al. , FourCastNet: A global data-driven high- resolution weather model using adaptive Fourier neural operators, arXiv preprint arXiv:2202.11214, (2022)

  63. [71]

    L. M. Pecora and T. L. Carroll , Synchronization in chaotic systems, Physical Review Letters, 64 (1990), p. 821

  64. [72]

    S. G. Penny, T. A. Smith, T.-C. Chen, J. A. Platt, H.-Y. Lin, M. Goodliff, and H. D. Abarbanel, Integrating recurrent neural networks with data assimilation for scalable data-driven state estimation, Journal of Advances in Modeling Earth Systems, 14 (2022), p. e2021MS002843

  65. [73]

    J. C. Robinson,Infinite-Dimensional Dynamical Systems: an Introduction to Dissipative Parabolic PDEs and the Theory of Global Attractors, vol. 28, Cambridge University Press, 2001

  66. [74]

    Sanz-Alonso, A

    D. Sanz-Alonso, A. Stuart, and A. Taeb , Inverse Problems and Data Assimilation, vol. 107, Cam- bridge University Press, 2023

  67. [75]

    Sanz-Alonso and A

    D. Sanz-Alonso and A. M. Stuart , Long-time asymptotics of the filtering distribution for partially observed chaotic dynamical systems, SIAM/ASA Journal on Uncertainty Quantification, 3 (2015), pp. 1200–1220

  68. [76]

    Sanz-Alonso and N

    D. Sanz-Alonso and N. W aniorek, Analysis of a computational framework for Bayesian inverse prob- lems: Ensemble Kalman updates and MAP estimators under mesh refinement, SIAM/ASA Journal on Uncertainty Quantification, 12 (2024), pp. 30–68

  69. [77]

    Stuart and A

    A. Stuart and A. R. Humphries , Dynamical Systems and Numerical Analysis, vol. 2, Cambridge University Press, 1998

  70. [78]

    Takeda and T

    K. Takeda and T. Sakajo , Uniform error bounds of the ensemble transform Kalman filter for infinite- dimensional dynamics with multiplicative covariance inflation , arXiv preprint arXiv:2402.03756, (2024)

  71. [79]

    S. Tang, T. Sapsis, and N. Azizan , Learning chaotic dynamics with embedded dissipativity, arXiv preprint arXiv:2410.00976, (2024)

  72. [80]

    Temam, Navier–Stokes Equations and Nonlinear Functional Analysis, SIAM, 1995

    R. Temam, Navier–Stokes Equations and Nonlinear Functional Analysis, SIAM, 1995. 40 D. SANZ-ALONSO AND N. WANIOREK

  73. [81]

    X. Tian, D. Holda w ay, and D. Kleist, Exploring the use of machine learning weather models in data assimilation, arXiv preprint arXiv:2411.14677, (2024)

  74. [82]

    M. K. Tippett, J. L. Anderson, C. H. Bishop, T. M. Hamill, and J. S. Whitaker , Ensemble square root filters, Monthly Weather Review, 131 (2003), pp. 1485–1490

  75. [83]

    X. T. Tong, A. J. Majda, and D. Kelly , Nonlinear stability of the ensemble Kalman filter with adaptive covariance inflation, Communications in Mathematical Sciences, 14 (2016), pp. 1283–1313

  76. [84]

    Tsuyuki and R

    T. Tsuyuki and R. Tamura , Nonlinear data assimilation by deep learning embedded in an ensemble Kalman filter, Journal of the Meteorological Society of Japan. Ser. II, 100 (2022), pp. 533–553

  77. [85]

    Y. Xiao, L. Bai, W. Xue, K. Chen, T. Han, and W. Ouyang , FengWu-4DVar: Coupling the data- driven weather forecasting model with 4D variational assimilation, arXiv preprint arXiv:2312.12455, (2023)

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.