Pith. sign in

REVIEW 4 major objections 6 minor 50 references

Model Error Covariance Estimation for Weak Constraint Data Assimilation

T0 review · 4 major / 6 minor · reviewed 2026-08-16 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper claims that standard regularization parameter selection methods—the L-curve, generalized cross-validation, and the chi-squared criterion—can estimate model error covariance hyperparameters in weak constraint 4D-Var, and that…

desk verdict Useful derivation, weak validation: the experiments never generate model error from the assumed covariance, so the headline quantitative claim is untested. read the letter →

arxiv 2504.17900 v1 pith:YMBNB7RH submitted 2025-04-24 stat.ME cs.NAmath.NAmath.OC

Sandra R. Babyale , Jodi Mead , Donna Calhoun , Patricia O. Azike This is my paper · ORCID
classification stat.MEcs.NAmath.NAmath.OC MSC 65K1065F22
keywords dataassimilationweakconstraint4D-VarmodelerrorcovarianceestimationregularizationparameterselectionL-curvegeneralizedcross-validationchi-squaredmethodrepresenter
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

State estimates from weak constraint 4D-Var depend heavily on the unknown model error covariance, which controls how much the analysis trusts the forecast versus the observations. This paper treats that covariance as the regularization matrix of a Tikhonov inverse problem and shows that three standard regularization parameter selection methods—the L-curve, generalized cross-validation, and the chi-squared criterion—can estimate its hyperparameters. Using the representer method, the problem is reduced from state space to data space, giving analytic expressions for the optimal state and for each selection criterion. In 1D transport experiments simulating wildfire smoke, the estimated variances are lower when the first guess is accurate and higher when the observations are trustworthy, and the non-isotropic covariance is preferred when data dominate. The contribution is a practical, data-driven way to set model error covariances without manual tuning.

What carries the argument

The representer method: the optimal weak-constraint 4D-Var state is written as the first guess plus a linear combination of $M$ representer functions, one per observation, so the infinite-dimensional PDE optimization collapses to an $M$-dimensional linear solve. The load-bearing pieces are the analytic state $\hat q = q_F + h^TP^{-1}r$ with $P=R+C_\epsilon$, the identity that the minimized weak-constraint cost satisfies $\hat J = h^TP^{-1}h$, and the leave-one-out identity $\hat q^{[k]}(x_k,t_k)-d_k = (\hat q(x_k,t_k)-d_k)/(1-(RP^{-1})_{kk})$ that makes GCV cheap. These reduce covariance estimation to choosing one or three scalar hyperparameters by minimizing a GCV surface or solving $\hat J=M$.

What would settle it

Repeat the experiments with a twin setup in which the true model error is drawn from a Gaussian field with known covariance of the form (3.22) and known $\sigma_f^2$, $l_f$, $\tau_f$, then check whether the L-curve, GCV, and chi-squared methods recover those values; failure to do so, or recovery for only one covariance shape, would falsify the claim that the methods estimate the actual model error covariance hyperparameters. A cheaper check is to compare the empirical covariance of (first guess minus truth) across many realizations with the estimated covariance.

Watch

Extended reading notes

Core claim

Framing weak constraint 4D-Var as a regularized inverse problem with the inverse model error covariance as the regularization matrix, the paper derives matrix expressions for the L-curve, GCV, and chi-squared criteria in the representer formulation. The central identity is the analytic optimal state $\hat q(x,t)=q_F(x,t)+h^TP^{-1}r(x,t)$ with $P=R+C_\epsilon$, which makes the reduced data misfit and model penalty explicit: $\hat J_{\rm data}=h^TP^{-1}W_d^{-1}P^{-1}h$ and $\hat J_{\rm mod}=h^TP^{-1}RP^{-1}h$. From these, GCV and chi-squared estimate hyperparameters (variance, spatial length scale, temporal scale) with only a handful of data-assimilation solves. In four simulated experiments the estimates track the experimental design: small variances when the first guess is good, larger variances and longer correlation scales when the data are reliable; the non-isotropic covariance improves RMSE in data-dominated cases. The central claim is that these selection methods recover model error covariance hyperparameters that reflect which source of information is more trustworthy, and that isotropic covariance suffices when the model is trusted whereas non-isotropic covariance is preferred when the data are reliable.

Load-bearing premise

The whole estimation inherits the assumption that model error is a zero-mean Gaussian field with a prescribed covariance kernel—yet the experiments generate model error by perturbing deterministic source-term parameters, so the criteria are applied under a misspecified error model.

Editorial extensions

If this is right

  • In weak constraint 4D-Var, the model error variance can be estimated by minimizing the GCV function or solving the chi-squared equation $h^TP^{-1}h=M$, with only a small number of data-assimilation runs (up to about five or seven in the isotropic test cases).
  • The L-curve can also be applied, but it requires evaluating a range of regularization parameters (around 100 solves), and in these experiments the corner is not always the maximum-curvature point; maximum curvature gave better estimates.
  • When the first guess is more accurate than the observations, an isotropic covariance is sufficient; when observations are more reliable, the non-isotropic Gaussian-exponential covariance produces lower RMSE in the assimilated state.
  • The same representer-based framework extends to jointly estimating initial and boundary-condition error covariances, and to higher-resolution grids through efficient covariance multiplication.
  • The estimated variances are lower in model-dominant experiments and higher in data-dominant experiments, meaning the criteria reproduce the intended balance without manual tuning.

Reading between the lines

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

  • A direct test would generate model error from the assumed Gaussian covariance with known hyperparameters and check whether the three methods recover them; the paper's twin experiments use a misspecified error model, so this would separate the selection methods' validity from the covariance-shape assumption.
  • The same representer reduction could estimate spatially varying or flow-dependent error covariances by choosing a richer parametric family and applying the multi-parameter GCV and chi-squared machinery, at the cost of a higher-dimensional search.
  • Because GCV and chi-squared need only a handful of assimilation solves, the approach is likely to transfer to operational settings with expensive forward and adjoint solvers, provided the covariance kernel can be evaluated implicitly.
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

4 major / 6 minor

Summary. The paper frames weak-constraint 4D-Var as a regularized inverse problem in which the inverse model error covariance plays the role of the regularization matrix, and uses the representer method to reduce the variational problem from state space to data space. It derives matrix expressions for three regularization parameter selection methods—the L-curve, generalized cross-validation (GCV), and the chi-squared method—for estimating hyperparameters of isotropic and non-isotropic model error covariances. The method is tested on a 1D wildfire smoke transport model with synthetic observations in four experiments that vary whether the first guess or the observations are more accurate. The authors claim that the methods successfully estimate model error covariance hyperparameters that reflect the relative reliability of the first guess versus the observational data, and that isotropic covariances suffice when the first guess is more accurate whereas non-isotropic covariances are preferable when the data are more reliable.

Significance. The methodological core—representer reduction to data space, analytic formulas for the three selection criteria (Lemmas A.1–A.2, Theorem A.3, Theorem 3.1), and the resulting data-space computational cost—is sound and potentially useful for weak-constraint 4D-Var problems with modest observation counts. The qualitative trend that estimated variances increase as the experimental design moves from model-dominated to data-dominated settings is plausible and consistent across the isotropic experiments. However, the validation currently lacks a controlled recovery experiment and contains post hoc filtering, so the strong quantitative claim in the abstract is not yet established. If the validation gap is closed, the paper would be a useful contribution to covariance estimation in variational data assimilation.

major comments (4)
  1. [Sec. 4.1.3 / Eq. (4.5) and Sec. 3.2] The numerical experiments never generate model error from the covariance model assumed in the derivation. The first guess is obtained by perturbing source-term parameters, so the model error field f = ∂q/∂t + u∂q/∂x − Q is a deterministic, smooth, space-time correlated field produced by parameter perturbations, not a zero-mean Gaussian process with covariance (3.18) or (3.22). The chi-squared identity h^T P^{-1} h ∼ χ²_M (Theorem A.3, Eq. 3.20) and the GCV leave-one-out derivation (Appendix B) both require the innovation to have covariance R + C_f under the chosen C_f, so those hypotheses are never active in the experiments. As a result, the reported σ_f², l_f, τ_f may be proxies for relative reliability rather than estimates of a true covariance, and the abstract's claim "successfully estimate hyperparameters" is not quantitatively supported. A controlled recovery experiment—where synthetic model errors are drawn from a known C_f and the estimates are compared against the true hyperparameters—is needed to support the central claim.
  2. [Sec. 4.3.1 / Table 2] The outlier-removal thresholds [0.35,0.7], [0.003,0.7], [0.8,10], and [0.5,6] are introduced post hoc, are experiment-specific, and are not pre-specified or justified by a statistical rule. Since the reported means and standard deviations are computed after removing these outliers, the discrepancy between Experiments 1–2 and 3–4 in Table 2 may be partly an artifact of the filtering. Please report the full distributions, the fraction of runs removed in each experiment, and a sensitivity analysis to threshold choice.
  3. [Sec. 4.4 / Table 4] The non-isotropic comparison rests on a single data vector d ∈ R^{30×1} (Sec. 4, opening paragraph), yet Sec. 4.4.1 draws comparative conclusions about GCV versus chi-squared behavior, including the opposite l_f and τ_f choices in Experiment 4. There is no ensemble or uncertainty quantification for any of the values in Table 4, so the qualitative claims in that section are not supported by the evidence shown. The non-isotropic experiments should be repeated over multiple datasets and the variability reported.
  4. [Sec. 3.2.1 / Eq. (3.21)] The chi-squared method is a moment-fitting condition by construction: selecting the hyperparameter such that h^T P^{-1} h = M forces the minimized cost to equal its expected value under the assumed Gaussian model. The paper should state plainly that for this criterion, "successful" estimation in the experiments is a consistency check with the same data used to build P, not an independent prediction. This does not invalidate the method, but it does affect how much weight the experiments can carry.
minor comments (6)
  1. [Abstract and Sec. 4.3.1] There are duplicated words: "and and" in the abstract and "are are" in Sec. 4.3.1.
  2. [Sec. 3.2.1] The sentence "It has been demonstrated in in [31]" contains a duplicated "in".
  3. [Eqs. (3.23) and (3.24)] The notation g(σ²_f, l_f τ_f) and the corresponding chi-squared expression omit a comma and should read g(σ²_f, l_f, τ_f).
  4. [Sec. 4.3.2, Experiment 2] The time "t = 05.93" should be "t = 5.93," and the word "osccilation" is a typo.
  5. [Sec. 4.1.2] The statement that equation (4.3) is run 10^5 times to compute the mean and standard deviation of the RMSE is ambiguous; please state the exact number of realizations and the criterion used to select the 500 retained columns.
  6. [Sec. 4.3.1] The L-curve section states that the maximum-curvature value was used instead of the corner; this choice should be justified and its effect on the results reported, since it is another post hoc selection decision.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity; the core derivations are self-contained and the chi-square method is transparently a moment fit.

full rationale

The paper's derivation chain is largely self-contained. The representer reduction (Eqs. 3.10-3.16), the reduced penalty functionals (Lemmas A.1-A.2 and Theorem A.3), and the GCV leave-one-out formula (Appendix B) are proven in the paper rather than imported as black boxes. The chi-square method is explicitly identified as a method of moments (Section 2.1.3), and its selection rule h^T P^{-1}h = M (Eq. 3.21) is a moment-matching estimator by design; the paper does not claim to predict a quantity independent of that fit. Self-citations to Mead for the chi-square distributional result and monotonicity, and to Bennett for representer lemmas, support standard or externally published results and are not the sole justification for the central claim. The main weaknesses are validation-related rather than circular: the twin experiments generate model error by perturbing source-term parameters (Eq. 4.5), not from the assumed Gaussian covariance (Eqs. 3.18 and 3.22), so no ground-truth covariance is available for comparison; and the post hoc outlier removal uses experiment-specific thresholds (Section 4.3.1). These are correctness risks for the claim that the hyperparameters are 'successfully estimated,' but they do not make any derived quantity equivalent to its inputs by construction. Score 2 reflects the minor reliance on self-cited chi-square machinery; no circular step was identified.

Assumptions & free parameters 5 free parameters · 7 assumptions · 0 invented entities

The framework rests on standard inverse-problem and representer machinery, with the key ad hoc elements being the parametric covariance ansatz and the experiment-specific outlier thresholds. The Gaussian process assumption for model error is not satisfied by the twin experiment design, which is the most consequential modeling choice.

free parameters (5)
  • sigma_f^2 (isotropic model error variance) = Experiment means: L-curve/GCV/chi-squared, e.g., exp1: 0.5291/0.5413/0.5332; exp2: 0.6825/0.0496/0.0044
    The central claim that the methods estimate covariance depends on this scalar variance selected from the data by each of the three methods; reported in Table 2.
  • sigma_f^2 (non-isotropic model error variance) = GCV/chi-squared: exp1: 0.000556/0.000194; exp3: 0.005569/0.668314; exp4: 4.457971/0.027128
    Estimated by multi-parameter GCV and chi-squared in Section 4.4.1, Table 4.
  • l_f (spatial correlation length) = GCV/chi-squared: exp1: 3.000118/3.000528; exp4: 12.000000/1.000003
    Estimated spatial scale in the non-isotropic covariance (3.22), fitted to the data.
  • tau_f (temporal correlation scale) = GCV/chi-squared: exp1: 5.000048/5.000171; exp4: 1.000000/15.594117
    Estimated temporal scale in (3.22), fitted to data.
  • Outlier removal thresholds for GCV and chi-squared estimates = exp1: [0.35,0.7]; exp2: [0.003,0.7]; exp3: [0.8,10]; exp4: [0.5,6]
    Experiment-specific bounds used to remove 'outliers' before computing Table 2 means; chosen post hoc without stated criteria, directly influencing the reported estimates.
assumptions (7)
  • domain assumption Model error is a zero-mean Gaussian process with covariance Cf of known parametric form (isotropic (3.18) or separable Gaussian/exponential (3.22))
    Used throughout Section 3.2; the GCV and chi-squared selection criteria are derived under this error model.
  • domain assumption Observation error covariance C_epsilon is known
    Stated in the abstract and Section 3.2: 'we assume that the data error is known'.
  • domain assumption Initial and boundary condition errors are known or exact
    Section 3.2 assumes Ci and Cb are known or exact; experiments set initial and boundary conditions exact (Section 4.1.1).
  • domain assumption The minimized weak constraint cost function follows a chi-squared distribution with M degrees of freedom
    Used by the chi-squared method in (3.20)-(3.21); attributed to Mead [30] and Bennett [6].
  • standard math The leave-one-out GCV formula applies with influence matrix R P^{-1}
    Theorem 3.1 proof in Appendix B uses the standard GCV leave-one-out lemma, assuming the representer expansion (3.16) is exact.
  • standard math The representer theorem representation (3.10)-(3.11) is exact for the linearized operator
    Section 3.1; relies on the RKHS representer theorem; requires L linearized.
  • ad hoc to paper Perturbing source term parameters in the first guess produces model errors that can be represented by an additive Gaussian covariance
    Section 4.1.3; the twin experiment generates discrepancies via perturbed alpha and k values (4.5), not via sampling from Cf. The methods are nonetheless applied as if Cf were the true error covariance.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Model Error Covariance Estimation for Weak Constraint Data Assimilation." pith.science (2026). https://pith.science/paper/YMBNB7RH

@misc{pith2026250417900,
  author       = {Pith},
  title        = {Pith review of: Model Error Covariance Estimation for Weak Constraint Data Assimilation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YMBNB7RH}},
  note         = {Machine review of arXiv:2504.17900}
}
read the original abstract

State estimates from weak constraint 4D-Var data assimilation can vary significantly depending on the data and model error covariances. As a result, the accuracy of these estimates heavily depends on the correct specification of both model and observational data error covariances. In this work, we assume that the data error is known and and focus on estimating the model error covariance by framing weak constraint 4D-Var as a regularized inverse problem, where the inverse model error covariance serves as the regularization matrix. We consider both isotropic and non-isotropic forms of the model error covariance. Using the representer method, we reduce the 4D-Var problem from state space to data space, enabling the efficient application of regularization parameter selection techniques. The Representer method also provides an analytic expression for the optimal state estimate, allowing us to derive matrix expressions for the three regularization parameter selection methods i.e. the L-curve, generalized cross-validation (GCV), and the Chi-square method. We validate our approach by assimilating simulated data into a 1D transport equation modeling wildfire smoke transport under various observational noise and forward model perturbations. In these experiments the goal is to identify the model error covariances that accurately capture the influence of observational data versus model predictions on assimilated state estimates. The regularization parameter selection methods successfully estimate hyperparameters for both isotropic and non-isotropic model error covariances, that reflect whether the first guess model predictions are more or less reliable than the observational data. The results further indicate that isotropic variances are sufficient when the first guess is more accurate than the data whereas non-isotropic covariances are preferred when the observational data is more reliable.

Figures

Figures reproduced from arXiv: 2504.17900 by the authors.

Figure 1
Figure 1. Illustrative example of model error variance estimation for 4D-Var with rep [PITH_FULL_IMAGE:figures/full_fig_p010_1.png] view at source ↗
Figure 2
Figure 2. PM2.5 concentration as a function of space for experiment 1 Experiment 2. This experiment was set up similarly to experiment 1, but the difference in this experiment is that it has two sources of PM2.5 at x = 33 and x = 40, and no flux boundary conditions were used [PITH_FULL_IMAGE:figures/full_fig_p016_2.png] view at source ↗
Figure 3
Figure 3. PM2.5 concentration estimates a function of space for experiment 2 Experiment 3. This experiment was designed with the simulated observational data more accurate than the first guess. Similar to experiment 1 a single source is located at x = 33 and periodic boundary conditions were used [PITH_FULL_IMAGE:figures/full_fig_p017_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: PM2.5 concentration estimates a function of time for experiment 3 Experiment 4 This experiment was set up similarly to experiment 3, but the key difference in this experiment is that it has two sources of PM2.5 located at x = 33 and x = 40, and no flux boundary conditi…
Figure 5
Figure 5. Figure 5: PM2.5 concentration estimates a function of time for experiment 4 4.3.3. Root mean square errors (RMSE) [PITH_FULL_IMAGE:figures/full_fig_p019_5.png]
Figure 6
Figure 6. Figure 6: PM2.5 concentration estimates as a function of space for experiment 1 Experiment 4 [PITH_FULL_IMAGE:figures/full_fig_p022_6.png]
Figure 7
Figure 7. Figure 7: PM2.5 concentration estimates as a function of time for experiment 4 4.4.3. Root mean square errors [PITH_FULL_IMAGE:figures/full_fig_p023_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

50 extracted references · 41 canonical work pages

  1. [1]

    A monte carlo implementation of the nonlinear filtering problem to produce ensemble assimilations and forecasts

    Jeffrey L Anderson and Stephen L Anderson. A monte carlo implementation of the nonlinear filtering problem to produce ensemble assimilations and forecasts. Monthly weather review , 127(12):2741–2758, 1999

    work page 1999

  2. [2]

    Parameter estimation and inverse problems

    Richard C Aster, Brian Borchers, and Clifford H Thurber. Parameter estimation and inverse problems. Elsevier, 2018

    work page 2018

  3. [3]

    A review of operational methods of variational and ensemble-variational 26 S

    Ross N Bannister. A review of operational methods of variational and ensemble-variational 26 S. R. BABYALE, J. MEAD, D. CALHOUN AND P. O. AZIKE data assimilation. Quarterly Journal of the Royal Meteorological Society , 143(703):607– 633, 2017

    work page 2017

  4. [4]

    Aerosol analysis and forecast in the european centre for medium- range weather forecasts integrated forecast system: 2

    A Benedetti, J-J Morcrette, O Boucher, A Dethof, RJ Engelen, M Fisher, H Flentje, N Huneeus, L Jones, JW Kaiser, et al. Aerosol analysis and forecast in the european centre for medium- range weather forecasts integrated forecast system: 2. data assimilation. Journal of Geo- physical Research: Atmospheres , 114(D13), 2009

    work page 2009

  5. [5]

    Inverse methods in physical oceanography

    Andrew F Bennett. Inverse methods in physical oceanography . Cambridge university press, 1992

    work page 1992

  6. [6]

    Inverse modeling of the ocean and atmosphere

    Andrew F Bennett. Inverse modeling of the ocean and atmosphere . Cambridge University Press, 2005

    2005

  7. [7]

    Generalized inversion of a global numerical weather prediction model

    Andrew F Bennett, Boon S Chua, and LM Leslie. Generalized inversion of a global numerical weather prediction model. Meteorology and Atmospheric Physics , 60:165–178, 1996

    work page 1996

  8. [8]

    Multi- parameter regularization techniques for ill-conditioned linear systems

    Claude Brezinski, Michela Redivo-Zaglia, Giuseppe Rodriguez, and Sebastiano Seatzu. Multi- parameter regularization techniques for ill-conditioned linear systems. Numerische Math- ematik, 94:203–228, 2003

    work page 2003

Show all 50 references
  1. [9]

    Data assimilation in the geosciences: An overview of methods, issues, and perspectives

    Alberto Carrassi, Marc Bocquet, Laurent Bertino, and Geir Evensen. Data assimilation in the geosciences: An overview of methods, issues, and perspectives. Wiley Interdisciplinary Reviews: Climate Change , 9(5):e535, 2018

    2018

  2. [10]

    A strategy for operational implementation of 4d-var, using an incremental approach

    PHILIPPE Courtier, J-N Th´ epaut, and Anthony Hollingsworth. A strategy for operational implementation of 4d-var, using an incremental approach. Quarterly Journal of the Royal Meteorological Society, 120(519):1367–1387, 1994

    1994

  3. [11]

    Topex/poseidon tides estimated using a global invese model

    GD Egbert. Topex/poseidon tides estimated using a global invese model. J. Geophys. Res. , 99:852, 1994

    1994

  4. [12]

    Using the l–curve for determining optimal regularization parameters

    Heinz W Engl and Wilhelm Grever. Using the l–curve for determining optimal regularization parameters. Numerische Mathematik , 69(1):25–31, 1994

    1994

  5. [13]

    Inverse methods and data assimilation in nonlinear ocean models

    Geir Evensen. Inverse methods and data assimilation in nonlinear ocean models. Physica D: Nonlinear Phenomena, 77(1-3):108–129, 1994

    1994

  6. [14]

    Sequential data assimilation with a nonlinear quasi-geostrophic model using monte carlo methods to forecast error statistics

    Geir Evensen. Sequential data assimilation with a nonlinear quasi-geostrophic model using monte carlo methods to forecast error statistics. Journal of Geophysical Research: Oceans , 99(C5):10143–10162, 1994

    1994

  7. [15]

    Rank issues

    Geir Evensen and Geir Evensen. Rank issues. Springer, 2009

    2009

  8. [16]

    Weak-constraint and long-window 4d-var

    Mike Fisher, Yannick Tr´ emolet, Harri Auvinen, David Tan, and Paul Poli. Weak-constraint and long-window 4d-var. ECMWF Technical Memoranda, 655:47, 2011

    2011

  9. [17]

    Tikhonov regularization and total least squares

    Gene H Golub, Per Christian Hansen, and Dianne P O’Leary. Tikhonov regularization and total least squares. SIAM journal on matrix analysis and applications , 21(1):185–194, 1999

    1999

  10. [18]

    Generalized cross-validation as a method for choosing a good ridge parameter

    Gene H Golub, Michael Heath, and Grace Wahba. Generalized cross-validation as a method for choosing a good ridge parameter. Technometrics, 21(2):215–223, 1979

    1979

  11. [19]

    Novel approach to nonlinear/non- gaussian bayesian state estimation

    Neil J Gordon, David J Salmond, and Adrian FM Smith. Novel approach to nonlinear/non- gaussian bayesian state estimation. In IEE proceedings F (radar and signal processing) , volume 140, pages 107–113. IET, 1993

    1993

  12. [20]

    The l-curve and its use in the numerical treatment of inverse problems

    Per Christian Hansen. The l-curve and its use in the numerical treatment of inverse problems. 1999

    1999

  13. [21]

    The elements of statistical learning, 2009

    Trevor Hastie, Robert Tibshirani, Jerome Friedman, et al. The elements of statistical learning, 2009

    2009

  14. [22]

    Data assimilation for numerical smoke prediction

    Edward J Hyer, Christopher P Camacho, David A Peterson, Elizabeth A Satterfield, and Pablo E Saide. Data assimilation for numerical smoke prediction. Landscape Fire, Smoke, and Health: Linking Biomass Burning Emissions to Human Well-Being , pages 105–125, 2023

    2023

  15. [23]

    Selecting the corner in the l-curve approach to tikhonov regularization

    Peter R Johnston and Ramesh M Gulrajani. Selecting the corner in the l-curve approach to tikhonov regularization. IEEE Transactions on biomedical engineering , 47(9):1293–1296, 2000

    2000

  16. [24]

    A new approach to linear filtering and prediction problems

    Rudolph Emil Kalman. A new approach to linear filtering and prediction problems. 1960

    1960

  17. [25]

    Atmospheric modeling, data assimilation and predictability

    Eugenia Kalnay. Atmospheric modeling, data assimilation and predictability . Cambridge uni- versity press, 2003

    2003

  18. [26]

    Variational algorithms for analysis and assim- ilation of meteorological observations: theoretical aspects

    Fran¸ cois-Xavier Le Dimet and Olivier Talagrand. Variational algorithms for analysis and assim- ilation of meteorological observations: theoretical aspects. Tellus A: Dynamic Meteorology and Oceanography, 38(2):97–110, 1986

    1986

  19. [27]

    Finite volume methods for hyperbolic problems , volume 31

    Randall J LeVeque. Finite volume methods for hyperbolic problems , volume 31. Cambridge university press, 2002

    2002

  20. [28]

    From stein’s unbiased risk estimates to the method of generalized cross validation

    Ker-Chau Li. From stein’s unbiased risk estimates to the method of generalized cross validation. The Annals of Statistics , pages 1352–1377, 1985. MODEL ERROR COVARIANCE ESTIMATION FOR WEAK CONSTRAINT 4D-VAR 27

    1985

  21. [29]

    AC Lorenc, SP Ballard, RS Bell, NB Ingleby, PLF Andrews, DM Barker, JR Bray, AM Clayton, T Dalby, D Li, et al. The met. office global three-dimensional variational data assimilation scheme. Quarterly Journal of the Royal Meteorological Society , 126(570):2991–3012, 2000

    2000

  22. [30]

    J.L. Mead. Parameter estimation: A new approach to weighting a priori information. J. Inv. Ill-posed Problems, 16(2):175–194, 2008

    2008

  23. [31]

    J.L. Mead. Chi-squared test for total variation regularization parameter selection. Inverse Problems & Imaging , 14(3):401–421, 2020

    2020

  24. [32]

    Mead and C.C

    J.L. Mead and C.C. Hammerquist. Chi-sqaured tests for the choice of the regularization param- eter in nonlinear inverse problems. SIAM Journal on Matrix Analysis and Applications , 34(3):1213–1230, 2013

    2013

  25. [33]

    Discontinuous parameter estimates with least squares estimators

    Jodi L Mead. Discontinuous parameter estimates with least squares estimators. Applied Math- ematics and Computation , 219(10):5210–5223, 2013

    2013

  26. [34]

    Efficient implementation of covariance multiplication for data assimilation with the representer method

    Hans E Ngodock. Efficient implementation of covariance multiplication for data assimilation with the representer method. Ocean Modelling, 8(3):237–251, 2005

    2005

  27. [35]

    Nonlinear conditional model bias estimation for data assimilation

    Jason A Otkin, Roland WE Potthast, and Amos S Lawless. Nonlinear conditional model bias estimation for data assimilation. SIAM Journal on Applied Dynamical Systems , 20(1):299– 332, 2021

    2021

  28. [36]

    Data assimilation with the weighted ensemble kalman filter

    Nicolas Papadakis, ´Etienne M´ emin, Anne Cuzol, and Nicolas Gengembre. Data assimilation with the weighted ensemble kalman filter. Tellus A: Dynamic Meteorology and Oceanog- raphy, 62(5):673–697, 2010

    2010

  29. [37]

    Four-dimensional data assimilation: Compari- son of variational and sequential algorithms

    F Rabier, P Courtier, and Martin Ehrendorfer. Four-dimensional data assimilation: Compari- son of variational and sequential algorithms. Quarterly Journal of the Royal Meteorological Society, 118(506):673–713, 1992

    1992

  30. [38]

    Rosemary A Renaut, Iveta Hnˇ etynkov´ a, and J. Mead. Regularization parameter estimation for large-scale tikhonov regularization using a priori information. Computational statistics & data analysis , 54(12):3430–3445, 2010

    2010

  31. [39]

    Inverse problems and data assimi- lation

    Daniel Sanz-Alonso, Andrew M Stuart, and Armeen Taeb. Inverse problems and data assimi- lation. arXiv preprint arXiv:1810.06191 , 2018

    2018 arXiv

  32. [40]

    A generalized representer theorem

    Bernhard Sch¨ olkopf, Ralf Herbrich, and Alex J Smola. A generalized representer theorem. In International conference on computational learning theory , pages 416–426. Springer, 2001

    2001

  33. [41]

    A review of innovation-based methods to jointly estimate model and observation error covariance matrices in ensemble data assimilation

    Pierre Tandeo, Pierre Ailliot, Marc Bocquet, Alberto Carrassi, Takemasa Miyoshi, Manuel Pulido, and Yicun Zhen. A review of innovation-based methods to jointly estimate model and observation error covariance matrices in ensemble data assimilation. Monthly Weather Review, 148(1...

    2020

  34. [42]

    Solutions of ill-posed problems

    Andre Nikolaevich Tikhonov and VIAK Arsenin. Solutions of ill-posed problems. (No Title) , 1977

    1977

  35. [43]

    Accounting for an imperfect model in 4d-var

    Yannick Tr’emolet. Accounting for an imperfect model in 4d-var. Quarterly Journal of the Royal Meteorological Society: A journal of the atmospheric sciences, applied meteorology and physical oceanography, 132(621):2483–2504, 2006

    2006

  36. [44]

    Model-error estimation in 4d-var

    Yannick Tr´ emolet. Model-error estimation in 4d-var. Quarterly Journal of the Royal Meteo- rological Society: A journal of the atmospheric sciences, applied meteorology and physical oceanography, 133(626):1267–1280, 2007

    2007

  37. [45]

    Variational data analysis with control of the forecast bias

    PA Vidard, Andrea Piacentini, and F-X Le Dimet. Variational data analysis with control of the forecast bias. Tellus A: Dynamic Meteorology and Oceanography , 56(3):177–188, 2004

    2004

  38. [46]

    Non-convergence of the l-curve regularization parameter selection method

    Curtis R Vogel. Non-convergence of the l-curve regularization parameter selection method. Inverse problems, 12(4):535, 1996

    1996

  39. [47]

    Spline models for observational data

    Grace Wahba. Spline models for observational data . SIAM, 1990

    1990

  40. [48]

    Representer theorem

    Grace Wahba and Yuedong Wang. Representer theorem. Wiley StatsRef: Statistics Reference Online, pages 1–11, 2019

    2019

  41. [49]

    Inverse methods in hydrogeology: Evolution and recent trends

    Haiyan Zhou, J Jaime G´ omez-Hern´ andez, and Liangping Li. Inverse methods in hydrogeology: Evolution and recent trends. Advances in Water Resources , 63:22–37, 2014

    2014

  42. [50]

    A general weak constraint applicable to operational 4dvar data assimilation systems

    Dusanka Zupanski. A general weak constraint applicable to operational 4dvar data assimilation systems. Monthly Weather Review , 125(9):2274–2292, 1997. 28 S. R. BABYALE, J. MEAD, D. CALHOUN AND P. O. AZIKE Appendix A. Reduced Posterior Penalty Functionals. In the weak- constra...

    1997

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.