WorksMathematics
Parameter Recovery for the Modified Leray-alpha Model via Continuous Data Assimilation
Abstract
In this study, we extend the recursive length-scale recovery method for the simplified Bardina model to the three-dimensional modified Leray-alpha model. The main difficulty is that the unknown parameter also occurs in the nonlinear convection term, so estimating the error in the parameter update requires control of second spatial derivatives of the velocity error. We combine continuous data assimilation with a recursive update of the squared length scale, using observed low Fourier modes and computed high Fourier modes. Under uniform regularity bounds on the reference solution and suitable conditions on the observations and the recovery procedure, we prove geometric convergence of both the parameter and velocity errors at the update times. When the update-window lengths are uniformly bounded, this convergence is exponential in time at the update times.
Introduction
The Navier–Stokes equations describe the motion of viscous incompressible fluids. In turbulent regimes, however, the range of dynamically relevant scales makes direct numerical simulation expensive. Regularized fluid models provide a way to describe the large-scale motion without resolving every small scale. Beginning with Leray’s regularization [23], this approach has led to several families of turbulence models, including the Navier–Stokes-, Leray-, modified Leray-, Clark-, and simplified Bardina models [7, 9, 10, 14, 18, 20, 22]. The parameter specifies a filtering length, and its choice influences the relation between the resolved velocity and the underlying regularized flow. Recovering this parameter from observations is therefore a natural problem in model calibration.
In this paper, we consider the three-dimensional modified Leray- model introduced in [20]. On the periodic box , its equations are
Here is the filtered velocity, is the associated unfiltered velocity, is the pressure, is the viscosity, and is the external force. We impose periodic boundary conditions and assume
The force and viscosity are known, whereas the initial state and the parameter are unknown. Only a finite number of low Fourier modes of are observed. We assume that known constants and satisfy
The theory of determining modes and nodes shows that sufficiently resolved partial observations determine the asymptotic behavior of the full solution under model-specific hypotheses [16, 17, 21]. Continuous data assimilation incorporates these observations into an evolution equation through a feedback term. We use the framework introduced by Azouani, Olson, and Titi [6], whose applications to turbulence models include [1, 2]. When some parameters are also unknown, the feedback system can be coupled with a parameter-recovery procedure. Such methods have been studied for viscosity recovery and sensitivity analysis [8, 11, 12, 24], identification of forcing [15], and simultaneous recovery of multiple parameters [13, 28]. Related developments include on-the-fly model discovery and relaxation-based estimation [25, 27], viscosity recovery for the steady Navier–Stokes equations [29], recovery for the primitive equations with a magnetic field [32], and data-driven operator inference [31].
For -regularization models, the effect of a mismatch in the filtering length has been investigated in [3, 4]. A recursive recovery procedure for the simplified Bardina model was subsequently developed in [5]. We adapt that procedure to recover and approximate the reference solution of the modified Leray- equation. Unlike the simplified Bardina model, whose convection term is , (1) contains . When the system is written in terms of , the unknown parameter therefore occurs not only in the time-derivative and viscous terms, but also in the convection term. To construct the approximate solution, let denote the orthogonal projection onto the spatial Fourier modes indexed by with . For a fixed approximate parameter , we use the same feedback system as in [4] with Fourier-mode observations:
The fields and satisfy the same periodic and zero-mean conditions as and ; is the corresponding pressure, is the identity, and is the nudging strength. The initial value of is arbitrary in the energy space defined below.
The parameter update depends on the unknown velocity, whereas only its low Fourier modes are observed. Following [5], we replace the unknown velocity with a field formed from the observed low Fourier modes and the high Fourier modes of the approximate solution. For the modified Leray- model, this replacement introduces an error involving second spatial derivatives, which cannot be controlled by a basic velocity error estimate alone. We combine a synchronization estimate for the velocity error with a time-integrated higher-order dissipation estimate to control the second spatial derivatives appearing in the parameter update. These estimates are then used to construct a recursive parameter update. Before each update window, we keep the parameter fixed for a relaxation period to allow the assimilated solution to approach the reference solution. Under suitable regularity assumptions on the reference solution and sufficient observational signal relative to the approximation error, we prove that both the estimated parameter and the approximate velocity converge geometrically to their corresponding true values at the update times. Furthermore, if the update intervals are uniformly bounded in length, this convergence is exponential in time at the update times.
The rest of the paper is organized as follows. § 2 presents the functional setting. § 3 introduces the parameter recovery procedure. § 4 establishes the estimates for the reference and data-assimilated systems. § 5 presents and proves the main theorem. § 6 concludes the paper.
Functional analytic framework
For , we denote by the usual Lebesgue spaces of three-dimensional vector fields. For each , we define the Hilbert space
where are the Fourier coefficients of , and the series is understood in the sense of distributions. We endow this space with the norm
So . We write
We denote by the Helmholtz–Leray orthogonal projection, given by
The same formula defines its extension to periodic distributions. For each , the Stokes operator is given by
In particular,
For the inner product and norms, we use the notation
Thus round parentheses denote the inner product, whereas denotes the appropriate Sobolev duality pairing, which extends this inner product. The zero-mean condition gives the Poincaré inequalities
We recall the following particular cases of the Gagliardo–Nirenberg inequalities [19, 30]:
These inequalities hold for every . We shall also use the Sobolev embeddings and .
For each , we denote the Helmholtz operator by . Its inverse is the Fourier multiplier operator with multiplier , and the following estimates hold:
The observations are represented by the low-mode projection introduced in the Introduction. For , this projection and its complement are given by
By orthogonality in each space , these projections satisfy the following relations:
We define the bilinear operator by
It extends continuously to , with
For , and also for and , we have
Throughout the paper, denotes a positive constant that may change from one occurrence to another, and may depend on the domain.
Parameter recovery algorithm
Throughout the analysis, we assume that and write . Applying to (1) gives the functional formulation
Similarly, for fixed , the assimilation system (2) becomes
Here the feedback field is divergence free.
For sufficiently regular , we introduce two operators that separate the terms with and without the squared filtering length:
With this notation, systems (8) and (9) can be written in the form
Estimate (6) yields
Fixing , a regular solution satisfies
on a bounded interval , where and .
For and , we define the corresponding energy by
For , the reference system admits a unique global regular solution: see [20] for autonomous forcing and [4] for the time-dependent forcing considered here. Taking the duality pairing of (8) with and using (7) and Young’s inequality together with (3), we obtain
Applying Gronwall’s inequality, we conclude that, for ,
Assume that a known, parameter-independent bound is available. Then (11) gives
For the parameter recovery estimates, we impose the following additional regularity assumptions on the reference solution.
Assumption 1. There exist constants , depending only on the reference solution, such that
(i)
(ii)
For every , condition (ii) and the embedding give a continuous representative of , which we use throughout. Thus the supremum in (13) agrees with the essential supremum.
As in [5], we keep and fixed throughout the recovery iteration. Let denote the successive parameter-update times, with held fixed on . On each interval , we use the approximate parameter and solve
The initial state is arbitrary.
Let , so that . Subtracting the two equations in (10), we obtain
Following [5], we approximate the reference solution by combining the observed low modes of with the computed high modes of :
Define
Using the same replacement in the projected error equation, we define
Since , (17) and (18) can be computed from , , and . Moreover,
We use exact observations in continuous time.
Definition 1. For each , we choose an intermediate time and denote the corresponding update interval and its length by
The interval is used for relaxation, and is used to compute the parameter update. We require the update signal to satisfy
Fix . An update interval is accepted if .
On each update interval, we first form a candidate value for the squared parameter:
then project this value onto , obtaining
We propose the following parameter updating algorithm.
Algorithm 1.
Initialization. Given the observations on , where , choose and . Let be the fixed relaxation time before each update window. Keep , , and the threshold fixed, and choose a finite list of trial lengths satisfying
Require the reference estimates from onward; set , .
Iteration. Given :
If , stop. Otherwise set .
Try the candidate window lengths in increasing order, restricted to . Set
On , solve (14) with the same for every trial. Compute (16)–(19); accept the first trial with .
If all eligible trials fail, stop without changing or .
Output. Return , or if no update was accepted.
Estimates for the reference and assimilation systems
For each , we introduce the quantities
where is the shortest trial length and is a sufficiently large constant depending only on the domain.
Lemma 1. Under Assumption 1, for , with , we have and
Moreover, for every ,
Proof. Applying Assumption 1 on successive unit intervals starting at , we obtain
For the nonlinear term, Hölder’s inequality and the Sobolev embeddings give, for a.e. ,
Since , we have
Integrating over and applying (25), we obtain
For , we have . Hence
where the constant in (22) is chosen sufficiently large. This proves (23).
For , set for . Each nonempty is contained in a unit interval starting at or after . Applying Assumption 1 on these intervals, we obtain
We have the following well-posedness result.
Proposition 1. Let , , , , and let be the reference solution. With fixed, (14) has a unique global solution satisfying, for every ,
Proof. Fix and , and write . We use the Galerkin construction in [4 Theorem 2]. For , let be the orthogonal projections onto the spaces in an increasing sequence of finite-dimensional Stokes eigenspaces such that and strongly in and . Let denote the corresponding Galerkin approximation, with . Since is an orthogonal projection commuting with ,
Testing the Galerkin equation with , using (7) and Young’s inequality, gives
Thus, for every , the approximations exist on and are uniformly bounded in . Applying to their equations and using (4) and (6), we obtain
Thus is bounded in . By the Aubin–Lions theorem, after extracting a subsequence,
The limit solves (14) with the given initial data. The embedding yields the continuity.
For uniqueness, let for two solutions with identical data. Applying the interpolation estimates and Young’s inequality, we have
Testing the difference equation with yields
Gronwall’s inequality implies . Since is arbitrary, the solution is global.
For a sufficiently large domain-dependent constant , we define
We require the observation resolution and nudging strength to satisfy
Using the quantities in (22), we also fix
The following estimate refines [4] by retaining the higher-order dissipation and keeping the constants uniform over all assimilation intervals. This improvement is useful in the parameter update.
Proposition 2. Let be a solution of (8) satisfying Assumption 1 and on . Let be a solution of (14) with fixed , and set . Under (27), with constants (26) and (28), we have
for a.e. . Moreover, for and with ,
The constants , , , and are independent of and once (27) holds.
Proof. Subtracting the reference equation from (9), we obtain
Here we have used the identity . Pairing with and using (7), we have
Using the interpolation inequalities, Young’s inequality, and , , and , we have
The Poincaré and Young inequalities give
Choosing large enough, we then obtain
Using the orthogonality of the low and high Fourier modes together with (5), we have
By (27), the low-mode contribution is absorbed by the feedback term, and the high-mode contribution is absorbed by part of the viscous dissipation. Consequently,
From (3), we have
This proves (29). Multiplying by the exponential integrating factor and applying (24), we obtain the first bound in (30). Integrating the differential inequality over and applying (23), we obtain
Since , we obtain the second bound in (30). □
Remark 1. For the synchronization estimates above, once satisfies (27), any fixed is allowed. Unlike condition (20) in [5], these estimates impose no upper bound on .
Synchronization estimates
We use the synchronization and dissipation estimates to control the error introduced by replacing the unobserved part of the reference solution. The following identity relates this error to the parameter update.
Lemma 2. Under Assumption 1, let be given by Proposition 1. For a.e. , there is a residual satisfying
where depends on , but is independent of .
Proof. Applying the low-mode projection to (15) and using the definitions in (17) and (18), we obtain the identity in (31), with residual
To estimate this residual, we write so that . For with , we have , where depends only on and . By bilinearity and integration by parts, using , we obtain
Combining these estimates and taking the supremum over unit test fields , we obtain
Finally, we use (13), the embedding , the inequality , and the identity . These bounds yield (31), with a constant uniform in .
For , substituting (31) into (20) gives
Applying the Cauchy–Schwarz inequality, we obtain
With as in Lemma 2, we fix the following constants before choosing the relaxation length :
We then define
The constants , , , are independent of and . The remaining assumptions require successful update windows and a sufficiently strong signal relative to the bound on .
Assumption 2. (H1) , with . The reference solution satisfies Assumption 1 and the initial energy bound used in (12).
(H2) and remain fixed and satisfy (27), with given by (26).
(H3) Fix the trial lengths and threshold independently of . There exists such that, for every , Algorithm 1, run without a terminal time, accepts a trial at every iteration. Hence and
(H4) With from (33),
Remark 2. Conditions (H3)–(H4) are additional assumptions: (H3) requires a successful update window at every iteration, while (H4) requires . Neither follows from Assumption 1. Unlike [5], our bound on has no explicit decay in ; the constant contains , which may grow with . Thus increasing does not necessarily improve (H4).
Theorem 1. Under Assumption 2, generate by (14), (20) and (21), from , , with . There exists , which may depend on , such that every fixed admits and satisfying
Moreover, there exist constants such that
where . In particular, and in as .
Proof. Set , , and
Since , the triangle inequality and (12) give
Moreover, (21) ensures , so
Applying the first estimate in (30) at the end of the assimilation interval, we obtain
We set and . Using and , we conclude that
Choose sufficiently large that for every . Iterating the recurrence and using gives
We increase , if necessary, so that . The first estimate in (30) then gives, on every update interval ,
This bound makes the coefficient in the residual estimate uniform over the update intervals. Denoting and using , we combine (32), Lemma 2, and (34), and apply Minkowski's inequality to obtain
Applying (30) over the update interval, since and , we have
Substituting these bounds into the preceding estimate for and using (33), we obtain
Combining this estimate with , we obtain the recurrence
Let denote the spectral radius of . As ,
The limiting matrix has eigenvalues and . Hence, by continuity of the spectral radius and (H4),
Thus for all sufficiently large . Choose so that this holds for every . By (H3) and the nonnegativity of , induction yields
For , the standard matrix-power estimate gives, for some ,
This proves the geometric bounds. Moreover, since , the preceding estimate gives and . By their definitions and , we obtain
Finally, since , we have
This gives exponential decay in at the update times. For a fixed geometric factor, increasing decreases this rate, so a longer relaxation interval does not by itself guarantee faster convergence.
Conclusions
In this study, we have constructed a recursive algorithm for recovering the length-scale parameter in the modified Leray- model. The synchronization estimate controls the velocity error, while the integrated dissipation estimate supplies the higher-order control needed for the analysis. With fixed , , a sufficiently large fixed relaxation length, uniform reference regularity, successful update windows, and the condition , the parameter and velocity errors converge geometrically at the update times. If the update windows have a uniform upper bound on their lengths, the errors also decay exponentially in time at the update times.
The convergence result remains conditional on the signal hypotheses. Deriving these conditions from the reference trajectory remains open. Martinez, Strikwerda, and Wan [26] study excitation and viscosity identifiability for the two-dimensional Navier–Stokes equations using spectral observations and a nudging-based sensitivity analysis. A natural direction for future work is to investigate whether a similar analysis can provide verifiable sufficient conditions for (H3)–(H4). Moreover, the present analysis assumes exact observations available continuously in time. Extensions to discrete-time observations and noisy data are left for future work.
Acknowledgments
Jing Tian’s work was partially supported by the NSF LEAPS-MPS Grant #2316894.
References
- [1]D. A. F. Albanez, H. J. Nussenzveig Lopes, and E. S. Titi, Continuous data assimilation for the three-dimensional Navier–Stokes-α model, Asymptot. Anal. 97 (2016), 139–164.
- [2]D. A. F. Albanez and M. J. Benvenutti, Continuous data assimilation algorithm for simplified Bardina model, Evol. Equ. Control Theory 7 (2018), 33–52.DOI
- [3]D. A. F. Albanez, M. J. Benvenutti, S. Little, and J. Tian, Parameter analysis in continuous data assimilation for various turbulence models, Commun. Nonlinear Sci. Numer. Simul. 151 (2025), 109073.
- [4]D. A. F. Albanez, M. J. Benvenutti, S. Little, and J. Tian, Parameter error analysis for the 3D modified Leray-α model: analytical and numerical approaches, Comput. Appl. Math. 44 (2025), 200.arxiv.org/abs/2411.16324
- [5]D. A. F. Albanez, M. J. Benvenutti, and J. Tian, Parameter recovery for the simplified Bardina model via continuous data assimilation, Ann. Phys. (N.Y.) 493 (2026), 170612.DOI
- [6]A. Azouani, E. Olson, and E. S. Titi, Continuous data assimilation using general interpolant observables, J. Nonlinear Sci. 24 (2014), 277–304.DOI
- [7]J. Bardina, J. H. Ferziger, and W. C. Reynolds, Improved subgrid-scale models for large-eddy simulation, AIAA Paper 80-1357, 1980.DOI
- [8]A. Biswas and J. Hudson, Determining the viscosity of the Navier–Stokes equations from observations of finitely many modes, Inverse Problems 39 (2023), 125012.
- [9]C. Cao, D. D. Holm, and E. S. Titi, On the Clark-α model of turbulence: global regularity and long-time dynamics, J. Turbul. 6 (2005), N20.DOI
- [10]Y. Cao, E. M. Lunasin, and E. S. Titi, Global well-posedness of the three-dimensional viscous and inviscid simplified Bardina turbulence models, Commun. Math. Sci. 4 (2006), 823–848.DOI
- [11]E. Carlson, J. Hudson, and A. Larios, Parameter recovery for the 2 dimensional Navier–Stokes equations via continuous data assimilation, SIAM J. Sci. Comput. 42 (2020), A250–A270.DOI
- [12]E. Carlson and A. Larios, Sensitivity analysis for the 2D Navier–Stokes equations with applications to continuous data assimilation, J. Nonlinear Sci. 31 (2021), 84.arxiv.org/abs/2007.01860
- [13]E. Carlson, J. Hudson, A. Larios, V. R. Martinez, E. Ng, and J. P. Whitehead, Dynamically learning the parameters of a chaotic system using partial observations, Discrete Contin. Dyn. Syst. 42 (2022), 3809–3839.arxiv.org/abs/2108.08354
- [14]A. Cheskidov, D. D. Holm, E. Olson, and E. S. Titi, On a Leray-α model of turbulence, Proc. R. Soc. A 461 (2005), 629–649.DOI
- [15]A. Farhat, A. Larios, V. R. Martinez, and J. P. Whitehead, Identifying the body force from partial observations of a two-dimensional incompressible velocity field, Phys. Rev. Fluids 9 (2024), 054602.DOI
- [16]C. Foias and G. Prodi, Sur le comportement global des solutions non-stationnaires des équations de Navier–Stokes en dimension 2, Rend. Sem. Mat. Univ. Padova 39 (1967), 1–34.
- [17]C. Foias and R. Temam, Determination of the solutions of the Navier–Stokes equations by a set of nodal values, Math. Comp. 43 (1984), 117–133.DOI
- [18]C. Foias, D. D. Holm, and E. S. Titi, The Navier–Stokes-α model of fluid turbulence, Physica D 152–153 (2001), 505–519.
- [19]C. Foias, O. Manley, R. Rosa, and R. Temam, Navier–Stokes Equations and Turbulence, Cambridge University Press, Cambridge, 2001.DOI
- [20]A. A. Ilyin, E. M. Lunasin, and E. S. Titi, A modified-Leray-α subgrid scale model of turbulence, Nonlinearity 19 (2006), 879–897.DOI
- [21]D. A. Jones and E. S. Titi, Upper bounds on the number of determining modes, nodes, and volume elements for the Navier–Stokes equations, Indiana Univ. Math. J. 42 (1993), 875–887.
- [22]W. Layton and R. Lewandowski, On a well-posed turbulence model, Discrete Contin. Dyn. Syst. Ser. B 6 (2006), 111–128.DOI
- [23]J. Leray, Sur le mouvement d’un liquide visqueux emplissant l’espace, Acta Math. 63 (1934), 193–248.DOI
- [24]V. R. Martinez, Convergence analysis of a viscosity parameter recovery algorithm for the 2D Navier–Stokes equations, Nonlinearity 35 (2022), 2241–2287.
- [25]V. R. Martinez, J. Murri, and J. P. Whitehead, Relaxation-based schemes for on-the-fly parameter estimation in dissipative dynamical systems, Inverse Problems 41 (2025), 055001.arxiv.org/abs/2408.14296
- [26]V. R. Martinez, S. Strikwerda, and X. Wan, Excitation and identifiability in the 2D Navier–Stokes equations, arXiv preprint arXiv:2609.31230 (2026).arxiv.org/abs/2609.31230
- [27]J. Newey, J. P. Whitehead, and E. Carlson, Model discovery on the fly using continuous data assimilation, J. Comput. Phys. 537 (2025), 114121.arxiv.org/abs/2411.13561
- [28]B. Pachev, J. P. Whitehead, and S. A. McQuarrie, Concurrent multiparameter learning demonstrated on the Kuramoto–Sivashinsky equation, SIAM J. Sci. Comput. 44 (2022), A2974–A2990.DOI
- [29]L. G. Rebholz, J. Reyes, and J. P. Whitehead, Continuous data assimilation in steady Navier–Stokes equations with unknown viscosity: robust and efficient solvers and fast parameter recovery, arXiv preprint arXiv:2609.02862 (2026).arxiv.org/abs/2609.02862
- [30]R. Temam, Navier–Stokes Equations and Nonlinear Functional Analysis, 2nd ed., SIAM, Philadelphia, 1995.
- [31]Y. Wang and L. Yan, Data-driven operator inference for parameter estimation in nonlinear partial differential equations, J. Comput. Phys. 544 (2026), 114442.DOI
- [32]Y. Zhao, W. Liu, J. Zhang, G. Lv, and Y. Wang, Parameter recovery for the three-dimensional primitive equations with magnetic field via continuous data assimilation, ZAMM 106 (2026), e70534.DOI