pith. sign in

arxiv: 2604.21821 · v1 · submitted 2026-04-23 · 🧮 math.NA · cs.NA· math.AP

Direct Problem for Gas Diffusion in Polar Firn with Variable Coefficients

Sophie Moufawad , Nabil Nassif , Faouzi Triki This is my paper

Pith reviewed 2026-05-09 20:42 UTC · model grok-4.3

classification 🧮 math.NA cs.NAmath.AP MSC 35K6565N30
keywords degenerate parabolic PDEweighted Sobolev spacesgas diffusionpolar firnfinite element methodexistence and uniquenessLions theorem
0
0 comments X

The pith

The degenerate parabolic PDE for gas diffusion in polar firn is well-posed with variable coefficients D(z) and f(z).

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

The paper proves existence and uniqueness for a parabolic model of gas trapping in firn that degenerates when both the diffusion coefficient and volume fraction are allowed to be non-constant functions that may reach zero at the bottom boundary. It introduces weighted Sobolev spaces to restore coercivity and continuity of the bilinear form, then invokes Lions' theorem on the resulting semi-variational formulation. The same spaces support a practical discretization via linear finite elements in space and implicit Euler in time, for which sufficient conditions again guarantee a unique discrete solution. A reader should care because earlier analyses fixed most coefficients and therefore could not represent realistic depth-dependent physical properties in deep ice.

Core claim

By defining appropriate weighted Sobolev spaces to handle the possible vanishing of D(z) and f(z) at the bottom of the firn, Lions' theorem yields existence and uniqueness for the semi-variational formulation of the degenerate parabolic PDE. A P1 finite element Galerkin discretization in space combined with an implicit Euler scheme in time produces a fully discrete system whose solution exists and is unique under sufficient conditions on the coefficients.

What carries the argument

Weighted Sobolev spaces that restore coercivity and continuity of the bilinear form despite degeneracy at one boundary, together with Lions' theorem applied to the semi-variational formulation.

If this is right

  • Gas concentration profiles remain uniquely determined by the model even when both coefficients vary with depth.
  • Standard P1 finite-element and backward-Euler time-stepping schemes can be applied without loss of well-posedness.
  • Numerical simulation of gas trapping can incorporate measured or reconstructed depth-dependent diffusion and porosity.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same weighted-space construction may extend to other parabolic models that degenerate when a coefficient vanishes at a boundary, such as certain porous-medium or filtration equations.
  • Because the analysis supplies a well-posed direct problem, it supplies the foundation needed to pose an inverse problem for recovering the unknown functions D(z) and f(z) from gas-concentration data.
  • The sufficient conditions on the coefficients for the discrete system can be verified numerically on standard firn profiles to confirm practical applicability.

Load-bearing premise

The variable coefficients D(z) and f(z) must obey the positivity and integrability conditions that make the weighted Sobolev spaces produce a coercive and continuous bilinear form.

What would settle it

An explicit choice of D(z) and f(z) that vanishes at the bottom boundary yet violates the weighted integrability requirements, for which no solution exists in the proposed spaces.

read the original abstract

We consider the mathematical model of gas trapping in deep polar ice (firns), which consists of a parabolic partial differential equation, that can degenerate at one boundary extreme. In [1], we considered all the coefficients to be constants, except the diffusion coefficient D(z) that is to be reconstructed. In this paper, we assume both the diffusion coefficient D(z) and the volume fraction f(z) are functions. The difficulty in this problem, both theoretically and computationally, arises from the fact that D(z) and f(z) may be zero at bottom of the firn. To handle such degeneracy, we defined appropriate weighted Sobolev spaces and used Lion's theorem to prove existence and uniqueness of the semi-variational formulation of the Firn PDE. A full discrete system is obtained through a P1 Finite element Galerkin procedure in space and an Euler-Implicit scheme in time. Sufficient conditions for the existence and uniqueness of the solution for the discrete system are obtained.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. The paper analyzes the direct problem for gas diffusion in polar firn modeled by a degenerate parabolic PDE with variable coefficients D(z) and f(z) that may vanish at the bottom boundary. It introduces weighted Sobolev spaces to address the degeneracy, applies Lions' theorem to establish existence and uniqueness for the semi-variational formulation, discretizes via P1 finite-element Galerkin in space and implicit Euler in time, and derives sufficient conditions for existence and uniqueness of the discrete solution.

Significance. If the weighted-space coercivity estimates and discrete conditions hold for physically relevant D(z) and f(z), the work provides a rigorous extension of constant-coefficient results to variable-coefficient firn models, supporting more accurate numerical simulations of gas trapping for paleoclimate reconstruction. The combination of functional-analysis existence proofs with a practical discretization scheme is a constructive contribution in numerical analysis for degenerate diffusion problems.

major comments (3)
  1. [Abstract] Abstract and introduction: the central claim that 'appropriate weighted Sobolev spaces' yield a coercive and continuous bilinear form (enabling Lions' theorem) is not supported by any explicit minimal assumptions on the decay or integrability of D(z) and f(z) near the degenerate boundary (e.g., D(z) ≥ c z^α or bounds on 1/(D f)). Without these, coercivity may fail for admissible physical profiles, rendering the existence/uniqueness result unverifiable.
  2. [Semi-variational formulation] Section on the semi-variational formulation: the application of Lions' theorem is asserted but no coercivity constant estimate or continuity constant is derived in terms of the variable coefficients; the proof sketch therefore does not confirm that the chosen weights compensate for possible faster vanishing of D(z) and f(z) than the weights allow.
  3. [Discrete system] Section on the discrete system: the 'sufficient conditions' for existence/uniqueness of the P1 Galerkin + implicit Euler scheme are stated to be obtained, yet they inherit the same unspecified requirements on D(z) and f(z); without explicit verification that the discrete bilinear form remains coercive under the same weights, the discrete claim is not load-bearing.
minor comments (2)
  1. [Introduction] Notation for the weighted spaces (e.g., the precise definition of the weight functions) should be introduced earlier and used consistently when stating the variational formulation.
  2. [Time discretization] The transition from the continuous to the discrete problem would benefit from a brief remark on how the implicit Euler time step interacts with the degeneracy at each time level.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the careful and constructive review. The comments correctly identify areas where additional explicit assumptions and estimates would strengthen the manuscript. We address each major comment below and will revise accordingly to make the results more verifiable for physical coefficient profiles.

read point-by-point responses

  1. Referee: [Abstract] Abstract and introduction: the central claim that 'appropriate weighted Sobolev spaces' yield a coercive and continuous bilinear form (enabling Lions' theorem) is not supported by any explicit minimal assumptions on the decay or integrability of D(z) and f(z) near the degenerate boundary (e.g., D(z) ≥ c z^α or bounds on 1/(D f)). Without these, coercivity may fail for admissible physical profiles, rendering the existence/uniqueness result unverifiable.

    Authors: We agree that the manuscript would benefit from explicit minimal assumptions on D(z) and f(z) to ensure the weighted spaces yield coercivity. The weighted Sobolev spaces are constructed to account for the possible vanishing of the coefficients at the boundary, but the precise integrability or decay conditions guaranteeing that the weights compensate the degeneracy were not listed. In the revised manuscript we will add these assumptions (for example, D(z) ≥ c z^α with α < 1 together with suitable bounds ensuring 1/(D f) belongs to the appropriate L^p space) in both the abstract/introduction and the formulation section. revision: yes

  2. Referee: [Semi-variational formulation] Section on the semi-variational formulation: the application of Lions' theorem is asserted but no coercivity constant estimate or continuity constant is derived in terms of the variable coefficients; the proof sketch therefore does not confirm that the chosen weights compensate for possible faster vanishing of D(z) and f(z) than the weights allow.

    Authors: The referee is right that the proof sketch does not supply explicit coercivity and continuity constants expressed in terms of D(z) and f(z). While the weighted formulation is designed so that the bilinear form satisfies the hypotheses of Lions' theorem, the constants were not derived. We will expand the semi-variational section with the missing estimates, showing that the coercivity constant remains positive and the continuity constant is finite under the added assumptions on the coefficients. revision: yes

  3. Referee: [Discrete system] Section on the discrete system: the 'sufficient conditions' for existence/uniqueness of the P1 Galerkin + implicit Euler scheme are stated to be obtained, yet they inherit the same unspecified requirements on D(z) and f(z); without explicit verification that the discrete bilinear form remains coercive under the same weights, the discrete claim is not load-bearing.

    Authors: We accept that the discrete existence and uniqueness result relies on the same coercivity properties as the continuous problem and therefore requires the same explicit conditions on D(z) and f(z). The manuscript states sufficient conditions but does not verify that the discrete bilinear form remains coercive in the weighted discrete spaces. In the revision we will add a discrete coercivity estimate that mirrors the continuous one, confirming that the P1 Galerkin plus implicit Euler scheme inherits the required properties under the stated assumptions. revision: yes

Circularity Check

0 steps flagged

No significant circularity; standard application of Lions' theorem to newly defined weighted spaces

full rationale

The derivation defines appropriate weighted Sobolev spaces to accommodate possible degeneracy of D(z) and f(z) at the boundary, then invokes Lions' theorem to obtain existence and uniqueness for the semi-variational formulation. This is an independent mathematical construction relying on external functional-analysis results rather than any self-referential definition, fitted parameter renamed as prediction, or load-bearing self-citation. The reference to prior work [1] merely contextualizes the constant-coefficient case and does not supply the coercivity or continuity arguments used here. The discrete Galerkin-Euler scheme inherits the same non-circular structure. No step reduces by construction to its own inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The paper rests on standard functional-analysis results and the authors' prior model; it introduces no fitted parameters or new physical entities.

axioms (1)
  • standard math Lions' theorem for existence and uniqueness of variational problems
    Invoked to establish existence and uniqueness of the semi-variational formulation in the weighted spaces.

pith-pipeline@v0.9.0 · 5474 in / 1511 out tokens · 46492 ms · 2026-05-09T20:42:35.551932+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

23 extracted references · 23 canonical work pages

  1. [1]

    Hardy-Type Inequalities , series =

    Opic, Bohum. Hardy-Type Inequalities , series =

    work page

  2. [2]

    Folland , title =

    Gerald B. Folland , title =. 1999 , publisher =

    work page 1999

  3. [3]

    2011 , publisher =

    Haim Brezis , title =. 2011 , publisher =

    work page 2011

  4. [4]

    Moufawad, S. M. and Nassif, N. R. and Triki, F. , title =. Communications in Analysis and Computation , volume =. 2024 , doi =

    work page 2024

  5. [5]

    and Headly, M

    Ahn, J. and Headly, M. and Wahlen, M. and Brook, E. J. and Mayewski, P. A. and Taylor, K. C. , title =. Journal of Glaciology , volume =

    work page

  6. [6]

    and Triki, F

    Ammari, H. and Triki, F. , title =. Communications in Partial Differential Equations , volume =

    work page

  7. [7]

    and Triki, F

    Ammari, H. and Triki, F. and Tsou, C.-H. , title =. European Journal of Applied Mathematics , volume =

    work page

  8. [8]

    and Triki, F

    Bonnetier, E. and Triki, F. and Tsou, C.-H. , title =. Journal of Mathematical Analysis and Applications , volume =

    work page

  9. [9]

    2010 , doi =

    Brezis, Haim , title =. 2010 , doi =

    work page 2010

  10. [10]

    2023 , note =

    D'Errico, John , title =. 2023 , note =

    work page 2023

  11. [11]

    Dunlavy, D. M. and Kolda, T. G. and Acar, E. , title =

    work page

  12. [12]

    Hager, W. W. and Zhang, H. , title =. Pacific Journal of Optimization , volume =

    work page

  13. [13]

    2006 , volume =

    Isakov, Victor , title =. 2006 , volume =

    work page 2006

  14. [14]

    and Seo, J

    Jang, J. and Seo, J. K. , title =. Physiological Measurement , volume =

    work page

  15. [15]

    Laube, J. C. and Newland, M. J. and Hogan, C. and Brenninkmeijer, C. A. M. and Fraser, P. J. and Martinerie, P. and Oram, D. E. and Reeves, C. E. and R. Newly detected ozone-depleting substances in the atmosphere , journal =. 2014 , doi =

    work page 2014

  16. [16]

    Moufawad, S. M. and Nassif, N. R. and Triki, F. , title =. 2023 , note =

    work page 2023

  17. [17]

    2013 , doi =

    Eemian interglacial reconstructed from a Greenland folded ice core , journal =. 2013 , doi =

    work page 2013

  18. [18]

    Nelder, J. A. and Mead, R. , title =. The Computer Journal , volume =

    work page

  19. [19]

    , title =

    Triki, F. , title =. Journal de Mathématiques Pures et Appliquées , volume =

    work page

  20. [20]

    and Martinerie, P

    Witrant, E. and Martinerie, P. and Hogan, C. and others , title =. Atmospheric Chemistry and Physics , volume =. 2012 , doi =

    work page 2012

  21. [21]

    and Martinerie, P

    Witrant, E. and Martinerie, P. , title =. Proceedings of the 18th Mediterranean Conference on Control and Automation , pages =

    work page

  22. [22]

    and Martinerie, P

    Witrant, E. and Martinerie, P. , title =. Proceedings of the 5th IFAC Symposium on System Structure and Control , volume =

    work page

  23. [23]

    Yeung, L. Y. and Murray, L. T. and Martinerie, P. and Witrant, E. and Hu, H. and Banerjee, A. and Orsi, A. and Chappellaz, J. , title =. Nature , volume =. 2019 , doi =

    work page 2019