A fourth-order compact solver for fractional-in-time fourth-order diffusion equations

Bingquan Ji; Hong-lin Liao; Jialing Zhong; Luming Zhang

arxiv: 1907.01708 · v1 · pith:TYJQWID3new · submitted 2019-07-03 · 🧮 math.NA · cs.NA

A fourth-order compact solver for fractional-in-time fourth-order diffusion equations

Jialing Zhong , Hong-lin Liao , Bingquan Ji , Luming Zhang This is my paper

Pith reviewed 2026-05-25 10:27 UTC · model grok-4.3

classification 🧮 math.NA cs.NA

keywords compact finite differencefourth-order subdiffusionCaputo derivativeL1 schemeirregular time meshstability and convergencefractional Gronwall inequality

0 comments

The pith

Reducing a fourth-order subdiffusion equation to a coupled second-order system enables a fourth-order compact spatial scheme combined with L1 time stepping on irregular meshes.

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

The paper develops a numerical method that achieves fourth-order spatial accuracy for fourth-order subdiffusion equations under Dirichlet boundary conditions. The approach begins by rewriting the fourth-order spatial operator as a coupled pair of second-order equations, then discretizes space with a compact averaged operator that uses only two grid points near the boundary. Time is handled by the L1 approximation to the Caputo derivative on a graded mesh that clusters points near t equals zero to capture the initial singularity. Stability follows from a complementary discrete convolution kernel and a discrete fractional Gronwall inequality applied to an error structure. Readers should care because these equations appear in models of viscoelasticity and anomalous diffusion where both high spatial order and proper treatment of the weak singularity improve practical accuracy without excessive computational cost.

Core claim

By first reducing the fourth-order subdiffusion equation with Dirichlet boundary conditions to a coupled system of second-order equations and then applying an averaged compact operator for the spatial discretization together with the L1 scheme on irregular time grids, the resulting method is shown to be stable and convergent, with the proof relying on a complementary discrete convolution kernel, a discrete fractional Gronwall inequality, and an error convolution structure.

What carries the argument

The averaged compact operator obtained after reduction to a second-order system, together with the L1 formula on irregular time meshes for the Caputo derivative.

If this is right

The spatial discretization remains compact, involving only two neighboring grid points for the boundary conditions.
The method achieves the designed fourth-order spatial accuracy.
The graded time mesh resolves the initial singularity while maintaining overall convergence.
Stability holds via the discrete fractional Gronwall inequality.
Numerical experiments confirm the theoretical orders of accuracy and efficiency.

Where Pith is reading between the lines

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

Similar reduction and compact averaging might be used for other spatial orders or different boundary conditions if the preservation properties hold.
The convolution-kernel technique for stability could apply to variable-coefficient or nonlinear fractional problems.
Combining graded meshes with high-order compact schemes may reduce computational cost in long-time simulations of anomalous diffusion.

Load-bearing premise

The fourth-order problem can be reduced to a coupled system of second-order equations while preserving the boundary conditions and regularity properties needed for the compact averaged operator and the subsequent error analysis to remain valid.

What would settle it

Running the scheme on a uniform time mesh and observing that the temporal convergence rate does not improve as expected when the initial singularity is present, or finding that the reduction step violates the boundary conditions and causes the error analysis to fail.

read the original abstract

A fourth-order compact scheme is proposed for a fourth-order subdiffusion equation with the first Dirichlet boundary conditions. The fourth-order problem is firstly reduced into a couple of spatially second-order system and we use an averaged operator to construct a fourth-order spatial approximation. This averaged operator is compact since it involves only two grid points for the derivative boundary conditions. The L1 formula on irregular mesh is considered for the Caputo fractional derivative, so we can resolve the initial singularity of solution by putting more grid points near the initial time. The stability and convergence are established by using three theoretical tools: a complementary discrete convolution kernel, a discrete fractional Gronwall inequality and an error convolution structure. Some numerical experiments are reported to demonstrate the accuracy and efficiency of our method.

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.

Desk Editor's Note private letter to a colleague

The paper gives a concrete fourth-order compact scheme for fourth-order subdiffusion by reducing to a coupled second-order system and pairing it with irregular-mesh L1 time discretization, but the reduction step needs explicit verification.

read the letter

The new piece is the specific combination: reduce the fourth-order subdiffusion equation to two coupled second-order problems, apply an averaged compact spatial operator that stays compact with the derivative boundary conditions, and discretize the Caputo derivative with L1 on an irregular mesh to handle the initial singularity. They then prove stability and O(τ^{2-α} + h^4) convergence using a complementary discrete convolution kernel, discrete fractional Gronwall inequality, and error convolution structure. Numerical experiments are included to check the rates. That combination for this exact problem class is not already in the cited literature, so the technical extension is real. The approach is sensible for applied modeling that needs higher spatial accuracy without losing the ability to cluster points near t=0. The reduction to second-order equations is the part that carries the most weight. It must preserve the original Dirichlet conditions and the solution regularity so that the compact operator and the subsequent error analysis apply without extra consistency errors. The abstract states this is the first step but does not show the auxiliary variables or how the boundary conditions are transferred. If that step introduces mismatches, the three theoretical tools would not automatically cover it. The rest of the analysis rests on standard external results, which is fine when the setup is clean. This paper is for numerical analysts working on high-order methods for fractional PDEs. A reader in that area gets a usable scheme and some analysis to build on. It is worth sending to peer review so the reduction and any hidden regularity assumptions can be checked in detail.

Referee Report

2 major / 1 minor

Summary. The manuscript proposes a fourth-order compact finite difference scheme for a fourth-order subdiffusion equation with Dirichlet boundary conditions. The fourth-order problem is reduced to a coupled system of second-order equations; an averaged compact operator is used for O(h^4) spatial accuracy (involving only two grid points for the derivative boundary conditions); the L1 formula on an irregular time mesh approximates the Caputo derivative to resolve the initial singularity; and stability plus convergence are proved via a complementary discrete convolution kernel, a discrete fractional Gronwall inequality, and an error convolution structure. Numerical experiments are reported to demonstrate accuracy and efficiency.

Significance. If the reduction step preserves the exact Dirichlet boundary conditions and the solution regularity (including the initial singularity) needed for the compact operator and the three theoretical tools, the work would supply a practical high-order solver for fractional fourth-order diffusion problems that handles the typical weak singularity via nonuniform meshes. The explicit use of the three standard tools for the stability/convergence analysis is a methodological strength, as is the compact stencil that avoids extra boundary points.

major comments (2)

[Abstract (method description) and the opening of the scheme construction] The reduction of the fourth-order problem to a coupled second-order system (stated as the first step of the method in the abstract) is load-bearing for the subsequent claims. The manuscript must explicitly verify that the auxiliary variables inherit the same Dirichlet boundary conditions without mismatch and that the fractional-order regularity (including the initial singularity) is unchanged, so that the averaged compact operator, L1 discretization on irregular mesh, and error convolution structure apply directly without additional consistency error.
[Abstract and the stability/convergence analysis section] The abstract asserts that stability and convergence follow from the complementary discrete convolution kernel, discrete fractional Gronwall inequality, and error convolution structure, yet supplies no derivation steps, no explicit constants in the error bound, and no check that the second-order reduction introduces no extra truncation error. This omission prevents verification that the stated O(τ^{2-α} + h^4) rate holds after the reduction.

minor comments (1)

[Scheme construction] Notation for the averaged operator and the auxiliary variables should be introduced with a clear diagram or table showing how the Dirichlet conditions are transferred to the second-order system.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and constructive comments. We address each major comment below and will revise the manuscript to strengthen the presentation where needed.

read point-by-point responses

Referee: [Abstract (method description) and the opening of the scheme construction] The reduction of the fourth-order problem to a coupled second-order system (stated as the first step of the method in the abstract) is load-bearing for the subsequent claims. The manuscript must explicitly verify that the auxiliary variables inherit the same Dirichlet boundary conditions without mismatch and that the fractional-order regularity (including the initial singularity) is unchanged, so that the averaged compact operator, L1 discretization on irregular mesh, and error convolution structure apply directly without additional consistency error.

Authors: We agree that explicit verification strengthens the argument. In Section 2 the auxiliary variable is defined via the Laplacian and the boundary conditions are stated to be compatible; however, to address the concern directly we will add a short paragraph confirming that both components satisfy the identical homogeneous Dirichlet conditions and that the time-fractional regularity (including the initial singularity) is preserved under the spatial reduction. Because the reduction is an exact algebraic equivalence at the continuous level, no additional consistency error is introduced, allowing the compact operator and the three theoretical tools to apply unchanged. revision: yes
Referee: [Abstract and the stability/convergence analysis section] The abstract asserts that stability and convergence follow from the complementary discrete convolution kernel, discrete fractional Gronwall inequality, and error convolution structure, yet supplies no derivation steps, no explicit constants in the error bound, and no check that the second-order reduction introduces no extra truncation error. This omission prevents verification that the stated O(τ^{2-α} + h^4) rate holds after the reduction.

Authors: The abstract is a high-level summary; the full derivations using the three tools appear in Sections 3 and 4. We will nevertheless add an explicit remark in Section 4 stating that the reduction is exact and introduces no extra truncation error, together with the leading constants from the discrete Gronwall inequality and the final error estimate, so that the O(τ^{2-α} + h^4) rate can be verified directly on the reduced system. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation relies on external tools

full rationale

The paper's derivation begins with a standard reduction of the fourth-order subdiffusion equation to a coupled second-order system, followed by an averaged compact spatial operator and L1 discretization on irregular time meshes to handle initial singularities. Stability and convergence are then proved using three cited external tools (complementary discrete convolution kernel, discrete fractional Gronwall inequality, and error convolution structure), which are presented as independent mathematical results rather than derived internally or via self-citation chains. No steps match the enumerated circularity patterns: there are no self-definitional reductions, fitted inputs renamed as predictions, load-bearing self-citations, uniqueness theorems imported from the authors' prior work, smuggled ansatzes, or renamings of known results. The chain is self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The paper introduces no new physical entities or fitted parameters. It relies on the standard domain assumption that the continuous problem possesses a sufficiently regular solution so that the truncation errors of the compact operator and the L1 formula can be bounded.

axioms (1)

domain assumption The continuous fourth-order subdiffusion problem admits a unique solution with the regularity required for the error estimates to hold.
Invoked implicitly when claiming convergence rates for the discrete scheme.

pith-pipeline@v0.9.0 · 5660 in / 1346 out tokens · 48621 ms · 2026-05-25T10:27:26.844122+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

27 extracted references · 27 canonical work pages

[1]

Om. P. Agrawal, A general solution for a fourth-order fractional diﬀusion- wave equation deﬁned in a bounded domain , Comput. Struct., 79 (16) (2001), 1497-1501

work page 2001
[2]

Cui , Compact ﬁnite diﬀerence method for the fractional diﬀusion e quation, J

M. Cui , Compact ﬁnite diﬀerence method for the fractional diﬀusion e quation, J. Comput. Phys., 228(20) (2009), 7792-7804

work page 2009
[3]

Chen , Nonlinear dynamics and chaos in a fractional-order ﬁnancia l system , Chaos

W. Chen , Nonlinear dynamics and chaos in a fractional-order ﬁnancia l system , Chaos. Soliton. Fract., 36 (5) (2008), 1305-1314

work page 2008
[4]

J. Guo, C. Li and H. Ding , Finite diﬀerence methods for time subdiﬀusion equation wit h space fourth-order, Commun. Appl. Math. Comput., 28 (2014), 96-108

work page 2014
[5]

Gao and Z

G. Gao and Z. Sun , A compact ﬁnite diﬀerence scheme for the fractional sub-diﬀ usion equations, J. Comput. Phys., 230 (3) (2011), 586-595

work page 2011
[6]

Golbabai and K

A. Golbabai and K. Sayevand , Fractional calculus - a new approach to the analysis of generalized fourth-order diﬀusion-wave equations , Appl. Math. Comput., 61 (8) (2011), 2227-2231

work page 2011
[7]

Hilfer , Applications of fractional calculus in physics , World Scientiﬁc, Singapore, 2000

R. Hilfer , Applications of fractional calculus in physics , World Scientiﬁc, Singapore, 2000. 18

work page 2000
[8]

Halpern, O

D. Halpern, O. E. Jensen and J. B. Grotberg , A theoretical study of surfactant and liquid delivery into the lung , J. Appl. Physiol., 85 (1) (1998), 333-352

work page 1998
[9]

Hu and L

X. Hu and L. Zhang , A compact ﬁnite diﬀerence scheme for the fourth-order fract ional diﬀusion-wave system , Comput. Phys. Commun., 182 (8) (2011), 1645-1650

work page 2011
[10]

Hu and L

X. Hu and L. Zhang , On ﬁnite diﬀerence methods for fourth-order fractional diﬀ usion- wave and subdiﬀusion systems , Appl. Math. Comput., 218 (9) (2012), 5019-5034

work page 2012
[11]

Jafari, M

H. Jafari, M. Dehghan and K. Sayevand , Solving a fourth-order fractional diﬀusion- wave equation in a bounded domain by decomposition method , Numer. Methods. Part. Diﬀer. Equ., 24 (4) (2008), 1115-1126

work page 2008
[12]

C. Ji, Z. Sun and Z. Hao , Numerical algorithms with high spatial accuracy for the fou rth- order fractional sub-diﬀusion equations with the ﬁrst Diri chlet boundary conditions , J. Sci. Comput., 66 (3) (2016), 1148-1174

work page 2016
[13]

V. I. Karpman , Stabilization of soliton instabilities by higher-order di spersion: fourth- order nonlinear Schr¨ odinger-type equations, Phys. Rev. E., 53 (2) (1996), 1336-1339

work page 1996
[14]

A. A. Kilbas, H. M. Srivastava and J. J. Trujillo , Theory and applications of fractional diﬀerential equations , Elsevier., 2006

work page 2006
[15]

H.-L. Liao, D. Li, and J. Zhang , Sharp error estimate of the nonuniform L1 formula for linear reaction-subdiﬀusion equations , SIAM J. Numer. Anal., 56 (2) (2018), 1112-1133

work page 2018
[16]

H. Liao, W. Mclean and J. Zhang , A discrete Gr¨ onwall inequality with application to numerical schemes for subdiﬀusion problems , SIAM J. Numer. Anal., 57(1) (2019), 218-237

work page 2019
[17]

Liao and Z

H. Liao and Z. Sun , Maximum norm error bounds of ADI and compact ADI methods for solving parabolic equations , Numer. Methods. Part. Diﬀer. Equ., 26 (1) (2010), 37-60

work page 2010
[18]

H. Liao, Y. Yan and J. Zhang , Unconditional convergence of a fast two-level linearized algorithm for semilinear subdiﬀusion equations , J. Sci. Comput., 80(1) (2019), 1-25

work page 2019
[19]

T. G. Myers, J. P. F. Charpin and S. J. Chapman , The ﬂow and solidiﬁcation of a thin ﬂuid ﬁlm on an arbitrary three-dimensional surface , Phys. Fluids., 14 (8) (2002), 2788-2803

work page 2002
[20]

T. G. Myers and J. P. F. Charpin , A mathematical model for atmospheric ice accretion and water ﬂow on a cold surface , Int. J. Heat. Mass. Transf., 47 (25) (2004), 5483-5500

work page 2004
[21]

Metzler and J

R. Metzler and J. Klafter , The random walk’s guide to anomalous diﬀusion: a frac- tional dynamics approach , Phys. Rep., 339 (1) (2000), 1-77

work page 2000
[22]

J. Ren, H. Liao, J. Zhang and Z. Zhang , Sharp H 1-norm error estimates of two time-stepping schemes for reaction-subdiﬀusion problems , arXiv:1811.08059v1, 2018

work page arXiv 2018
[23]

Ren and Z

J. Ren and Z. Sun , Numerical algorithm with high spatial accuracy for the frac tional diﬀusion-wave equation with Neumann boundary conditions , J. Sci. Comput., 56 (2) (2013), 381-408. 19

work page 2013
[24]

Stynes, E

M. Stynes, E. Oriordan and J. L. Gracia , Error analysis of a ﬁnite diﬀerence method on graded meshes for a time-fractional diﬀusion equation , SIAM J. Numer. Anal., 55 (2) (2017), 1057-1079

work page 2017
[25]

Sun and X

Z. Sun and X. Wu , A fully discrete diﬀerence scheme for a diﬀusion-wave syste m, Appl. Numer. Math., 56 (2) (2006), 193-209

work page 2006
[26]

Yao and Z

Z. Yao and Z. W ang , A compact diﬀerence scheme for fourth-order fractional sub - diﬀusion equations with Neumann boundary conditions , J. Appl. Anal. Comput., 8 (4) (2018), 1159-1169

work page 2018
[27]

Zhang and H

P. Zhang and H. Pu , A second-order compact diﬀerence scheme for the fourth-ord er fractional sub-diﬀusion equation , Numer. Algor., 76 (2) (2017), 573-598. 20

work page 2017

[1] [1]

Om. P. Agrawal, A general solution for a fourth-order fractional diﬀusion- wave equation deﬁned in a bounded domain , Comput. Struct., 79 (16) (2001), 1497-1501

work page 2001

[2] [2]

Cui , Compact ﬁnite diﬀerence method for the fractional diﬀusion e quation, J

M. Cui , Compact ﬁnite diﬀerence method for the fractional diﬀusion e quation, J. Comput. Phys., 228(20) (2009), 7792-7804

work page 2009

[3] [3]

Chen , Nonlinear dynamics and chaos in a fractional-order ﬁnancia l system , Chaos

W. Chen , Nonlinear dynamics and chaos in a fractional-order ﬁnancia l system , Chaos. Soliton. Fract., 36 (5) (2008), 1305-1314

work page 2008

[4] [4]

J. Guo, C. Li and H. Ding , Finite diﬀerence methods for time subdiﬀusion equation wit h space fourth-order, Commun. Appl. Math. Comput., 28 (2014), 96-108

work page 2014

[5] [5]

Gao and Z

G. Gao and Z. Sun , A compact ﬁnite diﬀerence scheme for the fractional sub-diﬀ usion equations, J. Comput. Phys., 230 (3) (2011), 586-595

work page 2011

[6] [6]

Golbabai and K

A. Golbabai and K. Sayevand , Fractional calculus - a new approach to the analysis of generalized fourth-order diﬀusion-wave equations , Appl. Math. Comput., 61 (8) (2011), 2227-2231

work page 2011

[7] [7]

Hilfer , Applications of fractional calculus in physics , World Scientiﬁc, Singapore, 2000

R. Hilfer , Applications of fractional calculus in physics , World Scientiﬁc, Singapore, 2000. 18

work page 2000

[8] [8]

Halpern, O

D. Halpern, O. E. Jensen and J. B. Grotberg , A theoretical study of surfactant and liquid delivery into the lung , J. Appl. Physiol., 85 (1) (1998), 333-352

work page 1998

[9] [9]

Hu and L

X. Hu and L. Zhang , A compact ﬁnite diﬀerence scheme for the fourth-order fract ional diﬀusion-wave system , Comput. Phys. Commun., 182 (8) (2011), 1645-1650

work page 2011

[10] [10]

Hu and L

X. Hu and L. Zhang , On ﬁnite diﬀerence methods for fourth-order fractional diﬀ usion- wave and subdiﬀusion systems , Appl. Math. Comput., 218 (9) (2012), 5019-5034

work page 2012

[11] [11]

Jafari, M

H. Jafari, M. Dehghan and K. Sayevand , Solving a fourth-order fractional diﬀusion- wave equation in a bounded domain by decomposition method , Numer. Methods. Part. Diﬀer. Equ., 24 (4) (2008), 1115-1126

work page 2008

[12] [12]

C. Ji, Z. Sun and Z. Hao , Numerical algorithms with high spatial accuracy for the fou rth- order fractional sub-diﬀusion equations with the ﬁrst Diri chlet boundary conditions , J. Sci. Comput., 66 (3) (2016), 1148-1174

work page 2016

[13] [13]

V. I. Karpman , Stabilization of soliton instabilities by higher-order di spersion: fourth- order nonlinear Schr¨ odinger-type equations, Phys. Rev. E., 53 (2) (1996), 1336-1339

work page 1996

[14] [14]

A. A. Kilbas, H. M. Srivastava and J. J. Trujillo , Theory and applications of fractional diﬀerential equations , Elsevier., 2006

work page 2006

[15] [15]

H.-L. Liao, D. Li, and J. Zhang , Sharp error estimate of the nonuniform L1 formula for linear reaction-subdiﬀusion equations , SIAM J. Numer. Anal., 56 (2) (2018), 1112-1133

work page 2018

[16] [16]

H. Liao, W. Mclean and J. Zhang , A discrete Gr¨ onwall inequality with application to numerical schemes for subdiﬀusion problems , SIAM J. Numer. Anal., 57(1) (2019), 218-237

work page 2019

[17] [17]

Liao and Z

H. Liao and Z. Sun , Maximum norm error bounds of ADI and compact ADI methods for solving parabolic equations , Numer. Methods. Part. Diﬀer. Equ., 26 (1) (2010), 37-60

work page 2010

[18] [18]

H. Liao, Y. Yan and J. Zhang , Unconditional convergence of a fast two-level linearized algorithm for semilinear subdiﬀusion equations , J. Sci. Comput., 80(1) (2019), 1-25

work page 2019

[19] [19]

T. G. Myers, J. P. F. Charpin and S. J. Chapman , The ﬂow and solidiﬁcation of a thin ﬂuid ﬁlm on an arbitrary three-dimensional surface , Phys. Fluids., 14 (8) (2002), 2788-2803

work page 2002

[20] [20]

T. G. Myers and J. P. F. Charpin , A mathematical model for atmospheric ice accretion and water ﬂow on a cold surface , Int. J. Heat. Mass. Transf., 47 (25) (2004), 5483-5500

work page 2004

[21] [21]

Metzler and J

R. Metzler and J. Klafter , The random walk’s guide to anomalous diﬀusion: a frac- tional dynamics approach , Phys. Rep., 339 (1) (2000), 1-77

work page 2000

[22] [22]

J. Ren, H. Liao, J. Zhang and Z. Zhang , Sharp H 1-norm error estimates of two time-stepping schemes for reaction-subdiﬀusion problems , arXiv:1811.08059v1, 2018

work page arXiv 2018

[23] [23]

Ren and Z

J. Ren and Z. Sun , Numerical algorithm with high spatial accuracy for the frac tional diﬀusion-wave equation with Neumann boundary conditions , J. Sci. Comput., 56 (2) (2013), 381-408. 19

work page 2013

[24] [24]

Stynes, E

M. Stynes, E. Oriordan and J. L. Gracia , Error analysis of a ﬁnite diﬀerence method on graded meshes for a time-fractional diﬀusion equation , SIAM J. Numer. Anal., 55 (2) (2017), 1057-1079

work page 2017

[25] [25]

Sun and X

Z. Sun and X. Wu , A fully discrete diﬀerence scheme for a diﬀusion-wave syste m, Appl. Numer. Math., 56 (2) (2006), 193-209

work page 2006

[26] [26]

Yao and Z

Z. Yao and Z. W ang , A compact diﬀerence scheme for fourth-order fractional sub - diﬀusion equations with Neumann boundary conditions , J. Appl. Anal. Comput., 8 (4) (2018), 1159-1169

work page 2018

[27] [27]

Zhang and H

P. Zhang and H. Pu , A second-order compact diﬀerence scheme for the fourth-ord er fractional sub-diﬀusion equation , Numer. Algor., 76 (2) (2017), 573-598. 20

work page 2017