A robust time-split linearized explicit/implicit technique for two-dimensional hydrodynamic model: an application to floods in Cameroon far north region

Eric Ngondiep

REVIEW 4 major objections 5 minor 61 references

A robust time-split linearized explicit/implicit technique for two-dimensional hydrodynamic model: an application to floods in Cameroon far north region

T0 review · 4 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read A symmetric time-split explicit/implicit algorithm for the two-dimensional shallow water equations is stable under a computable $\Delta x$-scaled time-step restriction and, on test problems, converges at fourth order in space and second…

desk verdict The convergence tests on Thacker solutions look real, but the stability theorem is vacuous as stated and the Cameroon flood application is unphysical; the scheme might be salvageable, but the paper in its current form should not be published. read the letter →

arxiv 2411.17740 v3 pith:REJQP7CF submitted 2024-11-24 math.NA cs.NA

Eric Ngondiep This is my paper · ORCID

classification math.NAcs.NA MSC 65M1265M06

keywords two-dimensionalshallowwaterequationstime-splitexplicit/implicitschemesymmetricoperatorsplittingfourth-orderspatialaccuracysecond-ordertemporalstabilitytime-steprestrictionCameroonfarnorthfloodshydrodynamicfloodforecasting

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper aims to replace expensive two-dimensional shallow-water solves with a three-stage split: an explicit sweep in $x$, an implicit sweep in $y$, and another explicit sweep in $x$. The paper claims a stability theorem (Theorem 3.1) giving an explicit, $\Delta x$-scaled restriction on the time step, and numerical tests on exact solutions suggest the scheme is fourth-order accurate in space and second-order in time. If the scheme behaves as claimed, it offers a fast high-order alternative for dam-break and overland flood simulation, and the paper applies it to the 2024 floods in Cameroon's far north region, locating the peak of inundation around 21 September. The claim matters because flood forecasting needs solvers that are fast enough to run in real time and accurate enough to resolve sharp moving water fronts.

What carries the argument

The central object is the symmetric operator composition $P(k)=P_1(k/2)P_2(k)P_1(k/2)$: the two outer explicit steps use the same operator, which makes the composite sequence symmetric and hence second-order in time, while the inner implicit step is unconditionally stable. The stability proof is carried by the matrix form of the fourth-order difference operator, $\delta^4_x=\frac{1}{12\Delta x}A$, where $A$ is a normal pentadiagonal matrix. A standard eigenvalue localization argument bounds the norm of $A$ by 18, and that single bound is what converts the stability inequality into the user-visible restriction (47) involving the computed solution's norms $\|u\|_{0,\infty}$ and $\|u^2+\frac12 gh\|_{0,\infty}$.

What would settle it

Run the full scheme (38)-(45) on a smooth periodic test and compare its largest stable time step with restriction (47) across a range of $\gamma$: a run that satisfies (47) with $\gamma$ near 18 and nevertheless grows in the $L^2$ norm would falsify the claimed sufficient condition. A more direct check is to evaluate the dropped term $\frac{k}{4}\delta^2_x(u\delta^{3\mp}_xE+h\delta^{3\mp}_xE)$ alongside $\delta^4_xE$ over many steps; if its norm is not small compared with the retained term uniformly as $k\to 0$, the approximation behind (58) and (66) fails.

Watch

Extended reading notes

Core claim

The paper's central claim is that the time-split linearized explicit/implicit scheme (38)-(45) is stable under the time-step restriction (47) and numerically achieves fourth-order spatial accuracy with second-order temporal convergence. The scheme is written as the symmetric composition $P(k)=P_1(k/2)P_2(k)P_1(k/2)$, where $P_1$ is an explicit one-dimensional operator built from fourth-order spatial differences and a quadratic linearization of the $x$-flux, and $P_2$ is an implicit operator obtained by averaging the explicit operator at the two intermediate levels for the $y$-flux and source term. The proof of Theorem 3.1 treats the implicit middle step as unconditionally stable and reduces the stability of the whole scheme to the two explicit steps; the fourth-order difference $\delta^4_x$ is represented as $\frac{1}{12\Delta x}A$ with a normal pentadiagonal matrix $A$ whose spectral radius is bounded by 18. From that bound the paper derives restriction (47), and the reported experiments on the two exact shallow-water solutions show $L^\infty$ errors decreasing by factors close to $3^4$ under mesh refinement and close to $3^2$ under time-step refinement. The same algorithm is then applied to the Logone basin to simulate the July-October 2024 flood season.

Load-bearing premise

The load-bearing assumption is that, for small time steps, the correction term $\frac{k}{4}\delta^2_x(\cdots)$ inside the explicit operator can be dropped from the stability estimate without hiding an instability; the proof relies on this approximation but does not bound the discarded term.

Editorial extensions

If this is right

Each time level costs one explicit sweep in $x$, one implicit sweep in $y$, and one explicit sweep in $x$, so the solver avoids inverting two-dimensional block systems while keeping fourth-order spatial accuracy.
On the exact shallow-water tests, the $L^\infty$ errors for water depth and both velocity components shrink by factors close to $3^4$ as the mesh is refined by 3 and by factors close to $3^2$ as the time step is refined by 3, matching the advertised orders.
In the Logone basin application the computed water depth and velocities reach their peak around 21 September 2024, which the paper identifies with the observed flood peak in the far north region.
Because the restriction (47) depends on norms of the computed solution, a user must monitor the admissible time step during a run; the numerical section reports that too-small initial depths or too-large initial velocities can let oscillations destroy the solution.

Reading between the lines

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

The proof of Theorem 3.1 discards the $\frac{k}{4}\delta^2_x(\cdot)$ correction term inside the explicit operator without a bound on that term, so a careful reader should treat restriction (47) as a validated guideline with a tuning parameter $\gamma$ rather than as a fully proven sufficient condition.
The same matrix-bounding argument does not use the detailed form of the shallow-water source terms, so the split-template should transfer to other hyperbolic balance laws; testing it on the full Saint-Venant system with wetting and drying would be a natural next step.
The Cameroon application is driven by 2013 discharge climatology, so a true operational forecast would require comparing the computed inundation extents against observed gauge or satellite data for 2024.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

Colleague,

The paper has one credible numerical result and two serious problems. On Thacker's analytical solutions, the split scheme shows clean fourth-order spatial and second-order temporal convergence. That part is reproducible and suggests the method does work on smooth problems.

But the stability theorem is not what it claims to be. Restriction (47) contains a free parameter gamma in (0,18] that never gets tied to the discrete operator. In the proof, gamma would have to be rho_max(A), but since it is not, you can satisfy (47) for any time step by taking gamma small enough. The numerical section uses gamma=18, 1, 12, 3, which confirms it is treated as a tuning knob. On top of that, the step from (58) to (59) and (65) to (67) drops the k/4 terms with only "for small k" as justification, and no bound on the neglected product is supplied. So Theorem 3.1 is not a proven sufficient condition; it is a heuristic dressed as a theorem.

The Cameroon flood case is worse. Reported velocities around 1.7e6 m/s and depths blowing up to 1e129 m are not flood forecasts; they are signs that the simulation has broken down. The paper itself admits small initial depth or large velocities destroy the solutions, but then still presents those numbers as part of the flood analysis. That section does not support the applied claim.

What is genuinely new is the specific combination of Strang splitting with a fourth-order explicit x-step and an implicit y-step. That is a reasonable component to exist, and the convergence tables give some evidence it performs as advertised on smooth test problems. The writing is clear and the scheme is described in enough detail to reproduce.

For a serious referee, I would want the stability theorem fixed or honestly reframed as a CFL heuristic, and the flood section either validated against observations or removed. In its current form, the paper's central theoretical claim and its practical application both fail. I would not cite it this year, and I would not send it to review as is.

That's my read.

Referee Report

4 major / 5 minor

Summary. The manuscript proposes a time-split linearized explicit/implicit finite-difference scheme for the two-dimensional shallow-water equations, with a stability restriction claimed in Theorem 3.1 (Eq. (47)). It reports fourth-order spatial and second-order temporal convergence on two Thacker analytical solutions, and applies the scheme to a 2024 flood event in the Far North region of Cameroon.

Significance. If the stability theorem and convergence rates were fully established, the scheme would offer a useful high-order time-split solver for shallow-water flows. The numerical convergence tables (Tables 1-4) are coherent and benchmark against external analytical solutions, which gives credible evidence of accuracy on smooth problems. The Cameroon flood application, however, produces nonphysical values, and the stability theorem as stated is vacuous because of the free parameter γ.

major comments (4)

[Theorem 3.1, Eq. (47)] The stability restriction (47) involves a free parameter γ with 0 < γ ≤ 18 that is never tied to the discrete operator. The proof itself derives the bounds (64) and (68) with ρ_max(A) — the spectral radius of the pentadiagonal matrix A — in the denominator, but the theorem substitutes an arbitrary γ. Consequently, for any positive time step k and any data, one can choose γ small enough (while keeping γ > 0) so that (47) holds; the condition therefore imposes no upper bound on k and cannot function as a stability restriction. The numerical tables use γ = 18, 1, 12, 3, confirming that γ is treated as tunable. The theorem should be restated with γ identified as ρ_max(A) or with an explicit upper bound derived from A.
[Theorem 3.1 proof, Eqs. (57)-(59) and (65)-(67)] The step from (57) to (59) and from (65) to (67) replaces a norm of δ⁴_x E − (k/4)δ²_x(...) by the norm of δ⁴_x E alone, justified only by 'for small values of the time step k' and an unquantified assertion that dropping the term 'does not compromise the result on stability.' No bound on the dropped term is proved, and the direction of the inequality is not preserved by this replacement. Since Theorem 3.1 claims a sufficient condition for stability, the missing control of the k² term is a load-bearing gap.
[Section 4, flood application, Figure 5] The flood simulation results are nonphysical: reported peak values include water depth 2×10² m, water depth up to 3.3×10¹²⁹ m, and velocities up to 1.7×10⁶ m/s and 10¹⁵³ m/s. The text itself notes that small initial depths or large initial velocities destroy the computed solutions, yet the section still presents the peak date of 21 September 2024 as a forecast. As reported, the application does not support the claim that the scheme is a reliable flood-prediction tool.
[Sections 2 and 3, stability of stage (39) and convergence order] The paper asserts that the implicit second stage (39) is unconditionally stable and that the scheme is temporally second-order accurate and spatially fourth-order convergent, but no proof is given for either statement: stability of the nonlinear implicit stage is stated without argument, and the convergence order is only inferred from the local truncation error estimates (28) and (36) together with the numerical tables. If the stability theorem is repaired, the unconditional stability of stage (39) should also be justified.

minor comments (5)

[Eq. (15)] Equation (15) uses the same notation (·,·)_∼ and ∥·∥∼ for both R^{Mx−3} and R^{My−3} without distinguishing the dimensions; this is confusing for the reader.
[Eqs. (58) and (66)] The symbol ≈ in equations (58) and (66) is informal; if the dropped terms are truly negligible, the proof should state the precise inequality that justifies the replacement.
[Eq. (4)] The Manning friction formulas in (4) have unusual powers of u and v; a reader cannot verify dimensional consistency from the text, and the relation to the standard Manning formula should be clarified.
[Figure 5 caption] In the caption of Figure 5, the mesh size in the y-direction is given as both 12.36 and 111; the discrepancy should be resolved.
[Throughout] There are numerous typos, including 'T ables' and 'Abowitz' in reference [47]; the manuscript needs a careful proofreading before resubmission.

Circularity Check

1 steps flagged · score 6.0 of 10

The stability restriction (47) is self-fulfilling: γ is free (0<γ≤18) and untied to ρmax(A), so any time step can be made to satisfy it; Tables 1-4 tune γ per run and then 'confirm' stability.

fitted input called prediction [Section 3, Theorem 3.1 and its proof around (60)-(68); Section 4, Tables 1-4]
"Theorem 3.1. ... k ≤ 48/γ min{ ‖β‖0/(√Mx−3‖|u|‖0,∞), ‖|u|‖0,∞/‖|u2+1/2gh|‖0,∞ } Δx, where β=1∈L2(Ω), 0<γ≤18. ... It follows from Lemma 3.3 that 0<‖|A|‖=ρmax(A)≤2(1+8)=18. ... satisfying restriction (47), with γ=18 ... with γ=1 ... with γ=12 ... with γ=3."

The proof obtains its final time-step bounds with ρmax(A) in the denominator and proves only the bracket 0<ρmax(A)≤18. The theorem, however, replaces ρmax(A) by a free parameter γ∈(0,18] and never identifies γ with ρmax(A) or with any other property of the discrete operator. Since γ may be chosen arbitrarily close to 0, the right-hand side of (47) can be made arbitrarily large; for any prescribed k, one can pick γ small enough that (47) holds. Thus (47) excludes no time step and cannot function as a stability restriction. The numerical section then chooses γ=18, 1, 12, and 3 in Tables 1-4 and labels each run as satisfying (47).

full rationale

The central convergence claim is not circular: Examples 1 and 2 are compared against Thacker's analytical solutions, an external benchmark, with L∞ errors and measured rates near 4 in space and 2 in time. The minor self-citation of the author's own reference [28] for standard fourth-order difference approximations does not drive that conclusion, because the accuracy result is independently tested. The serious issue is Theorem 3.1's stability restriction. The proof bounds ρmax(A) only by 0<ρmax(A)≤18, yet the theorem states the restriction with an arbitrary parameter γ∈(0,18] that is never linked to A or to the scheme. Because γ can be chosen after k, condition (47) is vacuously satisfiable and is therefore not a genuine restriction. The paper's own tables tune γ per experiment, and its appended limitation that 'condition (47) deals with the computed solutions. Thus, physical insight must be used' further confirms that the condition is not a priori checkable. Equations (58) and (66) also drop O(k) nonlinear terms with only the assertion that the truncation 'does not compromise the result on stability'; that is a proof gap, not a circularity. Overall, the stability component of the paper's claim reduces to a tuned free parameter, while the accuracy component remains independently validated. This is partial circularity rather than complete derivation-by-definition.

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

The theorem's time-step restriction depends on gamma, a free parameter tuned in the experiments, and on the unproved small-k approximation in equations (58) and (66). The physical model and boundary treatment are standard but not independently established. No new physical entities are introduced.

free parameters (1)

gamma (stability parameter) = 18, 1, 12, 3 in Tables 1 to 4
Appears in Theorem 3.1 restriction (47) with only 0 < gamma <= 18. The proof bounds rho_max(A) <= 18, but gamma is never tied to the actual matrix A. Choosing gamma small makes the bound arbitrarily loose, so the numerical tables effectively tune gamma to their runs.

assumptions (5)

domain assumption The 2D shallow water model (1) is an adequate description of overland and flood flow, with hydrostatic pressure and depth-averaged velocity.
Invoked in the introduction with references [49,51]; this is standard hydraulics but not proved in the paper. It matters because the numerical experiments and flood application inherit this assumption.
ad hoc to paper For small k, the dropped terms in equations (58) and (66) do not change the stability threshold of the explicit operator.
Used in the proof of Theorem 3.1 to obtain inequalities (59) and (67); no error bound or higher-order analysis is provided.
domain assumption The implicit second stage P2(k) is unconditionally stable despite the nonlinear source term with h^{-4/3} friction.
Stated in Section 3 without proof for the nonlinear system; this underpins the claim that only the explicit stages impose a time step restriction.
standard math Gershgorin's theorem and normal-matrix diagonalizability correctly bound the spectral radius of the difference matrix A.
Standard linear algebra, used in Lemma 3.3. The proof misstates C^T = -C for a matrix with nonzero diagonal, but the particular matrix A is skew-symmetric, so the bound is plausible.
domain assumption The boundary values in (42) to (45) are compatible with the split intermediate stages.
The paper sets boundary conditions for all intermediate variables by flat, following Remark 2.2; incompatible intermediate boundary conditions are a known source of splitting error.

how reviews work

0 comments

Cite this review

Pith. "Pith review of A robust time-split linearized explicit/implicit technique for two-dimensional hydrodynamic model: an application to floods in Cameroon far north region." pith.science (2026). https://pith.science/paper/REJQP7CF

@misc{pith2026241117740,
  author       = {Pith},
  title        = {Pith review of: A robust time-split linearized explicit/implicit technique for two-dimensional hydrodynamic model: an application to floods in Cameroon far north region},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/REJQP7CF}},
  note         = {Machine review of arXiv:2411.17740}
}

abstract

This paper deals with a time-split explicit/implicit approach for solving a two-dimensional hydrodynamic flow model with appropriate initial and boundary conditions. The time-split technique is employed to upwind the convection term and to treat the friction slope so that the numerical oscillations and stability are well controlled. A suitable time step restriction for stability and convergence accurate of the new algorithm is established using the $L^{\infty}(0,T; L^{2})$-norm. Under a time step requirement, some numerical examples confirm the theoretical studies and suggest that the proposed computational technique is spatial fourth-order accurate and temporal second-order convergent. An application to floods observed in Cameroon far north region is considered and discussed.

Figures

Figures reproduced from arXiv: 2411.17740 by the authors.

**Figure 1.** Dam-break, Water depth and Geometric configuration. [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗

**Figure 2.** Situation of floods, Logone river and Guere lake in Mayo-Da [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

Figures from the paper (6 more)

**Figure 2.** Figure 2: v [PITH_FULL_IMAGE:figures/full_fig_p016_2.png]

**Figure 1.** Figure 1: iv [PITH_FULL_IMAGE:figures/full_fig_p017_1.png]

**Figure 1.** Figure 1: iv [PITH_FULL_IMAGE:figures/full_fig_p018_1.png]

**Figure 3.** Figure 3: Graphs of water depth, x- and y-direction velocities and e [PITH_FULL_IMAGE:figures/full_fig_p024_3.png]

**Figure 4.** Figure 4: Graphs of water depth, x- and y-direction velocities and e [PITH_FULL_IMAGE:figures/full_fig_p025_4.png]

**Figure 5.** Figure 5: Graphs of water depth, x- and y-direction velocities for fl [PITH_FULL_IMAGE:figures/full_fig_p026_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

61 extracted references · 52 canonical work pages

[1]

Abarbaned, A

S. Abarbaned, A. Kumar. ”Compact high-order schemes for th e Euler equations”, J. Scientiﬁc Com- puting, 3(1988), 275-288

work page 1988
[2]

Ablowitz, B

M.J. Ablowitz, B. M. Herbst. ”On the numerical solution of the sine -Gordon equation. 1. Integrable discretizations and homoclinic manifolds”, J. Comput. Phys., 126(19 96), 299-314

work page
[3]

K. O. Aiyesimoju, R. J. Sobey. ”Process splitting of the boundar y conditions for the advection-dispersion equation”, Int. J. Numer. Methods Fluids, 9(1989), 235-244

work page 1989
[4]

R. T. Alqahtani, J. C. Ntonga, E. Ngondiep. ”Stability analysis and convergence rate of a two-step predictor-corrector approach for shallow water equations with s ource terms”, AIMS Mathematics, 8(4) (2023), 9265-9289

work page 2023
[5]

W. Bao, Q. Du. ”Computing the ground state solution of Bose-Ein stein condensates by a normalized gradient ﬂow”, SIAM J. Sci. Comput., 25(2004), 1674-1697

work page 2004
[6]

G. F. Carey, Y. Shen. ”Least-squres ﬁnite element approximat ion of Fisher’s reaction diﬀusion equation”, Numer. Meth. partial Diﬀerential Equations, 11(1995), 175-186

work page 1995
[7]

Demkowicz, T

L. Demkowicz, T. J. Oden, W. Rachowicz. ”A new ﬁnite element met hod for solving compressible Navier-Stokes equations based on an operator splitting method an d h-p adaptivity”, Comput. Methods Appl. Mech. Engng., 84(1990), 275-326

work page 1990
[8]

Descombes, M

S. Descombes, M. Massot. ”Operator splitting for nonlinear rea ction-diﬀusion systems with an entropic structure: singular perturbation and order reduction”, Numer. Math., 97(2004), 667-698

work page 2004

Show all 61 references

[9]

Fiedler, J.A

F.R. Fiedler, J.A. Ramirez. ”A numerical method for simulating disco ntinuous shallow ﬂow over an inﬁltrating surface”, Int. J. Numer. Meth. Fluids, 32 : 219-240 , (2000)

2000
[10]

Garcia, R

R. Garcia, R. A. Kahawaita. ”Numerical solution of the St. Vena nt equations with MacCormack ﬁnite- diﬀerence scheme”, Int. J. Numer. Meth. Fluids, 6(5) (1986), 25 9-274

1986
[11]

Goldman, T

D. Goldman, T. J. kaper. ”Nth-order operator splitting schem es and nonreversible systems”, SIAM J. Numer. Anal., 33(1996), 349-367

1996
[12]

Gottlieb, J

D. Gottlieb, J. S. Hesthaven. ”Spectral methods for hyperb olic problems”, J. Comput. Appl. Math., 128(2001), 83-131

2001
[13]

Gottlieb, E, tadmor

D. Gottlieb, E, tadmor. ”The CFL condition for spectral appro ximations to hyperbolic initial-boundary value problems”, Math. Comput., 56(1991), 565-588

1991
[14]

M. M. Gupta. ”High accuracy solutions of incompressible Navier- Stokes equations”, J. Comput. Phys., 93(1991), 343-357

1991
[15]

Henonin, B

J. Henonin, B. Russo, O. Mark, P. Gourbesville. ”Real-time urba n ﬂood forecasting and modelling-a state of the art”, J. Hydroinformatics 15(3) (2013), 717-736

2013
[16]

Hixon, E

R. Hixon, E. Turkel. ”Compact implicit MacCormack type schemes with high accuracy”, J. Comput. Phys., 158 51-70 (2000)

2000
[17]

Holden, K

H. Holden, K. H. Karlsen, N. H. Risebro. ”Operator splitting met hods for generalized Korteweg-de Vries equations”, J. Comput. Phys., 153(1999), 203-222

1999
[18]

L. A. Khan, P. L. -F. Liu. ”Numerical analyses of operator-sp litting algorithms for the two-dimensional advection-diﬀusion equation”, Comput. Meth. Appl. Mech. Engng., 152(1998), 337-359

1998
[19]

H. O. Kreiss. ”On diﬀerence approximations of the dissipative ty pe for hyperbolic diﬀerential equations”, Comm. Pure Appl. Math., 17(1964) 335-353 , MR 29#4210. 21

1964
[20]

J. Lee, B. Fornberg. ”A split step approach for the 3-D Maxwe ll’s equations”, J. Comput. Appl. Math., 158(2003), 485-505

2003
[21]

S. K. Lele. ”Compact ﬁnite diﬀerence schemes with spectral-like resolution”, J. Comput. Phys., 103(1992), 16-42

1992
[22]

R. J. LeVeque. ”Intermediate boundary conditions for time sp lit methods applied to hyperbolic partial diﬀerential equations”, Math. Comput., 47(1986), 37-54

1986
[23]

M. Li, T. Tang. ”A compact fourth-order ﬁnite diﬀerence sche me for the steady incompressible Navier- Stokes equations”, Int. J. Numer. Meth. Fluids, 20(1995), 1137 -1151

1995
[24]

M. Li, T. Tang. ”A compact fourth-order ﬁnite diﬀerence sche me for unsteady viscous incompressible ﬂows”, J. Scientiﬁc Computing, 16(2001), 29-45

2001
[25]

R. W. Maccormack. ”The eﬀect of viscosity in hypervelocity impa ct cratering”, AIAA, 69-354 , (1969)

1969
[26]

G. M. Muslu, H. A. Erbay. ”A split-step Fourier method for the c omplex modiﬁed Korteweg-de Vries equation”, Comput. Math. Appl., 45(2003), 503-514

2003
[27]

Ngondiep

E. Ngondiep. ”An eﬃcient numerical approach for solving three -dimensional Black-Scholes equation with stochastic volatility”, Math. Meth. Appl. Sci., (2024), 1-21, DO I 10. 1002/mma.10576

2024
[28]

Ngondiep

E. Ngondiep. ”An eﬃcient high-order two-level explicit/implicit nu merical scheme for two-dimensional time fractional mobile/immobile advection-dispersion model”, Int. J. Numer. Meth. Fluids, 96(8) (2024), 1305-1336

2024
[29]

Ngondiep

E. Ngondiep. ”Stability analysis of MacCormack rapid solver meth od for evolutionary Stokes-Darcy problem”, J. Comput. Appl. Math., 345(2019), 269-285

2019
[30]

Ngondiep

E. Ngondiep. ”A six-level time-split Leap-Frog/Crank-Nicolson approach for two-dimensional nonlinear time-dependent convection-diﬀusion-reaction equation”, Int. J . Comput. Meth., 20(08) (2023), 2250064, Doi: 10. 1142/S0219876222500645

2023
[31]

Ngondiep

E. Ngondiep. ”An eﬃcient three-level explicit time-split scheme f or solving two-dimensional unsteady nonlinear coupled Burgers equations”, Int. J. Numer. Methods Flu ids, 92(4) (2020), 266-284

2020
[32]

Ngondiep

E. Ngondiep. ”Unconditional stability over long time intervals of a two-level coupled MacCormack/Crank-Nicolson method for evolutionary mixed Stoke s-Darcy model”, J. Comput. Appl. Math., 409(2022), 114148, Doi: 10 . 1016/j.cam.2022. 114148

2022
[33]

Ngondiep

E. Ngondiep. ”Long time unconditional stability of a two-level hy brid method for nonstationary incom- pressible Navier-Stokes equations”, J. Comput. Appl. Math. 345( 2019), 501-514

2019
[34]

Ngondiep

E. Ngondiep. ”A high-order numerical scheme for multidimension al convection-diﬀusion-reaction equa- tion with time-fractional derivative”, Numer. Algorithms, 94(2023 ), 681-700

2023
[35]

Ngondiep

E. Ngondiep. ”A high-order combined ﬁnite element/interpolatio n approach for solving nonlinear multidimensional generalized Benjamin-Bona-Mahony-Burgers’ eq uations”, Math. Comput. Simul., 215(2024), 560-755

2024
[36]

Ngondiep, A

E. Ngondiep, A. N. Njomou, G. I. Mondinde. ”A predictor-corr ector approach to investigate and predict the dynamic of cytokine levels and human immune cell activation to Sta phylococcus Aureus”, Int. J. of Biomath., DOI: 10. 1142/S1793524524501122, (2024), 1-29

2024
[37]

Ngondiep

E. Ngondiep. ”Long time stability and convergence rate of MacC ormack rapid solver method for non- stationary Stokes-Darcy problem”, Comput. Math. Appl., 75, (20 18), 3663-3684
[38]

Ngondiep

E. Ngondiep. ”A novel three-level time-split MacCormack sche me for two-dimensional evolutionary linear convection-diﬀusion-reaction equation with source term”, I nt. J. Comput. Math., 98(1) (2021), 47-74. 22

2021
[39]

Ngondiep

E. Ngondiep. ”A robust numerical two-level second-order ex plicit approach to predict the spread of covid-2019 pandemic with undetected infectious cases”, J. Com put. Appl. Math., 403(2022), DOI:10.1016/j.cam.2021.113852

2022
[40]

Ngondiep

E. Ngondiep. ”A posteriori error estimates of MacCormack ra pid solver method for nonstationary in- compressible Navier-Stokes equations”, J. Comput. Appl. Math., 4 38(2024), 115569

2024
[41]

Ngondiep

E. Ngondiep. ”A fast three-step second-order explicit numer ical approach to investigating and forecast- ing the dynamic of corruption and poverty in Cameroon”, Heliyon, DO I: 10. 1016/ j.heliyon.2024.e38236, 10(19) (2024)

2024
[42]

Ngondiep

E. Ngondiep. ”A two-level fourth-order approach for time-f ractional convection-diﬀusion-reaction equa- tion with variable coeﬃcients”, Commun. Nonlinear Sci. Numer. Simul., 1 11(2022), 106444, Doi:

2022
[43]

1016/j.cnsns.2022. 106444

2022
[44]

S. K. Pandit, J. C. Kalita, D. C. Dalal. ”A transient higher order c ompact scheme for incompressible viscous ﬂows on geometries beyond rectangular”, J. Comput. Phy s., 225(2007), 1100-1124

2007
[45]

Sportisse

B. Sportisse. ”An analysis of operator splitting techniques in th e stiﬀ case”, J. Comput. Phys., 161(2000), 140-168

2000
[46]

Stoer, R

J. Stoer, R. Bulirsch. ”Introduction to numerical analysis, te xt in applied mathematics”, second edition, Spring-Verlag, (1991)

1991
[47]

G. Strang. ”On the construction and comparison of diﬀerence schemes”, SIAM J. Numer. Anal., 5(1968), 506-517

1968
[48]

T. R. Taha, M. J. Abowitz. ”Analytical and numerical aspects o f certain nonlinear evolution equations, II. Numerical, nonlinear Schr¨ odinger equation”, J. Comput. Phys ., 55(1984), 231-253

1984
[49]

W. C. Thacker. ”Some exact solutions to the nonlinear shallow wa ter wave equations”, J. Fluid Mech., 107(1981), 499-508

1981
[50]

C. B. Vreugdenhil. ”Numerical method for shaloww water ﬂow”, Kluwer Academic Publishers, Dordrecht (1994)

1994
[51]

P. S. Yu, T. C. Yang, C. M. Kuo, S. T. Chen. ”Development of an integrated computational tool to assess climate change impacts on water supply-demand and ﬂood inu ndation”, J. Hydroinformatics, 16(3) (2014), 710-730

2014
[52]

Zhang, T

W. Zhang, T. W. Cundy. ”Modeling of two-dimensional overland ﬂ ow”, Water Resour. Res., 25(1989), 2019-2035

1989
[53]

Zoppou, S

C. Zoppou, S. Roberts. ”Numerical solution of the two-dimens ional unsteady dam-break”, Appl. Math. Mode., 24(2000), 457-475

2000
[54]

Retrie ved, 2017-06-08

”Logone river/river, Africa”, Encyclopedia Britannica. Retrie ved, 2017-06-08

2017
[55]

”GRDC-Chari basin”, Der Logone in Bongor, 2009

2009
[56]

Retr ieve, 2014-06-10

”Cameroon: Floods-Oct 2013”, Reliefweb, 04 June 2014. Retr ieve, 2014-06-10

2013
[57]

”Unicef Cameroon ﬂash update N o3: Floods-Far North”, 25 September 2024

2024
[58]

”Cameroun: Extrˆ eme-Nord-Aper¸ cu des inondations (au 19 Septembre 2024)-Cameroon—ReliefWeb”

OCHA, ﬂood snapshot as of 19 September. ”Cameroun: Extrˆ eme-Nord-Aper¸ cu des inondations (au 19 Septembre 2024)-Cameroon—ReliefWeb”

2024
[59]

Regional Delegations of Basic and Secondary Education, 15 Sep tember
[60]

Regional Delegation of Public Health of the Far North, presente d on 13 September 2024 in Maroua, Far North

2024
[61]

23 Stability of time-split linearized explicit/implicit for 2d-hydrodynamic ﬂo w

Centre for humdata, based on data from WorldPop, FloodScan , HDX, ECMWF, on 19 September. 23 Stability of time-split linearized explicit/implicit for 2d-hydrodynamic ﬂo w. 0 1 2 3 4 5 6 7 0 0.5 1 1.5 2 2.5 analytical solution (green), computed solution (blue), error (red) exac...

2024

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.

[1] [1]

Abarbaned, A

S. Abarbaned, A. Kumar. ”Compact high-order schemes for th e Euler equations”, J. Scientiﬁc Com- puting, 3(1988), 275-288

work page 1988

[2] [2]

Ablowitz, B

M.J. Ablowitz, B. M. Herbst. ”On the numerical solution of the sine -Gordon equation. 1. Integrable discretizations and homoclinic manifolds”, J. Comput. Phys., 126(19 96), 299-314

work page

[3] [3]

K. O. Aiyesimoju, R. J. Sobey. ”Process splitting of the boundar y conditions for the advection-dispersion equation”, Int. J. Numer. Methods Fluids, 9(1989), 235-244

work page 1989

[4] [4]

R. T. Alqahtani, J. C. Ntonga, E. Ngondiep. ”Stability analysis and convergence rate of a two-step predictor-corrector approach for shallow water equations with s ource terms”, AIMS Mathematics, 8(4) (2023), 9265-9289

work page 2023

[5] [5]

W. Bao, Q. Du. ”Computing the ground state solution of Bose-Ein stein condensates by a normalized gradient ﬂow”, SIAM J. Sci. Comput., 25(2004), 1674-1697

work page 2004

[6] [6]

G. F. Carey, Y. Shen. ”Least-squres ﬁnite element approximat ion of Fisher’s reaction diﬀusion equation”, Numer. Meth. partial Diﬀerential Equations, 11(1995), 175-186

work page 1995

[7] [7]

Demkowicz, T

L. Demkowicz, T. J. Oden, W. Rachowicz. ”A new ﬁnite element met hod for solving compressible Navier-Stokes equations based on an operator splitting method an d h-p adaptivity”, Comput. Methods Appl. Mech. Engng., 84(1990), 275-326

work page 1990

[8] [8]

Descombes, M

S. Descombes, M. Massot. ”Operator splitting for nonlinear rea ction-diﬀusion systems with an entropic structure: singular perturbation and order reduction”, Numer. Math., 97(2004), 667-698

work page 2004