Pith. sign in

REVIEW 3 major objections 6 minor 1 cited by

Structure of undercompressive shock waves in three-phase flow in porous media

T0 review · 3 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read The undercompressive shock surface for capillary-pressure diffusion has the same four boundary types as the identity-diffusion surface, yet the diffusion matrix can remove the undercompressive shock from a Riemann solution.

desk verdict Solid analytic construction of the undercompressive surface for the identity matrix plus a plausible numerical analogue for capillary diffusion, but the topological-similarity claim for B != I is a numerical finding without a completeness certificate. read the letter →

arxiv 2412.04439 v1 pith:YW2DOLBS submitted 2024-12-05 math.AP

L. F. Lozano , I. Ledoino , B. J. Plohr , D. Marchesin This is my paper · ORCID
classification math.AP MSC 35C0635D3076S0576T30
keywords undercompressiveshockwavesthree-phaseflowinporousmediaviscousprofilecriterioncapillarypressureRiemannproblemwavemanifoldsaddle-saddleconnectionnon-strictlyhyperbolicconservationlaws
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 constructs the complete surface of undercompressive shock waves for three-phase flow in porous media, under two diffusion matrices: the identity matrix and the physical capillary-pressure matrix. For the identity matrix the construction is analytic, reducing along invariant lines to a scalar two-phase conservation law and producing explicit formulas for the boundary states. For the capillary matrix the construction is numerical, tracking saddle-to-saddle connections of the traveling-wave ODE. The central claim is that the two surfaces have the same topological structure, with the same four boundary types appearing in the same configurations for each umbilic-point type. The paper then shows, in a concrete Riemann problem, that the capillary matrix can eliminate the undercompressive shock from the solution, so the diffusion model changes which wave sequences are admissible.

What carries the argument

The load-bearing object is the undercompressive surface $T$ inside the wave manifold $W$, the three-dimensional set of shock triples $(U^-, U^+, \sigma)$ satisfying the shock jump condition after a blow-up removes the trivial solutions $U^-=U^+$. For $B(U)=I$, the construction uses the invariant lines along which the three-phase system reduces to a scalar two-phase conservation law; the paper derives the effective flux and explicit boundary-state formulas, making $T$ a ruled surface over the segments $[G,D]$, $[W,E]$, and $[O,B]$. For $B(U)\neq I$, the machinery is the traveling-wave ODE $dU/d\eta = \mathrm{Adj}(B(U))[F(U)-F(U^-)-\sigma(U-U^-)]$ together with a bisection algorithm that measures the signed distance between the stable and unstable manifolds of the two saddle equilibria along a transversal line, locating the saddle-to-saddle and saddle-to-saddle-node connections that form the boundary of $T$.

What would settle it

Run the same continuation for $B(U)\neq I$ seeded from a saddle-saddle connection near the opposite edge of the saturation triangle, away from the umbilic point, and compare the resulting boundary curves with the umbilic-seeded surface; the paper's claim predicts that no new boundary component appears. Alternatively, repeat the construction for an anisotropic capillary matrix with $c_{ow}\neq c_{og}$ and check whether the four boundary types persist.

Watch

Extended reading notes

Core claim

An undercompressive shock is a discontinuity whose viscous profile connects two saddle equilibria of the traveling-wave system, so it satisfies the viscous profile criterion rather than the classical shock inequalities. The paper establishes that the undercompressive surface $T$ inside the wave manifold has the same boundary structure for $B(U)=I$ and for the capillary-pressure matrix (10)-(11): slow characteristic, fast characteristic, undercompressive characteristic, and genuine undercompressive boundaries, arranged in the same patterns for umbilic points of type I and type II and for each viscosity-ratio regime. For $B(U)=I$, $T$ is a ruled surface lying over the invariant lines $[G,D]$, $[W,E]$, and $[O,B]$, with explicit formulas for the boundary states $D_0$, $D_1$, $D_2$, $Y_\Gamma$, and $Y^\Gamma$; for $B(U)\neq I$, the surface is found numerically and projects to two-dimensional regions of the saturation triangle. The same Riemann problem with states $L=(0.475708, 0.02608)^T$ and $R=(0.235578, 0.670876)^T$ produces different solutions: with $B(U)=I$ it contains an undercompressive shock, while with the capillary matrix the undercompressive shock is absent and a different fast shock segment becomes admissible.

Load-bearing premise

The numerical claim that the capillary-pressure surface has the same structure as the identity surface rests on the search finding the entire undercompressive surface, but the search is seeded from one saddle-saddle connection near the umbilic point and continued over a grid, with no proof that a disconnected component of the surface cannot be missed.

Editorial extensions

If this is right

  • Riemann solvers can reuse the same four boundary-type logic for both diffusion models, only replacing the analytic surface by the numerically computed one.
  • For the capillary matrix, undercompressive shocks occupy a two-dimensional region of the saturation triangle rather than one-dimensional invariant-line segments, so capillary diffusion broadens the set of left and right states that can be joined by an undercompressive shock.
  • The admissibility of nonlocal shock segments changes with the diffusion matrix: a segment that is only partly admissible for $B=I$ can become fully admissible for the capillary matrix.
  • There exist Riemann problems whose solution contains an undercompressive shock for $B=I$ but not for the capillary matrix, so the viscous profile criterion ties the wave sequence to the physical diffusion model.
  • For viscosity ratio $\nu_\Gamma\leq 1$ the undercompressive surface has a gap between $U$ and $B_0$, so undercompressive shock amplitude is bounded away from zero and the gap is filled by transitional rarefaction waves.

Reading between the lines

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

  • Because the boundary types are carried by saddle-to-saddle and saddle-to-saddle-node connections, which are structurally stable, the same four-boundary taxonomy is likely to persist for other diffusion matrices in the same physical class; this is an inference beyond the paper's two matrices.
  • A direct test of exhaustiveness would seed the numerical continuation from each of the 31 umbilic-region configurations and from boundary states away from the umbilic point; the paper's claim predicts the same boundary-type counts in every region.
  • If the capillary matrix can remove the undercompressive shock in a single Riemann problem, then in heterogeneous reservoir simulations models with $B(U)=I$ may over-predict non-classical waves; computing the Riemann solution for the same $L$, $R$ with the two matrices at finite grid resolution would make this quantitative.
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.

Referee Report

3 major / 6 minor

Summary. The paper studies undercompressive shock waves for a 2x2 system of conservation laws modeling three-phase flow in porous media with quadratic Corey relative permeabilities. For the identity diffusion matrix B(U)=I, the authors exploit invariant lines to reduce the system to scalar Buckley-Leverett equations, derive explicit formulas for the undercompressive interval endpoints and for the characteristic boundaries, and construct the undercompressive shock surface T as a ruled surface in the three-dimensional space of saturations and speed. For the capillary-pressure diffusion matrix B(U)≠I, they compute saddle-saddle connections numerically by continuation and bisection, and they claim that the resulting undercompressive surface has the same boundary structure as in the identity case. The paper closes with numerical simulations of a Riemann problem showing that the choice of B changes the admissibility of the nonlocal shock segment and can eliminate the undercompressive shock from the solution.

Significance. If the results hold, the paper provides the first systematic construction of the undercompressive shock surface for a physically motivated capillary diffusion matrix, alongside an explicit analytical construction for the identity matrix. The analytic part of the paper is a genuine strength: Sections 4--6 contain closed-form expressions for the endpoints S, F, B0, B1, B2, YΓ, and YΓ, and the reduction along invariant lines is derived from the governing equations rather than assumed. The paper also makes a concrete, falsifiable prediction: that the shape of the undercompressive surface is insensitive to the choice of diffusion matrix, while the solution of the Riemann problem can change qualitatively. However, the central structural claim for B≠I rests on numerical continuation from a single seed, with no completeness certificate, and one key analytic claim is justified by a figure rather than a proof. The absence of code or data also prevents independent verification of the numerical experiments.

major comments (3)
  1. [§7.3] The central claim that 'when calculating the undercompressive boundaries with B(U) ≠ I, we obtain the same boundaries as those discussed for B(U) = I' is supported only by 'extensive numerical experiments.' The construction in §7.1 starts from one saddle-saddle connection near the umbilic point, obtained from the quadratic approximation in [4,16], and then continues it over a grid using the bisection procedures of Algorithms 1--3. Such a continuation can only discover the connected component of the undercompressive surface T that contains the seed; nothing in the paper rules out additional disconnected components, or holes bounded by different boundary types, away from the umbilic neighborhood. If such a component existed, the boundary count and the claimed topological similarity in the four cases enumerated in §7.3 would fail. This is not an internal inconsistency, but the evidence presented does not determine the global structure of T. The authors should either provide a completeness argument for the continuation, demonstrate coverage by multiple independent seeds and sweeping strategies, or soften the conclusion to a conjecture. In addition, providing the code/data used for Figures 12--14 would allow independent checking.
  2. [§6, Claim 6.1] Claim 6.1, which states that for 0 < νG ≤ 1 there is no left state L on the invariant line [G,D) forming an undercompressive shock with a right state R in (U, D0], is justified by inspection of the speed diagram in Figure 9: the proof says 'According to Fig. 9' and identifies the Bethe-Wendroff point S and the states F and F1 graphically. This claim is load-bearing because it establishes the gap between U and D0 and thereby determines the boundary structure in Lemma 6.3 and in the analogous numerical case 2 of §7.3. A figure-based argument is not a proof, especially since the same figure is used to assert that no orbit connects M to R along the invariant line. The authors should supply an analytic proof, or at least a machine-checkable computation, of the ordering and the absence of an admissible connection.
  3. [§4.3 and Remark 6.3] The paper states that 'numerical solid evidence shows that the mixed double contact locus exists only when the umbilic point lies in one of the regions of type II' and then uses the position of XG to decide whether a compatibility-loss boundary appears (Remark 6.3, Figure 8(b)). This assertion is not proven, and it affects the boundary classification for the identity case as well as the interpretation of the non-identity case. If the 'only when' statement is not known analytically, it should be marked as a numerical observation, and its role in the boundary classification should be made explicit.
minor comments (6)
  1. [§3] The sentence 'the Rankine-Hugoniot condition constitutes two equations for the five variables U+, U+, and σ' contains a typo: it should read U-, U+, and σ.
  2. [§6, Definition 6.1] Definition 6.1 says 'the set of pairs of states pS and xM in Ω × R+', but pS and xM are triples in Ω × R+, not pairs. Rephrase to 'the set of points (U-, σ) and (U+, σ) in Ω × R+' or the equivalent.
  3. [§5, Corollary 5.1] The phrase 'For νΓ > 8 por (νΓ−)2/νΓ > 8' contains a typo: 'por' should be 'or'.
  4. [Appendix A] The algorithms rely on a 'Sotomayor line s', which is described only as 'a line conveniently placed in the phase space'. A precise definition or a reference for the Sotomayor line would make the numerical procedure reproducible.
  5. [Figure 12] The caption of Figure 12(b) mentions the z axis but does not state its meaning; the text identifies it as shock speed, so the caption should be updated accordingly.
  6. [§4.3] The notation νΓ− is introduced in Definition 4.2 and used later as (νΓ−)2/νΓ, but the connection between the superscript '−' and the difference μα−μβ could be made more prominent, since it is easy to confuse with a negative exponent.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the B=I surface is derived analytically from the flux and Rankine-Hugoniot relations, and the B≠I surface is computed by independent numerical continuation from the same ODE; the similarity claim is an empirical comparison, not an input.

full rationale

The paper's derivation chain is not circular. For B(U)=I, the undercompressive surface is constructed analytically: Theorem 5.1 and Theorem 5.2 derive the undercompressive intervals along the invariant lines from the reduced Buckley-Leverett flux (31), the Rankine-Hugoniot condition (35), and the characteristic speed formulas (36); Lemmas 6.1-6.3 then give the boundary types directly from Definition 6.1 and these theorems. The cited uniqueness result that for B=I only invariant-line undercompressive shocks exist is a parameter-free prior theorem ([9,34]) whose assumptions do not include the paper's central comparison for B≠I, so it counts as independent support rather than a self-referential premise. For B(U)≠I, the undercompressive surface is not fitted to the identity answer: Section 7.1 seeds the computation with a local saddle-saddle connection near the umbilic point using the quadratic-approximation results [4,16], and Algorithms 1-3 then continue the connection by bisection on the actual traveling-wave ODE (16) with the capillary matrix. The four boundary types (SCB, FCB, UCB, GUB) are not imported from the identity case as an ansatz; they are the generic ways a saddle-saddle connection can cease to exist (saddle-node transitions and domain-boundary cases), and the numerical procedure detects these cases from the ODE. The statement in Section 7.3 that the same boundaries are obtained is an observed outcome of those computations, not an assumption of the algorithm. Finally, the Riemann-solution comparison in Section 7.4 is corroborated by direct numerical simulation (RCD), so the 'disappearance' of the undercompressive shock under the capillary matrix is an externally checked result. The concern that the single-seed continuation might miss disconnected components of the undercompressive surface is a completeness/robustness issue for the numerical evidence, not a circularity: nothing in the construction defines the B≠I surface in terms of the B=I surface. Accordingly, no circular step is exhibited, and the correct circularity score is 0.

Assumptions & free parameters 5 free parameters · 6 assumptions · 0 invented entities

The paper introduces no new physical entities. Its central claim rests on several prior results from the same research group (structure of Hugoniot loci, quadratic umbilic approximation, structural stability of undercompressive shocks, invariant-line reduction), which are cited but not re-derived. The parameters mu_w, mu_o, mu_g, cow, and cog are model inputs chosen by hand for the numerical example, not fitted to data.

free parameters (5)
  • water viscosity mu_w = 1.0
    Chosen for the numerical example in Section 7.4; the structural claims are asserted across parameter ranges, so this single choice is load-bearing for the empirical part of the paper.
  • oil viscosity mu_o = 2.0
    Chosen for the numerical example in Section 7.4; the structural claims are asserted across parameter ranges, so this single choice is load-bearing for the empirical part of the paper.
  • gas viscosity mu_g = 0.75
    Chosen for the numerical example in Section 7.4; the structural claims are asserted across parameter ranges, so this single choice is load-bearing for the empirical part of the paper.
  • capillary constant cow = 1
    Appears in the capillary pressure matrix P1 (11); set to 1 in the Section 7.4 simulations, with no sensitivity study.
  • capillary constant cog = 1
    Appears in the capillary pressure matrix P1 (11); set to 1 in the Section 7.4 simulations, with no sensitivity study.
assumptions (6)
  • domain assumption For B=I, the only admissible non-Lax shocks are heteroclinic orbits lying on the invariant lines [G,D], [W,E], [O,B].
    Invoked in Section 4 to restrict the analytic construction of T to invariant lines; cited from [9] and [34] and not re-proven here.
  • domain assumption The capillary pressure matrix B(U)=Q(U)P1(U) is symmetric positive definite in the interior of Omega with det B=0 on the boundary.
    Used to define the rescaled traveling-wave ODE (16) and guarantee the saddle structure; quoted from Azevedo et al. [21].
  • standard math Near the umbilic point, the quadratic flux approximation admits undercompressive shocks, and these are structurally stable under perturbations of the viscosity matrix.
    Seeds the numerical search in Section 7.1; results from [4,16,20] are used without derivation.
  • ad hoc to paper The mixed double contact locus exists only when the umbilic point lies in type II regions.
    Stated in Section 4 as 'Numerical solid evidence shows...' without proof or citation; it shapes the region classification in Fig. 4(c).
  • domain assumption The Hugoniot locus of a state D on an invariant line consists of the three straight segments [O,W], [G,D], and [E,B].
    Used in the proof of Lemma 4.2 and in Claim 6.1; cited from Lemma 4.8 of [29].
  • domain assumption The viscous profile criterion is the correct shock admissibility rule.
    Underlies the definition of undercompressive shocks and all constructions in the paper.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Structure of undercompressive shock waves in three-phase flow in porous media." pith.science (2026). https://pith.science/paper/YW2DOLBS

@misc{pith2026241204439,
  author       = {Pith},
  title        = {Pith review of: Structure of undercompressive shock waves in three-phase flow in porous media},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YW2DOLBS}},
  note         = {Machine review of arXiv:2412.04439}
}
read the original abstract

Undercompressive shocks are a special type of discontinuities that satisfy the viscous profile criterion rather than the Lax inequalities. These shocks can appear as a solution to systems of two or more conservation laws. This paper presents the construction of the undercompressive shock surface for two types of diffusion matrices. The first type is the identity matrix. The second one is the capillarity matrix associated with the proper modeling of the diffusive effects caused by capillary pressure. We show that the structure of the undercompressive surface for the different diffusion matrices is similar. We also show how the choice of the capillarity matrix influences the solutions to the Riemann problem.

Figures

Figures reproduced from arXiv: 2412.04439 by the authors.

Figure 1
Figure 1. Saturation triangle GW O. The invariant lines [G, D], [W, E], and [O, B]. The state U represents the umbilic point of type IIO in this work; see Corollary 4.1. initial conditions Upx, 0q “ # UL, x ă 0, UR, x ą 0, (5) where UL and UR are constant. In some cases, we use L and R. Scale invariance is a property of Riemann problems (3), (5). It means that a coordinate change px, tq Ñ pcx, ctq, with c ą 0, changes neither… view at source ↗
Figure 2
Figure 2. Graphs for expressions λa (black curve) and λb (purple curve) above the white region. The dashed (resp. solid) line above the shaded region is the slow (resp. fast) characteristic speed λs (resp. λf ) defined in (37). The horizontal axis corresponds to a parametrization of an invariant line in terms of effective saturation s, and the vertical axis is characteristic speed. The states sU , sYΓ and sYΓ are given in (38… view at source ↗
Figure 3
Figure 3. Bifurcation curves in state space that contain the states [PITH_FULL_IMAGE:figures/full_fig_p018_3.png] view at source ↗
Figures from the paper (13 more)
Figure 4
Figure 4. Figure 4: Position of the umbilic point U. (a) Classification of U as type I or II given in Corollary 4.1. (b) Region associated with vertex G where the location of U imply the existence of the reduced states sYG and sYG . (c) Region for U in which XG is relevant for the invaria…
Figure 5
Figure 5. Figure 5: Typical example of a Hugoniot locus from [PITH_FULL_IMAGE:figures/full_fig_p023_5.png]
Figure 6
Figure 6. Figure 6: The surface of undercompressive shocks for the case [PITH_FULL_IMAGE:figures/full_fig_p032_6.png]
Figure 7
Figure 7. Figure 7: Undercompressive boundaries in Definition [PITH_FULL_IMAGE:figures/full_fig_p035_7.png]
Figure 8
Figure 8. Figure 8: Undercompressive boundaries in Definition [PITH_FULL_IMAGE:figures/full_fig_p036_8.png]
Figure 9
Figure 9. Figure 9: Speed diagram for HpRq, R P pU, D0s. The horizontal axis corresponds to a parametrization of effective saturation, and the vertical axis shows speeds. The blue (resp. red) line is the characteristic speed λs (resp. λf ), while the black line is the shock speed σ. Horiz…
Figure 10
Figure 10. Figure 10: Explanation to find connections between saddles. [PITH_FULL_IMAGE:figures/full_fig_p040_10.png]
Figure 11
Figure 11. Figure 11: Schematic procedure to find connections between saddle-node and saddle. [PITH_FULL_IMAGE:figures/full_fig_p041_11.png]
Figure 12
Figure 12. Figure 12: Surfaces of undercompressive shocks for umbilic point of type [PITH_FULL_IMAGE:figures/full_fig_p043_12.png]
Figure 13
Figure 13. Figure 13: Projection of the surface of undercompressive shocks on [PITH_FULL_IMAGE:figures/full_fig_p045_13.png]
Figure 14
Figure 14. Figure 14: Projection of the surface of undercompressive shocks on the plane [PITH_FULL_IMAGE:figures/full_fig_p046_14.png]
Figure 15
Figure 15. Figure 15: Solution of the Riemann problem for left state [PITH_FULL_IMAGE:figures/full_fig_p047_15.png]
Figure 16
Figure 16. Figure 16: Numerical solution of Riemann problem for left state [PITH_FULL_IMAGE:figures/full_fig_p048_16.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Displacement of three-phase flow for Heavy Oil: Riemann Solutions

    math.AP 2025-06 conditional novelty 6.0 of 10

    The paper classifies all Riemann solutions for three-phase flow with heavy oil, where oil viscosity dominates, for left states on the water-gas edge and right states across nearly the whole saturation triangle.

Reference graph

Works this paper leans on

54 extracted references · 51 canonical work pages · cited by 1 Pith paper

  1. [1]

    Abreu, Numerical modelling of three-phase immiscible flow in hetero- geneous porous media with gravitational effects, Math

    E. Abreu, Numerical modelling of three-phase immiscible flow in hetero- geneous porous media with gravitational effects, Math. Comput. Simu- lation 97 (2014) 234–259.doi:10.1016/j.matcom.2013.09.010

    work page doi:10.1016/j.matcom.2013.09.010 2014

  2. [2]

    Abreu, J

    E. Abreu, J. Douglas, F. Furtado, D. Marchesin, F. Pereira, Three-phase immiscible displacement in heterogeneous petroleum reservoirs, Math. Comput. Simulation 73 (1) (2006) 2–20.doi:10.1016/j.matcom.2006. 06.018

    work page doi:10.1016/j.matcom.2006 2006

  3. [3]

    Marchesin, B

    D. Marchesin, B. J. Plohr, Wave structure in WAG recovery, SPE Jour- nal 6 (02) (2001) 209–219.doi:10.2118/71314-PA

    work page doi:10.2118/71314-pa 2001

  4. [4]

    E. L. Isaacson, D. Marchesin, B. J. Plohr, Transitional waves for con- servation laws, SIAM J. Math. Anal. 21 (4) (1990) 837–866. doi: 10.1137/0521047. 54

    work page doi:10.1137/0521047 1990

  5. [5]

    Mehrabi, K

    M. Mehrabi, K. Sepehrnoori, M. Delshad, Solution construction to a class of Riemann problems of multiphase flow in porous media, Transp. PorousMedia132(2020)241–266. doi:10.1007/s11242-020-01389-x

    work page doi:10.1007/s11242-020-01389-x 2020

  6. [6]

    A. T. Corey, C. H. Rathjens, J. H. Henderson, M. R. J. Wyllie, Three- phase relative permeability, J. Pet. Technol. 8 (11) (1956) 63–65

    work page 1956

  7. [7]

    Isaacson, D

    E. Isaacson, D. Marchesin, B. Plohr, B. Temple, The Riemann problem near a hyperbolic singularity: the classification of solutions of quadratic riemann problems I, SIAM J. Appl. Math. 48 (5) (1988) 1009–1032. doi:10.1137/0148059

    work page doi:10.1137/0148059 1988

  8. [8]

    Castaneda, D

    P. Castaneda, D. Marchesin, F. Furtado, On singular points for con- vex permeability models, in: 14th International Conference: Theory, Numerics and Applications of Hyperbolic Problems, 2012, pp. 415–422

    work page 2012

Show all 54 references
  1. [9]

    Isaacson, D

    E. Isaacson, D. Marchesin, B. Plohr, J. B. Temple, Multiphase flow mod- els with singular Riemann problems, Mat. Apl. Comput 11 (2) (1992) 147–166

    1992

  2. [10]

    Matos, P

    V. Matos, P. Castañeda, D. Marchesin, Classification of the umbilic point in immiscible three-phase flow in porous media, in: Proceedings of the 14th International Conference on Hyperbolic Problems: Theory, Numerics, Applications (Padova, Italy), 2012, pp. 791–799

    2012

  3. [11]

    D. G. Schaeffer, M. Shearer, The classification of 2ˆ 2 systems of non- strictly hyperbolic conservation laws, with application to oil recovery, Comm. Pure Appl. Math. 40 (2) (1987) 141–178. doi:10.1002/cpa. 3160400202

    1987 doi

  4. [12]

    J. B. Bell, J. A. Trangenstein, G. R. Shubin, Conservation laws of mixed type describing three-phase flow in porous media, SIAM J. Appl. Math. 46 (6) (1986) 1000–1017.doi:10.1137/0146059

    1986 doi

  5. [13]

    Fayers, Extension of Stone’s method 1 and conditions for real char- acteristics in three-phase flow, SPE Reservoir Engineering 4 (04) (1989) 437–445

    F. Fayers, Extension of Stone’s method 1 and conditions for real char- acteristics in three-phase flow, SPE Reservoir Engineering 4 (04) (1989) 437–445. doi:10.2118/16965-PA

    1989 doi

  6. [14]

    Holden, On the strict hyperbolicity of the Buckley–Leverett equations for three-phase flow in a porous medium, SIAM J

    L. Holden, On the strict hyperbolicity of the Buckley–Leverett equations for three-phase flow in a porous medium, SIAM J. Appl. Math. 50 (3) (1990) 667–682. doi:10.1137/0150039. 55

    1990 doi

  7. [15]

    Holden, On the Riemann problem for a prototype of a mixed type conservation law, Comm

    H. Holden, On the Riemann problem for a prototype of a mixed type conservation law, Comm. Pure Appl. Math. 40 (2) (1987) 229–264.doi: 10.1002/cpa.3160400206

    1987 doi

  8. [16]

    A. V. Azevedo, D. Marchesin, Multiple viscous solutions for systems of conservation laws, Trans. Amer. Math. Soc. 347 (8) (1995) 3061–3077. doi:10.1090/S0002-9947-1995-1277093-8

    1995 doi

  9. [17]

    Matos, D

    V. Matos, D. Marchesin, Large viscous solutions for small data in sys- tems of conservation laws that change type, J. Hyperbolic Differ. Equ. 5 (02) (2008) 257–278.doi:10.1142/S0219891608001477

    2008 doi

  10. [18]

    A. G. Kulikovskii, Surfaces of discontinuity separating two perfect me- dia of different properties. Recombination waves in magnetohydrody- namics, J. Appl. Math. Mech. 32 (6) (1968) 1145–1152.doi:10.1016/ 0021-8928(68)90046-4

    1968

  11. [19]

    A. A. Mailybaev, D. Marchesin, Dual-family viscous shock waves in systems of conservation laws: a surprising example, in: Proc. Conf. on Analysis, Modeling and Computation of PDE and Multiphase Flow, 2004, pp. 125–135

    2004

  12. [20]

    Marchesin, A

    D. Marchesin, A. A. Mailybaev, Dual-family viscous shock waves in n conservation laws with application to multi-phase flow in porous media, Arch. Ration. Mech. Anal. 182 (2006) 1–24. doi:10.1007/ s00205-005-0419-9

    2006

  13. [21]

    A. V. Azevedo, D. Marchesin, B. Plohr, K. Zumbrun, Capillary instabil- ity in models for three-phase flow, Z. fur Angew. Math. Phys. 53 (2002) 713–746. doi:10.1007/s00033-002-8180-5

    2002 doi

  14. [22]

    Majda, R

    A. Majda, R. L. Pego, Stable viscosity matrices for systems of con- servation laws, J. Differ. Equ. 56 (2) (1985) 229–262. doi:10.1016/ 0022-0396(85)90107-X

    1985

  15. [23]

    E. L. Issacson, D. Marchesin, C. Frederico Palmeira, B. J. Plohr, A global formalism for nonlinear waves in conservation laws, Commun. Math. Phys. 146 (3) (1992) 505–552. 56

    1992

  16. [24]

    Marchesin, C

    D. Marchesin, C. F. B. Palmeira, Topology of elementary waves for mixed-type systems of conservation laws, J. Dynam. Differ. Equ. 6 (1994) 427–446

    1994

  17. [25]

    Azevedo, C

    A. Azevedo, C. Eschenazi, D. Marchesin, C. Palmeira, Topological reso- lution of Riemann problems for pairs of conservation laws, Quart. Appl. Math. 68 (2) (2010) 375–393

    2010

  18. [26]

    Guckenheimer, P

    J. Guckenheimer, P. Holmes, Nonlinear Oscillations, Dynamical Sys- tems, and Bifurcations of Vector Fields, Springer-Verlag, New York, 1986

    1986

  19. [27]

    ELI, Interactive Graphical Riemann Problem Solver., https://eli.fluid.impa.br/, Accessed : 2024-12-04

    2024

  20. [28]

    A. V. Azevedo, A. J. de Souza, F. Furtado, D. Marchesin, B. Plohr, The solution by the wave curve method of three-phase flow in virgin reservoirs, Transp. Porous Media 83 (1) (2010) 99–125.doi:10.1007/ s11242-009-9508-9

    2010

  21. [29]

    A. V. Azevedo, A. J. de Souza, F. Furtado, D. Marchesin, Uniqueness of the Riemann solution for three-phase flow in a porous medium, SIAM J. Appl. Math. 74 (6) (2014) 1967–1997.doi:10.1137/140954623

    2014 doi

  22. [30]

    Lozano, Diffusive effects in Riemann solutions for the three phase flow in porous media, Ph.D

    L. Lozano, Diffusive effects in Riemann solutions for the three phase flow in porous media, Ph.D. thesis, Instituto Nacional de Matemática Pura e Aplicada, IMPA (2018)

    2018

  23. [31]

    Andrade, A

    P. Andrade, A. de Souza, F. Furtado, D. Marchesin, Three-phase fluid displacements in a porous medium, J. Hyperbolic Differ. Equ. 15 (04) (2018) 731–753. doi:10.1142/S0219891618500236

    2018 doi

  24. [32]

    Shearer, D

    M. Shearer, D. G. Schaeffer, D. Marchesin, P. L. Paes-Leme, Solution of the Riemann problem for a prototype 2ˆ 2 system of non-strictly hyperbolic conservation laws, Arch. Ration. Mech. Anal. 97 (1987) 299–

    1987

  25. [33]

    H. B. Mederios, Stable hyperbolic singularities for three-phase flow mod- els in oil reservoir simulation, Acta Appl. Math. 28 (2) (1992) 135–159. doi:10.1007/BF00047553. 57

    1992 doi

  26. [34]

    A. J. de Souza, Stability of singular fundamental solutions under per- turbations for flow in porous media, Mat. Apl. Comput. 11 (2) (1992) 43

    1992

  27. [35]

    P. D. Lax, Hyperbolic systems of conservation laws II, Comm. Pure Appl. Math. 10 (4) (1957) 537–566.doi:10.1002/cpa.3160100406

    1957 doi

  28. [36]

    Smoller, Shock waves and reaction—diffusion equations, Vol

    J. Smoller, Shock waves and reaction—diffusion equations, Vol. 258, Springer Science & Business Media, 2012

    2012

  29. [37]

    Liu, The Riemann problem for general systems of conserva- tion laws, J

    T.-P. Liu, The Riemann problem for general systems of conserva- tion laws, J. Differ. Equ. 18 (1) (1975) 218–234. doi:10.1016/ 0022-0396(75)90091-1

    1975

  30. [38]

    I. M. Gel’fand, Some problems in the theory of quasilinear equation, Usp. Math. Nauk. 14 (1959) 87–158, Eng. Trans. in Amer. Math. Soc. Trans. Ser. 2, 29, 295-381 (1963)

    1959

  31. [39]

    Petrova, B

    Y. Petrova, B. J. Plohr, D. Marchesin, Vanishing adsorption limit of riemann problem solutions for the polymer model, Journal of Hyper- bolic Differential Equations 21 (02) (2024) 299–327. doi:10.1142/ S0219891624500103

    2024

  32. [40]

    Schecter, D

    S. Schecter, D. Marchesin, B. J. Plohr, Structurally stable Riemann solutions, J. Differ. Equ. 126 (2) (1996) 303–354.doi:10.1006/jdeq. 1996.0053

    1996

  33. [41]

    M. E. S. Gomes, Riemann problems requiring a viscous profile entropy condition, Adv. Appl. Math. 10 (3) (1989) 285–323. doi:10.1016/ 0196-8858(89)90016-X

    1989

  34. [42]

    B. L. Keyfitz, H. C. Kranzer, A system of non-strictly hyperbolic conser- vation laws arising in elasticity theory, Arch. Ration. Mech. Anal. 72 (3) (1980) 219–241

    1980

  35. [43]

    A. J. de Souza, D. Marchesin, Conservation laws possessing contact characteristic fields with singularities, Acta Appl. Math. 51 (1998) 353–

    1998

  36. [44]

    A. I. Volpert, V. A. Volpert, V. A. Volpert, Traveling wave solutions of parabolic systems, Vol. 140, American Mathematical Soc., 1994. 58

    1994

  37. [45]

    Matos, A

    V. Matos, A. V. Azevedo, J. C. Da Mota, D. Marchesin, Bifurcation under parameter change of Riemann solutions for nonstrictly hyperbolic systems, Z. fur Angew. Math. Phys. 66 (4) (2015) 1413–1452. doi: 10.1007/s00033-014-0469-7

    2015 doi

  38. [46]

    Ole˘ ınik, On the uniqueness of the generalized solution of the Cauchy problem for a nonlinear system of equations occurring in mechanics, Usp

    O. Ole˘ ınik, On the uniqueness of the generalized solution of the Cauchy problem for a nonlinear system of equations occurring in mechanics, Usp. Mat. Nauk 12 (1957) 169–176

    1957

  39. [47]

    Castañeda, E

    P. Castañeda, E. Abreu, F. Furtado, D. Marchesin, On a universal struc- tureforimmisciblethree-phaseflowinvirginreservoirs, Comput.Geosci. 20 (2016) 171–185. doi:10.1007/s10596-016-9556-5

    2016 doi

  40. [48]

    F. Asakura, Stone-marchesin model equations of three-phase flow in oil reservoir simulation, in: Proceedings of the 14th International Confer- ence on Hyperbolic Problems: Theory, Numerics, Applications (Padova, Italy), 2012, pp. 1–56

    2012

  41. [49]

    Wendroff, The Riemann problem for materials with nonconvex equa- tions of state i: Isentropic flow; ii general flow, J

    B. Wendroff, The Riemann problem for materials with nonconvex equa- tions of state i: Isentropic flow; ii general flow, J. Math. Anal. Appl. 38 (2) (1972) 454–466.doi:10.1016/0022-247X(72)90103-5

    1972 doi

  42. [50]

    P. L. Andrade, A. J. de Souza, F. Furtado, D. Marchesin, Oil dis- placement by water and gas in a porous medium: the Riemann prob- lem, Bull. Braz. Math. Soc. (N.S.) 47 (2016) 77–90. doi:10.1007/ s00574-016-0123-4

    2016

  43. [51]

    Lambert, A

    W. Lambert, A. Alvarez, I. Ledoino, D. Tadeu, D. Marchesin, J. Bruin- ing, Mathematics and numerics for balance partial differential-algebraic equations (pdaes), J. Sci. Comput. 84 (2) (2020) 29. doi:10.1007/ s10915-020-01279-w

    2020

  44. [52]

    L. F. L. Guerrero, D. Marchesin, Diffusive Riemann solutions for 3- phase flow in porous media, Proceeding Series of the Brazilian Society of Computational and Applied Mathematics 7 (1) (2020).doi:10.5540/ 03.2020.007.01.0372. 59

    2020

  45. [320]

    doi:10.1007/BF00280409

    doi

  46. [364]

    doi:10.1023/A:1005928309554

    doi

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.