{"id":"453b7f16-59fb-4e02-9c55-3cc1b1ad25b3","arxiv_id":"2508.21599","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"CDHD is a central flux plus high-order DOTRS dissipation for primitive-variable Euler updates, giving second-order accuracy with O(Δx^4) dissipation for transcritical flows.","lead":"This paper proposes a new numerical flux for primitive-variable simulations of transcritical flows, combining a central flux with a small upwind dissipation term. If correct, it offers a simple way to suppress spurious pressure oscillations while keeping formal second-order accuracy and low energy error.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Dissipation subdominance relies on dV=O(Δx^5); the PVRS sensor does not flag contact discontinuities, so cells across a contact use the primitive CDHD update where dV=O(1), making dissipation dominant and the scheme first-order there.","rationale":"The reader flagged the C^2/eigenstructure assumptions in Section 4.3, but those are not the most load-bearing: the O(dV) bound on H^± holds for any bounded Jacobian, so even without smooth eigenstructure the dissipation scaling H^±=O(dV) remains. The critical assumption is dV=O(Δx^5) in the cells where CDHD is applied. The PVRS sensor in (18) only detects shocks via pressure jumps, not contacts or under-resolved smooth gradients, so in those common cases dV is O(1) and the dissipation is no longer subdominant. The paper does not quantitatively validate the shock tube or a sharp-interface case, so this failure mode is unaddressed. The convergence test and smooth advection case are fine for their regime, but they do not exercise the failure mode. A targeted shock-tube comparison against an exact/reference solution would settle whether the hybrid framework actually preserves the claimed robustness for the full wave structure. This is a genuine gap, but not proof of failure; hence the verdict remains CONDITIONAL rather than REJECT.","tokens_in":18295,"tokens_out":13962,"duration_ms":159466,"concrete_test":"Rerun the Section 5.2 shock tube (pressures 60/6 MPa, densities 800/80 kg/m³) to t=5×10^-4 s and compare the CDHD hybrid density, velocity, and pressure profiles against an exact real-gas Riemann solution or a fully conservative WENO-5+Roe reference on a much finer grid (e.g., 4096 cells). Measure the contact width in cells and the L1 error around the contact. If the contact is smeared over more than ~2-3 cells or the L1 error is substantially higher than the conservative reference, the primitive update is operating at dV=O(1), confirming that the O(Δx^4) dissipation and second-order accuracy do not hold for the full hybrid framework.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central accuracy claim (Section 4.3, Eq. 41) depends on the interface jump dV = V_R - V_L being O(Δx^5) in every cell updated with the primitive CDHD scheme. This warrants the dissipation term L_v^diss = O(Δx^4) and a second-order overall truncation error. However, the hybrid sensor (18) buffers cells to the conservative update only when the pressure ratio exceeds 1+ε. It does not detect contact discontinuities, where pressure is continuous but density jumps. In a shock tube, cells adjacent to the contact have dV=O(1); then H^± = O(1) and L_v^diss = O(1/Δx), so the dissipation is dominant, not subdominant, and the scheme is first-order near the contact. The shock tube test (Section 5.2) presents only profiles with no exact or reference solution, leaving contact smearing unquantified. Likewise, the 2D droplet test uses a resolved tanh profile (η=0.05, ≈7 cells across the interface); sharper physical interfaces would fall into the same regime. Thus the claim that the hybrid framework robustly handles shock waves is not established for one of the standard wave structures.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a hybrid numerical method for transcritical real-gas flows. In smooth regions, a primitive-variable update is used, based on a new CDHD flux: a central difference of WENO-5 reconstructed interface states plus an upwind-biased dissipation term computed via Dumbser-Osher-Toro path integrals. Near shock waves, a PVRS-based sensor switches cells to a conservative WENO-5/Roe update. A truncation-error analysis argues that the central term is second-order accurate and the dissipation is O(dx^4) when the interface jump scales as O(dx^5), so the overall scheme is second-order in smooth regions. Numerical tests include 1D advection with convergence rates ~2.17, an energy-conservation comparison against first-order DOTRS, a 1D shock tube without a reference solution, and a 2D droplet advection problem.","tokens_in":18668,"tokens_out":5751,"duration_ms":62993,"significance":"If the claims hold, CDHD provides a simple primitive-variable flux with formal second-order accuracy and subdominant dissipation, which would be a useful building block for transcritical-flow simulation. The error analysis is explicit and the smooth-region convergence test is consistent with the claimed order. However, the validation is incomplete in several load-bearing respects, and the hybrid sensor has a gap for contact discontinuities that directly affects both the error analysis and the robustness claims. The paper is a plausible contribution to a specialized computational-fluid-dynamics venue, but the central claims need reinforcement before acceptance.","major_comments":[{"comment":"The error analysis for the dissipation term assumes dV = O(dx^5) in every cell where the primitive CDHD update is applied. The PVRS sensor (18) switches to the conservative update only when the pressure ratio exceeds 1+epsilon; it does not detect contact discontinuities, where pressure is continuous but density jumps. Across a contact, dV = O(1), so H^± = O(1) and L_v^diss = O(1/dx), making the dissipation dominant, not subdominant. This contradicts the statement in Section 4.1 that the primitive update is used only where the O(dx^5) scaling holds. The shock-tube test (Section 5.2) includes a contact but does not provide a quantitative reference solution, so the behavior near contacts is unverified. This gap directly affects the hybrid method's claimed robustness for standard wave structures.","section":"Section 3.3, Eq. (18) and Section 4.3, Eq. (41)"},{"comment":"The convergence study evaluates only the pure CDHD scheme on a smooth advection problem. No convergence study is presented for the hybrid scheme, nor for cases that involve discontinuities or cell switching. Since the sensor does not flag contacts, the hybrid scheme will apply the primitive update across contact interfaces where dV=O(1), reducing the local order to first order. This should degrade the global error or at least contaminate the solution near contacts, but the paper does not quantify this effect. The claim of second-order accuracy is therefore not supported for the practical hybrid scheme outside pure smooth regions.","section":"Section 5.1.3, Table 1"},{"comment":"The energy-conservation claim is based on a comparison against a first-order DOTRS scheme on a smooth advection test. The abstract states that CDHD reduces the energy conservation error by two orders of magnitude compared to 'traditional primitive-variable approaches,' but only one baseline is shown, and it is a deliberately diffusive first-order scheme. No comparison is made to other common primitive-variable methods (e.g., pressure-evolution or double-flux formulations). Furthermore, the hybrid scheme's energy behavior in shock-containing flows is not assessed; switching between non-conservative and conservative updates can introduce conservation defects at the switch points, and this is not measured in the shock-tube test.","section":"Section 5.1.2, Figure 3"}],"minor_comments":[{"comment":"Typo: 'Wisdom line' should be 'Widom line'.","section":"Section 3, paragraph 1"},{"comment":"The notation in Eqs. (15) and (16) is confusing: u_{i+1/2} is used without prior definition, and the index placement of i+1/2 in the density terms is inconsistent with the text. Please clarify.","section":"Section 3.3, Eqs. (15)-(16)"},{"comment":"The figure lacks a legend; it is unclear which curve corresponds to density, momentum, and total energy. The caption mentions all three but the plot appears to have three curves without labels.","section":"Figure 4"},{"comment":"The 2D droplet interface uses η=0.05 on a 151×151 mesh, giving roughly 3-4 cells across the tanh profile. This is a well-resolved, smooth interface rather than a sharp-interface challenge. The statement that the scheme preserves a 'sharp interface' should be tempered.","section":"Section 5.3"},{"comment":"The statement that the cell average is v(x_i)+O(dx^2) is standard for a smooth function, but the subsequent argument for WENO-5 accuracy relies on the cell-average reconstruction; this should be stated explicitly to avoid the impression that the O(dx^2) cell-average error limits the interface reconstruction order.","section":"Section 4.1, Eq. (24)"}],"recommendation":"major_revision","confidential_remarks":"The contact-discontinuity issue is the main technical obstacle. If the authors can either modify the sensor to detect contacts (or more generally any O(1) jump in any primitive variable) or clearly restrict the high-order claim to cases where the sensor guarantees O(dx^5) jumps, the method would be more defensible. As it stands, the hybrid framework's robustness claim is not fully established for standard shock-tube wave structures. The paper is borderline; with additional reference solutions, a hybrid convergence test, and a broader energy-error comparison, it could become acceptable for a specialized journal."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here is my read. The paper proposes CDHD, a central flux plus DOTRS-based dissipation for the primitive-variable Euler system, with the dissipation shown to be O(Δx^4) in smooth regions. The analysis is clear and the convergence test backs it up: observed rates around 2.17 on fine meshes. The key insight, that two-point central differencing caps accuracy at second order regardless of WENO-5 interface states, is correct and worth stealing. The specific combination is new as far as I know.\n\nThe main selling point, two orders of magnitude better energy conservation, is weaker than it looks. They compare only against first-order DOTRS. That is a strawman. A second-order primitive-variable method would be the relevant baseline; otherwise the claim is misleading. The shock-tube and 2D tests are qualitative. The shock-tube has no reference solution, so contact smearing is invisible. The 2D droplet uses a resolved tanh interface (about 7 cells), so it does not test the regime that matters.\n\nThe bigger issue is the hybrid sensor. It keys on pressure ratio, so it flags shocks but not contact discontinuities. Across a contact, pressure is continuous and density jumps; dV = O(1). The dissipation term then scales as O(1/Δx), not O(Δx^4). So near contacts the scheme reverts to first-order diffusion, and the claim that the hybrid robustly handles the standard wave structure is not established. The paper shows no reference solution or error metric for the contact in the shock tube, so we cannot tell how bad it is. This is a real gap, not a nitpick.\n\nI would send this to peer review. The error analysis and the smooth-flow behavior are solid, and the method is useful for its niche. But the referee should ask for a quantitative contact test, a saner energy comparison, and a sharper interface case. If those come back, the paper can be published after moderate revision.","headline":"CDHD is a solid second-order primitive-variable flux for smooth transcritical flows; the energy-conservation claim is overstated and the hybrid's contact-discontinuity behavior is unexamined.","tokens_in":19096,"tokens_out":3081,"would_cite":true,"duration_ms":32436,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper shows that adding a minimal, high-order upwind dissipation to a central primitive-variable flux stabilizes transcritical flow simulations without sacrificing second-order accuracy.","keywords":["transcritical flow","primitive-variable formulation","hybrid scheme","pressure oscillations","energy conservation","WENO reconstruction","real-gas equation of state","DOTRS dissipation"],"falsifier":"Run the 1D transcritical advection case with a significantly sharper temperature transition (say η = 0.01 instead of 0.1) at CFL = 0.8, and measure the L1 norm of the dissipation term as the grid is refined; if the dissipation no longer scales as roughly Δx^4, or if density oscillations or growing energy error appear over ten periods, the smooth-jump assumption fails.","tokens_in":18198,"feed_emoji":"🌊","tokens_out":6587,"duration_ms":66231,"temperature":0.7,"pith_summary":"Transcritical fluid flows, where thermodynamic properties swing sharply across the Widom line, make standard conservative schemes generate spurious pressure oscillations and fail. The paper proposes the Central Differential flux with High-Order Dissipation (CDHD), a primitive-variable flux that combines a central advection term with a tiny upwind-biased dissipation. Because the dissipation is built from high-order reconstructed interface states, it scales as O(Δx^4) in smooth regions and stabilizes the simulation without degrading the second-order accuracy of the central term. In a hybrid framework that switches to a conservative WENO-5/Roe update near shocks, the method keeps sharp fronts, suppresses oscillations, and reduces energy conservation error by about two orders of magnitude compared with a first-order primitive-variable solver.","feed_headline":"New flux cuts energy error 100x in transcritical flow","feed_subtitle":"A hybrid CDHD scheme keeps second-order accuracy while suppressing pressure oscillations near the Widom line.","key_machinery":"The central object is the CDHD flux operator, written as L_v = L_v^central + L_v^diss. L_v^central approximates -B(V)∂_x V by a two-point central difference built from WENO-5 reconstructed one-sided interface states, which is second-order accurate. L_v^diss is a Dumbser–Osher–Toro Riemann solver (DOTRS) path-integral fluctuation, H^± = ∫₀¹ B^±(Ψ(s)) ds (V_R − V_L), evaluated along the straight-line path Ψ(s) between reconstructed states; in smooth regions this term scales as O(Δx^4) and provides stabilization without changing the leading-order error.","core_discovery":"The central claim is that spurious pressure oscillations in transcritical simulations can be suppressed without giving up accuracy by writing the primitive-variable spatial operator as L_v = L_v^central + L_v^diss. The central term is a two-point central difference of WENO-5 reconstructed one-sided interface states, which is second-order accurate; the dissipative term is a Dumbser–Osher–Toro path-integral fluctuation evaluated along the straight-line segment between the reconstructed states. Since WENO-5 reconstruction makes the interface jump O(Δx^5) in smooth regions, the dissipation contributes only O(Δx^4), so the total truncation error remains O(Δx^2). Embedding this in a hybrid scheme","pith_inferences":["The same central-plus-path-integral-dissipation construction could be applied to other non-conservative formulations (multicomponent or multiphase flows), but the O(Δx^4) dissipation scaling is only guaranteed when the interface jump is genuinely high-order small.","Near the Widom line, where B(V) and a² vary rapidly, the smoothness assumptions behind the error analysis can be violated; users may need to monitor the dissipation norm and possibly strengthen it locally.","The paper does not quantify pressure-equilibrium error in the 2D droplet test; measuring it as a function of interface sharpness would be a clean follow-up.","To reach orders beyond two while keeping dissipation subdominant, the reconstruction order and the central-difference order must be increased together; otherwise the dissipation could become the leading error."],"forward_implications":["Formal second-order accuracy holds in smooth transcritical regions; least-squares convergence rates on the three finest meshes are r = (2.17, 2.17, 2.21) for density, momentum, and total energy.","Energy conservation error remains below 0.04% over ten convection periods, about two orders of magnitude smaller than a first-order primitive-variable scheme.","Omitting the dissipation term leads to rapid growth of oscillations and blow-up; the O(Δx^4) dissipation is necessary and sufficient for stabilization.","The hybrid switch to conservative WENO-5/Roe near shocks captures the correct wave structure in transcritical shock tubes without limiters or filters.","The scheme is extendable: replacing the two-point central gradient with a higher-order central or split-flux differencing would raise the overall order without changing the dissipation design."],"supporting_citations":[{"why":"Supplies the WENO-5 reconstruction whose fifth-order interface states give the O(Δx^5) jump scaling used in the error analysis.","marker":"[50]"},{"why":"Supplies the Dumbser–Osher–Toro extension to non-conservative systems that defines the path-integral dissipation H^±.","marker":"[54]"},{"why":"Provides the path-integrated dissipation machinery adapted to real-gas Euler equations with highly nonlinear equations of state.","marker":"[55]"},{"why":"Supplies the PVRS approximate Riemann solver used as the shock-detection sensor in the hybrid framework.","marker":"[49]"},{"why":"Supplies the conservative Roe-type flux for real gases used in cells flagged as near shocks.","marker":"[51]"},{"why":"Provides the Peng–Robinson cubic equation of state used for all thermodynamic evaluations.","marker":"[23]"},{"why":"Provides the real-gas internal-energy and speed-of-sound relations that close the Euler equations with the chosen equation of state.","marker":"[52]"}],"fun_headline_variants":["New flux cuts energy error 100x in transcritical flows","Flux scheme tames pressure spikes without losing accuracy","Hybrid flux kills oscillations, keeps 2nd order in transcritical flow","CDHD: 100x energy improvement, zero spurious pressure"],"cache_read_input_tokens":2688,"weakest_assumption_plain":"The scheme assumes that in smooth transcritical regions the WENO-5 interface jump is genuinely fifth-order small and the thermodynamic Jacobian stays smooth along the straight-line path between states; near the Widom line, where properties change extremely fast, that assumption can be violated, so the dissipation term would no longer be safely subdominant.","fun_headline_variants_meta":{"raw":{"variants":["New flux cuts energy error 100x in transcritical flows","Flux scheme tames pressure spikes without losing accuracy","Hybrid flux kills oscillations, keeps 2nd order in transcritical flow","CDHD: 100x energy improvement, zero spurious pressure"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000136,"raw_usage":{"total_tokens":972,"prompt_tokens":725,"completion_tokens":247,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":469,"completion_tokens_details":{"reasoning_tokens":174}},"tokens_in":469,"tokens_out":247,"duration_ms":3720,"temperature":1.0,"reasoning_tokens":174,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T14:08:07.287483+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the 1D transcritical advection case with a significantly sharper temperature transition (say η = 0.01 instead of 0.1) at CFL = 0.8, and measure the L1 norm of the dissipation term as the grid is refined; if the dissipation no longer scales as roughly Δx^4, or if density oscillations or growing energy error appear over ten periods, the smooth-jump assumption fails.","supporting_citations":[{"cited_title":"Efficient implementation of weighted eno schemes","cited_arxiv_id":null,"evidence_quote":"Supplies the WENO-5 reconstruction whose fifth-order interface states give the O(Δx^5) jump scaling used in the error analysis."},{"cited_title":"Toro, Cristóbal E","cited_arxiv_id":null,"evidence_quote":"Supplies the Dumbser–Osher–Toro extension to non-conservative systems that defines the path-integral dissipation H^±."},{"cited_title":"A simple extension of the osherriemannsolvertonon-conservativehyperbolicsystems","cited_arxiv_id":null,"evidence_quote":"Provides the path-integrated dissipation machinery adapted to real-gas Euler equations with highly nonlinear equations of state."},{"cited_title":"A simple extension of roe’s scheme for real gases.Journal of Computational Physics, 329:16–28, 2017","cited_arxiv_id":null,"evidence_quote":"Supplies the conservative Roe-type flux for real gases used in cells flagged as near shocks."},{"cited_title":"A new two-constant equa- tion of state","cited_arxiv_id":null,"evidence_quote":"Provides the Peng–Robinson cubic equation of state used for all thermodynamic evaluations."},{"cited_title":"Thermo- dynamic modeling based on a generalized cubic equation of state for kerosene/lox rocket combustion","cited_arxiv_id":null,"evidence_quote":"Provides the real-gas internal-energy and speed-of-sound relations that close the Euler equations with the chosen equation of state."}],"review_version":1}