{"id":"2b64b959-a47b-4684-bffa-dd653324176e","arxiv_id":"2502.00990","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"A minimally implicit Runge-Kutta scheme for RRMHD treats only the electric-field source term implicitly, avoiding extra primitive-variable recoveries while remaining stable at high conductivity.","lead":"The authors propose a minimally implicit Runge-Kutta scheme for the resistive relativistic magnetohydrodynamics (RRMHD) equations, keeping only the stiff electric-field term implicit. They test it on two smooth 1D problems and argue it costs about the same as explicit methods while avoiding IMEX's repeated primitive-variable recoveries.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The stability of the MIRK schemes is asserted from an omitted linear-stability analysis; the coefficient choices and the headline 'stable evolutions' claim stand or fall on that unshown calculation, so the derivation must be supplied.","rationale":"The abstract's strongest claim is stability at explicit-like cost. All of the coefficient values that make the implicit inversion trivial (c1=0 or c1=-1/sqrt(2), c3=1, c6=1/2) are justified exclusively by the phrase 'We perform a linear stability analysis...' with no equations shown. A reader cannot check whether the eigenvalue bound holds, whether the zero-eigenvalue multiplicity argument is correct, or whether the claim 'independently of the velocity field' is true. The numerical tests provide some independent evidence: the current-sheet and CP Alfven runs are stable, and the explicit method blows up. However, both tests are one-dimensional and smooth, and the current-sheet test is effectively decoupled from hydrodynamics, so they do not exercise the full coupled system. The CP Alfven test also exhibits unexplained first-order convergence for the second-order method, which indicates that the full coupled behavior is not yet understood. Because the omitted calculation is precisely the item that would rule out a hidden instability at high conductivity, requiring it is the natural condition for accepting the headline claim. The reader's weakest assumption identifies the same issue, so I agree with the conditional verdict and recommend no change pending the stability derivation and, ideally, a runtime or recovery-count comparison with IMEX methods.","tokens_in":17181,"tokens_out":5773,"duration_ms":58305,"concrete_test":"Reconstruct the von Neumann analysis for both MIRK schemes: linearize Eqs. (9)-(11) with Ohm's law about a constant state with v=(v,0,0) and B=(0,B_y,B_z), using second-order centered spatial differences, and form the amplification matrix for the first-order scheme (Eq. (16)) and the second-order scheme (Eqs. (25)-(26)). Compute the spectral radius for sigma in {0, 10, 10^3, 10^8}, v in {0, 0.1, 0.9}, and all wavenumbers k*Dx in [0, pi], at CFL=0.8 and CFL=0.3. If the maximum |lambda| exceeds 1+O(Dt) in any case, the asserted stability is false; if it does not, the missing derivation is recovered and the conditional objection is resolved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central practical claim is that MIRK methods are able to deal with stiff terms producing stable numerical evolutions at explicit-like cost. The theoretical support for this is the linear stability analysis promised in Section 4: after Eqs. (16) and (22) the text states that in the infinite-conductivity limit one can choose c3=1 (and c6=1/2) keeping numerical stability, that c1=0 (first order) or c4=(1-c1)^2/(2c1) with c1=-1/sqrt(2) (second order) sets a zero eigenvalue of multiplicity at least three, and that the remaining eigenvalue is bounded by 1 in absolute value independently of the velocity field. None of the corresponding eigenvalue calculation, the linearized system, or the amplification matrix is shown. This is load-bearing because these coefficient constraints are the only mechanism preventing the implicit solve from amplifying errors in the sigma-to-infinity limit, and the finite-sigma, coupled-hydrodynamic behavior is then validated only by two smooth 1D tests. The manuscript itself flags the limits of that validation: Section 5.2 reports first-order convergence for the second-order MIRK method and states that the authors were not able to find the reason for this reduction in the order of convergence, and Section 6 concedes that more complex and non-smooth simulations are needed. If the omitted stability analysis has a flaw, or if its conclusions do not carry over to finite sigma or to multiple dimensions, the headline stability and cost advantage is not established.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript introduces first- and second-order minimally implicit Runge-Kutta (MIRK) schemes for the resistive relativistic magnetohydrodynamic (RRMHD) equations, following Komissarov's augmented-system formulation. In these schemes only the electric-field components appear implicitly, and the resulting algebraic solve is analytic, so the method is claimed to retain a computational cost comparable to explicit schemes while remaining stable at high conductivity. The paper derives the two schemes, motivates the coefficient choices through a linear stability analysis in the infinite-conductivity limit, and validates the methods on two one-dimensional smooth tests: the self-similar current sheet and circularly polarized Alfvén waves. The current-sheet test agrees well with the exact solution, and the Alfvén-wave test approaches the ideal MHD limit at sigma = 10^8, though the measured convergence order of the second-order method is first order in that test.","tokens_in":17489,"tokens_out":3587,"duration_ms":35549,"significance":"If the central claims hold, the MIRK approach is a useful contribution to numerical RRMHD: it avoids the repeated primitive-variable recoveries required by IMEX schemes, and the analytic inversion of the implicit operator makes it easy to retrofit into existing explicit codes. The paper is transparent about the algebraic form of both schemes, and the two tests are checked against independent exact or reference solutions, which is a genuine strength. The potential significance is, however, moderated by the fact that the method is validated only on smooth one-dimensional problems and that the stability assertion rests on an unshown analysis. The practical claim of 'stable numerical evolutions at explicit-like cost' is plausible but not fully established by the evidence presented.","major_comments":[{"comment":"The coefficient choices c3=1, c6=1/2, c4=(1-c1)^2/(2c1), and c1=-1/sqrt(2) are said to follow from a linear stability analysis of the evolution system in the infinite-conductivity limit, but neither the linearized system, the amplification matrix, nor the eigenvalue calculation is shown. This is load-bearing because the claimed stability of the schemes and the selection of all ci coefficients depend on that analysis. The authors should provide the full amplification matrix and the eigenvalue bounds, including the statement that the remaining eigenvalue is bounded by 1 in absolute value independently of the velocity field, and should state any assumptions on the discretization of the spatial derivatives used in the analysis.","section":"Section 4, after Eqs. (16) and (22)"},{"comment":"Table 2 shows that the second-order MIRK method converges only with order approximately 1.0 in the CP Alfvén wave test (p values near 0.87-0.98 for the finer resolutions), while the method is presented as second-order. The text states that the authors were not able to find the reason for this reduction. This is a central issue: the method's advertised order is not realized in the coupled hydrodynamic test. The authors should either identify the source of the order reduction (e.g., the Kreiss-Oliger dissipation term, the CFL value, or the recovery procedure) or explicitly present the second-order method as first-order in practice for coupled evolutions.","section":"Section 5.2, Table 2"},{"comment":"The paper claims in the abstract that the MIRK methods 'are able to deal with stiff terms producing stable numerical evolutions,' but the stability demonstration is limited to two smooth one-dimensional tests with CFL numbers at or below 0.8 (and 0.3 for the Alfvén wave test). Section 5.1 itself shows that at large CFL values oscillations appear, which are attributed to the explicitly treated source terms, and Section 6 concedes that non-smooth and more complex simulations are needed. The stability claim should therefore be qualified to the tested regime, and the conditions under which the method is stable (including restrictions on CFL and on the explicitly treated source terms) should be stated precisely.","section":"Section 5.1 and Section 5.2"}],"minor_comments":[{"comment":"The legends in Figures 6 and 7 label the reference solution as 'Exact solution v=0' even though the simulations are performed with vx=0.1; the legend should be corrected to avoid confusion about which solution is being compared.","section":"Figures 6 and 7"},{"comment":"The text says that for the convergence study the power of the factor Delta x in the Kreiss-Oliger term is increased by one unit and the coefficient is set to 0.1, but the exact modified form of the dissipation term is not written; please specify the expression used in the convergence runs.","section":"Section 5.2, after Eq. (36)"},{"comment":"The statement that 'MIRK methods have the same computational cost as explicit ones' is stronger than what is demonstrated, since no wall-clock timing or operation-count comparison is reported and the implicit solve, although analytic, still introduces additional algebraic operations; the cost claim should be stated as 'the same order of cost per time step'.","section":"Section 6"},{"comment":"The text refers to 'terms in orange' in Eq. (9) and the corresponding explanation, but in a monochrome printout or in some rendering of the manuscript this color coding is not visible; please add an explicit mathematical designation of the implicitly treated terms.","section":"Section 1 and Section 4"}],"recommendation":"major_revision","confidential_remarks":"The manuscript addresses a timely and practical problem, and the proposed MIRK construction is elegant. However, the two load-bearing issues—the omitted linear stability analysis and the first-order convergence of the second-order method in the coupled test—must be resolved before the central claims can be accepted. The paper fits the scope of Applied Mathematics and Computation, but the numerical validation is currently too narrow to support the strong statements made in the abstract."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a sensible and potentially useful paper, but the central stability claim is asserted from an omitted calculation, and that calculation is load-bearing. I would send it to review, not desk reject, but I would insist the analysis be shown.\n\nWhat is new: MIRK treats only the sigma W E term in the electric-field evolution implicitly, keeping everything else explicit. The implicit solve is analytic, so no extra primitive-variable recoveries are needed beyond what explicit methods already do. That is a real difference from the IMEX approach of Palenzuela et al. (2009) and from the locally implicit DG of Dumbser and Zanotti (2009). The construction is clearly explained, the algebraic update formulas are written out explicitly, and the two tests compare against exact or ideal-limit solutions. The current-sheet test uses an exact erf solution; the Alfven test checks the ideal MHD limit at sigma = 10^8. That is genuine evidence, not curve fitting. The paper also honestly flags its own limits: first-order convergence for the second-order method in the Alfven test, and the need for non-smooth multidimensional tests.\n\nSoft spots, in order of importance. First and biggest: the linear stability analysis that fixes c1, c3, c4, and c6 is asserted but not shown. Section 4 says one can choose c3 = 1, imposes c1 = 0 (first order) or the c4 relation (second order), and claims the remaining eigenvalue is bounded by 1 independent of velocity, but no linearized system, amplification matrix, or eigenvalue calculation appears. This is the entire theoretical basis for stability at high conductivity, so it needs to be in the paper. Second, the unexplained order reduction: the second-order MIRK converges only first-order in the hydrodynamic Alfven test. The authors say they could not find the reason. That matters because the method is advertised as second-order. Third, the cost advantage over IMEX is claimed in the abstract but never measured: there is no runtime or recovery-count comparison. Fourth, the CFL scans show oscillations from the explicitly treated source terms; the paper acknowledges this, but it does bound the practical benefit. None of these kill the idea; they all point to a promising first report rather than a fully verified method.\n\nThe paper is for people building explicit RRMHD codes who want a cheap stiff-source fix. It deserves a serious referee. My recommendation: send to review with a request that the stability derivation be included, the order reduction be explained or at least analyzed, and at least one runtime or recovery-count comparison with IMEX be added.","headline":"A promising, clearly explained RRMHD stiff-source scheme whose headline stability claim rests on an omitted linear-stability derivation; worth refereeing, but the missing analysis and the unexplained order reduction need to be addressed.","tokens_in":18035,"tokens_out":1544,"would_cite":true,"duration_ms":15537,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35B35","35F50","35L60","35Q35","35Q75","65M06","65M12","76W05"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper argues that the stiff Ohmic source in resistive relativistic MHD can be advanced by a minimally implicit Runge-Kutta update whose implicit step is a single algebraic division, giving explicit-like cost and no added recovery…","keywords":["Resistive relativistic magnetohydrodynamics","Minimally implicit Runge-Kutta (MIRK)","Stiff source terms","Implicit-explicit (IMEX) methods","Primitive variable recovery","Finite-difference methods","Self-similar current sheet","Circularly polarized Alfvén waves"],"falsifier":"Assemble the amplification matrix of the MIRK update for the full RRMHD system with the stated coefficients at $\\sigma\\to\\infty$ or at a large finite $\\sigma$, for generic velocity directions, and check whether every eigenvalue lies in the unit disk for $\\Delta t$ up to the CFL limit; any admissible velocity and step size with spectral radius above 1 would directly contradict the claimed stability, since the paper does not display this calculation.","tokens_in":16991,"feed_emoji":"⚡","tokens_out":13944,"duration_ms":117817,"temperature":0.7,"pith_summary":"This paper proposes Minimally Implicit Runge-Kutta (MIRK) schemes that make the stiff Ohmic source term in resistive relativistic magnetohydrodynamics (RRMHD) cost about the same as an explicit update. The key move is to evaluate implicitly only the electric- and magnetic-field components multiplied by the conductivity, which are conserved variables themselves, so the implicit step is a single algebraic division instead of an iterative primitive-variable recovery. First- and second-order versions are constructed, with coefficients chosen from wave-propagation PIRK methods and a linear stability analysis in the ideal infinite-conductivity limit. The schemes are tested on a self-similar current sheet and on a circularly polarized Alfvén wave, and the authors argue that existing explicit codes can be adapted to them with minimal changes. The central claim, read sympathetically, is that physical resistivity can be added to relativistic MHD simulations without the extra cost or convergence risk of implicit-explicit (IMEX) recovery loops.","feed_headline":"One division per update tames stiff resistivity in relativistic MHD","feed_subtitle":"Minimally implicit Runge-Kutta keeps explicit-method cost and skips the expensive primitive-variable recovery loop.","key_machinery":"The central object is the MIRK (Minimally Implicit Runge-Kutta) update for the electric field: in the evolution equation $\\partial_t E^j = S_E^j - \\bar\\sigma [E^j + (v\\times B)^j - (v_l E^l)v^j]$, only the bracketed $E$- and $B$-components are evaluated at the new stage, while $\\bar\\sigma = \\sigma W$ and $v$ stay at their explicit stage values. For the first-order scheme the update becomes $E^j|_{n+1} = (E^j|_n + \\Delta t [S_E^j|_n + \\bar\\sigma|_n E^l|_n (v^j|_n v_l|_n - \\delta^j_l) - \\bar\\sigma|_n (v|_n \\times B|_{n+1})^j])/(1 + \\Delta t\\bar\\sigma|_n)$, an explicit-looking step with effective time step $\\Delta t/(1+\\Delta t\\bar\\sigma)$; the second-order scheme applies the same idea in two stages with $\\bar\\sigma$ evaluated at the old and first-stage values. The remaining coefficients are fixed by PIRK coefficients for wave-like behavior and by a linear stability analysis in the $\\sigma\\to\\infty$ limit: the ideal limit requires a zero eigenvalue of multiplicity at least three, so $E\\to -v\\times B$, and the other eigenvalue must stay inside the unit disk for any $v$, which selects $c_1 = -1/\\sqrt{2}$ for the second-order method.","core_discovery":"The paper's central claim is that the stiffness of Ohm's law in resistive relativistic MHD can be handled by treating the conductivity-weighted electric and magnetic field components implicitly while keeping every other term explicit; because those components are already conserved variables, the implicit stage reduces to solving a diagonal linear system, i.e. a division. The authors present a first-order scheme and a second-order two-stage scheme, with coefficients selected by PIRK wave-propagation coefficients and by a linear stability analysis in the ideal $\\sigma\\to\\infty$ limit that fixes the remaining parameters, including $c_1=-1/\\sqrt{2}$ for the second-order method. They validate the schemes on a self-similar current sheet at $\\sigma=10^3$ and on a circularly polarized Alfvén wave at $\\sigma=10^8$, recovering the exact and ideal-MHD solutions, and they argue that the computational cost stays comparable to explicit methods with no additional primitive-variable recovery loops.","pith_inferences":["Extending beyond the paper: because the implicit solve is a single diagonal division, the same MIRK update should transfer almost unchanged to multidimensional and shock-dominated RRMHD, but the decisive test would be a 2D or 3D problem with non-smooth data, which the paper explicitly leaves to future work.","The paper does not use the term asymptotic-preserving, but its effective time step $\\Delta t/(1+\\Delta t\\bar\\sigma)$ formally drives $E$ toward $-v\\times B$ as $\\bar\\sigma\\to\\infty$; this suggests the method could serve as a building block for schemes that switch between resistive and ideal regimes, provided the explicitly treated source terms are also controlled.","The observed instability of the circularly polarized Alfvén wave test at CFL=0.8, attributed by the authors to the explicitly evolved source terms, implies that the practical barrier to large timesteps may sit in $S_E$, $S_B$, and $S_Y$ rather than in the conductivity term; a natural extension would be to apply the same minimal-implicitness idea to those terms, which the paper does not pursue."],"forward_implications":["Existing explicit RRMHD codes can incorporate the stiff Ohm term by replacing the electric-field update with the MIRK formula, a division by $1+\\Delta t\\bar\\sigma$, without changing the rest of the time-stepping structure.","The number of primitive-variable recoveries per timestep stays the same as in an explicit scheme, which removes the main extra cost and the convergence risk that IMEX methods introduce.","At high conductivity the electric field is driven toward the ideal value $-v\\times B$ between stages, so the ideal MHD limit is recovered as $\\bar\\sigma\\to\\infty$.","The same minimally implicit construction, making implicit only the terms $G_i(U)U^i$ that multiply a large parameter while evaluating $G_i$ explicitly, extends to other hyperbolic balance laws with stiff sources."],"supporting_citations":[{"why":"Supplies the augmented RRMHD system with the scalar fields $\\phi$ and $\\psi$ that the new MIRK scheme is built on.","marker":"[12]"},{"why":"Provides the IMEX Runge-Kutta treatment whose extra primitive-variable recoveries and computational cost motivate the MIRK alternative, and supplies the two test problems.","marker":"[11]"},{"why":"Offers an alternative high-order locally implicit scheme for RRMHD and supplies the primitive-variable recovery procedure used in the Alfvén wave test.","marker":"[15]"},{"why":"Provides the first-order PIRK method for wave-like equations that sets $c_2=0$ in the first-order MIRK scheme.","marker":"[16]"},{"why":"Provides the second-order PIRK method whose coefficients set $c_2$ and $c_5$ in the two-stage MIRK scheme.","marker":"[17]"},{"why":"A private discussion invoked to support the statement that large-CFL oscillations also occur with IMEX methods.","marker":"[18]"}],"fun_headline_variants":["One division per update for stiff relativistic MHD","Minimally implicit Runge-Kutta avoids primitive recovery","Stable resistive MHD at explicit-method cost","Taming Ohm's law stiffness in one division","Skip the recovery loop: new RRMHD scheme"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The scheme's stability rests on a linear stability analysis in the perfect-conductor limit that the paper asserts without displaying, and on that analysis continuing to hold at finite conductivity and for the full set of coupled fluid and field equations.","fun_headline_variants_meta":{"raw":{"variants":["One division per update for stiff relativistic MHD","Minimally implicit Runge-Kutta avoids primitive recovery","Stable resistive MHD at explicit-method cost","Taming Ohm's law stiffness in one division","Skip the recovery loop: new RRMHD scheme"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000207,"raw_usage":{"total_tokens":1404,"prompt_tokens":952,"completion_tokens":452,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":568,"completion_tokens_details":{"reasoning_tokens":377}},"tokens_in":568,"tokens_out":452,"duration_ms":5141,"temperature":1.0,"reasoning_tokens":377,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T18:34:58.097188+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Assemble the amplification matrix of the MIRK update for the full RRMHD system with the stated coefficients at $\\sigma\\to\\infty$ or at a large finite $\\sigma$, for generic velocity directions, and check whether every eigenvalue lies in the unit disk for $\\Delta t$ up to the CFL limit; any admissible velocity and step size with spectral radius above 1 would directly contradict the claimed stability, since the paper does not display this calculation.","supporting_citations":[{"cited_title":"Cordero-Carri´ on, P","cited_arxiv_id":null,"evidence_quote":"Provides the second-order PIRK method whose coefficients set $c_2$ and $c_5$ in the two-stage MIRK scheme."},{"cited_title":"palenzuela","cited_arxiv_id":null,"evidence_quote":"A private discussion invoked to support the statement that large-CFL oscillations also occur with IMEX methods."}],"review_version":1}