{"id":"be339977-e3e9-4dd3-b3a2-13a556fb71b2","arxiv_id":"2502.03223","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":8,"one_line_summary":"SACRA-2D is a new axisymmetric relativistic hydrodynamics code with the HLLC solver and adaptive mesh refinement, validated by benchmarks showing improved accuracy over the TVDLF solver.","lead":"This paper presents SACRA-2D, a new computer program for simulating dense astrophysical objects like neutron stars and black holes, assuming the system is symmetric around an axis. It shows the code can run simulations more accurately and for longer times than previous axisymmetric codes, which matters for studying events like star collapse and neutron-star mergers.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Cartoon-method off-plane closure at FMR boundaries is exact in continuum but lacks direct 3D cross-validation; self-convergence tests cannot rule out systematic axisymmetric error.","rationale":"The reader's weakest assumption focuses on the cartoon method's treatment of y-derivatives at FMR boundaries. I concur that this is the most load-bearing technical element: if the off-plane closure is wrong, every dynamical-spacetime result is compromised. However, I do not agree that Eq. (14) is an approximation that can 'fail' in the sense of being inexact for axisymmetric data; the identities are exact in the continuum, and the paper's justification about non-zero x is about avoiding division by zero and small-x amplification. The real soft spot is that the numerical evaluation of these identities uses finite-difference x-derivatives of interpolated buffer-zone data, with 1/x amplification factors, and no test in the paper isolates this contribution. All full-GRHD benchmarks are axisymmetric self-consistency checks: they demonstrate convergence and conservation but cannot detect a systematic, axisymmetry-preserving error in the off-plane closure that converges to a wrong answer. The trumpet and Bondi tests anchor the individual sectors against analytic solutions, but there is no independent 3D cross-code comparison of the coupled metric+hydrodynamics evolution. That gap is precisely what the concrete test addresses. The paper's evidence is otherwise strong: sixth-order convergence in metric variables and waveforms (modulo the acknowledged late-time loss at coarse refinement boundaries), second-order hydrodynamics convergence, near-machine-precision conservation for mass and angular momentum in the tested regimes, and a clear HLLC improvement over TVDLF in the neutron-star surface and instability tests. The absence of public code and data is a reproducibility concern but not, by itself, a correctness argument. My read therefore leaves the reader's CONDITIONAL verdict unchanged: the code is likely sound, but the cartoon closure deserves direct external validation before the central claim is taken as established.","tokens_in":33158,"tokens_out":7795,"duration_ms":72843,"concrete_test":"Set up the same Brill-Lindquist head-on collision (b=M/2, as in Sec. III.A.3) in SACRA-2D and in a well-established 3D BSSN code (e.g., Einstein Toolkit or SACRA-MPI) with the same puncture gauge and comparable resolution. Compare the (l=2,m=0) Psi4 waveform, the Hamiltonian constraint, and the apparent-horizon mass over t in [0,100]M. If the difference exceeds SACRA-2D's own low-vs-high convergence error band, the cartoon off-plane closure, particularly the Eq. (14) buffer-cell identities at the finest FMR boundary, is implicated. To isolate 1/x sensitivity, repeat the SACRA-2D run with the finest level placed at two different coordinate offsets and check that the waveform changes by less than the convergence error.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that SACRA-2D solves axisymmetric GRHD in dynamical spacetime rests on the cartoon method's off-plane closure in Sec. II.B.2. For interior points, off-plane layers are filled by interpolation plus rotation (Eq. 12), and y-derivatives are finite-differenced. For buffer cells x in [N+1,N+4] at each FMR level, Eq. (14) replaces d_xy by algebraic identities containing 1/x. These identities are exact for continuous axisymmetric tensor fields; the 'far from the symmetric axis' remark is about avoiding small-x error amplification, not an approximation, so the paper is internally consistent. The load-bearing gap is different: the numerical implementation evaluates these identities with finite-difference x-derivatives of interpolated buffer data, and 1/x factors can amplify truncation error at fine-level boundaries where x is only ~L/2 (e.g., L=2.15M in the SMS collapse, L=8.65 km in the NS tests). More importantly, every full-GRHD benchmark is an axisymmetric self-consistency check (convergence, conservation, ringdown-frequency fits). A systematic error in the off-plane closure that preserves axisymmetry and converges with resolution would pass all these tests while still biasing the spacetime evolution. Thus the pivotal unvalidated premise is not a demonstrated inconsistency but the absence of any direct comparison of the off-plane closure against an independent 3D evolution of the same axisymmetric problem.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents SACRA-2D, a new MPI+OpenMP parallelized axisymmetric general-relativistic hydrodynamics code. The spacetime is evolved with the BSSN formalism augmented by Z4c constraint propagation in Cartesian coordinates on the y=0 plane, using the cartoon method for off-plane closure and a two-to-one fixed mesh refinement hierarchy. The hydrodynamic sector uses finite-volume shock capturing with both TVDLF and HLLC Riemann solvers, a hybrid piecewise-polytropic EOS, and an artificial atmosphere. The authors validate the metric sector with trumpet and spinning black hole evolutions and a head-on black hole collision, validate the hydrodynamics sector with a shock tube and Bondi accretion, and then present full dynamical-spacetime tests: a stable rotating neutron star, unstable neutron star migration, collapse of an unstable rotating neutron star to a black hole, and collapse of a supermassive star. They report sixth-order convergence for metric variables and gravitational waveforms, second-order convergence for hydrodynamics, excellent baryon-mass and angular-momentum conservation, and better performance of HLLC over TVDLF, especially in resolving contact discontinuities and reducing artificial surface heating.","tokens_in":33476,"tokens_out":7565,"duration_ms":77007,"significance":"If validated, SACRA-2D is a valuable tool for long-term axisymmetric GRHD simulations, including neutron-star stability, stellar collapse, black-hole-torus systems, and modified-gravity studies, where 3D simulations are prohibitively expensive. The benchmark suite is unusually broad and includes several external analytic anchors: the trumpet solution, Bondi accretion, the 1D shock-tube analytic solution, quasinormal-mode frequencies for black hole ringdown, and prior published results for neutron star migration and supermassive star collapse. The reported conservation of baryon mass and angular momentum to the 1e-10 level or better in HLLC runs is a strong quantitative result, as is the sixth-order convergence observed in the metric sector. The explicit comparison of HLLC versus TVDLF, showing reduced artificial surface heating and sharper instability fingers, is a concrete and useful contribution. The main gap, discussed below, is the absence of a direct 3D cross-validation of the cartoon method's off-plane closure in a coupled matter+spacetime evolution.","major_comments":[{"comment":"The spinning-black-hole test in Section III.A.2 provides some metric-sector validation of the cartoon method for a rotating spacetime, but it is a stationary vacuum configuration and does not exercise the off-plane closure in the presence of dynamical matter and strong hydrodynamics.","section":"II.B.2 (Cartoon method), Eq. (14)"},{"comment":"The waveform convergence statement in Section III.A.3 is also only demonstrated for t_ret ≲ 40M; the loss of convergence at later times due to reflection at coarse refinement boundaries should be mentioned in the same qualified way.","section":"III.A.1 and Fig. 1; Abstract and Summary"},{"comment":"The SMS collapse comparison with Ref. [45] is similarly qualitative ('the ejecta mass is ~1% of the total mass, and this result agrees with that of [45]').","section":"III.C.2 and Fig. 9"}],"minor_comments":[{"comment":"The text states the neutron star has angular frequency Ω = 6.28 rad/s, but with this value the rotational period is 1 s, which is inconsistent with the statement that the 250 ms evolution covers roughly 250 rotational periods. The unit should presumably be rad/ms or the value should be corrected.","section":"III.C.1"},{"comment":"The sentence 'the structure of the neutron star remains intact across the refinement boundary as shown by the green solid line in Fig. 10' appears to reference the wrong figure: the stable-neutron-star snapshots are in Fig. 8, while Fig. 10 shows the migration test. Please correct the cross-reference.","section":"III.C.1 and Fig. 8"},{"comment":"The caption lists 'blue, green, red, and cyan lines' but the figure text describes only three resolutions, N = 200, 300, and 400. The color list and the number of curves should be made consistent.","section":"Fig. 2 caption"},{"comment":"The summary refers to 'Z4c constraint transport' but the formulation in Section II.B is constraint propagation via the Z4c damping terms; the wording should be aligned with the body of the paper.","section":"IV. Summary"},{"comment":"The phrase 'providing shreds of evidence' is understandable but informal for a journal article; consider replacing it with 'evidence' or 'clues'.","section":"I. Introduction"}],"recommendation":"major_revision","confidential_remarks":"This is a solid code paper with a broad benchmark suite. The main issue is the absence of a direct 3D cross-validation of the cartoon off-plane closure in a full-GRHD setting; the spinning-black-hole test helps but is not a fully dynamical matter test. This is fixable with one targeted comparison or a quantitative error estimate. The self-citation pattern is not objectionable because the benchmarks include several independent analytic and prior-code results. The paper fits the scope of the journal and, once the validation gap and the unqualified convergence claims are addressed, should be suitable for publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a serious code paper. SACRA-2D is a new axisymmetric GRHD code combining the cartoon method, HLLC, fixed mesh refinement, and adaptive time stepping. The benchmark suite is broad and the reported convergence (sixth-order for metric variables, second-order for hydrodynamics) and conservation (baryon mass and angular momentum to ~1e-10 or better) line up. The HLLC-vs-TVDLF comparison is a real improvement: in the rotating-neutron-star test, TVDLF produces an artificial surface atmosphere about five orders of magnitude denser than HLLC. That is the most convincing single result in the paper.\n\nActual newness: the cartoon method itself is old (Shibata 2000; Alcubierre et al. 2001), but the combination with HLLC in full-GRHD dynamical spacetime, including FMR with flux correction, is new. The implementation details are clearly described: nine-point interpolation for off-plane layers, tetrad transformation for HLLC, Berger-Oliger time stepping, and conservative restriction. The authors also honestly report the late-time convergence loss in the head-on collision due to coarse-grid reflections.\n\nSoft spots: the paper does not ship code or data, which matters for a code paper. Several full-GRHD tests are single-resolution runs. The SMS collapse and the rotating-NS migration are compared against the authors' own previous results rather than an independent code. These are not fatal; the external benchmarks (trumpet solution, Bondi accretion, shock tube, QNM frequencies) anchor the code's correctness.\n\nThe bigger gap is the one your stress-test flags. Every full-GRHD validation is axisymmetric, so a systematic error in the off-plane closure could pass all tests. I want to push back on one framing: Eq. (14) is an exact continuum identity, not an approximation, so the code is internally consistent. However, the numerical implementation uses interpolated buffer data and finite differences with 1/x factors that can amplify truncation error near fine-level boundaries. A direct comparison against a 3D code on the same axisymmetric spacetime would close this. It is a validation gap, not a demonstrated flaw.\n\nRecommendation: send it to peer review. A serious referee can ask for code/data release and one 3D cross-validation run. This is exactly the kind of work that deserves referee time.","headline":"A serious new axisymmetric GRHD code with a strong benchmark suite; the main gap is the lack of a direct 3D cross-validation of the cartoon off-plane closure, but the work deserves peer review.","tokens_in":34036,"tokens_out":2537,"would_cite":true,"duration_ms":23358,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"SACRA-2D, a new axisymmetric general-relativistic hydrodynamics code, claims sixth-order-accurate spacetime evolution and near-machine-precision conservation of mass and angular momentum, with the HLLC solver beating TVDLF on contact…","keywords":["axisymmetric general-relativistic hydrodynamics","cartoon method","HLLC Riemann solver","BSSN formalism","Z4c constraint propagation","fixed mesh refinement","neutron star collapse","gravitational waves"],"falsifier":"Compare the same axisymmetric system, say the unstable rotating neutron-star migration or the head-on black-hole collision, between SACRA-2D and a full three-dimensional GRHD code at matched resolution; if the 2D-3D difference in the $\\Psi_4$ waveform or in Hamiltonian constraint violation exceeds the 3D code's own convergence error, the cartoon method's off-plane y-derivative treatment in the refinement-boundary region is the likely cause.","tokens_in":1597,"feed_emoji":"🕳️","tokens_out":1801,"duration_ms":73584,"temperature":0.7,"pith_summary":"This paper presents SACRA-2D, a new code for general-relativistic hydrodynamics in dynamical spacetimes that assumes axial symmetry to reduce the computational grid from three spatial dimensions to two. The authors claim that by combining the cartoon method for geometry, a finite-volume scheme with the HLLC Riemann solver for matter, and two-to-one fixed mesh refinement, the code achieves sixth-order convergence in metric variables and gravitational waveforms and second-order convergence in hydrodynamics. They further claim that baryon mass and angular momentum are conserved to near machine precision and that HLLC markedly reduces artificial surface heating compared with the TVDLF solver. If these claims hold, SACRA-2D provides a cheap and accurate tool for long-term axisymmetric studies of neutron-star stability, collapse, and black-hole-torus systems that would be prohibitively expensive in full 3D.","feed_headline":"Axisymmetric code simulates stellar collapse in full relativity","feed_subtitle":"SACRA-2D cuts the simulation grid to two dimensions while conserving mass and spin to about one part in 10^10.","key_machinery":"The central mechanism is the cartoon method, which imposes axial symmetry on a Cartesian grid by solving Einstein's equations only on the x-z plane and filling three extra layers in the off-plane y-direction using interpolation plus a rotation law for the geometric variables. Around this sit the BSSN formalism with Z4c constraint propagation for the spacetime sector, the finite-volume high-resolution shock-capturing scheme with the HLLC Riemann solver (using a local tetrad transformation) for the matter sector, and a two-to-one fixed mesh refinement with Berger-Oliger adaptive time stepping and flux correction across refinement boundaries. The HLLC solver restores the contact discontinuity in the approximate Riemann fan, and the flux correction across mesh-refinement boundaries is what keeps mass and angular momentum conserved to near machine precision.","core_discovery":"The paper's central claim is that a fully general-relativistic hydrodynamics code confined to two spatial dimensions can reproduce the dynamics of axisymmetric neutron stars, black holes, and their collapse with a combination of sixth-order-accurate metric evolution, second-order hydrodynamics, and near-machine-precision conservation of baryon mass and angular momentum. The authors demonstrate the claim through benchmark tests covering vacuum spacetimes, fixed-spacetime hydrodynamics, stable and unstable neutron stars, black-hole formation, and supermassive-star collapse, with gravitational waveforms matching analytical quasinormal-mode frequencies. A key quantitative claim is that the HLLC solver resolves the stellar surface contact discontinuity about five orders of magnitude better than TVDLF, reducing artificial atmosphere density and long-term error.","pith_inferences":["Inference: the HLLC advantage should be at least as important in magnetohydrodynamics, where contact discontinuities and current sheets are even more sharply degraded by diffusive solvers; the authors list MHD with HLLD as planned future work.","Inference: a direct head-to-head with a full 3D GRHD code on the same axisymmetric initial data would provide the cleanest test of the cartoon method's off-plane approximation, especially in the outer refinement levels.","Inference: because the code is axisymmetric and comparatively cheap, it could serve as a quick survey tool for alternative theories of gravity before investing in 3D runs, an application the paper mentions as future work.","Inference: the near-machine-precision conservation of angular momentum suggests the code may be useful for studying disk evolution and collapsar scenarios over dynamical timescales inaccessible to 3D codes."],"forward_implications":["Long-term axisymmetric simulations of neutron-star migration and collapse become practical at a fraction of the 3D computational cost.","The HLLC solver's sharper contact resolution reduces artificial atmosphere densities around neutron stars by about five orders of magnitude relative to TVDLF, improving long-term fidelity.","Black-hole formation, apparent-horizon properties, and ringdown waveforms can be computed in axisymmetry, with final black-hole mass and spin matching the initial system to about 0.03%.","The supermassive-star-collapse test shows the FMR structure can span radii from about 2M to 1100M and follow a black-hole-plus-disk remnant of a $\\sim 1.5\\times10^5\\,M_\\odot$ star.","Strong scaling of about 70% efficiency up to thousands of cores makes parameter surveys feasible for axisymmetric relativistic astrophysics."],"supporting_citations":[{"why":"Establishes the cartoon method that reduces axisymmetric 3D general relativity to the x-z plane with off-plane layers.","marker":"[97–99]"},{"why":"Provides the BSSN formalism used to evolve the metric and its conformal variables.","marker":"[84, 85]"},{"why":"Supplies the Z4c constraint propagation and damping that keep constraints under control.","marker":"[86, 87]"},{"why":"Defines the HLLC approximate Riemann solver, including the contact-discontinuity construction.","marker":"[12, 13]"},{"why":"Gives the tetrad-transformed HLLC implementation for general-relativistic hydrodynamics that the code adapts.","marker":"[17, 18]"},{"why":"The parent SACRA-MPI code whose numerical algorithms and grid structure SACRA-2D adapts to 2D.","marker":"[82, 83]"},{"why":"Supplies the two-to-one fixed mesh refinement and Berger-Oliger adaptive time stepping.","marker":"[88]"},{"why":"Provides the robust primitive-variable recovery procedure used by the code.","marker":"[116]"},{"why":"Provides the trumpet-coordinate Bondi solution used as a smooth GRHD convergence test.","marker":"[141]"},{"why":"Serves as a comparison for the unstable neutron-star migration test with an adiabatic equation of state.","marker":"[145]"}],"fun_headline_variants":["SACRA-2D: 2D relativity code with near-perfect mass and spin conservation","HLLC solver gives 2D GRHD code a five-order accuracy boost","SACRA-2D cuts GRHD cost by 1/2 while keeping full relativity","Axisymmetric GRHD code simulates stellar collapse in full relativity"],"cache_read_input_tokens":36096,"weakest_assumption_plain":"The code's accuracy rests on the assumption that the cartoon method's interpolated off-plane layers and the algebraic y-derivative formulas at the refinement-boundary region correctly represent a three-dimensional axisymmetric spacetime; if that representation is wrong, errors in the y-derivatives contaminate the Einstein equations and the whole evolution.","fun_headline_variants_meta":{"raw":{"variants":["SACRA-2D: 2D relativity code with near-perfect mass and spin conservation","HLLC solver gives 2D GRHD code a five-order accuracy boost","SACRA-2D cuts GRHD cost by 1/2 while keeping full relativity","Axisymmetric GRHD code simulates stellar collapse in full relativity"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001442,"raw_usage":{"total_tokens":5770,"prompt_tokens":861,"completion_tokens":4909,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":477,"completion_tokens_details":{"reasoning_tokens":4819}},"tokens_in":477,"tokens_out":4909,"duration_ms":29346,"temperature":1.0,"reasoning_tokens":4819,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-09T05:27:49.760296+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compare the same axisymmetric system, say the unstable rotating neutron-star migration or the head-on black-hole collision, between SACRA-2D and a full three-dimensional GRHD code at matched resolution; if the 2D-3D difference in the $\\Psi_4$ waveform or in Hamiltonian constraint violation exceeds the 3D code's own convergence error, the cartoon method's off-plane y-derivative treatment in the refinement-boundary region is the likely cause.","supporting_citations":[{"cited_title":"High accuracy binary black hole simulations with an extended wave zone","cited_arxiv_id":"0910.3803","evidence_quote":"Serves as a comparison for the unstable neutron-star migration test with an adiabatic equation of state."}],"review_version":1}