{"id":"c44053e4-e796-411f-94a9-7569397a4ed7","arxiv_id":"2411.19288","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Midplane-based relative constants of motion give accurate guiding-center databases with radial electric field, and Er-induced electric frequency shifts carry a reference-point bias.","lead":"The paper extends the VisualStart guiding-center orbit code to include a steady radial electric field and tests it on KSTAR, JT-60U, and ITER models. It shows that midplane-based coordinates are more accurate than global constants of motion when the electric field is present, and it warns that electric frequency shifts depend on where on an orbit they are measured.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Superiority of midplane-relative CoM over absolute CoM is benchmarked only against a deliberately naive meshing algorithm; the paper concedes an accurate implementation could match it, so the categorical 'superior' claim is not established.","rationale":"The reader's weakest_assumption focuses on the unique-midplane and double-counting premise. That is a real limitation, but the paper acknowledges it explicitly and scopes it to edge/divertor cases (Section 3.4, 3.8, Appendix A). The single most load-bearing issue for the paper's headline claim is different: the comparative claim of 'superiority' over absolute CoM is supported only by a benchmark against a deliberately unsophisticated absolute-CoM mesher. The manuscript's own text concedes that a more accurate absolute-CoM implementation is possible in principle. This does not invalidate the relative-CoM method or its practical utility, but it means the abstract's categorical 'superior' is stronger than the evidence demonstrates. The correct response is to require the authors to either qualify the claim (e.g., 'superior for a given implementation effort' or 'superior with the simple mesh used here') or provide a fair benchmark with a boundary-fitted absolute-CoM mesh. Since this condition is compatible with the reader's existing CONDITIONAL verdict, no change in verdict is needed, but the rationale for the condition should be updated to include this missing fair comparison. I credit the paper for detailed derivations, internal consistency checks (Jacobian identities, Eq. (37) variants, convergence tests), and explicit statements of limitations; the concern is about scope of a claim, not about correctness of the method or integrity of the authors.","tokens_in":62406,"tokens_out":7541,"duration_ms":73072,"concrete_test":"Re-run the Appendix B benchmark (JT-60U, Er0=30 kV/m, uniform Nref and Tref) using a boundary-fitted absolute-CoM mesh that explicitly resolves the u=0 constraint (u^2/2 = E - Phi/rho0 - mu B >= 0), e.g., by splitting cells along the u=0 curve and computing exact cell volumes. Compare N(psiP), Gamma_tor, Gamma_pol and the velocity histogram fgc(K, alpha) at the same resolutions (24x48x48 and 64x128x128) against the relative-CoM results in the left column of Fig. B.2. If the absolute-CoM errors drop to the relative-CoM level and the u near 0 deficit disappears, then the 'superior' claim is not supported; if errors persist at equal resolution and cost, the claim stands.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The abstract's central claim is that in the presence of Er, midplane-based relative CoM coordinates are 'not only equivalent but superior' to conventional constants of motion, 'allowing to attain high numerical accuracy and efficiency with a relatively simple mesh.' The numerical evidence for this superiority is Appendix B, which compares relative CoM slicing with absolute CoM slicing. However, the absolute CoM implementation uses intentionally simple AND/OR filters (Table B.1) that discard or mis-size cells near the u=0 boundary, causing up to ~40% density errors at moderate resolution (Fig. B.2(o)). The paper itself states: 'in principle, all this could be done accurately if one invests more effort into the meshing algorithm, which we did not do because we already have a simpler, more elegant way to obtain accurate results' (Appendix B.1). Thus the demonstrated 'superiority' is a comparison against a self-imposed naive implementation, not against the coordinate system itself. If a boundary-fitted absolute CoM mesh (respecting the curved u=0 surface) were implemented, it could plausibly achieve comparable accuracy at comparable cost. The load-bearing part of the claim is therefore not 'equivalent' (which is supported) but 'superior' without qualification. This is an internal admission of a missing fair benchmark, not an outside disagreement with consensus. The unique-midplane assumption flagged by the reader is valid but acknowledged and scoped; the superiority claim is asserted categorically in the abstract yet rests on an asymmetric comparison.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper describes an extension of the VisualStart guiding-center (GC) code that includes a stationary, toroidally symmetric radial electric field Er. The authors sample GC orbit space on the magnetic midplane using relative constants of motion (midplane kinetic energy, pitch, radius) and compare this with slicing using absolute constants of motion (energy, invariant pitch, canonical toroidal momentum). The method is exercised on KSTAR, JT-60U, and ITER equilibria, with a Maxwellian-like CoM distribution used for weighting. The paper also analyzes the Er-induced modification of individual GC orbits, including the shift of the trapped-passing boundary, the reference-point bias in electric frequency shifts, and the validity of approximating the electric Doppler shift by bulk rotation. The central claim is that midplane-based relative CoM are equivalent but superior to absolute CoM in the presence of Er, allowing high accuracy with a simple mesh.","tokens_in":62718,"tokens_out":5225,"duration_ms":43447,"significance":"The potential value of the work is substantial: if robust, it provides a practical recipe for constructing GC orbit databases in rotating tokamak plasmas, which is relevant for low-frequency instabilities and resonance analyses. The paper ships extensive numerical validation: convergence tests in time step (Fig. 6), resolution studies for relative vs absolute CoM (Appendix B), analytic orbit-width estimates agreeing within 2-30% (Section 5.4), a quantitative continuity check (Eq. (53)), and reconstruction of the input Er from computed flow fields (Section 4.3). The demonstration that relative and absolute CoM are equivalent when the relative-CoM mesh is accurate is well supported, and the reference-point bias analysis in Section 5.5 identifies a subtle and important ambiguity. However, the paper's headline claim of superiority over absolute CoM is not established by the benchmark presented, as detailed in the major comments.","major_comments":[{"comment":"The load-bearing claim that midplane-based relative CoM are 'not only equivalent but superior' to absolute constants of motion is not established by the evidence provided. The benchmark in Appendix B compares the relative-CoM slicing against an absolute-CoM implementation that uses deliberately simple AND/OR filters (Table B.1) and does not compute boundary cell sizes near the u=0 constraint (B.1) accurately; the paper itself concedes in Appendix B.1 that 'in principle, all this could be done accurately if one invests more effort into the meshing algorithm'. Thus the documented performance gap reflects the choice of a naive comparator, not an intrinsic property of the coordinate system. The 'equivalent' part of the claim is supported; the unqualified 'superior' part is not. The authors should either soften the claim to state that relative CoM are equivalent and much simpler to implement, or add a benchmark with a boundary-fitted absolute-CoM mesh that respects the curved u=0 surface.","section":"Abstract; §3.6; Appendix B.1"},{"comment":"The method's accuracy and efficiency presuppose that each relevant GC orbit crosses a unique, single-valued magnetic midplane z_mid(R) exactly twice. Section 3.4 acknowledges that additional midplanes exist (e.g., in the JT-60U divertor region, Fig. A.2) and Section 3.8 warns that stagnation and potato orbits near the trapped-passing boundary may be missed or corrupted by the midplane polynomial fit and grid. The unqualified claim in the abstract of attaining 'high numerical accuracy and efficiency with a relatively simple mesh' in the presence of Er therefore holds only for configurations satisfying this premise. The conditions under which the method is applicable should be stated in the abstract or conclusions, and it would be useful to quantify how large a fraction of orbit space is affected in the JT-60U edge region where the extra midplane exists.","section":"§3.4; §3.8; Abstract"}],"minor_comments":[{"comment":"There is a typo in the third paragraph: 'vales up to ±10 kV/m' should read 'values up to ±10 kV/m'.","section":"§4.1"},{"comment":"The marker density is referred to as Nmk in the text of Section 4.3 but as Nmpc in the figure captions and elsewhere; please unify the notation.","section":"§4.3"},{"comment":"The notation 'Maxw. − →' in Eq. (47c) is nonstandard and likely a LaTeX artifact; please replace it with a conventional equality or arrow.","section":"Eq. (47)"},{"comment":"The caption of Fig. 15 contains 'Top[59]' in the line listing launch-point coordinates, which appears to be a stray citation tag; please correct or remove.","section":"Figure 15"},{"comment":"Given the paper's length and its role as a methods-reference for VisualStart, a short 'code and data availability' statement (whether the code will be released, and on what platform) would be helpful.","section":"General"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is very long and mixes two distinct contributions: (i) a numerical method for Er-aware phase-space sampling, and (ii) a physical analysis of Er effects on orbits and frequencies. The (ii) part is interesting but somewhat tangential; the main methodological claim needs to be rephrased as suggested. Also, the paper's reference to its own earlier work is extensive, which is fine, but the absence of a comparison with any other independent GC-sampling code weakens the 'superiority' assertion. I recommend major revision with a focus on the benchmark fairness."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"You should know two things before reading this. First, the genuinely new material is good: the extension of VisualStart's midplane-based relative-CoM sampling to nonzero Er, and the identification of the \"reference point bias\" in electric frequency shifts (Section 5.5), which is a real subtlety that affects any Er-scan in relative CoM space. Second, the abstract's categorical claim that midplane-based coordinates are \"superior\" to absolute constants of motion is not established by the evidence. The benchmark in Appendix B compares against an absolute-CoM mesher the authors themselves call deliberately simple, and they concede in Appendix B.1 that a more careful meshing algorithm could match it. The body is more careful than the abstract, but the headline overreaches.\n\nWhat the paper does well: the numerical work is extensive and mostly convincing. Time-step convergence tests, the continuity check in Eq. (53), the reconstruction of the input Er from flow fields, and analytic orbit-width estimates within 2-30% all behave as they should. The paper also flags its own limitations clearly: the unique-midplane assumption (which fails in the JT-60U divertor), missed or corrupted stagnation/potato orbits near the trapped-passing boundary, grid aliasing noise, and the fact that orbit weights and Er are free inputs with no self-consistency enforced. For a methods paper, that is a commendable level of candor.\n\nThe soft spots: the superiority claim is the main one, and it is fixable -- soften the abstract or run the absolute-CoM comparison with a boundary-fitted mesh. The reader's worry about the unique-midplane assumption is valid but scoped; the authors acknowledge the failure cases and exclude them. The analytic verification in Appendix C is partly circular (same conservation laws as the integrator), but the independent checks against uniform-flow expectations in ITER and the continuity equation give me reasonable confidence the results are not just self-consistent artifacts. No code or input data is released, so independent reproduction is not possible from the paper alone; that is a real gap for a methods paper meant to feed into ITER IMAS workflows.\n\nWho this is for: anyone building GC orbit databases for low-frequency instability analysis in rotating tokamak plasmas. It is a serious, useful methods contribution and deserves a serious referee. The referee should push on the superiority claim, but the central technical content -- relative-CoM sampling with Er and the reference-point bias analysis -- is sound enough to engage with. Recommendation: yes, send it to peer review. The overstatement is correctable, and the underlying work is honest and reproducible in principle.","headline":"Solid methods paper with genuinely new content; the abstract overstates the 'superior' claim against absolute CoM, but the body is honest and the core material holds up.","tokens_in":63255,"tokens_out":3438,"would_cite":true,"duration_ms":29584,"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":"A midplane-based coordinate system for guiding-center orbits is both simpler and more accurate than constants of motion when a radial electric field is present.","keywords":["tokamak plasmas","guiding center orbits","radial electric field","constants of motion","magnetic midplane","wave-particle resonances","phase space sampling","drift orbit database"],"falsifier":"Run the same orbit database construction on the paper's worst-case test equilibrium with a full-orbit or random-launch Monte Carlo sampler that does not assume midplane crossings, including orbits near the trapped-passing boundary and any external divertor midplane, and compare flux-surface-averaged density and flow moments.","tokens_in":62193,"feed_emoji":"🧲","tokens_out":8765,"duration_ms":74291,"temperature":0.7,"pith_summary":"This paper argues that for building guiding-center drift-orbit databases in tokamaks with a radial electric field $E_r$, the best sampling coordinates are not the standard constants of motion but simpler coordinates measured at the magnetic midplane crossing. Because the valid phase-space boundary is a straight line in these relative coordinates, a coarse rectangular mesh gives accurate densities, flows, and temperatures, while conventional absolute-constant-of-motion meshes under-resolve the curved trapped-passing boundary and lose low-energy particles. The method is demonstrated on three realistic tokamak-like test cases with $E_r$ up to 30 kV/m, and orbit widths and velocity modulations are verified against analytical estimates. The paper also uncovers a practical subtlety: $E_r$ modulates the mirror force in a way that makes single-launch-point electric frequency shifts ambiguous for passing orbits, so resonance analyses need orbit-averaged or multi-launch-point frequencies. A sympathetic reader would take away that midplane-based phase-space slicing is the robust choice for rotating tokamak plasmas, and that electric Doppler shifts require care.","feed_headline":"Midplane sampling beats constants of motion with Er","feed_subtitle":"A simple mesh in midplane coordinates stays accurate where energy-momentum slicing loses trapped particles.","key_machinery":"The load-bearing object is the magnetic midplane, defined by $B \\cdot \\nabla B = 0$ at the guiding-center position. Each ordinary guiding-center orbit crosses this surface twice, so the entire orbit space can be parameterized by the coordinates at the crossing: initial kinetic energy $K_I$, pitch $\\alpha_I$ (or $\\Lambda_I \\equiv \\mu B_0/K_I$), and midplane radius $X_I$. The Jacobian for this parameterization is computed explicitly, and the volume element carries a factor $1/2$ to correct for double counting of the two crossings. The second essential mechanism is the term $E_\\nabla = (u/\\omega_B)\\,\\mathbf{v}_E \\cdot \\nabla B$ in the parallel equation of motion, the electric modulation of the mirror force, which accelerates and decelerates the guiding center periodically and produces the reference-point bias in electric frequency shifts.","core_discovery":"The paper's central claim is that when an ambient radial electric field $E_r$ is present, sampling the guiding-center orbit space with midplane-based relative constants of motion — kinetic energy $K$, pitch coordinate $\\Lambda$ (or pitch angle $\\alpha$), and midplane radius $X$ — is not only equivalent but numerically superior to slicing with conventional absolute constants of motion $\\{E, \\mu, P_\\zeta\\}$. The superiority comes from the shape of the valid-domain boundary: in relative coordinates the boundary $u = 0$ (where the parallel velocity vanishes) is a straight line, so rectangular mesh cells and their volumes are evaluated accurately; in absolute coordinates the same boundary is a curved line that depends on the potential profile $\\Phi(X)$, so a simple implementation either discards or miscounts boundary cells near deeply trapped and barely passing orbits. In the paper's worst-case test with $E_{r0} = 30$ kV/m, absolute-CoM slicing underestimated the mid-radius density by roughly 40% at moderate resolution and left a spurious deficit of particles near $u = 0$, while relative-CoM meshes converged quickly. The paper also establishes that $E_r$ modulates the mirror force through a parallel electric acceleration term that averages to zero over a poloidal transit but makes electric transit-frequency shifts reference-point dependent; for passing orbits the toroidal component of this bias cancels when launch points are summed.","pith_inferences":["The same midplane-relative-CoM trick could be adapted to other axisymmetric or quasisymmetric magnetic configurations whenever a unique $B \\cdot \\nabla B = 0$ surface exists; in configurations with multiple midplanes, the double-counting factor would need to be replaced by a sum over crossing points.","The reference-point bias implies that any numerical resonance code that scans $E_r$ while launching at one fixed phase-space point may mis-estimate the electric Doppler shift; a direct test would compare single-launch-point frequency scans against launch-point-averaged frequencies for the same orbit.","Because the $1/2$ double-counting and the Jacobian assume exactly two midplane crossings, orbit databases near the trapped-passing boundary, stagnation orbits, and divertor-region orbits are the natural failure candidates; local mesh refinement or a hybrid full-orbit check would bound the error.","For sharp $E_r$ layers, the guiding-center model's neglect of gyroaveraging becomes questionable, and a full-orbit simulation with the same fields would show whether the midplane-based method's accuracy persists where orbit squeezing is strong."],"forward_implications":["A rectangular mesh over midplane kinetic energy, pitch, and radius gives converged moments (density, flows, temperature) with $E_r$, whereas absolute-CoM meshes at similar resolution can lose a large fraction of the density in the paper's worst-case test.","$E_r$ scans of transit frequencies should be interpreted as launch-point-dependent; single-launch-point scans for passing orbits can report toroidal shifts of either sign with magnitude up to $|\\Delta u_{\\rm orb}|/2$, so resonance analyses should use orbit-averaged or multi-launch-point frequencies.","The method captures the $E_r$-induced shift of the trapped-passing boundary, which in turn explains the multi-peaked toroidal flow structure seen in the peaked-temperature test case.","Under weakly varying $E_r$, the electric precession Doppler shift of a trapped orbit agrees with the relaxed plasma toroidal rotation speed to about 10%, supporting the common assumption that precessional resonances can be analyzed in the rotating plasma frame.","With $E_r$ present, orbit weights should be assigned using orbit-averaged kinetic energy $\\langle K \\rangle$ and radius $\\langle r \\rangle$ to minimize bias from the electric modulation of the mirror force."],"supporting_citations":[{"why":"Supplies the guiding-center equations of motion and the conserved magnetic moment used throughout the orbit computations.","marker":"[5]"},{"why":"Establishes the constants-of-motion representation of unperturbed orbit surfaces that the paper uses as basis functions for distribution weights.","marker":"[7]"},{"why":"Previous description of the code and its midplane-based sampling workflow, which the present paper extends to include $E_r$ and uses as a benchmark.","marker":"[16]"},{"why":"Supplies the KSTAR plasma equilibrium, profiles, and the estimate of $E_r$ up to 30 kV/m used as a working example.","marker":"[45]"},{"why":"Provides the omnigenity and equilibrium condition involving $B \\cdot \\nabla B$ that motivates defining the magnetic midplane by $B \\cdot \\nabla B = 0$.","marker":"[46]"},{"why":"Supply the JT-60U equilibrium and experimental context of neutral-beam-driven Alfvénic instabilities used for test cases and orbit analyses.","marker":"[52, 53, 54]"}],"fun_headline_variants":["Midplane coords beat energy-momentum in Er fields","Midplane sampling outdoes constants of motion with Er","Er makes midplane coords superior to absolute invariants","In Er fields, midplane mesh beats energy-momentum slicing","Midplane coordinates win with radial electric field"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"Every guiding-center orbit that matters crosses the magnetic midplane exactly twice, with a single midplane height for each radius; if not, the double-counting factor and the mesh omit or mis-weight orbits.","fun_headline_variants_meta":{"raw":{"variants":["Midplane coords beat energy-momentum in Er fields","Midplane sampling outdoes constants of motion with Er","Er makes midplane coords superior to absolute invariants","In Er fields, midplane mesh beats energy-momentum slicing","Midplane coordinates win with radial electric field"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000512,"raw_usage":{"total_tokens":2567,"prompt_tokens":1099,"completion_tokens":1468,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":715,"completion_tokens_details":{"reasoning_tokens":1388}},"tokens_in":715,"tokens_out":1468,"duration_ms":9621,"temperature":1.0,"reasoning_tokens":1388,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T10:19:49.400983+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the same orbit database construction on the paper's worst-case test equilibrium with a full-orbit or random-launch Monte Carlo sampler that does not assume midplane crossings, including orbits near the trapped-passing boundary and any external divertor midplane, and compare flux-surface-averaged density and flow moments.","supporting_citations":[{"cited_title":"Landreman and P .J","cited_arxiv_id":null,"evidence_quote":"Provides the omnigenity and equilibrium condition involving $B \\cdot \\nabla B$ that motivates defining the magnetic midplane by $B \\cdot \\nabla B = 0$."}],"review_version":1}