{"id":"1065f8d4-02e1-4936-98d4-62c4c4b3f00f","arxiv_id":"2507.20244","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Across gyrokinetic simulations of five tokamaks, negative triangularity destabilizes micro-tearing modes through faster magnetic drifts, reversing its usual transport benefit in high-beta, high-shear regimes.","lead":"This paper uses gyrokinetic simulations of five tokamaks to show that negative triangularity plasma shaping makes micro-tearing magnetic turbulence stronger than positive triangularity when pressure, shear, and temperature gradients are high. The finding warns that spherical tokamaks with negative triangularity could lose their transport advantage unless magnetic shear is kept low.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Linear kyρi=0.2 regime maps do not establish nonlinear MTM dominance; the paper's own DEMO nonlinear run is a counterexample, so the 'all NT scenarios' conclusion overreaches.","rationale":"The reader's weakest-assumption analysis identifies exactly the same load-bearing concern: the paper builds its regime maps from linear simulations at a single wavenumber kyρi = 0.2 and then generalizes to nonlinear transport. My independent reading confirms this is the most consequential soft spot. I also looked for independent support that could rescue the claim. The mechanism section (Sec. VI) is a genuine strength: the artificial modification of vDy(z) shows that the poloidally averaged magnetic drift velocity controls the linear growth rate, and the nonlinear drift-swap in pMTM-1 directly demonstrates that NT magnetic drifts, not FLR effects, are responsible for stronger MTM transport. The SMART nonlinear runs, the flux spectra, and the field-line diffusion diagnostics provide a consistent, concrete demonstration that NT can become MTM-dominated and transport much worse in a spherical-tokamak-like regime. The vulnerability is not internal inconsistency in the linear simulations but unsupported extrapolation from linear dominance at one ky to saturated transport. The authors are partially aware of this: they explicitly acknowledge in Section V that DEMO nonlinear transport remains ITG-dominated despite linear MTM instability at kyρi = 0.2, and they offer a plausible reason (higher-ky ITG modes dominate nonlinearly). That honesty is a credit, but the abstract and the conclusions bullet remove the caveat and state that all NT scenarios enter an MTM-dominated regime. This matters because the headline scientific message is a transport statement: MTMs, 'when present, make transport much worse than in positive triangularity.' The DEMO counterexample shows that linear presence of MTMs does not guarantee that statement at the transport level. If conventional tokamaks with NT at high beta remain ITG-dominated nonlinearly, as DEMO suggests, then the 'worse transport' claim is only evidenced for SMART-like STs, and the regime boundaries in Figs. 5-9 are transport regimes only after an additional, unvalidated step. The proposed nonlinear test on TCV or DIII-D at a map-predicted MTM corner would settle whether the single counterexample is an isolated DEMO quirk or a general failure of the single-ky linear proxy. No ad hominem is intended; the critique is on the inference. The verdict remains CONDITIONAL: the central mechanism is credible and the ST warning is supported, but the broad 'all NT scenarios' generalization needs either nonlinear confirmation or an explicit caveat.","tokens_in":25551,"tokens_out":3606,"duration_ms":44360,"concrete_test":"Run a nonlinear GENE flux-tube simulation for the NT TCV (or DIII-D) scenario at a point inside the MTM-dominated region of Fig. 5 (or Fig. 6), e.g. β = 0.5%, s_hat = 2.5, ω_Te/ω_Ti = 1.5, using the resolution and Lx ~ 250ρi described in Sec. V and γ_ExB = 0.05-0.1 cs/R. If the saturated electron electromagnetic heat flux Qe,em is not the dominant channel (Qe,em/Qe,es not much greater than 1) and ⟨D_fl⟩ stays below unity in the Fig. 16 sense, the linear kyρi = 0.2 maps do not define the actual transport regime and the 'all NT scenarios' generalization fails; if Qe,em does dominate and field lines stochastize, the caveat is limited to DEMO and the current conclusion stands.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing step is the inference from the linear regime maps (Figs. 5-8) to the conclusion bullet that 'all NT scenarios entered an MTM dominated regime for β ≳ 0.3%, s_hat ≳ 2.5, ω_Te/ω_Ti > 1.' Those maps are single-wavenumber linear scans at kyρi = 0.2. The paper itself provides a direct counterexample to this inference. Section V reports that for DEMO at β = 0.6-0.8% the kyρi = 0.2 mode is MTM (Fig. 7), yet the nonlinear fluxes in Fig. 12 are dominated by the electrostatic ion heat flux, i.e., ITG, and the paper explains that linear ky scans (not shown) have ITG at higher ky, so 'MTMs are not strong enough linearly to dominate nonlinear transport and remain weak.' Thus linear MTM instability at the reference wavenumber is not sufficient for MTM-dominated transport. Since no nonlinear run is presented for TCV or DIII-D in the high-β, high-s_hat corners where Figs. 5-6 and Fig. 9 predict an MTM-dominated NT regime, the blanket claim rests on an unvalidated linear-to-nonlinear mapping. The physical mechanism claim (faster poloidally averaged magnetic drifts in NT) is supported by the drift-swap experiments in Sec. VI and survives this concern, but the transport-level conclusion does not.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This manuscript uses local flux-tube GENE simulations to compare negative- and positive-triangularity versions of five devices (TCV, DIII-D, MAST-U, SMART, and EU-DEMO). It claims that NT geometry is more susceptible to micro-tearing modes, that the destabilization is caused by faster poloidally averaged magnetic drifts in NT, and that at sufficiently large beta, magnetic shear, and electron-to-ion temperature gradient ratio, all NT scenarios enter an MTM-dominated regime with transport several times worse than in PT. The linear regime maps are built from single-wavenumber scans at kyρi = 0.2; the supporting nonlinear evidence consists of beta scans for DEMO and SMART plus two pure-MTM scenarios with magnetic-drift and FLR profile swaps. The paper notes that nonlinear MTM simulations are numerically challenging and that only a limited set of nonlinear runs was possible.","tokens_in":25798,"tokens_out":6178,"duration_ms":75522,"significance":"If the transport-level conclusion were established, it would matter directly for NT reactor design: NT spherical tokamaks could suffer MTM-driven confinement degradation at high beta, and NT pedestal formation might be limited by the same mechanism. The manuscript's genuine strengths are the multi-machine scope, the direct nonlinear SMART demonstration of MTM dominance with field-line stochasticity diagnostics, and a clean mechanistic test in which vDy and FLR profiles are swapped between NT and PT (Figs. 17-19). The drift-based mechanism is well supported by the simulations. However, the paper's central transport-level claim is currently supported mainly by single-wavenumber linear scans and is contradicted by the paper's own nonlinear DEMO results, so the conclusion as stated overreaches. The paper does not ship machine-checked proofs or code, but the direct gyrokinetic simulations and the explicit drift-swap experiments are reproducible in principle and represent a useful contribution to the MTM literature.","major_comments":[{"comment":"The paper's own nonlinear DEMO results contradict the inference from the linear regime maps. In Fig. 7 the kyρi = 0.2 mode in NT DEMO is identified as MTM for β ≳ 0.6%, but Fig. 12 shows that for the same β values the electrostatic electron and ion heat fluxes dominate, with Qe,em remaining small and Qi,es the largest channel. The text in Section V states that MTMs 'are not strong enough linearly to dominate nonlinear transport and remain weak.' Linear MTM dominance at a single wavenumber is therefore not sufficient for MTM-dominated transport, and the conclusion bullet that 'All NT scenarios entered an MTM dominated regime when β ≳ 0.3%, s_hat ≳ 2.5, ωTe/ωTi > 1' cannot be inferred from the linear maps alone. This is load-bearing for the abstract's claim that MTMs 'make transport much worse than in positive triangularity.'","section":"Section V, Figs. 7 and 12"},{"comment":"The reconciliation of the linear and nonlinear DEMO results relies on unshown ky scans: the text says that linear ky scans of NT DEMO with β ≳ 0.6% (not shown here) show ITG modes at kyρi > 0.2. These scans are the only evidence that the kyρi = 0.2 MTM is unrepresentative of the full wavenumber spectrum, and they are used to explain why the nonlinear transport remains ITG-dominated. Since this point is central to limiting the regime maps, the scans should be presented in the paper or included as supplementary material; without them the reader cannot verify the reconciliation.","section":"Section V, 'linear ky scans ... (not shown here)'"},{"comment":"The nonlinear confirmation is limited to DEMO and SMART at their reference shears and with externally imposed flow shear (γExB = 0.1 cs/R for SMART and 0.05 cs/R for DEMO), and for SMART with fixed ωTe/ωTi = 1. No nonlinear simulation is reported for TCV, DIII-D, or MAST-U in the high-β, high-s_hat corners where Figs. 5-6 and 9 predict MTM dominance in NT. The 'all NT scenarios' conclusion therefore extrapolates beyond the nonlinear evidence. To support the transport-level claim, the authors should either add nonlinear cases in at least one of the predicted MTM-dominated corners or restrict the conclusion to linear susceptibility and to the machines for which nonlinear runs are presented.","section":"Section V, Figs. 11-13"},{"comment":"The drift-mechanism test uses the parametrization vDy(z) = σ vDy^NT(z) + vD0, which varies the magnitude and offset of the NT drift profile but not its shape independently. The statement that the growth rate depends only on ⟨vDy⟩ and not on how the profile changes with poloidal angle is therefore established only within this one-parameter family of profiles. This is a caveat rather than a fatal flaw, because the nonlinear drift-swap experiment in Fig. 19 points in the same direction, but the linear text in Section VI.A.1 is somewhat stronger than the test actually demonstrates.","section":"Section VI.A, Eq. (4)"}],"minor_comments":[{"comment":"The sentence 'this simplifies th way geometry enters' contains a typo; it should read 'the way geometry enters'.","section":"Section VI.A.1"},{"comment":"The phrase 'The the solid circles show no difference' contains a duplicated article; it should read 'The solid circles show no difference.'","section":"Section VI.A.2"},{"comment":"The sentence 'The reminder of the paper is structured as follows' should say 'remainder of the paper.'","section":"Section III"},{"comment":"The text referring to an MTM 'in figure 1(b)' appears to be a figure-reference error: panel 1(b) is the parallel vector potential for the ITG row, while the corresponding MTM quantity is in panel 1(e).","section":"Section II, Fig. 1"},{"comment":"The mode-identity distinction in the frequency colormaps relies on color intensity alone (light blue for TEM, darker blue for MTM). An explicit mode-boundary overlay or hatching would make the MTM-dominated regions much easier to read and would reduce the risk of misclassification in regions with continuous color variation.","section":"Section IV, Figs. 5-8"}],"recommendation":"major_revision","confidential_remarks":"The paper fits the scope of physics.plasm-ph and the reliance on Ref. [20] for methodology is natural and appropriate. I do not see a circularity problem in the drift mechanism: the mechanism is tested by direct profile swaps rather than inferred from a model that assumes it. The main issue is the gap between single-wavenumber linear regime maps and the transport-level conclusion, which the paper's own nonlinear DEMO run exposes. That gap is fixable by adding the missing ky scans, adding or referencing a nonlinear case in a regime where MTM dominance is actually observed, and adjusting the abstract and conclusion wording to distinguish linear susceptibility from nonlinear transport dominance."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"You should know two things about arXiv:2507.20244. First, it contains a genuinely new result: negative triangularity makes micro-tearing modes more unstable than positive triangularity, and the authors identify a concrete mechanism — faster poloidally averaged magnetic drifts in NT — supported by targeted numerical experiments. Second, the paper overreaches in its headline conclusion: the regime maps that supposedly show \"all NT scenarios enter an MTM-dominated regime\" are single-wavenumber linear scans, and the paper's own nonlinear DEMO runs directly contradict that blanket statement.\n\nThe mechanism section is the strongest part. The artificial drift-profile scans in Fig. 17 collapse onto a single curve relating growth rate to averaged drift velocity, which is a clean result. Swapping NT drifts into a PT geomtry nonlinearly produces MTM-dominated fluxes, while swapping FLR profiles does essentially nothing. That is good, falsifiable evidence for the drift mechanism. The multi-machine comparison (TCV, DIII-D, DEMO, SMART, MAST-U) is also a step beyond Ref. [25] and gives the paper a coherent phenomenological picture rather than a single-device anecdote.\n\nThe soft spots are proportional. The main one is the linear-to-nonlinear inference. The regime maps (Figs. 5–9) use kyρi = 0.2 only, and the paper itself states that for DEMO at β = 0.6–0.8% the kyρi = 0.2 mode is MTM but nonlinear transport is ITG-dominated because higher-ky ITG modes win nonlinearly. So linear MTM dominance at one wavenumber is not sufficient for MTM-dominated transport. That means the abstract bullet — \"all NT scenarios entered an MTM dominated regime for β ≳ 0.3%, ŝ ≳ 2.5 ...\" — is too strong without the nonlinear caveat. The SMART nonlinear runs do support a reversal at high β, but they use a hand-chosen E×B shear and fix ωTe/ωTi = 1, which weakens MTM drive and makes convergence easier. That is not fatal, but it should be stated more visibly as a limitation. I also note the artificial equilibria flip only δ and sδ while keeping everything else fixed; that is a standard isolation approach and is fine, though it means the PT/NT pairs are not self-consistent experimental equilibria. No public input files or code were released; the paper would be stronger with them.\n\nOverall, the central physical claim about faster magnetic drifts in NT holds up. The transport-level prediction needs a guarded statement. This paper deserves a serious referee: it is important for NT fusion power plant design and for spherical tokamak planning. I would recommend accept with major revision — require the authors to soften or qualify the regime-map conclusion, add the DEMO nonlinear counterexample into the abstract, and either provide input files or justify why they are withheld. I would bring this to the reading group; it is worth debating.\n\nRecommendation: send to peer review with expectation of heavy revision.","headline":"Multi-machine evidence that negative triangularity destabilizes micro-tearing modes via faster average magnetic drifts, but the overbroad regime-map claim needs a nonlinear caveat before it is fully convincing.","tokens_in":26387,"tokens_out":1370,"would_cite":true,"duration_ms":19996,"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":"Negative triangularity makes micro-tearing modes stronger, and when those modes dominate, heat transport is several times worse than in positive triangularity.","keywords":["negative triangularity","micro-tearing modes","gyrokinetic simulation","electromagnetic turbulence","magnetic drift","tokamak transport","spherical tokamak","magnetic shear"],"falsifier":"A concrete check is to run a well-resolved nonlinear flux-tube simulation at a parameter point where the linear maps predict NT MTM dominance but where no nonlinear run has been done — for example, NT SMART with $\\beta$ between $0.3\\%$ and $0.6\\%$ and $\\omega_{T_e}/\\omega_{T_i}>1$ — and observe whether the saturated electromagnetic electron heat flux actually exceeds the electrostatic ion heat flux. A second, experimental falsifier would be to measure the electron temperature gradient and confinement in an NT spherical tokamak as $\\beta$ is raised toward the predicted onset: if the temperature profile stiffens and confinement degrades relative to PT, the regime is real; if it does not, the linear-to-nonlinear mapping used for the onset boundaries is wrong.","tokens_in":25256,"feed_emoji":"🔥","tokens_out":9602,"duration_ms":103045,"temperature":0.7,"pith_summary":"Negative triangularity (NT) shaping has a well-documented ability to reduce electrostatic turbulence and improve confinement, but this paper argues that the benefit does not extend to micro-tearing modes (MTMs), the small-scale electromagnetic instabilities that break magnetic field lines into stochastic islands. Using linear and nonlinear flux-tube gyrokinetic simulations of five tokamak scenarios (TCV, DIII-D, MAST-U, SMART, and EU-DEMO), the authors find that NT geometry makes MTMs unstable more easily than positive triangularity (PT) geometry. When MTMs dominate, NT heat fluxes are several times larger than PT heat fluxes, reversing the usual ordering. The paper traces this to faster poloidally averaged magnetic drifts in NT geometry. A sympathetic reader would therefore take the paper as establishing a boundary on where NT is advantageous: conventional tokamaks usually sit far from the MTM onset and keep the NT benefit, while spherical tokamaks, with their naturally high $\\beta$ and magnetic shear, can cross into an MTM-dominated regime and lose it.","feed_headline":"Negative triangularity worsens micro-tearing turbulence","feed_subtitle":"Across five tokamaks, the shaping advantage vanishes at high plasma pressure and shear, especially in spherical tokamaks.","key_machinery":"The load-bearing object is the micro-tearing mode (MTM), an electromagnetic micro-instability that creates thin current layers, shears the magnetic field, and forms gyroradius-scale magnetic islands whose overlap stochastizes field lines and lets electrons escape radially. The comparative argument runs through the poloidally averaged binormal magnetic drift velocity $\\langle v_{Dy}\\rangle = \\int J v_{Dy}\\,dz / \\int J\\,dz$, which is faster in NT than in PT; linear scans in which the drift profile is artificially rescaled or shifted show that MTM growth rates track $\\langle v_{Dy}\\rangle$ rather than the profile shape, and nonlinear simulations in which NT drifts are imposed on PT geometry recover the NT-level electromagnetic heat flux. Mode identity is established with the parity of the parallel vector potential $P(A_\\parallel)$: $P(A_\\parallel)<1$ and negative real frequency identify an MTM, separating it from ITG and TEM. A local analytic equilibrium parametrization is used to flip triangularity $\\delta$ and its shear $s_\\delta$ while holding all other profiles fixed, isolating geometry as the variable.","core_discovery":"On the paper's own terms, the central discovery is that negative triangularity is more susceptible to micro-tearing modes than positive triangularity, and that when MTMs are present they make transport much worse in NT. The claim is supported by a coherent multi-machine picture: at sufficiently large plasma pressure $\\beta$, magnetic shear $\\hat{s}$, and electron-to-ion temperature gradient ratio $R/L_{T_e} / R/L_{T_i}$, every NT scenario in the study becomes MTM-dominated, while the corresponding PT scenario remains dominated by electrostatic ITG or TEM turbulence. The qualitative explanation is that the magnetic drifts are faster in NT geometry, measured through the poloidally averaged binormal magnetic drift velocity $\\langle v_{Dy}\\rangle$, and faster drifts make MTMs more unstable. Nonlinear simulations confirm the ordering in the spherical-tokamak case: NT SMART has heat fluxes about 1.5 times larger than PT once $\\beta \\gtrsim 0.3\\%$, with a strong electromagnetic electron heat flux and strongly diffusing magnetic field lines. The paper also notes one place where linear extrapolation fails: for NT DEMO at $\\beta \\sim 0.6$–$0.8\\%$, linear MTMs exist but nonlinear transport remains ITG-dominated.","pith_inferences":["This reader's extension: if the mechanism is the poloidally averaged magnetic drift, then shaping changes beyond triangularity that raise $\\langle v_{Dy}\\rangle$ — stronger elongation, larger Shafranov shift, or particular current-density profiles — should also destabilize MTMs; this could be tested directly with the same drift-swapping simulations used here.","This reader's extension: the DEMO discrepancy suggests that the regime maps, built at a single wavenumber $k_y\\rho_i=0.2$, should be treated as necessary rather than sufficient for nonlinear MTM dominance; the boundaries could be sharpened by nonlinear checks at a few points where linear MTMs are marginal.","This reader's extension: the proposed H-mode link implies a concrete experimental signature — in NT plasmas approaching pedestal conditions, electron temperature profiles should stiffen or flatten exactly where linear MTMs are predicted, and heating scans should show a confinement saturation that PT plasmas do not.","This reader's extension: integrated modeling chains for NT power plants would need electromagnetic transport self-consistently, since linear stability alone can both overstate (DEMO) and understate (SMART) the MTM transport level."],"forward_implications":["Conventional-aspect-ratio tokamaks (TCV, DIII-D, DEMO at their nominal parameters) keep the NT transport benefit, because their local $\\beta$ is too low to enter the MTM-dominated window identified here ($\\hat{s} \\gtrsim 2.5$, $\\beta \\gtrsim 0.3\\%$, $R/L_{T_e}/R/L_{T_i}>1$).","Spherical tokamaks such as SMART and MAST-U operate near that window, so NT shaping there can degrade rather than improve confinement; lowering magnetic shear (for instance by avoiding double-null configurations) or steepening the density gradient is predicted to suppress MTMs and restore the NT benefit.","When MTMs dominate, the electron electromagnetic heat flux becomes the main loss channel and the average magnetic field-line diffusion coefficient rises by more than two orders of magnitude, so NT performance in that regime is set by magnetic stochasticity rather than by electrostatic turbulence.","The same mechanism could help explain why NT plasmas resist H-mode pedestal formation: during pedestal buildup the conditions are exactly the high-$\\beta$, high-shear, flat-density-gradient conditions that trigger MTMs, and the required sustaining heat flux is unrealistically high.","If the paper's regime boundaries are correct, reactor design for NT must actively avoid the MTM window rather than assume the electrostatic NT benefit carries over unchanged."],"supporting_citations":[{"why":"Establishes the earlier result that NT stabilizes ITG/TEM via faster magnetic drifts and finite-Larmor-radius effects, which this paper extends to MTMs.","marker":"[20]"},{"why":"The only prior study of the triangularity-MTM interplay; this paper broadens it to five machines and nonlinear simulations.","marker":"[25]"},{"why":"Shows with linear gyrokinetic simulations that MTMs can dominate in MAST and are destabilized by beta, collisionality, and electron temperature gradient.","marker":"[36]"},{"why":"One of the first nonlinear gyrokinetic simulations with dominant MTMs, showing magnetic stochastization and enhanced electron transport.","marker":"[27]"},{"why":"Nonlinear simulations showing MTMs can dominate in standard-aspect-ratio tokamaks and confirming the role of beta and electron temperature gradient.","marker":"[28]"},{"why":"Provides a pure-MTM scenario used here and demonstrates the importance of electrostatic potential and magnetic drifts in destabilizing MTMs.","marker":"[30]"},{"why":"Defines the field-line diffusion coefficient used to quantify magnetic stochasticity in the nonlinear simulations.","marker":"[47]"},{"why":"Provides the local equilibrium parametrization used to specify shaping and flip triangularity while holding other profiles fixed.","marker":"[38]"}],"fun_headline_variants":["Negative triangularity worsens micro-tearing","Micro-tearing turbulence intensifies with NT shape","NT plasma geometry amplifies micro-tearing modes","Shaping effect: NT accelerates micro-tearing"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that linear MTM dominance at a single wavenumber, $k_y\\rho_i=0.2$, can be used to map out where turbulence will be MTM-dominated; the paper itself reports one nonlinear case (NT DEMO at $\\beta\\simeq0.6$–$0.8\\%$) where this fails, so the regime boundaries inherit that uncertainty.","fun_headline_variants_meta":{"raw":{"variants":["Negative triangularity worsens micro-tearing","Micro-tearing turbulence intensifies with NT shape","NT plasma geometry amplifies micro-tearing modes","Shaping effect: NT accelerates micro-tearing"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000807,"raw_usage":{"total_tokens":3607,"prompt_tokens":1070,"completion_tokens":2537,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":686,"completion_tokens_details":{"reasoning_tokens":2479}},"tokens_in":686,"tokens_out":2537,"duration_ms":27379,"temperature":1.0,"reasoning_tokens":2479,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T13:41:33.812539+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A concrete check is to run a well-resolved nonlinear flux-tube simulation at a parameter point where the linear maps predict NT MTM dominance but where no nonlinear run has been done — for example, NT SMART with $\\beta$ between $0.3\\%$ and $0.6\\%$ and $\\omega_{T_e}/\\omega_{T_i}>1$ — and observe whether the saturated electromagnetic electron heat flux actually exceeds the electrostatic ion heat flux. A second, experimental falsifier would be to measure the electron temperature gradient and confinement in an NT spherical tokamak as $\\beta$ is raised toward the predicted onset: if the temperature profile stiffens and confinement degrades relative to PT, the regime is real; if it does not, the linear-to-nonlinear mapping used for the onset boundaries is wrong.","supporting_citations":[{"cited_title":"Aucone, P","cited_arxiv_id":null,"evidence_quote":"Establishes the earlier result that NT stabilizes ITG/TEM via faster magnetic drifts and finite-Larmor-radius effects, which this paper extends to MTMs."},{"cited_title":"Marinoni, S","cited_arxiv_id":null,"evidence_quote":"The only prior study of the triangularity-MTM interplay; this paper broadens it to five machines and nonlinear simulations."},{"cited_title":"Maeyama, T.-H","cited_arxiv_id":null,"evidence_quote":"Shows with linear gyrokinetic simulations that MTMs can dominate in MAST and are destabilized by beta, collisionality, and electron temperature gradient."},{"cited_title":"Balestri, J","cited_arxiv_id":null,"evidence_quote":"One of the first nonlinear gyrokinetic simulations with dominant MTMs, showing magnetic stochastization and enhanced electron transport."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Nonlinear simulations showing MTMs can dominate in standard-aspect-ratio tokamaks and confirming the role of beta and electron temperature gradient."},{"cited_title":"Merlo and F","cited_arxiv_id":null,"evidence_quote":"Provides a pure-MTM scenario used here and demonstrates the importance of electrostatic potential and magnetic drifts in destabilizing MTMs."},{"cited_title":"Parisi, W","cited_arxiv_id":null,"evidence_quote":"Defines the field-line diffusion coefficient used to quantify magnetic stochasticity in the nonlinear simulations."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the local equilibrium parametrization used to specify shaping and flip triangularity while holding other profiles fixed."}],"review_version":1}