Pith. sign in

REVIEW 3 major objections 6 minor 46 references

Turbulence and Transport in Spectrally Accelerated full-f Gyrokinetic Simulations

T0 review · 3 major / 6 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Spectral full-f gyrokinetic simulations reproduce TEM-driven edge turbulence and separatrix power to within 10% of experiment.

desk verdict A solid numerical-consistency study of the spectral full-f GENE-X scheme against grid-based TEM turbulence; the physical validation is not yet independent and the abstract oversells the Psep match. read the letter →

arxiv 2505.04832 v1 pith:5536EBY4 submitted 2025-05-07 physics.plasm-ph

B. J. Frei , P. Ulbl , C. Pitzal , W. Zholobenko , F. Jenko This is my paper · ORCID
classification physics.plasm-ph PACS 52.35.Ra52.65.-y52.55.Fa
keywords gyrokineticsfull-fsimulationspectralvelocity-spaceHermite-Laguerreexpansionedgeturbulencescrape-offlayertrappedelectronmodeseparatrixpower
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper argues that a spectral, velocity-space version of a full-f gyrokinetic turbulence code reproduces the same edge and scrape-off-layer turbulence as the established grid-based version, at far lower computational cost. Using the TCV-X21 L-mode scenario, it shows that the spectral expansion captures trapped-electron-mode (TEM) fluctuations, their frequencies and phase shifts, the resulting particle and heat fluxes, and the power crossing the separatrix. If correct, this establishes the spectral approach as a first-principles solver that can predict edge transport in tokamaks with the fidelity of full-f gyrokinetics but with an order-of-magnitude fewer velocity-space degrees of freedom.

What carries the argument

The machinery is a global spectral discretization of velocity space: the gyrocenter distribution function is expanded in scaled Hermite polynomials in parallel velocity and Laguerre polynomials in magnetic moment, so that the gyrokinetic Vlasov equation becomes evolution equations for the spectral coefficients, which are velocity moments. Truncation plus a diagonal damping term closes the system, and the quasineutrality, Ampère, and Ohm equations are written directly in terms of moments. This reduces the velocity-space degrees of freedom by about an order of magnitude while retaining enough kinetic structure for TEMs.

What would settle it

Run the spectral simulation at resolutions above (16,8) and a grid-based reference at resolutions above (80,60); if the separatrix power moves away from the experimental 120 kW by more than the statistical uncertainty, the convergence claim fails. Alternatively, measure the density-fluctuation frequency spectrum in TCV-X21 directly: the paper predicts a strongly electron-diamagnetic-propagating band near 0.5 MHz at kyρs ≈ 0.4, so its absence would contradict the TEM interpretation.

Watch

Extended reading notes

Core claim

The central claim is that a spectral full-f gyrokinetic formulation in velocity space is not just a cheap surrogate for profiles: it faithfully reproduces the TEM-driven turbulent state. For spectral resolutions (4,2), (8,4), and (16,8), the fluctuation spectra, mode propagation in the electron diamagnetic direction, phase shifts between density and temperature fluctuations and the potential, and the flux spectra match the high-fidelity grid-based simulation; the (16,8) case gives a separatrix-crossing power of 131.7 kW, within 10% of the experimental TCV value of 120 kW and close to the grid-based 125.2 kW. The paper also verifies the radial force balance and decomposes the radial electric field in the edge and scrape-off layer, and contrasts the TEM-dominated gyrokinetic result with a Braginskii-like fluid model that misses TEMs entirely.

Load-bearing premise

The paper treats the (80,60) grid-based simulation as the true turbulent reference; if that reference is under-resolved or shares the same systematic modeling error, the close match does not by itself establish physical accuracy.

Editorial extensions

If this is right

  • The spectral full-f approach can predict edge and scrape-off-layer turbulence and transport at a fraction of the cost of grid-based full-f simulations, making routine TEM-resolving edge simulations more feasible.
  • If the agreement with the grid-based reference holds, separatrix power can be predicted to about 10% in this L-mode scenario, which is strong evidence for predictive transport modeling.
  • Braginskii-like fluid models that neglect trapped-electron physics can misattribute or under-predict transport; in TCV-X21 they miss the dominant TEM channel by an order of magnitude in separatrix power.
  • Resolution sensitivity is weak, so a small fixed set of spectral coefficients may be sufficient across edge conditions as long as the dominant instabilities remain similar.
  • The verified radial force balance means the long-wavelength radial electric field and flows are consistent with the turbulence in the full-f simulation, a prerequisite for credible transport predictions.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A natural test of the claim is to apply the spectral approach to a second experimental scenario with different turbulence (for example ITG- or pedestal-dominated) and check whether the same few spectral coefficients remain sufficient.
  • The radial-force-balance and Er decomposition could be used as routine quality diagnostics in full-f simulations; a simulation that does not satisfy the balance is probably not converged.
  • The finding that the fluid diamagnetic flux overestimates the kinetic one by roughly a factor of three suggests that fluid models may need kinetic corrections beyond pressure anisotropy, even where TEMs are weak.
  • The spectral method's efficiency may open the door to electromagnetic or multi-ion edge simulations that are currently too expensive in grid-based full-f codes.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 6 minor

Summary. This manuscript reports an extended validation and physics analysis of the spectrally accelerated full-f gyrokinetic code GENE-X on the TCV-X21 L-mode scenario. Building on the earlier spectral implementation paper [Frei et al., arXiv:2411.09232], the authors compare spectral simulations at three velocity-space resolutions against grid-based GENE-X simulations: normalized OMP gradients, radial force balance, the decomposition of the radial electric field in the edge and SOL, turbulence spectra, phase shifts between fluctuations, turbulent particle and heat fluxes, power crossing the separatrix, divertor heat flux, and diamagnetic flux contributions. They also contrast the gyrokinetic results with drift-reduced Braginskii simulations using GRILLIX, finding that the fluid model misses the TEM-driven transport that dominates in the GK simulations. The central claim is that the spectral approach quantitatively reproduces the grid-based results, including a separatrix power Psep = 131.7 kW for the (16,8) spectral case, close to the experimental TCV value of about 120 kW.

Significance. If the central comparison is sound, the paper makes a useful contribution: it demonstrates that a small spectral velocity-space basis is sufficient to capture not only mean profiles but also the fluctuation spectra, phase shifts, fluxes, and power balance that are usually viewed as demanding full velocity-space resolution. The manuscript is generally well organized, and the authors deserve credit for including detailed diagnostics, an explicit force-balance verification, a treatment of diamagnetic fluxes, a comparison against a lower-fidelity fluid model, and a public data record for the grid-based simulations. The significance is, however, conditional: the primary validation is a comparison between two discretizations of the same gyrokinetic model within the same code family, and the single experimental number used as an external anchor is a time- and surface-averaged power without a stated statistical uncertainty. The paper would be strengthened by an explicit statement of this limitation and by correcting the overbroad 'within 10%' phrasing.

major comments (3)
  1. [Abstract and Section 6.2, Table 1] The abstract and Section 6.2 state that the separatrix power agrees 'within 10%' with both grid-based results and experimental measurements, but Table 1 supports this only for the (16,8) spectral case. Relative to the grid-based 125.2 kW, the (4,2) case gives 142.4 kW (about 14% high) and the (8,4) case gives 144.9 kW (about 16% high); only the (16,8) case is within 10%. Relative to the experimental 120 kW, the (8,4) case is about 21% high and the (4,2) case about 19% high. The claim should be qualified to the (16,8) resolution or replaced by a stated range, and the statistical uncertainty of the 0.1 ms time average should be reported.
  2. [Section 3 versus Table 1] There is an unresolved inconsistency between the stated fidelity anchor and the reference actually reported. Section 3 says the grid-based simulations use (Nv_parallel, Nmu) = (80,60) and are treated as the highest-fidelity reference, while Table 1 lists only an '(80,24) Grid' case. As written, the reader cannot tell whether the black curves in Figures 5-10 and the 125.2 kW row come from the (80,60) case or the (80,24) case. If the actual reference is (80,24), the paper's stated justification for treating it as converged rests on an unpublished thesis reference and a qualitative statement that Nmu could be reduced to about 20. A documented velocity-space convergence test, or at minimum a clear statement of which grid resolution underlies every comparison, is needed before the agreement can be interpreted as a validation of the spectral method at the claimed fidelity level.
  3. [Section 3 and Section 6.2] The external anchor for physical accuracy is a single number, Psep = 120 kW, and the simulations are run with Dirichlet boundary conditions that impose experimental density and temperature profiles at the inner and outer radial boundaries. This setup is reasonable, but it means the agreement with experiment is not a fully independent check of the model: the boundary conditions already contain experimental profile information, and the separatrix power is a single scalar without a reported experimental or simulation uncertainty. I recommend adding a sentence that explicitly identifies which predictions are genuinely parameter-free and independent of the imposed boundary profiles, and which comparisons should be read as numerical self-consistency checks between the two discretizations.
minor comments (6)
  1. [Section 4.2, after Eq. (12)] The sentence 'Similarly, B_phi and B_phi are the toroidal and poloidal components' should read 'B_phi and B_theta'.
  2. [Section 2.2] The collision operator is called 'Lernard-Bernstein Daugherty' in Section 2.2; the standard spelling is 'Lenard-Bernstein/Dougherty' as used in Section 2.1.
  3. [Section 6.3, Figure 10] The divertor heat-flux comparison is explicitly qualitative, and the (4,2) case is excluded due to spurious oscillations in q_parallel_alpha. This is a reasonable choice, but the figure and text should make clear that the 'good agreement' statement in the conclusions refers only to the (8,4) and (16,8) cases and to the ion channel; the electron peak is about 15% higher and narrower than the grid-based result, and the falloff length is about 20% shorter.
  4. [Table 1 caption] The caption reads 'Psep in from TCV [19]'; it should read 'Psep = 120 kW from TCV [19]' or similar.
  5. [Section 4.3] The discussion of the Er decrease around rho_pol less than about 0.85 attributes it to the inner Dirichlet boundary condition on u_parallel_i; this is a useful caution, but it would be helpful to state explicitly that this feature is therefore not a physics prediction for TCV but a boundary-condition effect.
  6. [Section 6.4] In the diamagnetic-flux analysis, the conclusion that these fluxes are negligible is based on the (8,4) spectral run only; the text should state whether the same conclusion is expected to hold for the (16,8) run that is used for the main Psep comparison.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: spectral-vs-grid comparison is discretization consistency, and the experimental Psep comparison is an external anchor.

full rationale

The paper's central claim is that the spectrally accelerated GENE-X approach reproduces grid-based GENE-X turbulence and transport. This is a numerical self-consistency comparison between two discretizations of the same full-f gyrokinetic model, not a derivation in which an output is defined as an input. The spectral equations (7)-(11) are obtained by projecting the GK Vlasov equation and Maxwell equations onto a Hermite-Laguerre basis; nothing in that construction is fitted to the turbulence statistics being compared. The power-balance result in Table 1 is anchored to an external benchmark: Psep = 120 kW from TCV [19], and no parameter is tuned to match that value in either the spectral or grid-based runs. The analytical estimates in Sections 4.3-4.4 and Appendix A are derived from the same GK equations and then checked against the simulation; they are consistency diagnostics rather than independent predictions whose agreement would constitute a circularity. Self-citations to [18], [7], [27], and [29] are present, but the load-bearing spectral-vs-grid comparisons in Figures 5-10 are between two independent numerical discretizations and do not reduce to a self-citation chain: the spectral method would still be tested against the grid method even if every cited paper were removed. The noted inconsistency between the (80,60) reference resolution stated in Section 3 and the (80,24) grid case reported in Table 1 is a fidelity/convergence concern, not a circularity; an under-resolved reference would weaken the validation but would not make any predicted quantity equal to an input by construction. No step in the paper, by its own equations or by a fitted parameter renamed as a prediction, reduces to its own inputs.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The central validation rests on the grid-based GENE-X reference, the LBD collision model, and experimental boundary conditions, none of which are independently derived or externally benchmarked in this paper beyond the single P_sep comparison.

free parameters (3)
  • Reference temperature tau_alpha for spectral basis = 20 eV (Tref)
    Scales the Hermite-Laguerre basis (Section 2.2 and companion paper [18]); a numerical conditioning choice, not fit to the target data.
  • Spectral diagonal damping coefficient = not stated
    Added in (8) to mitigate finite spectral resolution (Section 2.2); the value is not given and its effect on turbulent amplitudes is not quantified.
  • GRILLIX heat-flux limiters = alpha_e = alpha_i = 1.0
    Chosen in Section 7 to reproduce the Landau-fluid closure; changing them changes the fluid comparison results.
assumptions (5)
  • domain assumption The long-wavelength gyrokinetic model (Equations 1-6) is a valid description of the TCV-X21 edge plasma
    The paper adopts this model without deriving or validating it against lower-level theory; it is the foundation of both the grid and spectral simulations.
  • domain assumption The Lenard-Bernstein/Dougherty collision operator is sufficient; the omitted thermal force (0.71 grad_parallel T_e) and velocity-dependent collision frequency are not important for the central turbulence comparison
    Section 4.4 acknowledges the LBD operator lacks the 0.71 grad_parallel T_e term and different resistivity prefactor, which affects the SOL E_r estimate; the impact on the edge TEM turbulence is not assessed.
  • domain assumption Dirichlet boundary conditions matching experimental profiles at the inner and outer radial boundaries do not force the separatrix power to match experiment
    Section 3 sets distribution functions to experimental Maxwellians at boundaries, so the P_sep comparison is not fully ab initio; the paper does not quantify how much of the P_sep agreement is constrained by these boundary conditions.
  • domain assumption The (80,60) grid-based simulation is a converged, high-fidelity reference
    Section 3 states N_mu could be reduced to about 20 without significantly compromising accuracy, but no full resolution scan is reported.
  • domain assumption Averaging over 0.1-0.2 ms in quasi-steady state yields converged turbulence statistics
    All spectra, fluxes, and P_sep values are time averages; no convergence test with longer windows or error bars is provided.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Turbulence and Transport in Spectrally Accelerated full-f Gyrokinetic Simulations." pith.science (2026). https://pith.science/paper/5536EBY4

@misc{pith2026250504832,
  author       = {Pith},
  title        = {Pith review of: Turbulence and Transport in Spectrally Accelerated full-f Gyrokinetic Simulations},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5536EBY4}},
  note         = {Machine review of arXiv:2505.04832}
}
read the original abstract

We investigate edge and scrape-off layer (SOL) turbulence and transport using the spectrally accelerated full-f gyrokinetic (GK) code GENE-X, recently introduced in [B. J. Frei et al., arXiv:2411.09232 (2024)]. Extending previous work on the TCV-X21 scenario, we show that the velocity-space spectral approach not only reproduces outboard midplane profiles but also captures key features of trapped electron mode (TEM)-driven turbulence and transport, including fluctuation spectra, turbulent fluxes, phase shifts, and power crossing the separatrix, in close agreement with grid-based results. This agreement remains robust when increasing spectral resolutions. We further analyze the radial force balance (accurately satisfied) and the structure of the radial electric fields and poloidal flows in the edge and SOL. Finally, we contrast our results with Braginskii-like fluid models, which inherently neglect TEMs. These results confirm the spectral full-f GENE-X approach as an efficient and first-principles tool for predicting edge and SOL turbulence.

Figures

Figures reproduced from arXiv: 2505.04832 by the authors.

Figure 2
Figure 2. OMP parallel (P∥α), perpendicular (P⊥α) and total (Pα) pressure profiles for the electrons (red) and ions (blue). The profiles are averaged over the toroidal direction and 0.1 ms at quasi-steady state obtained from the (16, 8) spectral simulation. Similar profiles are obtained with (8, 4) and (4, 2). balance the radial ion pressure gradient, if the poloidal and toroidal plasma flows are small. Otherwise, finite polo… view at source ↗
Figure 1
Figure 1. OMP normalized gradients of density n (top), electron temperature Te (middle), and ion temperature Ti (bottom) shown as a function of the normalized flux-surface label ρpol. The colored lines are the results from the spectral simulations [18] and the dashed black are the gradients obtained from the grid-based simulation [7]. projection of the ion momentum equation, reads [30] Er = 1 qini (∇ · Πi) · ∇r − UiθBϕ + UiϕB… view at source ↗
Figure 3
Figure 3. Poloidal flow contribution to the force balance at the OMP position estimated from (solid black) the force balance given in (14) and (red line) self-consistently computed from the spectral (16, 8) simulation using (15). The radial force balance is accurately satisfied. The quantities are evaluated on a single poloidal plane and average over 0.1 ms in steady-state. The black non-solid line indicate the different cont… view at source ↗
Figures from the paper (10 more)
Figure 4
Figure 4. Figure 4: Composition of the radial electric field Er (solid black line) at the OMP from the spectral (16, 8) simulation. The solid blue line represent the estimate of Er from force balance (17), with contributions from the total ion pressure gradient (dashed￾dotted) and intrins…
Figure 6
Figure 6. Figure 6: Temporal Fourier spectra of |ϕˆ 1(ky, ω)| 2 computed on the flux surface ρpol = 0.92 obtained from the (16, 8) spectral simulation. The spectrum maxima (calculated using a rolling average) obtained in the (4, 2) and (8, 4) spectral [18] and grid￾based [7] simulations a…
Figure 5
Figure 5. Figure 5: Amplitude of the Fourier component, ϕˆ 1, shown as a function of the normalized poloidal wavenumber kyρs (bottom x-axis) and of the poloidal mode number m (top x-axis) on different flux surfaces, ρpol = 0.79 (top), ρpol = 0.89 (center), and ρpol = 0.99 (bottom). The sp…
Figure 7
Figure 7. Figure 7: Histograms of the phase shifts, α(·, ϕˆ 1), of (from top to bottom) the density nα, parallel temperature T∥α, perpendicular temperature T⊥α, and total temperature Tα, with the electrostatic potential ϕ1 as a function of the normalized poloidal wavenumber kyρs (bottom a…
Figure 8
Figure 8. Figure 8: Fourier spectra of the turbulent radial particle flux Γα (top) and heat flux Qα (bottom) calculated on the ρpol = 0.92 flux surface for ions (left column) and electrons (right column). The spectra are plotted as functions of the normalized poloidal wavenumber kyρs (bot…
Figure 9
Figure 9. Figure 9: Fourier spectra calculated on the ρpol = 0.92 flux-surface of the convective part Qconv α (top row), the parallel Qcond ∥α (middle row) and perpendicular Qcond ⊥α (bottom row) components of the conductive part Qcond α for the ions (left column) and electrons (right col…
Figure 10
Figure 10. Figure 10: Right divertor heat flux q div ∥α (defined in (33)) for ions (top) and electrons (bottom) as a function of the OMP-mapped distance from the separatrix obtained from the spectral (solid colored lines) and grid-based (dashed back lines) simulations. be attributed to the…
Figure 11
Figure 11. Figure 11: shows snapshots of the radial compo￾nents of the electron heat flux Qe, associated with the E × B velocity, (26), and the diamagnetic heat flux QDe given in (37). The heat fluxes exhibit distinct spatial structures. While Qe features clear turbulent structures on the …
Figure 12
Figure 12. Figure 12: Spectra of the electron E × B turbulent Qe (blue line and left y-axis) and diamagnetic QDe (associated with VDe) (red line and right y-axis) heat fluxes evaluated on the ρpol = 0.99 flux-surface. For comparison, we also show the fluid (associated with a local Maxwelli…
Figure 13
Figure 13. Figure 13: Relative electrons (left) and ions (right) temperature fluctuation amplitudes, σTα /Tα, measured at the OMP. The colored lines represent the results from spectral simulations, while the dashed black lines correspond to grid￾based simulations, and the dotted black line…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

46 extracted references · 44 canonical work pages

  1. [1]

    Ikeda K 2007 Nuclear Fusion 47 E01

    work page 2007

  2. [2]

    2013 Nuclear Fusion 53 073019

    Zohm H, Angioni C, Fable E, Federici G, Gantenbein G, Hartmann T, Lackner K, Poli E, Porte L, Sauter O et al. 2013 Nuclear Fusion 53 073019

    work page 2013

  3. [3]

    2022 Plasma Physics and Controlled Fusion 64 034005

    Litaudon X, Jenko F, Borba D, Borodin D V, Braams B J, Brezinsek S, Calvo I, Coelho R, Donn´ e A J H, Embr´ eus O et al. 2022 Plasma Physics and Controlled Fusion 64 034005

    work page 2022

  4. [4]

    Brizard A J and Hahm T S 2007 Reviews of modern physics 79 421

    work page 2007

  5. [5]

    Zeiler A, Drake J F and Rogers B 1997 Physics of Plasmas 4 2134

    work page 1997

  6. [6]

    Frei B J, Jorge R and Ricci P 2020 Journal of Plasma Physics 86 905860205

    work page 2020

  7. [7]

    Ulbl P, Body T, Zholobenko W, Stegmeir A, Pfennig J and Jenko F 2023 Physics of Plasmas 30 107986

    work page 2023

  8. [8]

    Frei B J, Hoffmann A C D and Ricci P 2022 Journal of Plasma Physics 88 905880304

    work page 2022

Show all 46 references
  1. [9]

    Mandell N R, Hakim A, Hammett G W and Francisquez M 2020 Journal of Plasma Physics 86 905860109

    2020

  2. [10]

    Michels D, Stegmeir A, Ulbl P, Jarema D and Jenko F 2021 Comput. Phys. Commun. 264 107986

    2021

  3. [11]

    Hager R, Ku S, Sharma A Y, Chang C S, Churchill R M and Scheinberg A 2022 Physics of Plasmas 29

    2022

  4. [12]

    Bottino A, Stier A, Boesl M, Hayward-Schneider T, Bergmann A, Coster D, Brunner S, Di Giannatale G and Villard L 2025 Plasma Physics and Controlled Fusion 67 025008

    2025

  5. [13]

    2007 Plasma Physics and Controlled Fusion 49 B173

    Grandgirard V, Sarazin Y, Angelino P, Bottino A, Crouseilles N, Darmet G, Dif-Pradalier G, Garbet X, Ghendrih P, Jolliet S et al. 2007 Plasma Physics and Controlled Fusion 49 B173

    2007

  6. [14]

    Tamain P, Bufferand H, Ciraolo G, Colin C, Galassi D, Ghendrih P, Schwander F and Serre E 2016 Journal of Computational Physics 321 606

    2016

  7. [15]

    2019 Physics of Plasmas 26 052517

    Stegmeir A, Ross A, Body T, Francisquez M, Zholobenko W, Coster D, Maj O, Manz P, Jenko F, Rogers B Net al. 2019 Physics of Plasmas 26 052517

    2019

  8. [16]

    Giacomin M, Ricci P, Coroado A, Fourestey G, Galassi D, Lanti E, Mancini D, Richart N, Stenger L and Varini N 2021 J. Comput. Phys. 463 111294

    2021

  9. [17]

    Pitzal C, Stegmeir A, Zholobenko W, Zhang K and Jenko F 2023 Physics of Plasmas 30

    2023

  10. [18]

    Frei B J, Ulbl P, Trilaksono J and Jenko F 2024 arXiv preprint arXiv:2411.09232

    2024 arXiv

  11. [19]

    2022 Nuclear Fusion 62 096001

    Oliveira D S, Body T, Galassi D, Theiler C, Laribi E, Tamain P, Stegmeir A, Giacomin M, Zholobenko W, Ricci P et al. 2022 Nuclear Fusion 62 096001

    2022

  12. [20]

    Dif-Pradalier G, Grandgirard V, Sarazin Y, Garbet X and Ghendrih P 2009 Physical Review Letters 103 065002

    2009

  13. [21]

    2021 Plasma Physics and Controlled Fusion 63 034001

    Zholobenko W, Body T, Manz P, Stegmeir A, Zhu B, Griener M, Conway G D, Coster D, Jenko F, Team A U et al. 2021 Plasma Physics and Controlled Fusion 63 034001

    2021

  14. [22]

    Scott B and Smirnov J 2010 Physics of Plasmas 17

    2010

  15. [23]

    Dougherty J P 1964 The Physics of Fluids 7 1788

    1964

  16. [24]

    Ulbl P, Michels D and Jenko F 2022 Contributions to Plasma Physics 62 e202100180

    2022

  17. [25]

    2022 Physics of Plasmas 29

    Michels D, Ulbl P, Zholobenko W, Body T, Stegmeir A, Eich T, Griener M, Conway G D, Jenko F, Team A U et al. 2022 Physics of Plasmas 29

    2022

  18. [26]

    Gradshteyn I S and Ryzhik I M 2014 Table of integrals, series, and products (Academic press)

    2014

  19. [27]

    Frei B J, Hoffmann A C D, Ricci P, Brunner S and Tecchioll Z 2023 Journal Of Plasma Physics 89 905890414

    2023

  20. [28]

    Ulbl P, Body T, Zholobenko W, Stegmeir A, Pfennig J and Jenko F 2023 TCV-X21-GENEX: influence of collisions on the validation of global gyrokinetic simulations URL https://zenodo.org/record/7894731

    2023

  21. [29]

    thesis Technische Universit¨ at M¨ unchen

    Ulbl P 2023 Collision Models for Gyrokinetic Simulations of Edge Turbulence in Fusion Plasmas Ph.D. thesis Technische Universit¨ at M¨ unchen

    2023

  22. [30]

    Hazeltine R D and Meiss J D 2013 Plasma confinement (Courier Corporation)

    2013

  23. [31]

    Kim Y B, Diamond P H and Groebner R J 1991 Physics of Fluids B: Plasma Physics 3 2050

    1991

  24. [32]

    2023 Physics of Plasmas 30

    Plank U, Brida D, Conway G, Happel T, Hubbard A, P¨ utterich T, Angioni C, Cavedon M, Dux R, Eich Tet al. 2023 Physics of Plasmas 30

    2023

  25. [33]

    Connor J W, Hastie R J and Helander P 2006 Plasma physics and controlled fusion 48 885

    2006

  26. [34]

    2023 Nuclear Materials and Energy 34 101351

    Zholobenko W, Pfennig J, Stegmeir A, Body T, Ulbl P, Jenko F, Team A U et al. 2023 Nuclear Materials and Energy 34 101351

    2023

  27. [35]

    Dannert T and Jenko F 2005 Physics of Plasmas 12

    2005

  28. [36]

    2013 Nuclear fusion 53 093031

    Eich T, Leonard A W, Pitts R A, Fundamenski W, Goldston R J, Gray T K, Herrmann A, Kirk A, Kallenbach A, Kardaun O et al. 2013 Nuclear fusion 53 093031

    2013

  29. [37]

    2017 Nuclear Fusion 57 116023

    Chang C S, Ku S, Loarte A, Parail V, Koechl F, Romanelli M, Maingi R, Ahn J W, Gray T, Hughes J et al. 2017 Nuclear Fusion 57 116023

    2017

  30. [38]

    2024 Nuclear Fusion 64 106066

    Zholobenko W, Zhang K, Stegmeir A, Pfennig J, Eder K, Pitzal C, Ulbl P, Griener M, Radovanovic L, Plank U et al. 2024 Nuclear Fusion 64 106066

    2024

  31. [39]

    Helander P and Sigmar D J 2005 Collisional transport in magnetized plasmas vol 4 (Cambridge university press)

    2005

  32. [40]

    Zhang K, Zholobenko W, Stegmeir A, Eder K and Jenko F 2024 Nuclear Fusion 64 036016

    2024

  33. [41]

    Ulbl P, Stegmeir A, Told D, Merlo G, Zhang K and Jenko F 2025 arXiv preprint arXiv:2504.00475

    2025 arXiv

  34. [42]

    thesis EPFL Lausanne URL https: //infoscience.epfl.ch/handle/20.500.14299/210187

    Brunner S 1997 Global approach to the spectral problem of microinstabilities in tokamak plasmas using a gyroki- netic model Ph.D. thesis EPFL Lausanne URL https: //infoscience.epfl.ch/handle/20.500.14299/210187

    1997

  35. [43]

    Catto P J and Tsang K T 1978 The Physics of Fluids 21 1381

    1978

  36. [44]

    2004 Physics of Plasmas 11 2637

    Ernst D R, Bonoli P T, Catto P J, Dorland W, Fiore C L, Granetz R S, Greenwald M, Hubbard A E, Porkolab M, Redi M H et al. 2004 Physics of Plasmas 11 2637

    2004

  37. [45]

    Chen L, Berger R L, Lominadze J G, Rosenbluth M N and Rutherford P H 1977 Physical Review Letters 39 754

    1977

  38. [46]

    Zhao C, Zhang T and Xiao Y 2017 Physics of Plasmas 24

    2017

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.