pith. sign in

arxiv: 2605.23015 · v1 · pith:6TV6QNLOnew · submitted 2026-05-21 · ⚛️ physics.plasm-ph

Pre L-H Transition Radial Electric Field and Transport Validations of Edge and Scrape-off Layer Gyrokinetic Simulations at ASDEX Upgrade

B. J. Frei , C. Angioni , G. Lo-Cascio , W. Zholobenko , P. Ulbl , R. Bilato , F. Jenko , the ASDEX Upgrade Team This is my paper

Pith reviewed 2026-05-25 05:05 UTC · model grok-4.3

classification ⚛️ physics.plasm-ph
keywords gyrokinetic simulationL-H transitionradial electric fieldedge turbulenceASDEX Upgradetokamak plasmaGENE-X
0
0 comments X

The pith

Full-f gyrokinetic simulations reproduce the radial electric field well deepening toward the L-H transition in ASDEX Upgrade.

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

The paper performs stepwise validation of edge and scrape-off layer gyrokinetic simulations for multiple L-mode time slices approaching the L-H transition in a hydrogen discharge at ASDEX Upgrade. As boundary conditions are increased with rising ECRH power, the simulations show rising particle and heat fluxes and a deepening Er well that matches experimental measurements. Turbulence-driven poloidal flows are identified as the dominant contribution to the Er well through force balance decomposition. The inclusion of an edge density source for neutral gas ionization proves essential for matching experimental density profiles, Er, and ion heat fluxes.

Core claim

The central claim is that full-f gyrokinetic simulations using the GENE-X code, including X-point geometry, achieve excellent agreement with measured Er profiles and well depth at successive time slices approaching the L-H transition, with the edge density source being essential to reproduce relevant density profiles and fluxes.

What carries the argument

The GENE-X full-f gyrokinetic code applied to edge and scrape-off layer with X-point geometry, using force balance decomposition to identify turbulence-driven poloidal flows as dominant in the Er well.

Load-bearing premise

The modeling choice to introduce an edge density source representing neutral gas ionization is both necessary and sufficient to reproduce the experimental density profiles, Er, and ion heat fluxes without other missing physics dominating the result.

What would settle it

Running the simulations without the edge density source and observing failure to match the experimental density profiles, Er well depth, or ion heat fluxes at the time slices would falsify the claim that this source is essential.

Figures

Figures reproduced from arXiv: 2605.23015 by B. J. Frei, C. Angioni, F. Jenko, G. Lo-Cascio, P. Ulbl, R. Bilato, the ASDEX Upgrade Team, W. Zholobenko.

Figure 1
Figure 1. Figure 1: Experimental evolution of the plasma stored energy WMHD (gray solid line). The time traces of the experimental input heating power, PH, composed of the ECRH power (solid black line) and NBI blips (dashed black lines), are also shown on the right y-axis. Vertical colored dashed lines (see 1) indicate the selected L-mode time slices. The red shaded area marks the H-mode phase for t ≳ 4.85 s. Colors reference… view at source ↗
Figure 2
Figure 2. Figure 2: Initial (dashed) and quasi-steady state (solid) toroidally averaged OMP electron density (ne) profiles, at t = 4.8 s, without (dark blue) and with (red) density source Sn (see table 1). 4. Quasi-Steady State Particle and Energy Balance At quasi-steady state (QSS), particle and heat fluxes must establish in order to balance the presence of volumetic sources and sinks. In the present case, the particle and e… view at source ↗
Figure 3
Figure 3. Figure 3: Total surface integrated radial (top) particle Γe and (bottom) heat Qtot = P α Qα fluxes as a function of ρpol obtained in the different simulations at quasi-steady-state. The fluxes are averaged over 0.1 ms. The localized particle source, Sn, is also shown and scaled such that its maximum equals Sn = 1.6 × 1022 s−1 . At QSS, the time derivative in Eq. (6) vanishes when averaging over time. Thus, applying … view at source ↗
Figure 4
Figure 4. Figure 4: OMP density ne profiles plotted as a function of the normalized flux surface label ρpol. The GENE-X results are shown by the colored lines (table 1) and the shaded colored indicate the standard deviations associated with the toroidal and time averages performed at QSS. Experimental measurements (from TS and ECE diagnostics) are also shown. The OMP density profiles are obtained by interpolating the data at … view at source ↗
Figure 5
Figure 5. Figure 5: Same as figure 4 for the OMP electron Te (top) and ion Ti (bottom) temperature profiles. Experimental measurements (from TS, ECE, and CX diagnostics) are also shown. consistently from the GK quasineutrality condition and includes contributions from both long wavelength neoclassical (NC) and short wavelength turbulent components [24] [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: OMP Er profiles at different time slices approaching the L-H transition (from top left to bottom right). Solid colored lines show GENE-X simulation results (1). Experimental data include raw measurements at the corresponding time slices ( , , , markers) and time-averaged fitted profiles from CXRS and reflectometry (dotted black lines and taken from [12]). At t = 4.8 s, the results without ( ) and with ( ) … view at source ↗
Figure 7
Figure 7. Figure 7: Decomposition of Er at t = 4.8 s (or Pnet/PLH = 0.99) according to the radial force balance, Eq. (12). The GENE-X results without (top) and with (bottom) the density source Sn are shown. The diamagnetic (black dotted line), toroidal flow (black dashed line), and their combined (black solid line) contributions are overlaid. The SOL estimation, Er ∼ −∂rT∥e/e [24], is indicated by the gray dashed line, while … view at source ↗
Figure 8
Figure 8. Figure 8: Individual contributions to Er,min in the radial force balance equation, 12, and plotted as a function of Er,min: ∂rPi/(qini) + UiϕBθ ( ), and −UiθBϕ ( ). For comparison, the experimental Er,min, calculated from figure 6, are also shown ( ). Pnet/PLH increases from right to left (see table 1). The importance of poloidal flows in setting the depth of the Er well can be better visualized in figure 8 where th… view at source ↗
Figure 9
Figure 9. Figure 9: Poloidal flows Uiθ self-consistently calculated using Eq. (13) (solid lines) at the OMP obtained at t = 4.8 s with ( ) and without ( ) Sn. The NC predictions, UNC iθ [31] and calculated using GENE-X profiles, are also shown for comparison. Positive (negative) velocity indicates the electron (ion) diamagnetic direction. the maximum Er shear, E′ r = dEr/dr, for both layers obtained from the GENE-X simulation… view at source ↗
Figure 11
Figure 11. Figure 11: Radially resolved frequency power spectrum of Er (in kHz) at Pnet/PLH = 0.99, shown without (top) and with (bottom) density source. The linear estimate of the GAM frequency fGAM [39] is indicated by dotted red lines, and the separatrix position by green dashed lines. 7. Turbulence Characterization We characterize edge turbulence at Pnet/PLH = 0.99, close to the L–H transition. Previous linear flux￾tube GE… view at source ↗
Figure 12
Figure 12. Figure 12: Dispersion relations ω(ky) at Pnet/PLH = 0.99, shown with (top) and without (bottom) a density source at ρpol = 0.994 inside the Er well. Here, f = ω/(2π) and the binormal wavenumber ky is normalized to the averaged sound Larmor radius ρs on the flux-surface. The spectra are computed over a 0.2 ms time window and averaged over the toroidal direction. The linear frequencies of the eDW and CTEM, ω ∗ e and ω… view at source ↗
Figure 13
Figure 13. Figure 13: Evolution of the total ion and electron powers, Pi (top) and Pe (bottom) respectively, approaching the L-H transition obtained from GENE-X (dotted lines with colored ), nonlinear local flux-tube GENE (dashed line with , data from [12]), and experiments inferred from ASTRA (dashed line with , data from [12]). The ρpol = 0.98 flux surface (same as in [12]) are used to calculate Pi and Pe. See table 1 for th… view at source ↗
Figure 14
Figure 14. Figure 14: shows the radial profiles of the different contributions to Pα at Pnet/PL−H = 0.99 (t = 4.8 s). Both with and without Sn, Pα is dominated by turbulent E × B transport; however, this contribution is significantly reduced when Sn is included. As a result, the diamagnetic contribution P diam α becomes non-negligible. In particular, P diam i ≃ 0.2 MW, corresponding to approximately 30% of the total ion power … view at source ↗
Figure 15
Figure 15. Figure 15: Total ion (top) heat and (bottom) particle fluxes, Qi and Γi respectively, in gyro-Bohm units (QgyB and ΓgyB) obtained in GENE-X approaching the L-H transition. The markers indicate the results from nonlinear flux-tube GENE (from [12]). See table 1 for the color code. all time slices. In particular, when the turbulent ion heat flux Qes i (the dominant contribution to Qi) is normalized to the gyro-Bohm hea… view at source ↗
read the original abstract

This work presents a stepwise validation of the evolution of the radial electric field (Er) and heat transport during the pre L-H transition phase using full-f gyrokinetic simulations of the edge and scrape-off layer in the ASDEX Upgrade (AUG) tokamak, including X-point geometry. Several L-mode time slices up to the L-H transition from a dedicated hydrogen discharge, featuring stepwise increases in ECRH input power, are selected [N. Bonanomi \textit{et al.}, Phys. Plasmas 31, 072302 (2024)] and simulated with the \texttt{GENE-X} code. As the edge boundary conditions are progressively increased between the time slices, particle and heat fluxes rise, and the radial electric field Er well deepens. A detailed validation of the Er profiles and of the Er well depth shows excellent agreement with experimental measurements at the successive time slices approaching the L-H transition. A force balance decomposition identifies turbulence-driven poloidal flows as the dominant contribution within the Er well. Edge turbulence is governed by a competition between electron drift waves and trapped-electron modes. The introduction of an edge density source, modeling neutral gas ionization, is shown to be essential to reproduce experimentally relevant density profiles, Er, and edge ion heat fluxes, which are dominated by both turbulent and diamagnetic contributions. This stepwise validation constitutes an important milestone toward predictive, first-principles gyrokinetic simulations of the L-H transition power threshold.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. This paper reports stepwise full-f gyrokinetic simulations with the GENE-X code of the edge and scrape-off layer (including X-point) in ASDEX Upgrade for several L-mode time slices from a hydrogen discharge with increasing ECRH power up to the L-H transition. It claims excellent agreement between simulated and measured radial electric field (Er) profiles and well depth, identifies turbulence-driven poloidal flows as the dominant contribution to the Er well via force-balance decomposition, shows edge turbulence as a competition between electron drift waves and trapped-electron modes, and demonstrates that an edge density source (modeling neutral ionization) is essential to match experimental density profiles, Er, and ion heat fluxes (which include both turbulent and diamagnetic contributions).

Significance. If the quantitative validation holds, this constitutes a meaningful advance toward first-principles prediction of the L-H power threshold by showing that a gyrokinetic code can capture the evolution of Er and transport across successive time slices in realistic geometry with increasing power. The explicit identification of the density source as necessary and the decomposition of Er contributions provide concrete, testable elements that strengthen the result.

major comments (2)
  1. [Abstract] Abstract (final paragraph): the assertion that the edge density source 'is shown to be essential' to reproduce density profiles, Er, and ion heat fluxes is load-bearing for the central validation claim, yet the abstract supplies no quantitative comparison (e.g., profiles or flux values) of the simulation without the source; a dedicated sensitivity test or table contrasting the two cases is required to confirm that other missing physics is not being compensated.
  2. [Abstract] Abstract: the repeated claim of 'excellent agreement' for Er profiles and well depth is not accompanied by any quantitative metrics (RMS deviation, correlation coefficient, or error bars) or sensitivity tests on boundary conditions; without these in the results section the strength of the stepwise validation cannot be assessed.
minor comments (1)
  1. [Abstract] The abstract refers to 'GENE-X code' and specific discharge parameters but does not list the numerical resolution, time-stepping details, or exact boundary-condition values used; these should be stated explicitly in the methods section for reproducibility.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments. We address each point below and revise the manuscript to incorporate quantitative comparisons and metrics as requested.

read point-by-point responses

  1. Referee: [Abstract] Abstract (final paragraph): the assertion that the edge density source 'is shown to be essential' to reproduce density profiles, Er, and ion heat fluxes is load-bearing for the central validation claim, yet the abstract supplies no quantitative comparison (e.g., profiles or flux values) of the simulation without the source; a dedicated sensitivity test or table contrasting the two cases is required to confirm that other missing physics is not being compensated.

    Authors: We agree that a dedicated quantitative comparison strengthens the claim. The main text already shows results with the source and notes its necessity, but we will add an explicit sensitivity test (new figure panels and table) contrasting density, Er, and ion heat flux profiles with and without the source. This will be referenced in the abstract. revision: yes

  2. Referee: [Abstract] Abstract: the repeated claim of 'excellent agreement' for Er profiles and well depth is not accompanied by any quantitative metrics (RMS deviation, correlation coefficient, or error bars) or sensitivity tests on boundary conditions; without these in the results section the strength of the stepwise validation cannot be assessed.

    Authors: We acknowledge that quantitative metrics improve assessment of agreement. In the revised results section we will report RMS deviations and correlation coefficients for Er profiles across time slices, include error bars on experimental and simulated data, and add a short discussion of boundary condition sensitivity. revision: yes

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The paper's central claim is a validation exercise: GENE-X full-f gyrokinetic simulations are run on successive experimental time slices from an AUG discharge (with boundary conditions taken directly from measured profiles and power ramps), and the resulting Er profiles and well depths are compared to independent experimental measurements. The edge density source is explicitly presented as a required modeling choice to reach experimentally relevant density profiles, after which Er and ion heat flux agreement is reported; this does not reduce any claimed prediction to a fitted parameter by construction, nor does any equation or self-citation chain make the Er validation tautological. No self-definitional, fitted-input-renamed-as-prediction, or uniqueness-imported steps appear in the provided abstract or described chain. The derivation remains self-contained against external experimental benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review; no explicit free parameters, axioms, or invented entities are quantified. The edge density source is presented as a modeling addition required to match data, but its functional form and any associated parameters are not detailed.

pith-pipeline@v0.9.0 · 5836 in / 1249 out tokens · 29834 ms · 2026-05-25T05:05:14.885525+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

48 extracted references · 48 canonical work pages

  1. [1]

    Shimada M, Campbell D, Mukhovatov V, Fujiwara M, Kirneva N, Lackner K, Nagami M, Pustovitov V, Uckan N, Wesley Jet al.2007Nuclear Fusion47S1–S17

    work page

  2. [2]

    Goldston R J 2011Nuclear Fusion52013009

    work page

  3. [3]

    Wagner F, Becker G, Behringer K, Campbell D, Eberhagen A, Engelhardt W, Fussmann G, Gehre O, Gernhardt J, Gierke G vet al.1982Physical Review Letters491408

    work page

  4. [4]

    Zohm H 1996Plasma Physics and Controlled Fusion38 105

    work page

  5. [5]

    Doyle E, Houlberg W, Kamada Y, Mukhovatov V, Osborne T, Polevoi A, Bateman G, Connor J, Cordey J, Fujita T et al.2007Nuclear Fusion47S18

    work page

  6. [6]

    Diamond P H, Itoh S I, Itoh K and Hahm T S 2005Plasma Physics and Controlled Fusion47R35–R161

    work page

  7. [7]

    Ryter F, Orte L B, Kurzan B, McDermott R, Tardini G, Viezzer E, Bernert M, Fischer R, Team A Uet al.2014 Nuclear Fusion54083003

    work page 2014

  8. [8]

    Ryter F, Cavedon M, Happel T, McDermott R, Viezzer E, Conway G, Fischer R, Kurzan B, P¨ utterich T, Tardini Get al.2015Plasma Physics and Controlled Fusion58 014007

    work page

  9. [9]

    Sauter P, P¨ utterich T, Ryter F, Viezzer E, Wolfrum E, Conway G, Fischer R, Kurzan B, McDermott R, Rathgeber Set al.2011Nuclear Fusion52012001

    work page

  10. [10]

    Schmitz L, McKee G R, Rhodes T L, Groebner R J, Doyle E J, Peebles W A and Petty C C 2012Physical Review Letters108155002

    work page

  11. [11]

    Jenko F, Dorland W, Kotschenreuther M and Rogers B 2000 Physics of plasmas71904

    work page 2000

  12. [12]

    2024Physics of Plasmas31

    Bonanomi N, Angioni C, Conway G, Happel T, Plank U, Schneider P, Staebler G, Team A U, Team E Met al. 2024Physics of Plasmas31

    work page

  13. [13]

    Michels D, Ulbl P, Zholobenko W, Body T, Stegmeir A, Eich T, Griener M, Conway G D, Jenko F, Team A U et al.2022Physics of Plasmas29

    work page

  14. [14]

    Frei B, Ulbl P, Trilaksono J and Jenko F 2025Computer Physics Communications109817

    work page

  15. [15]

    Zholobenko W, Jenko F, Zhang K, Ulbl P, Eder K, Stegmeir A, Angioni C, and Manz P 2026Phys. Rev. Lett.– URL https://link.aps.org/doi/10.1103/b2s6-b5c1

    work page doi:10.1103/b2s6-b5c1

  16. [16]

    Willensdorfer M, Wolfrum E, Fischer R, Schweinzer J, Sertoli M, Sieglin B, Veres G, Aumayr F, Team A U et al.2012Review of scientific instruments83

    work page

  17. [17]

    2017Review of Scientific Instruments88

    Cavedon M, P¨ utterich T, Viezzer E, Dux R, Geiger B, McDermott R M, Meyer H, Stroth U, Team A Uet al. 2017Review of Scientific Instruments88

    work page

  18. [18]

    2023Physics of Plasmas30

    Plank U, Brida D, Conway G, Happel T, Hubbard A, P¨ utterich T, Angioni C, Cavedon M, Dux R, Eich Tet al. 2023Physics of Plasmas30

    work page

  19. [19]

    Michels D, Stegmeir A, Ulbl P, Jarema D and Jenko F 2021 Comput. Phys. Commun.264107986

    work page 2021

  20. [20]

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

    work page

  21. [21]

    Ulbl P, Michels D and Jenko F 2022Contributions to Plasma Physics62e202100180

    work page

  22. [22]

    Zholobenko W, Body T, Manz P, Stegmeir A, Zhu B, Griener M, Conway G D, Coster D, Jenko F, Team A U et al.2021Plasma Physics and Controlled Fusion63 034001

    work page

  23. [23]

    Sarazin Y, Grandgirard V, Abiteboul J, Allfrey S, Garbet X, Ghendrih P, Latu G, Strugarek A and Dif-Pradalier G 2010Nuclear Fusion50054004

    work page

  24. [24]

    Frei B, Ulbl P, Pitzal C, Zholobenko W and Jenko F 2025 Nuclear Fusion65116026

    work page 2025

  25. [25]

    Ulbl P, Body T, Zholobenko W, Stegmeir A, Pfennig J and Jenko F 2023Physics of Plasmas30107986

    work page

  26. [26]

    Plank U, McDermott R, Birkenmeier G, Bonanomi N, Cavedon M, Conway G, Eich T, Griener M, Grover O, Schneider Pet al.2022Plasma Physics and Controlled Fusion65014001

    work page

  27. [27]

    Conway G D, Poli E, Happel T, Team A Uet al.2010 Plasma and Fusion Research5S2005

    work page 2010

  28. [28]

    Zholobenko W, Stegmeir A, Griener M, Conway G, Body T, Coster D, Jenko F, Team A Uet al.2021Nuclear Fusion61116015

    work page

  29. [29]

    Viezzer E, P¨ utterich T, Angioni C, Bergmann A, Dux R, Fable E, McDermott R, Stroth U, Wolfrum E, Team A U et al.2013Nuclear Fusion54012003

    work page

  30. [30]

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

    work page

  31. [31]

    Kim Y B, Diamond P H and Groebner R J 1991Physics of Fluids B: Plasma Physics32050

    work page

  32. [32]

    Landreman M, Parra F I, Catto P J, Ernst D R and Pusztai I 2014Plasma Physics and Controlled Fusion56045005

    work page

  33. [33]

    Kolesnikov R, Wang W, Hinton F, Rewoldt G and Tang W 2010Plasma Physics and Controlled Fusion52042002

    work page

  34. [34]

    Cavedon M, Happel T, Hennequin P, Dux R, H¨ ofler K, Plank U, P¨ utterich T, Stroth U, Viezzer E, Wolfrum E et al.2024Plasma Physics and Controlled Fusion66 025011

    work page

  35. [35]

    The particle, conductive parallel and perpendicular, convective and total fluxes are shown from top to bottom, respectively

    Viezzer E, P¨ utterich T, McDermott R, Conway G, Cavedon Validation of pre L-H transition 20 0 2 4 1e20 ky e / (m 2 s 1) ky i / (m 2 s 1) 0 10 kyQcond e / (kW m 2) kyQcond i / (kW m 2) 0 20 40 kyQcond e / (kW m 2) kyQcond i / (kW m 2) 0 20 40 kyQconv e / (kW m 2) kyQconv i / (kW m 2) 10 1 100 ky s 0 50 100 kyQe / (kW m 2) 10 1 100 ky s kyQi / (kW m 2) Fig...

    work page 2052

  36. [36]

    Schirmer J, Conway G, Zohm H, Suttrop W, Team A U et al.2006Nuclear fusion46S780

    work page

  37. [37]

    Conway G, Team A Uet al.2008Plasma Physics and Controlled Fusion50085005

    work page

  38. [38]

    Conway G, Angioni C, Ryter F, Sauter P, Vicente J and Team) A U 2011Physical review letters106065001

    work page

  39. [39]

    Sugama H and Watanabe T H 2006Journal of plasma physics72825

    work page

  40. [40]

    Conway G D, Smolyakov A I and Ido T 2021Nuclear Fusion 62013001

    work page

  41. [41]

    Frei B J, Hoffmann A C D, Ricci P, Brunner S and Tecchioll Z 2023Journal Of Plasma Physics89905890414

    work page

  42. [42]

    Palermo F, Poli E, Bottino A, Biancalani A, Conway G and Scott B 2017Physics of Plasmas24

    work page

  43. [43]

    Bonanomi N, Angioni C, Crandall P, Di Siena A, Maggi C, Schneider P, Contributors J, Team A U, Team E Met al. Validation of pre L-H transition 21 50 100 R / LN 100 200 R / LTe 0.92 0.94 0.96 0.98 1.00 1.02 pol 50 100 R / LTi Figure D1.Normalized density (top) and temperature (bottom) normalized gradient,R/L f , associated with the OMP profiles (see figure...

    work page

  44. [44]

    Bonanomi N, Angioni C, Plank U, Schneider P, Maggi C, Contributors J, Team A U, Team E Met al.2021Physics of Plasmas28

    work page

  45. [45]

    Pereverzev G V and Yushmanov e P 2002

    work page 2002

  46. [46]

    Fable E, Angioni C, Ivanov A, Lackner K, Maj O, Yu S, Pautasso G, Pereverzev Get al.2013Nuclear Fusion53 033002

    work page

  47. [47]

    Staebler G, Kinsey J and Waltz R 2007Physics of Plasmas 14

    work page

  48. [48]

    2019Physics of Plasmas26052517

    Stegmeir A, Ross A, Body T, Francisquez M, Zholobenko W, Coster D, Maj O, Manz P, Jenko F, Rogers B Net al. 2019Physics of Plasmas26052517

    work page