List of GKV namelist

Table 1 List of run/gkvp_namelist

Group

Name

Parameter

&cmemo

memo

Memo

&calct

calc_type

  • “linear” – for linear runs

  • “lin_freq” – for linear runs with frequency check \(k_x = 0\)

  • “nonlinear” – for nonlinear runs

&calct

z_bound

  • “zerofixed” – Fixed boundary in \(z\)

  • “mixed” – Outflow boundary in \(z\) only for \(\tilde{f}_{\mathrm{s}\bm{k}}\)

&calct

z_filt

  • “on” – Enable 4th-order filtering in \(z\) on \(d\tilde{f}_{\mathrm{s}\bm{k}}/dt\)

  • “off” — Disable filtering

&calct

z_calc

  • “cf4” – 4th-order central finite difference for \(d\tilde{f}_{\mathrm{s}\bm{k}}/dz\) ( \(\texttt{nzb}=2\) )

  • “up5” – 5th-order upwind finite difference for \(d\tilde{f}_{\mathrm{s}\bm{k}}/dz\) ( \(\texttt{nzb}=3\) )

&calct

art_diff

Coefficient of artificial diffusion for z_calc=”cf4”

&calct

init_random

Switch whether phases of initial Fourier modes are randomized

&calct

num_triad_diag

Number of triad transfer diagnostics, which should be consistent with the number of “&triad mxt=*, myt=*/”.

&calct

vp_coord

  • “0” — the magnetic moment mu is the perpendicular velocity-space coordinate

  • “1” — the perpendicular velocity vp is the perpendicular velocity-space coordinate

&triad

mxt=*, myt=*/

Diagnosed mode number of triad transfer analysis. Add lines of “&triad mxt=*,myt=*/” as desire.

&equib

equib_type

  • “analytic” – Analytic helical field with the metrics in cylinder

  • “s-alpha” – s-alpha model with alpha = 0 (cylindrical metrics)

  • “s-alpha-shift” – s-alpha model with Shafranov shift

  • “circ-MHD” – Concentric circular field with consistent metrics

  • “vmec” – Tokamak/stellarator field from the VMEC code

  • “eqdsk” – Tokamak field (MEUDAS/TOPICS or G-EQDSK) via IGS code

  • “slab” – Shearless slab geometry

  • “ring” – Ring dipole geometry

&run_n

inum

Current run number

&run_n

ch_res

  • .true. – Change perpendicular resolutions (editing gkvp_f0.48_set.f90 is required.)

  • .false. – Disable changing resolution

&files

f_log

Data directory for log data

&files

f_hst

Data directory for time-series data

&files

f_phi

Data directory for field quantity data

&files

f_fxv

Data directory for distribution function data

&files

f_cnt

Data directory for continue data

&runlm

e_limit

Elapsed time limit [sec]

&times

tend

End of simulation time [L_ref/v_ref]

&times

dtout_fxv

Time spacing for data output [L_ref/v_ref]

&times

dtout_ptn

Time spacing for data output [L_ref/v_ref]

&times

dtout_eng

Time spacing for data output [L_ref/v_ref]

&times

dtout_dtc

Time spacing for time-step-size adaption [L_ref/v_ref]

&deltt

dt_max

Maximum time step size [L_ref/v_ref]

&deltt

adapt_dt

  • .true. – Enable time-step-size adaption

  • .false. – Time step size is fixed to be dt = dt_max

&deltt

courant_num

Courant number for time-step-size adaption

&deltt

time_advnc

  • “rkg4” – Explicit time integration by 4th-order Runge-Kutta-Gill method

  • “imp_colli” – 2nd-order operator split + 2nd-order implicit collision solver + 4th-order RKG method for collisionless physics

  • “auto_init” –

    • If collision restricts linear time step size, time_advnc=”imp_colli”.

    • Otherwise, time_advnc=”rkg4”

&physp

R0_Ln

Normalized density gradient, L_ref/L_ne, L_ref/L_ni, …

&physp

R0_Lt

Normalized temperature gradient, L_ref/L_te, L_ref/L_ti, …

&physp

nu

Bias factor for LB collision model, e.g., 1.d0, 0.5d0, 2.d0, …

Note

NOTE that after gkvp_f0.40, collision frequencies are consistently calculated by (Nref, Tref, Lref) in &nu_ref, and nu is just used as a bias factor only for LB case. Also, nu is not used in multi-species collisions (full).

&physp

Anum

Mass number, m_e/m_ref, m_i/m_ref, …

&physp

Znum

Atomic number, |e_e/e_ref|, |e_i/e_ref|, …

&physp

fcs

Charge fraction |e_e*n_e/(e_ref*n_ref)|, |e_i*n_i/(e_ref*n_ref)|, …

Note

NOTE that fcs = 1.0 for electron in the recommended setting (n_ref = n_e).

&physp

sgn

Sign of charge, e_e/ |e_e|, e_i/ |e_i|, …

&physp

tau

Normalized temperature, T_e/T_ref, T_i/T_ref, …

Note

NOTE that T_i/T_ref = 1.0 for the first ion species in the recommended setting (T_ref = T_i of first ion).

&physp

dns1

Initial perturbation amplitude, (L_ref/rho_ref)* \(\tilde{n}_\mathrm{e}\) /n_ref, (L_ref/rho_ref)* \(\tilde{n}_\mathrm{i}\) /n_ref, …

&physp

tau_ad

T_i/T_e for single species ITG-ae (sgn=+1), T_e/T_i for single species ETG-ai (sgn=-1)

&physp

lambda_i

Ratio of (Debye_length / rho_ref)**2 = epsilon_0 * B_ref**2 / (m_ref * n_ref)

&physp

beta

Local beta value evaluated by mu_0*n_ref*T_ref/B_ref**2

&physp

ibprime

  • “1” – Enable a grad-p (finite beta-prime) contribution on the magnetic drift kvd for equib_type = “eqdsk” and “vmec”

  • “0” – Ignore it

&physp

vmax

Velocity domain size in the unit of each thermal speed [v_ts]

&physp

nx0

Radial mode number assigned for the initial perturbation

Note

NOTE that if nx0 exceeds nx, nx0 is reset to nx. A sufficiently large value, thus, gives perturbations for entire kx-modes.

&rotat

mach

Not yet implemented

&rotat

uprime

Not yet implemented

&rotat

gamma_e

Equilibrium \(\bm{E} \times \bm{B}\) flow shearing rate \(\gamma_E\)

&nperi

n_tht

The length of fluxtube, \(z\) -domain \(= \pm\) N_tht* \(\pi\)

&nperi

kymin

Minimum field-line-label (or poloidal) wave number [1/rho_ref]

&nperi

m_j

Mode connection number for pseudo-periodic boundary in fluxtube, kxmin = |2*pi*s_hat*kymin/m_j|

&nperi

del_c

Mode connection phase in fluxtube model (Since it is arbitrary, del_c = 0.d0 in standard.)

&confp

eps_r

Inverse aspect ratio at the center of fluxtube, a*rho_0/L_ref

&confp

eps_rnew

Model factor for equib_type = “analytic”

&confp

q_0

Safety factor at the center of fluxtube, q(rho_0)

&confp

s_hat

Magnetic shear at the center of fluxtube, s(rho_0)

&confp

lprd

\(\vdots\)

malpha

factor for equib_type = “analytic”

&ring

ring_a

ring_a = a / R0, which specify a flux tube of the ring dipole. [There is a ring current at R=a. The field line passing through (R,Z)=(R0,0) is picked up as a flux-tube domain. The reference length is set to be R0 (not the ring current at R=a). The reference magnetic field strength is B0 at (R,Z)=(R0,0).]

&ring

kxmin

Minimum wavenumber in kx, valid only when equib_type == “ring”

&vmecp

s_input

Minor radial position of the local flux-tube analysis in Stellarator (VMEC) equilibrium?

&vmecp

nss

Number of radial grids on METRIC data (=nrho in BZX)

&vmecp

ntheta

ntheta = (number of poloidal grids on METRIC = ntht in BZX) = 2*global_nz

&vmecp

nzeta

= 0

&bozxf

f_bozx

File location of METRIC data produced by BZX code

&igsp

s_input

Reference radial flux surface, rho_0, in Tokamak (MEUDAS/TOPICS or G-EQDSK) equilibrium

&igsp

mc_type

  • “0” – Axisymmetric coordinates

  • “1” – Boozer coordinates

  • “2” – Hamada coordinates

&igsp

q_type

  • “1” – Use consistent q-value on g-eqdsk equilibrium (Recommended)

  • “0” – Use inconsistent, but given q_0 value in &confp.

&igsp

nss

Number of radial grids on METRIC data

&igsp

ntheta

ntheta = (number of poloidal grids on METRIC = ntht in BZX) = 2*global_nz

&igsf

f_igs

File location of METRIC data produced by IGS code

&nu_ref

Nref

Local electron density at the center of fluxtube, \(n_\mathrm{e}(\rho_0)\) [m^-3]

&nu_ref

Lref

Major radius at the magnetic axis, \(R_a\) [m]

&nu_ref

Tref

Main ion temperature at the center of fluxtube \(T_i(\rho_0)\) [keV]

&nu_ref

col_type

  • “LB” – Lenard-Bernstein model collision operator

  • “lorentz” – Lorentz model collision operator

  • “full” – Sugama model collision operator for multiple plasma species

&nu_ref

iFLR

  • “1” – Enable the FLR (gyrophase-averaging and classical diffusion) terms (for LB and full)

  • “0” – Disable it (DK-limit)

&nu_ref

icheck

  • “0” – for production runs

  • “1” – Debug test with Maxwellian Annihilation (should be used with iFLR = 0)

Note

Note that inum=%%% and f_**="%%DIR%%/..." will be automatically set by the shoot script. In the &physp group, species-dependent names R0_Ln – dns1 are the array of length nprocs . The &vmecp and &bozxf groups are active only when equib_type = "vmec" . Similarly, the &igsp and &igsf groups are active only when equib_type = "eqdsk" .