Example on frontera: /scratch3/03532/mgrudic/STARFORGE_RT/turbsphere/M2e4_R10_S0_T1_B0.01_Res271_n2_sol0.5_42/ DG: Clean (not live run) example on Stampede2: /work2/05917/tg852163/stampede2/stampede2/GMC_sim/Runs/Physics_ladder/M2e4_turbsphere_example

the initialization run is in output_stirring, its params file is params.txt, and the SF runs (with and without continued driving) are in output_SF_*

The stirring run GIZMO binary was compiled with flags:

SINGLE_STAR_STARFORGE_DEFAULTS
MAGNETIC
COOLING
SINGLE_STAR_FB_JETS
SINGLE_STAR_FB_RAD
SINGLE_STAR_FB_WINDS
SINGLE_STAR_FB_SNE
BOX_PERIODIC
GRAVITY_NOT_PERIODIC
ADAPTIVE_TREEFORCE_UPDATE=0.0625
OPENMP=2
STARFORGE_GMC_TURBINIT

and the value of RT_SPEEDOFLIGHT_REDUCTION was reduced by a factor of 3 in allvars.h (purely for speed, this is optional and should not be used for the eventual SF run). To perform the stirring run, start with the desired SPHERE IC and scale the turbulence driving parameters in params.txt appropriately (e.g. rescale all wavelengths to the cloud size, tune vRMS to get the desired Mach number, scale the coherence time to the cloud crossing time). The equilibrium turbulence properties will be converged at very modest resolution, so it’s easiest to tune at low resolution before running the real thing. It is highly recommended to pre-tune the turbulence parameters at low resolution before running your full-resolution stirring run.

Then to run the SF run, simply disable STARFORGE_GMC_TURBINIT, and optionally add TURB_DRIVING if you want driving to continue. The SF run should ideally be treated as a flag 2 snapshot-restart of the stirring run, as otherwise the radiation field will be re-initialized to the ISRF instead of the statistical equilibrium attained during the stirring.