CosmicGrain Analysis Scripts

This folder contains some analysis and visualization scripts used during CosmicGrain's development. A number were used in preparing the zoom itself and in selecting the target halo (Halo 569) from the parent box, others are the ones that directly produced the figures in the CosmicGrain methods paper, and are the best starting point if you want to reproduce a published result or adapt the analysis pipeline for your own runs.

Shared utilities (required by most scripts below)

• halo_utils.py — core shared library for identifying Halo 569 and computing R200c/M200c via spherical overdensity. Primary API:

  ref  = get_halo569_reference(output_dir)
  halo = get_halo569(groups_dir, snap_num, ref)
  

  See the module docstring for full unit conventions (comoving vs. physical kpc, code mass units, etc.) — read this before adapting any script below to a new halo or run.

• get_target_halo_particles.py — defines load_target_halo(), used by the age-coded histogram script to extract all particle types within a halo.

Figure-producing scripts

Script	Produces	Usage
plot_halo_projection_full_ladder.py	Dust surface density projections across the physics ladder (Fig. showing all S0–S10 runs)	python plot_halo_projection_full_ladder.py --snap-pattern "../{rung}_output_1024/snapdir_{num}/snapshot_{num}.0.hdf5" --group-pattern "../{rung}_output_1024/groups_{num}/fof_subhalo_tab_{num}.0.hdf5" --snap-num 047 --rungs S0 S1 S2 S3 S4 S5 S6 S7 S8 S9 S10 --axis z --view ism --depth-frac 0.5 --bar-quantity gas_compare --gas-compare-rung S10 --vmin-dust 1e-5 --npix 512 --dust-adaptive-k 8 --dust-adaptive-min 0.5 --gas-adaptive --out ladder_ism_gascompare.png — see --help for the --bar-quantity alternatives (mdust, dz, dg) and full smoothing/adaptive-kernel options
run_radial_evolution.py	Cumulative radial dust distribution at multiple redshifts	python run_radial_evolution.py ../S10_output_1024/ --redshifts 0 0.5 1 2 3 --rmax-factor 1.0 --outdir radial_evolution --summary dust_radial_evolution.pdf
plot_mdust_mstar_all_halos.py	Dust mass vs. stellar mass evolutionary track, vs. observations/SIMBA	python plot_mdust_mstar_all_halos.py ../S10_output_1024/ --simba-download --simba-dir ./simba/ --output mdust_mstar_S10_1024.png
plot_dust_histograms_agecoded.py	Age-coded histograms of grain radius, mass, velocity, age	python plot_dust_histograms_agecoded.py --catalog ../groups_049/fof_subhalo_tab_049.0.hdf5 --snapshot ../snapdir_049/snapshot_049 --out dust_histograms_agecoded.png --rmax 300
plot_gsd_comparison.py	Grain-size distribution vs. MRN/Weingartner & Draine/THEMIS models	python plot_gsd_comparison.py ../S10_output_1024/ (add --snap 049, --r-ism 20, --n-bins 45 etc. as needed)
plot_dust_evolution.py	5-panel evolution figure: M_dust/M_star, SFR, dM_dust/dt, D/G, D/Z vs. redshift	python plot_dust_evolution.py ../S10_output_1024/ (or pass multiple run dirs + --labels for a convergence comparison)
plot_radial_dgr.py	Radial D/G and D/Z profiles across the physics ladder at z=0	python plot_radial_dgr.py --res 1024 (add --runs S0 S4 S10 or --r-max 50 as needed)
plot_dz_vs_metallicity.py	Galaxy-integrated D/Z vs. gas-phase metallicity, vs. observations/other sims	python plot_dz_vs_metallicity.py --res 1024 (or --res 512 1024 2048 for a convergence-mode plot)

All scripts expect to be run from inside this scripts/ directory (so that halo_utils.py and cosmicgrain.mplstyle resolve via relative import), and all support --help for the full list of options.

Unit conventions (apply across all scripts above)

• Positions in snapshots/catalogs: comoving kpc/h → physical kpc via * a / h
• Masses: 1e10 Msun/h (Gadget code units)
• HubbleParam (h) is read from f["Parameters"].attrs, not f["Header"].attrs
• GrainRadius is already in nm in the HDF5 output (unit conversion applied on write) — do not apply an additional conversion factor