Skip to content

Absorption spectra

Compare C IV, Si IV, and O VI absorption along one ray through the TNG50 cutout. Cloudy tables supply the ion fractions; the snapshot supplies the elemental abundances.

1. Download the configuration

Inputs required

Download the galaxy cutout and ionization tables from Prepare.

Download absorption-metals.yaml into your working directory.

2. Configure and run

absorption-metals.yaml 
# Metal absorption using Cloudy C17.03 FG11 v2 tables. [AI-Codex]
dataset_type: pointcloud_voronoi
driver_type: raytracer
device: cpu
log_level: info

pointcloud_voronoi:
  hilbert_bits: 8
  loader: gadget
  cloudy:
    enabled: true
    ion_table_path: ./data/cloudy-v2/grid_ions.hdf5
    emission_table_path: ./data/cloudy-v2/grid_emissivities.hdf5
    redshift_interp: true
  gadget:
    path: ./tng50-halo625-gas-stars-v2.hdf5
    fields: [PositionX, PositionY, PositionZ, Density, Temperature,
             StarFormationRate, VelocityX, VelocityY, VelocityZ,
             cloudydensity_CIV, cloudydensity_SiIV, cloudydensity_OVI]
    zoom_box: [[0.4945, 0.5055], [0.4945, 0.5055], [0.4919, 0.5029]]
  field_filters:
    _non_star_forming:
      conditions:
        - {field: StarFormationRate, op: "==", value: 0.0}
      else: zero
    cloudydensity_CIV: {inherit: _non_star_forming}
    cloudydensity_SiIV: {inherit: _non_star_forming}
    cloudydensity_OVI: {inherit: _non_star_forming}

raytracer:
  max_step: 0.01
  hubble_flow: 0.0
  outputpath: absorption-metals.zr
  overwrite: true
  operators:
    sightlines:
      output:
        cells: true
      rays:
        - {id: 0, start: [0.57, 0.5, 0.01], end: [0.57, 0.5, 0.99]}
      tau:
        lines: [CIV, SiIV, OVI]
        line_density_fields:
          CIV: cloudydensity_CIV
          SiIV: cloudydensity_SiIV
          OVI: cloudydensity_OVI
        per_species: true
        lambda_min: 1028.0
        lambda_max: 1552.0
        nbins: 52400

The ray looks along +z, about 40 physical kpc from the map center. The dot marks its position on the visualization tutorial's temperature map: the ray travels into the image. This reuses the earlier projection; you do not need to render it again.

Temperature map with a dot marking the absorption sightline

The map shows density-weighted mean temperature along the viewing direction, rather than the depth of individual absorbers.

line_density_fields assigns each transition its own ion number density. per_species: true saves their optical depths separately for plotting. THOR multiplies each table ion fraction by the element number density from GFM_Metals. These absorption lines assume ground-state ions.

The field filters exclude star-forming gas, whose effective temperature is unsuitable for this equilibrium calculation. Hubble flow is off.

thor absorption-metals.yaml

A successful run ends with Thor finished. and writes absorption-metals.zr. Rerunning replaces that output (overwrite: true).

3. Inspect the result

Download thorplot.py and plot_absorption_metals.py into the same directory, then run:

uv run --python 3.11 --with zarr --with numpy --with matplotlib plot_absorption_metals.py

The output image absorption-metals.png compares the three ions.

C IV, Si IV, and O VI absorption through the TNG50 cutout C IV, Si IV, and O VI absorption through the TNG50 cutout

Each curve shows normalized flux \(e^{-\tau}\) against velocity relative to its own rest wavelength: C IV 1548.187 Å, Si IV 1393.755 Å, and O VI 1031.912 Å. The galaxy's bulk velocity has not been subtracted.

The narrow C IV and Si IV components contrast with the broader O VI profile. The ions weight the gas differently through their ion fractions and elemental abundances; their thermal widths also depend on temperature and atomic mass. No instrumental broadening or noise has been added.

4. Things to try

Trace the absorbing gasConnect gas density and velocity to the absorption features.

Gas density and velocity along the ray, with gas column overlaid on the metal spectra Gas density and velocity along the ray, with gas column overlaid on the metal spectra

The same run records the density, velocity and path length of each crossed cell (output.cells: true). No second simulation is needed.

Download plot_absorption_gas.py beside thorplot.py, then run:

uv run --python 3.11 --with zarr --with numpy --with matplotlib plot_absorption_gas.py

In absorption-gas.png, the top panel shows density and velocity along the ray. The lower panel overlays gas mass column per velocity bin in gray on the absorption spectra, using the right axis for its units. Positive velocity means redshift; gas moving toward the observer has negative velocity.

The gray profile includes all traced gas. Ion fractions, elemental abundances and thermal broadening determine how that gas appears in each absorption line.

Move the sightline

Change 0.57 to 0.61 in both ray endpoints, then run THOR and the plotting command again. The new ray passes about 63 kpc from the map center, farther left on the map. Compare which components remain visible in each ion.

Continue in the Cookbook

Extend the single sightline to a grid of spectra and maps of absorption strength.

For mock-observation workflows, see Sightline absorption.