Temperature-dependent UV cross sections

Most laboratory UV cross sections are measured at or below room temperature, which raises reliability concerns for hot atmospheres. VULCAN can interpolate photoabsorption cross sections in temperature, layer by layer (Section 2.5 of Tsai et al. 2021 ¹).

Motivation

Heays et al. (2017) ² noted that for many molecules a temperature increase of a few hundred kelvin only broadens the cross section slightly and leaves its wavelength integral unchanged. However, for molecules with prominent transitions between excited vibrational states (CO\(_2\) in particular) both the absorption threshold and the cross section depend strongly on temperature, which in turn affects the photolysis rate and the shielding of other species.

Available data

Temperature-dependent photoabsorption cross sections are included for:

For H\(_2\)O the noisy data above 216 nm are fitted log-linearly as a conservative estimate.

Interpolation scheme

The cross section of a given species is allowed to vary across the atmosphere with the local temperature. Interpolation is linear in temperature and logarithmic in the cross section. Because data are sparse, linear-in-\(T\) interpolation tends to underestimate the cross section, so the implementation is a conservative lower bound on how strongly photolysis increases with temperature.

Implementation in VULCAN

Temperature dependence is activated by listing species in config.T_cross_sp (empty by default). In op.ReadRate.make_bins_read_cross:

• For each such species, per-temperature files <sp>_cross_<T>K.csv are read from thermo/photo_cross/<sp>/, and the room-temperature file is registered as the 300 K entry.
• For every layer, the routine finds the two bracketing tabulated temperatures \(T_\mathrm{low}\) and \(T_\mathrm{high}\) around the local \(T_z\), interpolates \(\log_{10}\sigma\) linearly between them, and stores the result in cross_T[sp] (shape \(n_z \times n_\mathrm{bin}\)) and the per-branch cross_J_T[(sp, i)].
• Outside the wavelength range covered by the T-dependent data, the standard room-temperature cross section cross[sp] is used. Temperatures below the lowest / above the highest tabulated value fall back to the nearest available cross section.

These layer-resolved arrays then feed compute_tau (optical depth) and compute_J (photolysis rates), where the T-dependent branch is taken for any species in T_cross_sp.