Read an Astrodust model from cross sections to polarized emission

Freeze the question and the paper version

Use v1, submitted on 25 August 2022. The equation and figure numbers below refer to that version. The draft’s printed date is 29 August. The associated Dataverse model is version 2.0. A manuscript version and a dataset version identify different objects; record both.

The paper tests a combined Astrodust-plus-PAH description against adopted diffuse-interstellar constraints. Astrodust supplies a common composite material for the aligned larger grains; the small PAH population contributes additional spectral behavior. “One component” in the paper’s discussion does not mean that every grain has one size, one temperature, or no additional small-grain population.

The fitting objective in equation 27 uses relative residuals and wavelength weighting. It is not a noise-calibrated posterior over all material choices. A best-fit curve alone leaves other compositions possible. Start by identifying the fixed assumptions and the observations that constrain them.

Reading: Hensley & Draine, arXiv:2208.12365v1 (2022) · Astrodust+PAH Milky Way model, Dataverse version 2.0

Follow a signed cross section into Figure 6

Sections 2.1–2.2 distinguish cross sections for different grain orientations, a random-orientation extinction average and a signed polarization difference. A polarization cross section is not automatically a positive fraction. Keep the sign until a sky basis and observable have been specified. Also keep $C_{\mathrm{abs}}=C_{\mathrm{ext}}-C_{\mathrm{sca}}$: total extinction cannot be inserted into the heating integral.

The modified picket fence approximation, MPFA, combines a small set of grain orientations. Its Rayleigh-limit behavior is useful, but it is not exact at every grain size relative to wavelength. Figure 6’s lower panel compares the approximation with direct values at a specified angle; the upper panel separates contributions by size bin.

Read those panels together. The unwanted short-ultraviolet polarization feature is associated with particular sizes, not merely a plotting residual. It tests the assumption that the aligned grains all share the adopted composition and shape. Replacing signed cross sections by absolute values would hide part of that test.

Reading: Hensley & Draine, arXiv:2208.12365v1 (2022)

Why the temperature distribution comes before the size sum

Equation 8 averages thermal emission over a grain’s temperature distribution before summing sizes and materials. For equilibrium large grains the distribution can be narrow. For small grains and PAHs, discrete photon heating makes it broad. Replacing every distribution by one steady root would remove a physical emission mechanism.

Equations 9–10 define a fixed mMMP stellar/ultraviolet radiation shape scaled by $U$, plus an unscaled $2.725\,\mathrm K$ cosmic microwave background. Turning up $U$ is therefore not the same as multiplying every radiation component. Nor does it represent the frequency-dependent reddening inside our teaching cloud.

The source field extends to $0.0912\,\mu\mathrm m$, while the native text wavelength grid starts at $0.1\,\mu\mathrm m$. That small-looking mismatch matters to an energy audit. Missing heating wavelengths cannot be recreated by extending a table without a stated material model.

When comparing printed equations 8 and 11, write down your own per-hydrogen size weight before manipulating them: their explicit normalization notation differs. Preserve the v1 context rather than silently editing the equation or guessing an author correction. The audit below uses the native bin counts with an explicit definition.

Reading: Hensley & Draine, arXiv:2208.12365v1 (2022)

Worked native check: add volume, then weight alignment

The complete CC0 tar contains $167$ size rows. Let $N_i$ be the number of grains in native bin $i$ per hydrogen nucleus and $a_i$ its equal-volume radius. Convert micrometres to centimetres before computing $V_i=4\pi a_i^3/3$. The per-hydrogen volume is the sum $\sum_iN_iV_i$; no extra bin-width factor belongs on counts that are already per bin.

The actual sums give Astrodust volume $3.922306\times10^{-27}\,\mathrm{cm^3/H}$ and PAH volume $5.511864\times10^{-28}\,\mathrm{cm^3/H}$. Applying the source densities gives mass ratios to hydrogen mass of $0.00642172$ and $0.000658700$ respectively. These are not ratios to gas mass including helium.

For alignment, weight the native function $f_i$ by grain volume. The result $0.698305$ is consistent with the printed volume-weighted value $0.70$. This checks weights, units and source rounding. It does not independently infer alignment from observations or solve radiative-torque alignment.

After the offline replay, inspect model-unit-audit.json and native-polarization-audit.json in docs/astronomy/ac35. The former comes from the complete original tar with publisher MD5 and received SHA checks. Selected complete FITS HDUs have exact range and SHA records; the whole FITS was not downloaded and its full-file MD5 was not independently checked.

Reading: Astrodust+PAH Milky Way model, Dataverse version 2.0 · Dust absorption, heat and polarization: standalone science · Hensley & Draine, arXiv:2208.12365v1 (2022)

Read Figures 11, 12 and 18 with their denominators

Figure 11 compares polarized emission scaled to a maximum per-hydrogen geometry with adopted WMAP and Planck constraints. Figure 12 normalizes the polarization fraction at $353\,\mathrm{GHz}$ to study its frequency dependence. A normalized curve discards one amplitude; a maximum-geometry reference is not a recovered three-dimensional magnetic field.

The selected native FITS output gives a signed polarized-to-random-orientation-total reference of $0.191718$ at $353\,\mathrm{GHz}$ and $\log_{10}U=0.2$, consistent with the printed $0.192$. Keep those reference geometries in the label. It is not the arbitrary-direction Stokes fraction of a real cloud.

Spinning dust and vibrational emission are distinct components. The text file model_irem.dat excludes spinning dust, which has its own file. Figure 12 discusses unpolarized spinning dust in its long-wavelength decline; Figure 18 explicitly omits that contribution. Comparing those figures without this distinction would assign different denominators to the same apparent curve.

Figure 18 changes $U$ while keeping the size and alignment functions fixed. It asks how thermal weighting changes polarization for that family. It does not demonstrate how stronger radiation evolves alignment. The observing-facility forecasts in the v1 discussion are statements made in 2022, not a checked description of the observing program in 2026.

Reading: Hensley & Draine, arXiv:2208.12365v1 (2022) · Astrodust+PAH Milky Way model, Dataverse version 2.0 · Dust absorption, heat and polarization: standalone science

Keep the energy diagnostic that did not close

Run audit_model_archive.py through the scientific replay. It checks source composition sums, positive absorption, dimensions and the native mass weights. Separately, it constructs the printed radiation field and compares absorbed and emitted power over the available wavelength coverage.

The emitted-to-covered-absorbed ratio is about $0.9880$ to $0.9944$ across the native field grid. That is not unity, and the output does not label it a completed full-energy-conservation check. Missing ultraviolet coverage, temperature-distribution integration and other numerical or model differences have not been separated enough to assign one cause.

Do not rescale the emission to make this diagnostic pass. The failure is useful: it shows exactly where a source-model audit stops short of reproducing the original calculation. By contrast, our own bounded equilibrium-cloud producer has a closed declared heating integral, separate tail bounds and independent transfer references. Its pass cannot be transferred to the native model.

A productive follow-up question is now precise: which new wavelength coverage, material cross section or original temperature-distribution calculation would close that comparison? That is a stronger research starting point than claiming a full grain-model reproduction from a successful file reader.

Reading: Dust absorption, heat and polarization: standalone science · Hensley & Draine, arXiv:2208.12365v1 (2022)