[Paper Review] Dust Extinction and Emission in a Clumpy Galactic Disk. An Application of the Radiative Transfer Code TRADING

S. Bianchi|ArXiv.org|Jul 18, 2008

Galaxies: Formation, Evolution, PhenomenaPhysics and Astronomy70 references70 citations

TL;DR

This paper presents TRADING, a Monte Carlo radiative transfer code that self-consistently models dust extinction and emission in a clumpy galactic disk using an adaptive grid to resolve dense clouds and diffuse components. It reproduces the SED of NGC 891 from UV to submm, showing that 60% of dust emission is heated by old stellar populations, with significant cold emission from clouds heated by diffuse radiation, challenging prior models that overestimated UV heating contributions.

ABSTRACT

AIMS: I present the Monte Carlo radiative transfer code TRADING (Transfer of RAdiation through Dust In Galaxies). The code computes self-consistently the extinction of radiation in a dusty medium (including absorption and scattering) and the dust emission. METHODS: A binary-tree adaptive grid is used for the description of the dust distribution. Dust radiation is computed at thermal equilibrium or under stochastic heating condition, for a distribution of grains of different radii and materials. The code is applied to the case of a clumpy galactic disk, including both diffuse dust and a distribution of spherical clouds modelled on the GMCs of local galaxies. Diffuse and localised sources of starlight are used, with independent spectra. RESULTS: A model is provided for the edge-on galaxy NGC 891. The SED of the galaxy from the UV to the submm/mm range can be well reproduced by: a bulge/disk configuration of old stars together with an extended dust disk, as suggested by the analysis of optical/near-infrared images; a clumpy dust distribution of the same mass as the diffuse dust disk, together with a UV emitting component, half of which in the form of a diffuse disk and half in sources embedded in clouds. In total, it is found that about 35% of the bolometric radiation is absorbed (and emitted) by dust; and that absorption of starlight from the old population contributes to about 60% of the dust emission. A significant component of the dust emission from clouds is due to absorption of diffuse radiation. Radial profiles of dust emission in a clumpy disk are almost independent of the wavelength, with the exception of the wavelength range on the Wien side of the thermal equilibrium peak.

Motivation & Objective

• To develop a self-consistent radiative transfer code capable of modeling both dust extinction and emission in inhomogeneous galactic dust distributions.
• To investigate the impact of clumpy dust structures—specifically, molecular cloud-like regions—on the global spectral energy distribution (SED) of spiral galaxies.
• To determine the relative contributions of diffuse and embedded stellar sources to dust heating, particularly in edge-on systems like NGC 891.
• To assess whether standard models that treat diffuse and clumped dust components independently overestimate the role of UV radiation in heating dust.
• To explore how adaptive grid resolution improves the accuracy of dust emission and extinction predictions in complex, multi-scale galactic environments.

Proposed method

• TRADING employs a Monte Carlo radiative transfer technique to simulate photon propagation through a 3D dust distribution with both diffuse and clumpy components.
• An adaptive binary-tree grid is used to resolve high-density regions (e.g., molecular clouds) with higher spatial resolution than low-density regions.
• Dust emission is computed under both thermal equilibrium and stochastic heating conditions for a distribution of grain sizes and compositions, using the Draine & Li (2007) dust model.
• Starlight from separate diffuse and localized sources (old and young stars) is simulated with independent spectra, and their interaction with dust is tracked.
• Self-absorption and scattering are included in the radiative transfer, ensuring energy conservation and accurate SED construction.
• The code generates images across UV to submm wavelengths and integrates them to produce the total SED for comparison with observations.

Experimental results

Research questions

• RQ1 How does the inclusion of clumpy dust structures affect the predicted SED of a spiral galaxy like NGC 891?
• RQ2 What fraction of dust emission in a clumpy disk is powered by diffuse interstellar radiation rather than directly by embedded young stars?
• RQ3 To what extent do old stellar populations contribute to dust heating in edge-on galaxies, compared to young stars?
• RQ4 How does adaptive grid resolution improve the accuracy of dust emission and extinction modeling compared to regular grids?
• RQ5 Why do some models overpredict MIR emission and underpredict submm emission compared to observations?

Key findings

• The SED of NGC 891 from UV to submm is well reproduced by a model combining a diffuse dust disk (τ_V ≈ 1) and a clumpy dust component of similar mass.
• Approximately 35% of the total bolometric luminosity is absorbed and re-emitted by dust, with 60% of this emission powered by radiation from the old stellar population.
• A significant fraction of dust emission from clouds arises from heating by diffuse interstellar radiation, not just embedded UV sources, leading to cold submm emission not captured in earlier models.
• The radial profiles of dust emission in the clumpy model are nearly wavelength-independent across the MIR to submm range, except on the Wien side of the thermal peak (20–100 μm).
• The model explains the observed K-band extinction and 850 μm morphology better than previous models, due to self-consistent treatment of diffuse and clumped components.
• The discrepancy between model and observations in the MIR may stem from unresolved ISM complexity or an oversimplified ISRF, suggesting the need for more detailed ISM structure in future models.