Is the Galactic submillimeter dust emissivity underestimated?

¹ Max-Planck-Institut für Extraterrestrische Physik, Postfach 1312, 85741 Garching, Germany e-mail: dasyra@mpe.mpg.de
² University of Crete, Physics Department, PO Box 2208, 71003 Heraklion, Crete, Greece
³ National Observatory of Athens, I. Metaxa & Vas. Pavlou str., Palaia Penteli, 15236 Athens, Greece
⁴ Foundation for Research and Technology-Hellas, PO Box 1527, 71110 Heraklion, Crete, Greece

Received: 26 November 2004
Accepted: 9 March 2005

Abstract

We present detailed modeling of the spectral energy distribution (SED) of the spiral galaxies NGC 891, NGC 4013 and NGC 5907 in the far-infrared (FIR) and sub-millimeter (submm) wavelengths. The model takes into account the emission of the diffuse dust component, which is heated by the UV and optical radiation field produced by the stars, as well as the emission produced locally in star forming HII complexes. Radiative transfer simulations in the optical bands are used to constrain the stellar and dust geometrical parameters and the dust mass. We find that the submm emission predicted by our model cannot account for the observed fluxes at these wavelengths. Two scenarios that could account for the “missing” submm flux are examined. In the first scenario dust additional to that derived from the optical wavelengths is embedded in the galaxy in the form of a thin disk. This additional dust disk, which is not detectable in the optical and which is associated with the young stellar population, gives rise to additional submm emission, and makes the total flux match the observed values. The other scenario examines the possibility that the average emissivity at submm wavelengths of the dust grains found both in a diffuse component and in denser environments (e.g. molecular gas clouds) is higher than the values widely used in Galactic environments. This enhanced emissivity reproduces the observed FIR and submm fluxes with the dust mass equal to that derived from the optical observations. In the second scenario, we treat the submm emissivity as a free parameter and calculate its nominal value by fitting our model to the observed SED. We find a dust emissivity which is ~ times the often-used values for our Galaxy. Both scenarios can equally well reproduce the observed 850 μm surface brightness for all three galaxies. However, we argue that the scenario of having more dust embedded in a second disk is not supported by the near infrared observations. At 2.16 μm, the model images with a second dust disk reveal a prominent dust lane which is not present in the observations. Thus, the enhanced emissivity at submm wavelengths is a real possibility and the Galactic submillimeter dust emissivity may be underestimated.