EPoS Contribution
EPoS Contribution
What is the relation between the dust properties and temperature in molecular clouds?

Brandon Kelly
University of California, Santa Barbara, Santa Barbara, UnitedStates
Recent infrared observatories, including Herschel and Planck, provide unprecedented views into the nature of astrophysical dust across a range of environments. Determing the dust properties is critical to access physical conditions, such as temperatures and column densities, in star-forming regions. However, under commonly employed methods, the dust SED, as represented by a power--law modified black body, is subject to a strong degeneracy between the derived temperature and spectral index. This degeneracy can lead to a spurious anti-correlation between temperature and the spectral index when one fits these parameters independently for each source or pixel. I will discuss a novel hierarchical Bayesian method for fitting dust infrared spectral energy distributions to observed fluxes. I will discuss how our method is able to overcome the degeneracy between temperature and the spectral index by fitting both the distribution of the SED parameters simultaneously with the parameters for individual data points. Tests using simple models show that our method is capable of recovering the source parameters and correlations, indicating that this approach is substantially more accurate than traditional methods. Finally, I will discuss the application of our method to the star-forming Bok Globule CB244, and contrast the results with those obtained using traditional methods. Our method finds that there is a weak correlation between temperature and the spectral index, while the traditional methods find evidence for a strong anti-correlation, suggesting that the anti-correlation between these quantities seen previously is a spurious result caused by unaccounted for noise. Furthermore, we find that the spectral index is more strongly correlated with the column density than temperature, which may be driven by line-of-sight temperature and grain property variations in the dense ISM. Under the standard assumptions of optically thin line-of-sight averaged dust emission, our method allows for a better determination of the dust properties at long wavelengths, therefore facilitating the physical interpretation of the data. Moreover, our method represents an important new tool for the analysis of dusty systems and is broadly applicable to a range of astrophysical sources, such as GMCs, the diffuse ISM, and dust enshrouded ULIRGS/starburst systems.
R. Shetty, Univ. of Heidelberg, Germany; A. Stutz, MPIA, Germany; Jens Kauffmann, NASA JPL, USA; A. Goodman, Harvard-Smithsonian CfA, USA; R. Launhardt, MPIA, Germany