Estimating Interstellar Medium Dust Temperature And Spectral Index In The Far-infrared And Submillimeter

Dust temperature and spectral index are evidenced to be anti-correlated from observations in the far-infrared and millimeter wavelengths and from laboratory experiments. However, uncertainties in flux measurements combined with calibration errors and other source of systematic errors, affect the results of the spectral energy distribution (SED) fit. An inverse correlation between dust temperature and spectral index naturally arises from the spectral model assumed for the fit combined with data noise and systematic uncertainties. When the spectral coverage do not sample the whole SED but only a limited range of it, it is even more difficult to get reliable results on dust physical properties. We developed a method to fit the inverse relationship between the temperature and spectral index with Bayesian statistics taking properly into account both the statistics and the systematic errors. We simulate observations of one-component Interstellar Medium (15 K < T < 25 K), and of two-components sources both warm (HII regions) and cold (cold cores) in the Herschel PACS and SPIRE spectral bands (70-500 um). We also include some ancillary simulated data from Planck-HFI, IRAS and MIPS to better sample the SEDs.