2014MNRAS.440.3588T

Herschel far-infrared continuum data obtained as part of the Hi-GAL survey have been used, together with the GLIMPSE 8µm and MIPSGAL 24µm data, to attempt the first 3D-decomposition of dust emission associated with atomic, molecular and ionized gas at 15arcmin angular resolution. Our initial test case is a 2x2 square degrees region centred on (l, b) = (30°, 0°), a direction that encompasses the origin point of the Scutum-Crux Arm at the tip of the Galactic Bar. Coupling the IR maps with velocity maps specific for different gas phases (Hi 21cm, 12CO and 13CO, and radio recombination lines), we estimate the properties of dust blended with each of the gas components and at different Galactocentric distances along the line of sight (LOS). A statistical Pearson's coefficients analysis is used to study the correlation between the column densities estimated for each gas component and the intensity of the IR emission. This analysis provides evidence that the 2x2 square degree field under consideration is characterized by the presence of a gas component not accounted for by the standard tracers, possibly associated with warm H2 and cold Hi. We demonstrate that the IR radiation in the range 8 < λ < 500µm is systematically dominated by emission originating within the Scutum-Crux Arm. By applying an inversion method, we recover the dust emissivities associated with atomic, molecular and ionized gas. Using the DustEM model, we fit the spectral energy distributions for each gas phase, and find average dust temperatures of Td,H i = 18.82±0.47K, Td,H2 = 18.84±1.06K and Td,H ii = 22.56±0.64K, respectively. We also obtain an indication for polycyclic aromatic hydrocarbons depletion in the diffuse ionized gas. We demonstrate the importance of including the ionized component in 3D-decompositions of the total IR emission. However, the main goal of this work is to discuss the impact of the missing column density associated with the dark gas component on the accurate evaluation of the dust properties, and to shed light on the limitations of the inversion method approach when this is applied to a small section of the Galactic plane and when the working resolution allows sufficient de-blending of the gas components along the LOS.