The Initial Conditions of Stellar Protocluster Formation: III. The Herschel counterparts of the Spitzer Dark Cloud catalogue

Galactic plane surveys of pristine molecular clouds are key for establishing a Galactic-scale view of the earliest stages of star formation. For this reason Peretto & Fuller (2009) built an unbiased sample of IRDCs in the 10 deg < |l| < 65 deg, |b|<1 deg region of the Galactic plane using Spitzer 8micron extinction. However, in absorption studies, intrinsic fluctuations in the mid-infrared background can be mis-interpreted as foreground clouds. The main goal of the study presented here is to disentangle real clouds in the Spitzer Dark Cloud (SDC) catalogue from artefacts due to fluctuations in the mid-infrared background. We constructed H_2 column density maps at ~18" resolution using the 160micron and 250micron data from the Herschel Galactic plane survey Hi-GAL. We also developed an automated detection scheme that confirms the existence of a SDC through its association with a peak on these Herschel column density maps. Detection simulations, along with visual inspection of a small sub-sample of SDCs, have been performed to get better insight into the limitations of our automated identification scheme. Our analysis shows that 76(+/-19)% of the catalogued SDCs are real. This fraction drops to 55(+/-12)% for clouds with angular diameters larger than ~1 arcminute. The contamination of the PF09 catalogue by large spurious sources reflect the large uncertainties associated to the construction of the 8micron background emission, a key stage towards the identification of SDCs. A comparison of the Herschel confirmed SDC sample with the BGPS and ATLASGAL samples shows that SDCs probe a unique range of cloud properties, reaching down to more compact and lower column density clouds than any of these two (sub-)millimetre Galactic plane surveys.