2015A&A...574L...5P

Dust is a good tracer of cold dark clouds but its column density is difficult to quantify. We want to check whether the far-infrared and submillimeter high-resolution data from Herschel PACS and SPIRE and cameras combined with ground-based telescope bolometers allow us to retrieve the whole dust content of cold dark clouds. We compare far-infrared and submillimeter emission across L183 to the 8 µm absorption map from Spitzer data and fit modified blackbody functions towards three different positions. We find that none of the Herschel/SPIRE channels follow the cold dust profile seen in absorption. Even the ground-based submillimeter telescope observations, although more closely following the absorption profile, cannot help to characterize the cold dust without external information such as the dust column density itself. The difference in dust opacity can reach up to a factor of ∼3 in prestellar cores of high extinction. In dark clouds, the amount of very cold dust cannot be measured from its emission alone. In particular, studies of dark clouds based only on Herschel data can miss a large fraction of the dust content. This has an impact on core and filament density profiles, mass and stability estimates.