2013A&A...558A..44M

Mapping of the near-infrared (NIR) scattered light is a recent method for the study of interstellar clouds, complementing other, more commonly used methods, like dust emission and extinction. Our goal is to study the usability of this method on larger scale, and compare the properties of a filamentary structure using infrared scattering and other methods. We also study the radiation field and differences in grain emissivity between diffuse and dense areas. We have used scattered NIR J, H, and K band surface brightness observations with WFCAM instrument to map a filament TMC-1N in Taurus Molecular Cloud, covering an area of 1°x1° corresponding to ~(2.44pc)². We have converted the data into an optical depth map and compared the results with NIR extinction and Herschel observations of sub-mm dust emission. We have also modelled the filament with 3D radiative transfer calculations of scattered light. We see the filament in scattered light in all three NIR bands. We note that our WFCAM observations in TMC-1N show notably lower intensity than previous results in Corona Australis using the same method. We show that 3D radiative transfer simulations predict similar scattered surface brightness levels as seen in the observations. However, changing the assumptions about the background can change the results of simulations notably. We derive emissivity, the ratio of FIR dust emission to column density, by using optical depth in the J band, τ_J, obtained from NIR extinction map as an independent tracer of column density. We obtain a value 0.0013 for the ratio τ₂₅₀/τ_J^Nicer. This leads to opacity or dust emission cross-section σ_e(250µm) values 1.7-2.4x10^–25cm²/H, depending on assumptions of the extinction curve, which can change the results by over 40%. These values are twice as high as obtained for diffuse areas, at the lower limit of earlier results for denser areas. We show that NIR scattering can be a valuable tool in making high resolution maps. We conclude, however, that NIR scattering observations can be complicated, as the data can show comparatively low-level artefacts. This suggests caution when planning and interpreting the observations.