Comparing submillimeter polarized emission with near-infrared polarization of background stars for the Vela C molecular cloud

Fabio P. Santos, Peter A. R. Ade, Francesco E. Angile, Peter Ashton, Steven J. Benton, Mark J. Devlin, Bradley Dober, Laura M. Fissel, Yasuo Fukui, Nicholas Galitzki, Natalie N. Gandilo, Jeffrey Klein, Andrei L. Korotkov, Zhi-Yun Li, Peter G. Martin, Tristan G. Matthews, Lorenzo Moncelsi, Fumitaka Nakamura, Calvin B. Netterfield, Giles Novak, Enzo Pascale, Frederick Poidevin, Giorgio Savini, Douglas Scott, Jamil A. Shariff, Juan Diego Soler, Nicholas E. Thomas, Carole E. Tucker, Gregory S. Tucker, Derek Ward-Thompson

Published 2016-05-27Version 1

We present the first large-scale quantitative combination of near-infrared (near-IR) interstellar polarization data from background starlight with polarized emission data at submillimeter (sub-mm) wavelengths for a molecular cloud. Sub-mm data for the Vela C molecular cloud were obtained in Antartica by the Balloon-borne Large Aperture Submillimeter Telescope for Polarimetry (BLASTPol). The near-IR data consist of more than 6700 detections in the $I$-band, distributed in and around the cloud in the range of visual extinctions between $2$ and $20\,$mag. The main goal was to determine the polarization efficiency ratio $R_{\mathrm{eff}}$, defined as $p_{500}/(p_{I}/\tau_{V})$, where $p_{500}$ and $p_{I}$ are polarization fractions at $500\,\mu$m and $I$-band, respectively, and $\tau_{V}$ is the optical depth. To ensure that the same column density of material is producing both polarization from emission and from extinction, we introduce a new method to select stars that are located in the near-background, the Gaussian-logistic (GL) technique. The polarization efficiency ratio is critically affected by stars with contamination from the diffuse background Galactic material, emphasizing the need for a careful selection. Accounting for the statistical and systematic uncertainties from the GL method, we estimate an average $R_{\mathrm{eff}}$ value of $2.4\pm0.8$, which can be used to test the predictions of dust grain models designed for molecular clouds when such predictions become available. $R_{\mathrm{eff}}$ appears to be relatively flat as a function of the cloud depth for the range of visual extinctions probed.