EVIDENCE FOR ENVIRONMENTAL CHANGES IN THE SUBMILLIMETER DUST OPACITY

The submillimeter opacity of dust in the diffuse interstellar medium (ISM) in the Galactic plane has been quantified using a pixel-by-pixel correlation of images of continuum emission with a proxy for column density. We used multi-wavelength continuum data: three Balloon-borne Large Aperture Submillimeter Telescope bands at 250, 350, and 500 mu m and one IRAS band at 100 mu m. The proxy is the near-infrared color excess, E(J - K-s), obtained from the Two Micron All Sky Survey. Based on observations of stars, we show how well this color excess is correlated with the total hydrogen column density for regions of moderate extinction. The ratio of emission to column density, the emissivity, is then known from the correlations, as a function of frequency. The spectral distribution of this emissivity can be fit by a modified blackbody, whence the characteristic dust temperature T and the desired opacity sigma(e)(1200) at 1200 GHz or 250 mu m can be obtained. We have analyzed 14 regions near the Galactic plane toward the Vela molecular cloud, mostly selected to avoid regions of high column density (N-H > 10(22) cm(-2)) and small enough to ensure a uniform dust temperature. We find sigma(e)(1200) is typically (2-4) x 10 (25) cm(2) H-1 and thus about 2-4 times larger than the average value in the local high Galactic latitude diffuse atomic ISM. This is strong evidence for grain evolution. There is a range in total power per H nucleon absorbed (and re-radiated) by the dust, reflecting changes in the strength of the interstellar radiation field and/or the dust absorption opacity. These changes in emission opacity and power affect the equilibrium T, which is typically 15 K, colder than at high latitudes. Our analysis extends, to higher opacity and lower temperature, the trend of increasing sigma(e)(1200) with decreasing T that was found at high latitudes. The recognition of changes in the emission opacity raises a cautionary flag because all column densities deduced from dust emission maps, and the masses of compact structures within them, depend inversely on the value adopted.