1999ApJ...523..306K

We have carried out high-resolution (∼3') H I and CO line observations along one-dimensional cuts through the Galactic worm GW 46.4+5.5. By comparing the H I data with IRAS data, we have derived the distributions of I₁₀₀ excess and τ₁₀₀ excess, which are, respectively, the 100 µm intensity and 100 µm optical depth in excess of what would be expected from H I emission. In two observed regions, we were able to make a detailed comparison of the infrared excess and the CO emission. We have found that τ₁₀₀ excess has a very good correlation with the integrated intensity of CO emission, W_CO, but I₁₀₀ excess does not. There are two reasons for the poor correlation between I₁₀₀ excess and W_CO: first, there are regions with enhanced infrared emissivity without CO, and second, dust grains associated with molecular gas have a low infrared emissivity. In one region, these two factors completely hide the presence of molecular gas in the infrared. In the second region, we could identify the area with molecular gas, but I₁₀₀ excess significantly underestimates the column density of molecular hydrogen because of the second factor mentioned above. We therefore conclude that τ₁₀₀ excess, rather than I₁₀₀ excess, is an accurate indicator of molecular content along the line of sight. We derive τ₁₀₀/N(H)=(1.00±0.02)x10^–5 (10²⁰ cm^–2)^–1, and X≡N(H₂)/W_CO≃0.7x10²⁰ cm^–2 (K.km.s^–1)^–1. Our results suggest that I₁₀₀ excess could still be used to estimate the molecular content if the result is multiplied by a correction factor ξ_c≡I₁₀₀/N(H)_HI_/I100_/N(H)H₂_ (≃2 in the second region), which accounts for the different infrared emissivities of atomic and molecular gas. We also discuss some limitations of this work, which could stem from using single-temperature model and the IRAS 60 µm intensity in estimating the dust optical depth along the line of sight.