Previous abstract Next abstract
In order to understand some curious effects discovered in analyzing our deep JHK near-infrared survey of the background stars probing the IC 5146 dark cloud complex (Lada, Lada, Clemens, \& Bally 1993), we have constructed a simple model of the dust extinction through a molecular cloud. The effect noticed involved a correlation between the dispersion of the E(H-K) based estimate of A$_V$, when the stellar estimates of E(H-K) were binned into arcmin sized bins, with the mean A$_V$ computed for those bins. The sense of the correlation is that the dispersion of the extinction rises with the extinction in a nearly linear fashion. Further, the dispersion of the dispersion also rises with extinction. Our model was constructed to try to understand the origin of this unexpected behavior. The model consists of a Poisson generator to populate a bin with stars and various extinction generating functions to add extinction to each star. Additionally, measurement noise and varying amounts of foreground star contamination are added to simulate the actual observations. Remarkably, this simple model is able to rule out several cloud structure models, including uniform extinction across an arcmin sized bin and the case of dense clumplets (rocks) embedded in a low extinction medium. We show that a power law parameterization of the extinction variation with position across a bin is able to fully reproduce the observations for a fairly robust set of power law indices. We also show that foreground star contamination plus any simple extinction model cannot reproduce the observations, while foreground star contamination does not appreciably affect the power law extinction model for foreground stellar fractions less than 30 - 50\% of the total stellar content.
Wednesday program listing