The Effect of Noise on the Dust Temperature - Spectral Index Correlation

We investigate how uncertainties in flux measurements affect the results from modified blackbody SED fits. We show that an inverse correlation between the dust temperature T and spectral index (beta) naturally arises from least squares fits due to the uncertainties, even for sources with a single T and beta. Fitting SEDs to noisy fluxes solely in the Rayleigh-Jeans regime produces unreliable T and beta estimates. Thus, for long wavelength observations (lambda >~ 200 micron), or for warm sources (T >~ 60 K), it becomes difficult to distinguish sources with different temperatures. We assess the role of noise in recent observational results that indicate an inverse and continuously varying T - beta relation. Though an inverse and continuous T - beta correlation may be a physical property of dust in the ISM, we find that the observed inverse correlation may be primarily due to noise.