An empirical relation that correlates the ultraviolet spectral slope (beta) to the ratio of far-infrared (FIR) to far-ultraviolet (FUV) flux (infrared excess or IRX) is known to exist for galaxies where star formation strongly dominates the ultraviolet emission (starburst galaxies). Since the IRX is a measure of the FUV dust attenuation (AFUV), this fortuitous result enables one to determine AFUV from rest-frame ultraviolet measurements alone---at least for starburst galaxies. The main focus of this dissertation is to evaluate the applicability of the IRX-beta correlation to a wide range of galaxies in the local and high redshift universe
We conduct a thorough review of ultraviolet spectral slope characteristics, empirical attenuation relations, and theoretical interpretations. We examine the broad-band photometric estimate (betaph) of beta and develop methods for correcting it for foreground Galactic extinction. Using a sample of nearby galaxies imaged in the ultraviolet with the Ultraviolet Imaging Telescope, we explore the limits of the IRX-beta relation. We find that 55% of the sample agree with the IRX-beta relation and confirm that starburst and HII galaxies agree with the IRX-beta relation as well as Seyfert galaxies. We derive a new estimate for the total FIR flux (lambda > 40mum) (FIR3) using data on galaxies observed at wavelengths long-wards of 100mum. For the systems with smaller than expected values of IRX for a given value of betawe show the contribution of unobserved very cold dust is not the cause. We then consider the decoupling of FUV and FIR emission and the effects of old and intermediate age stellar populations. We suggest that the age dependency on extinction continues to evolve such that intermediate age and old stellar populations may have progressively lower extinctions and this may enhance the problems of measuring beta photometrically. We find that galaxies with a Young Mass Fraction (YMF ∝ specific star formation rate) < 10 -2.5 do not have enough young stars to dominate the near-UV (NUV) bandpass and for this reason the photometric estimate of beta is significantly worse than the spectroscopic method for systems with YMF < 10-2 . Many of the egregious outliers to the IRX-beta relation can be excluded be selecting only those with NUV-V < 1.7. The systems we highlight as problematic will contribute only ∼30% of the integrated UV luminosity in a UV survey, and with minimal corollary optical data many of these systems can be excluded entirely
In regard to high redshift galaxies, we find that Lyman Break Galaxies have YMF < 10-2 and NUV-V " 1.7 which suggests that they should follow the IRX-beta reddening correlation. We investigate the ultraviolet extinction properties of Lyman Break Galaxies by using X-rays as a proxy for FIR emission. We derive the bolometric to X-ray correlation for a local sample of normal and starburst galaxies and use it, in combination with several UV reddening schemes, to predict the 2--8 keV X-ray luminosity for a sample of Lyman-break galaxies in the HDF/CDF-N. We find that the mean X-ray luminosity, as predicted from the Meurer UV reddening relation for starburst galaxies, agrees extremely well with the Brandt stacking analysis. (Abstract shortened by UMI.)
