Star formation rate in galaxies from UV, IR, and Hα estimators

Infrared (IR) luminosity of galaxies originating from dust thermal emission can be used as an indicator of the star formation rate (SFR). Inoue et al. (\cite{inoue00}, IHK) have derived a formula for the conversion from dust IR luminosity to SFR by using the following three quantities: the fraction of Lyman continuum luminosity absorbed by gas (f), the fraction of UV luminosity absorbed by dust (epsilon ), and the fraction of dust heating from old (ga 10⁸ yr) stellar populations (eta ). We develop a method to estimate those three quantities based on the idea that the various way of SFR estimates from ultraviolet (UV) luminosity (2000 Å luminosity), Hα luminosity, and dust IR luminosity should return the same SFR. After applying our method to samples of galaxies, the following results are obtained in our framework. First, our method is applied to a sample of star-forming galaxies, finding that f ~ 0.6, epsilon ~ 0.5, and eta ~ 0.4 as representative values. Next, we apply the method to a starburst sample, which shows larger extinction than the star-forming galaxy sample. With the aid of f, epsilon , and eta , we are able to estimate reliable SFRs from UV and/or IR luminosities. Moreover, the Hα luminosity, if the Hα extinction is corrected by using the Balmer decrement, is suitable for a statistical analysis of SFR, because the same {correction factor for the Lyman continuum extinction (i.e. 1/f)} is applicable to both normal and starburst galaxies over all the range of SFR. The metallicity dependence of f and epsilon is also tested: Only the latter proves to have a correlation with metallicity. As an extension of our result, the local (z=0) comoving density of SFR can be estimated with our dust extinction corrections. We show that all UV, Hα , and IR comoving luminosity densities at z=0 give a consistent SFR per comoving volume ( ~ 3x 10^-2h M_sun yr^-1 Mpc^-3). Useful formulae for SFR estimate are listed.

Tables 1 and 2, and Appendix A are only available in electronic form at http://www.edpsciences.org