A Comparison of Independent Star Formation Diagnostics for an Ultraviolet‐selected Sample of Nearby Galaxies

Abstract

We present results from a decimetric radio survey undertaken with the Very Large Array (VLA) as part of a longer term goal to inter-compare star formation and dust extinction diagnostics, on a galaxy by galaxy basis, for a representative sample of nearby galaxies. For our survey field, Selected Area 57, star formation rates derived from 1.4 GHz luminosities are compared with earlier nebular emission line and ultraviolet (UV) continuum diagnostics. We find broad correlations, over several decades in luminosity, between Hα, the UV continuum and 1.4 GHz diagnostics. However, the scatter in these relations is found to be larger than observational errors, with offsets between the observed relations and those expected assuming constant star-formation histories and luminosity-independent extinction models. We investigate the physical origin of the observed relations, and conclude the discrepancies between different star-formation diagnostics can only be partly explained by simple models of dust extinction in galaxies. These models cannot by themselves explain all the observed differences, introducing the need for temporally varying star-formation histories and/or more complex models of extinction, to explain the entire dataset.

A Multiwavelength Approach to the Star Formation Rate Estimation in Galaxies at Intermediate Redshifts

“…Interestingly, all the galaxies at higher redshift (z $ 0:8) exhibit underestimated SFR UV . These results are consistent with the work of Sullivan et al (2001), who found that the UV/H ratio decreases with increasing SFR for galaxies in the redshift range 0 < z < 0:5, whereas for low-SFR objects the UV luminosities lead to higher SFRs than H . Previous observations of distant galaxies, using uncorrected H fluxes, had already revealed that the UV flux underestimates the SFR by a factor of a few (Glazebrook et al 1999;Yan et al 1999;Moorwood et al 2000).…”

“…As a result of the advent of the Hubble Space Telescope (HST) and of the 8-10 m class telescopes, the rest-frame ultraviolet (UV) star formation history per unit comoving volume initially derived up to z $ 1 (Lilly et al 1996) was extended up to z $ 4 (Madau et al 1996;Steidel et al 1999). Naturally, all these works rely on the use of the available star formation rate (SFR) diagnostics: nebular emission lines like H (Kennicutt 1992a(Kennicutt , 1998Gallego et al 1995;Tresse & Maddox 1998;Glazebrook et al 1999;Moorwood et al 2000;Hopkins, Connolly, & Szalay 2000) and [O ii] (Hogg et al 1998;Aragó n-Salamanca et al 2002), UV continuum luminosities (Lilly et al 1996;Madau et al 1996;Treyer et al 1998;Cowie, Songalia, & Barger 1999;Steidel et al 1999;Sullivan et al 2000Sullivan et al , 2001, far-infrared (FIR) luminosities (Rowan-Robinson et al 1997;Blain et al 1999;Flores et al 1999;Chary & Elbaz 2001), and 1.4 GHz radio luminosities (Condon 1992;Cram et al 1998). Different cosmic star formation histories have been advocated from these various studies, which can only be reconciled if the effect of dust extinction is accounted for.…”

“…Hopkins et al (2001) and Sullivan et al (2001) report an increase of the F(H )/F(H ) ratio, used to measure extinction, with the SFR of galaxies. They suggest to use the fit to this correlation in order to account for dust extinction when H is not available.…”

“…In this work we extend to higher redshifts the comparison between different SFR estimators that have been used for samples of local galaxies (e.g., Sullivan et al 2001). For that purpose, we compare classical SFR indicators such as the nebular emission lines [O ii] and H , or the UV continuum, which we correct for dust extinction using the Balmer decrement, with the FIR and radio luminosities, which are unaffected by extinction.…”

Star Formation and Dust Extinction in Nearby Star Forming and Starburst Galaxies

“…Only spatially resolved studies would allow to remove this uncertainty. Hopkins et al (2001) and Sullivan et al (2001) report a positive correlation between the dust extinction traced by the Balmer decrement and the star formation rate of the galaxies traced by their total FIR or Hα luminosities. In Fig.…”