paper

Exploring Systematic Effects in the Relation Between Stellar Mass, Gas Phase Metallicity, and Star Formation Rate

arXiv:1606.08850 · doi:10.3847/0004-637X/827/1/35

Abstract

There is evidence that the well-established mass-metallicity relation in galaxies is correlated with a third parameter: star formation rate (SFR). The strength of this correlation may be used to disentangle the relative importance of different physical processes (e.g., infall of pristine gas, metal-enriched outflows) in governing chemical evolution. However, all three parameters are susceptible to biases that might affect the observed strength of the relation between them. We analyze possible sources of systematic error, including sample bias, application of signal-to-noise ratio cuts on emission lines, choice of metallicity calibration, uncertainty in stellar mass determination, aperture effects, and dust. We present the first analysis of the relation between stellar mass, gas phase metallicity, and SFR using strong line abundance diagnostics from Dopita et al. (2013) for ~130,000 star-forming galaxies in the Sloan Digital Sky Survey and provide a detailed comparison of these diagnostics in an appendix. Using these abundance diagnostics yields a 30-55% weaker anti-correlation between metallicity and SFR at fixed stellar mass than that reported by Mannucci et al. (2010). We find that, for all abundance diagnostics, the anti-correlation with SFR is stronger for the relatively few galaxies whose current SFRs are elevated above their past average SFRs. This is also true for the new abundance diagnostic of Dopita et al. (2016), which gives anti-correlation between metallicity and SFR only in the high specific star formation rate (sSFR) regime, in contrast to the recent results of Kashino et al. (2016). The poorly constrained strength of the relation between stellar mass, metallicity, and SFR must be carefully accounted for in theoretical studies of chemical evolution.

arXiv comment21 pages, 18 figures. Accepted to the Astrophysical Journal. Updated to match published version (minor changes)

References in corpus (62)

Cited by in corpus (45)