We study whether integrated optical spectroscopy of a disk galaxy can be used to infer the mean or characteristic gas-phase oxygen abundance in the presence of systematic effects such as spatial abundance variations, contributions to the integrated emission-line spectrum from diffuse ionized gas, and dust attenuation. Our sample consists of 14 nearby disk galaxies with integrated spectrophotometry, and observations of more than 250 individual H II regions culled from the literature. We consider both theoretical and empirical strong-line abundance calibrations based on the R_(23) ≡ ([O II] + [O III])/Hβ parameter. We find that the integrated oxygen abundance correlates well with the gas-phase abundance measured at a fixed galactocentric radius, as determined by the H II region abundance gradient. The typical scatter in the correlation is ±0.1 dex, independent of the abundance calibration or whether the observed integrated emission-line fluxes, the reddening-corrected fluxes, or the emission-line equivalent widths are used. Integrated abundances based on the observed fluxes or equivalent widths, however, are susceptible to additional systematic effects of order 0.05-0.1 dex, at least for the range of reddenings and stellar populations spanned by our sample. Unlike the integrated R_(23) parameter, we find that the integrated [N II]/Hα and [S II]/Hα ratios are enhanced with respect to line ratios typical of H II regions, consistent with a modest contribution from diffuse ionized gas emission. We conclude that the R_(23) parameter can be used to reliably measure the gas-phase abundances of distant star-forming galaxies.
展开▼