H ii regions

Necessary input data for empirical calibrations of strong-line methods are obviously the direct abundances obtained from detailed studies of high-metallicity H ii regions focusing on the inner zones of giant spiral galaxies. Early works on the abundances of high-metallicity H ii regions from auroral lines include those of Kinkel & Rosa (1994) on the region Searle 5 in M101, the anchor point at high metallicity of the early Pagel et al. (1979) calibration, Castellanos et al. (2002) on CDT1 in NGC 1232, and Kennicutt et al. (2003) on Searle 5 and H1013 in M101. The availability of 8-m-class telescopes in recent years has expanded the field considerably. There are now direct abundance determinations for about a dozen H ii regions in M51 (Bresolin et al. 2004; Garnett et al. 2004b), and a larger sample in five southern spirals has been studied by Bresolin et al. (2005).

Figure 17.2 shows the relation between R23 and the nebular abundance derived from the availability of auroral lines, essentially [S iii] A6312 and [N ii] X5755, as described earlier. The data points have been taken from the literature, emphasizing with larger symbols the extension to low R23 values (higher abundance) derived from observations with 8-m-class telescopes. The double-valued nature of the R23 indicator is apparent; the degeneracy can be removed through the availability of emission-line diagnostics that are monotonic with the oxygen abundance (e.g. [N ii] A6583/Ha). The scatter in the data shows the sensitivity of R23 to additional parameters besides the chemical abundance, such as the ionization parameter and the effective temperature of the ionizing stars (Perez-Montero & Diaz 2005). Note that nebular abundances still hover around the Solar value even at the smallest observed R23, and only a handful of points appear to be more metal-rich than the Solar value. In addition, the curves in Figure 17.2 clearly illustrate the well-known discrepancy between R23 calibrations obtained empirically, here taken from the P-method of Pilyugin & Thuan (2005), from auroral line observations and those obtained from photoionization-model grids; here Kewley & Dopita (2002) is taken

log R23

Figure 17.2. This abundance versus R23 diagram shows observational data for extragalactic H11 regions, for which the O/H abundances have been derived from the detection of auroral lines (the data points constitute an incomplete collection drawn from various sources in the literature). The high-metallicity samples of Bresolin et al. (2004,2005) are data shown with enlarged symbols (full circles and triangles). The small full circles are data from the M101 sample studied by Ken-nicutt et al. (2003). The curves show the calibrations of the R23 indicator derived empirically by Pilyugin & Thuan (2005) and from photoionization-model grids by Kewley & Dopita (2002). In both cases two curves are drawn, corresponding to two different values of the ionization parameter.

log R23

Figure 17.2. This abundance versus R23 diagram shows observational data for extragalactic H11 regions, for which the O/H abundances have been derived from the detection of auroral lines (the data points constitute an incomplete collection drawn from various sources in the literature). The high-metallicity samples of Bresolin et al. (2004,2005) are data shown with enlarged symbols (full circles and triangles). The small full circles are data from the M101 sample studied by Ken-nicutt et al. (2003). The curves show the calibrations of the R23 indicator derived empirically by Pilyugin & Thuan (2005) and from photoionization-model grids by Kewley & Dopita (2002). In both cases two curves are drawn, corresponding to two different values of the ionization parameter.

as a representative example. This discrepancy amounts to approximately 0.3 dex, and its complete explanation still eludes us.

Was this article helpful?

0 0

Post a comment