The physical structure of Magellanic Cloud HII regions

Abstract

Abstract. Based on a new data set of optical and infrared spectra described in Vermeij et al. (2001), we analyse the gasphase elemental abundances of a sample of H  regions in the Large and Small Magellanic Cloud. The combined optical and infrared data set gives us access to all the ionization stages of astrophysically important elements such as sulfur and oxygen. We self-consistently determine the electron temperatures and densities for the O + , S ++ and O ++ ionization zones, and use these parameters in the derivation of the ionic fractions. We discuss the uncertainties on these ionic fractions. The different relations between the electron temperatures as proposed by Garnett (1992) and Thuan et al. (1995) are confronted with our results. We find our electron temperatures to be consistent with these relations, although the relation between T e [S ] and T e [O ] might be slightly steeper than predicted. We investigate the reliability of the Ionization Correction Factors (ICFs) used in the derivation of the full elemental abundances of sulfur and neon. We conclude that the prescription for the ICF used to derive the sulfur abundance as given by Stasińska (1978) for α = 3 is accurate for O + /O > 0.20. No conclusions could be drawn for neon. Avoiding the use of ICFs as much as possible, we then proceed to derive the full elemental abundances. We calculate a grid of general photoionization models to compare our results with the "bright-line" abundance diagnostics for oxygen (R23) and sulfur (S23(4)). The reliability of the newly proposed S234 parameter (Oey & Shields 2000) which includes emission lines from S + , S ++ and S +3 is checked. We find a very good agreement between the S234 models and our analysis results. Finally, we compare the heavy element-to-oxygen ratios of our sample objects to those of giant H  regions in a large sample of lowmetallicity blue dwarf galaxies (Izotov & Thuan 1999) and with the results from Kobulnicky & Skillman (1996 for the irregular galaxies NGC 1569 and NGC 4214.

Nitrogen line spectroscopy of O-stars

“…around Y He = 0.08−0.094 corresponding to [He] = 10.90−10.97 (Russell & Dopita 1990;Maeder & Meynet 2001;Vermeij & van der Hulst 2002;Peimbert 2003;Tsamis et al 2003), and agrees quite well with our minimum values for the derived helium abundance. We found only five stars with considerably enriched nitrogen close to this value, three (super-)giants and two dwarfs, but note also the attributed uncertainty in helium content.…”

Chemical Composition and Mixing in Giant HiiRegions: NGC 3603, 30 Doradus, and N66

“…Models of Tsamis & Péquignot (2005) for 30 Dor predict that $92% of sulfur is in S iii and S iv, while 8% is due to other ionization stages, mostly S ii. Peimbert (2003) and Vermeij & van der Hulst (2002) find consistent results observationaly, with S ii/S iii ranging from 3% to 8.5%. In N66, the contribution of S ii is approximately 15% that of S iii (Peimbert et al 2000;Vermeij & van der Hulst 2002).…”

“…Analyzing 15 positions, we find a comparable dispersion, 0.11 dex, from the minimum to the maximum Ne/H values. In the same object, Vermeij & van der Hulst (2002) find sulfur abundances ranging from 6.63 to 6.86 and argon abundances ranging from 6.18 to 6.63, which is a smaller dispersion than we measured (but with fewer positions). Our results in NGC 3603 and N66 also imply that the neon abundance and, to a lesser extent, the sulfur abundance show little dispersion in these regions.…”

“…Finally, we find (Ne iii /H)/(S iii/H) ¼ 9:8 AE 2:0 in N66, compared to 10.3 in the optical (Vermeij & van der Hulst 2002). Hence the MIR spectra seem to probe ionized gas with a degree of ionization equal to or lower than that of the gas probed in the optical.…”

“…16). Vermeij & van der Hulst (2002) analyzed four positions in 30 Dor and found neon abundances ranging between 7.84 and 7.94. Analyzing 15 positions, we find a comparable dispersion, 0.11 dex, from the minimum to the maximum Ne/H values.…”