We present a detailed comparison of the ionizing spectral energy distributions (SEDs) predicted by four modern stellar atmosphere codes, TLUSTY, CMFGEN, WMbasic, and FASTWIND. We consider three sets of stellar parameters representing a late O-type dw
arf (O9.5 V), a mid O-type (O7 V) dwarf, and an early O-type dwarf (O5.5 V). We explore two different possibilities for such a comparison, following what we called evolutionary and observational approaches: in the evolutionary approach one compares the SEDs of stars defined by the same values of Teff and logg; in the observational approach the models to be compared do not necessarily have the same Teff and logg, but produce similar H and HeI-II optical lines. We find that there is a better agreement, in terms of Q(H0), the ratio Q(He0)/Q(H0), and the shape of the SEDs predicted by the four codes in the spectral range between 13 and 30 eV, when models are compared following the observational approach. However, even in this case, large differences are found at higher energies. We then discuss how the differences in the SEDs may affect the overall properties of surrounding nebulae in terms of temperature and ionization structure. We find that the effect over the nebular temperature is not larger than 300-350 K. Contrarily, the different SEDs produce significantly different nebular ionization structures. This will lead to important consequences on the establishment of the ionization correction factors that are used in the abundance determination of HII regions, as well as in the characterization of the ionizing stellar population from nebular line ratios.
We present some results of an on-going project aimed at studying a sample of Galactic HII regions ionized by a single massive star to test the predictions of modern generation stellar atmosphere codes in the H Lyman continuum. The observations collec
ted for this study comprise the optical spectra of the corresponding ionizing stars, along with imaging and long-slit spatially resolved nebular observations. The analysis of the stellar spectra allows to obtain the stellar parameters of the ionizing star, while the nebular observations provide constraints on the nebular abundances and gas distribution. All this information is then used to construct tailored photoionization models of the HII regions. The reliability of the stellar ionizing fluxes is hence tested by comparing the photoionization model results with the observations in terms of the spatial variation across the nebula of an appropriate set of nebular line ratios.
