Astron. Astrophys. 364, 102-136 (2000)

6. The model atmosphere analysis

The high-resolution spectra were analyzed with the model atmosphere technique. Stellar parameters were estimated from Strömgren photometry, from spectrophotometry, from H profiles, from IUE ultraviolet colours and (for nine stars) from () colours. The estimates obtained by these different methods and the values of the parameters that we adopted are given in Table 7.

Table 7. Comparison of the stellar parameters obtained by different methods.

Table 7. (continued)

Table 7. (continued)

Table 7. (continued).
Notes:
 If   8000 K, H is almost independent of the gravity and  is then given in parentheses. The gravities used to derive  from the ,  & () grid are also given in parentheses and were not used to obtain the mean gravity.
) RR Lyrae variable at phase 0.42 (Sect. 10.7).
) The quoted errors in these  correspond to the range in  between Columns 6 & 8 in Table 6.
1)  taken from SFD map (Table 6, Column 6).
2)  derived from Strömgren colours using the Moon (1985) code (Table 6, Column 8).
3)  is the mean of the values given in Columns 6 & 8 of Table 6.
4)  was adjusted to obtain the best fit between the model and the observed Energy Distribution.
5)  was derived by comparing the observed and theoretical () colours for the ,  & [M/H] shown.
6)  was derived by comparing the observed and theoretical () vs.  for the ,  & [M/H] shown.

Having fixed the model atmosphere, we computed abundances from both the equivalent widths and line profiles for each star observed at KPNO, except for BD +00 0145 and HD 14829. For these two stars and for the BHB stars that were observed at ESO (whose spectra only extended over 50 Å) we estimated the abundances from the equivalent widths alone.

We used model atmospheres and fluxes that were computed by Castelli with an updated version of the ATLAS9 code (Kurucz 1993a). We adopted models for stars with an -element enhancement []=+0.4 (which is generally appropriate for halo stars). The symbol "a" near the metallicity in Column 6 of Table 7 indicates when these models were used. The convective models ( 8750 K) were calculated with the no-overshooting approximation. More details about these models can be found in Castelli et al. (1997) and in Castelli (1999). The synthetic grids of Strömgren indices, Johnson () indices and H profiles were also computed by Castelli from the above models. Grids of models, fluxes and colours are available either at the Kurucz website (http://kurucz.harvard.edu ) or upon request.

We derived abundances from the equivalent widths using the WIDTH code (Kurucz 1993a), modified so that we could derive the Mg abundance from the measured equivalent width of the doublet Mg II  4481 Å. The SYNTHE code (Kurucz 1993b), together with the atomic line lists from Kurucz & Bell (1995), were used to compute the synthetic spectra.

© European Southern Observatory (ESO) 2000

Online publication: December 15, 2000
helpdesk.link@springer.de