Abundance analysis from absorption lines

It is very useful to complement the compositional analysis of stars by a like analysis of the interstellar medium. This can be done by making use of absorption lines which the latter removes from the UV spectrum of hot, bright stars (Fig. 8.8). Measured abundances only concern gases lying between the source star and the observer. Matter contained in dust grains escapes detection. 

Thus, in the galaxies observed, oxygen could be as abundant as in the interstellar medium near the Sun or as low as 1/10 solar. When the R23 method is applied to cB58, a similar ambiguity obtains (Teplitz et al. 2000). The results from the analysis of the interstellar absorption lines described above ( 4.2) resolve the issue by showing that the upper branch solution is favoured (we have no reason to suspect that the neutral and ionised ISM have widely different abundances). It remains to be established whether this is also the case for other LBGs. 

Another important source for potential systematic uncertainty stems from the fact that the Li abundance is not directly observed but rather, inferred from an absorption line strength and a model stellar atmosphere. Its determination depends on a set of physical parameters and a model-dependent analysis of a stellar spectrum. Among these parameters, are the metallicity characterized by the iron abundance (though this is a small effect), the surface gravity which for hot stars can lead to an underestimate of up to 0.09 dex if log g is overestimated by 0.5, though this effect is negligible in cooler stars. Typical uncertainties in log g are 0.1 0.3. The most important source for error is the surface... 

Ayres et al. ([42] ) derived the O abundance from weak CO absorptions but did not find support for a lower O abundance. They recommend A(O) = 8.85, much closer to the value in [9] (A(O) = 8.93). However, to obtain the absolute O abundance using the CO molecule, Ayres et al. assumed that C/O = 0.5. Oxygen is about twice as abundant as carbon, and carbon is largely tied into the CO molecule. An analysis of the CO abundance thus provides only the lower limit to the total C abundance (as C is present also in other gases such as C, CH etc), and the O abundance can only be derived if the total C/O ratio and the C abundance is known. [42] find A(C) =8.54, and with an assumed C/O of 0.5, their corresponding O abundance is A(O) = 8.85. However, using the same CO lines [37] find a lower O abundance with their 3D models than reported by [42]. Given the problem that the O abundances from CO requires various assumptions about the distribution of C and the C/O ratio, these abundance determinations appear even more uncertain. 

---