The purpose of the computations presented in this paper was to sum up the diagnostic capabilities provided by observations of the profiles of the infrared triplet lines of ionized calcium. It has first been stressed that the local continuum level in the region of the Ca II IRT in the solar spectrum cannot be correctly described without taking into account the cumulative absorption by the hydrogen Paschen lines. As a consequence, the determination of the continuum of stellar observations in this region has to be carried out with special caution. It was then shown that the observed solar flux profiles of the LTE wings of the Ca II IRT can be well represented by computations based on two of the most popular empirical model solar photospheres (MACKKL and Holweger). Calculations based on theoretical solar photospheres proved significantly less successful. This confirms and illustrates what was already found and exploited by Smith, Drake and collaborators, i.e. that the wing profiles of the Ca II IRT provide a good test for the adequacy of a model temperature distribution for a given star. The values of the hydrogen damping constant of these lines leading to the best fits of the solar flux observations are consistent with the values determined by Smith & Drake (1988) from high quality solar intensity observations. It also turned out that the computed profiles are not very sensitive to the exact temperature variation law of the hydrogen damping constant which can very well be represented in the commonly used van der Waals approximation.
The differential sensitivity of the line profiles to the basic stellar atmospheric parameters was investigated through computations of the line with photospheres from the homogeneous GBEN grid of theoretical model atmospheres. The contribution of the Paschen lines cannot be ignored for dwarfs of effective temperature higher than 5800 K; for giant stars, it has to be taken into account down to 5500 K, or even cooler for the brightest ones. Otherwise, these computations generally confirm conclusions already reached by Smith & Drake (1987, 1990) and extended to cooler stars by Erdelyi-Mendes & Barbuy (1991). The wing profiles are remarkably insensitive to temperature for solar type dwarf stars with ; this property breaks off however for more luminous or metal-poor stars. Their behaviour with variations of the stellar gravity is not as simple as sometimes advocated: the sensitivity of the wing profiles to gravity is only effective for . By contrast, the variations with changes in the metallicity are much smoother and the depression in the wings of this line may be used as an indicator of the calcium abundance provided that the temperature and gravity of the star have been previously determined with reasonable accuracy and that the available observed spectra have a rather high signal-to-noise ratio. Relations such as those proposed by Diaz et al. (1989) or Zhou (1991) keep a statistical interest for studies of homogeneous stellar populations. However, for the determination of stellar parameters of individual stars, it is usually not difficult to find more sensitive spectroscopic indicators, especially for dwarf and subgiant stars.
The central depth of the chromospheric core or the line is a good indicator of the level of chromospheric activity of a star. A simple average relation has been found between it and the Mount-Wilson activity indicator . However, owing to the intrinsic variability of chromospheric activity, systematic studies of this phenomenon must ensure sufficient time sampling to cover such phenomena as rotational modulation (periods of the order of a month) or activity cycles (periods of the order of 10 years), which require great instrumental stability and reproductibility. In that respect, specialized spectrophotometers, such as the Mount-Wilson instrument, are best suited and much more efficient.
© European Southern Observatory (ESO) 2000
Online publication: December 17, 1999