In order to validate and calibrate automatic pipelines to derive stellar parameters from spectra, a
set of reference stars - benchmarks - are needed. This set should cover a wide range of parameters
(including dwarf and giant stars) and metallicities (including thin/thick disc and halo
The sample of benchmark stars must (1) have well defined reference stellar parameters, (2)
excellent data: high signal to noise spectra, high resolution and extended wavelength coverage.
(1) Reference Stellar Material
To satisfy condition (1), we need to have parameters that are as independent as possible from
spectroscopy. As such, our favourite candidates should have:
- ANGULAR DIAMETERS: Effective temperature can be derived from the Stephan Boltzmann relation for those stars for which we have a direct measurement of their angular diameter and their parallax.
- ASTEROSEISMOLOGY: Surface gravity can best be derived from Newton's law for those stars for which we have a mass measurement.
- SPECTROSCOPY: The best derivations of metallicity and elemental abundances come from analysis of high quality spectra. A combination of different methods is required in order to avoid biases towards a given method.
(2) Spectral Library
In order to link the parameters of different spectroscopic surveys, benchmark stars should have a
wavelength range that covers all the surveys.
Since very few stars satisfy the conditions of (1), we need to find compromises for collecting
the best candidates. There is a lot to do in this direction, such as new observations of angular
diameters, improved input physics for metallicity determinations and the inclusion of more seismic
targets for the determination of surface gravities.
Additionally, a library for (2) that would cover all spectroscopic surveys wavelength range is highly
desirable, although not available yet.
For a recent summary, see the article written for the "Reconstruction the Milky Way's History:
Spectroscopic surveys, Asteroseismology and Chemodynamical models" conference:
Papers on stellar parameters of benchmark stars:
I) Heiter et al 2015: http://adsabs.harvard.edu/abs/2015A%26A...582A..49H
II) Blanco-Cuaresma et al 2014: http://adsabs.harvard.edu/abs/2014A%26A...566A..98B
III) Jofre et al 2014: http://adsabs.harvard.edu/abs/2014A%26A...564A.133J
IV) Jofre et al 2015: http://adsabs.harvard.edu/abs/2015A%26A...582A..81J