Determining the effective temperature of cool stars from near-infrared spectra

Sammanfattning: Context. Effective temperature is one of the fundamental stellar parameters. A traditional spectroscopic way to determine it is from optical spectra using excitation balance of Fe-lines. With technological advances and the advent of the next generation telescopes there will be an emphasis on the relatively unexplored near-IR (NIR) wavelength region (1-5 μm). A method to determine the effective temperature based on NIR spectra is needed. Aims. In a proactive attempt, this thesis explores the NIR for the potential of such a method. Reference iron-linelists are created and optimised for the determination of stellar parameters of cool stars (mainly giants). The K-giant benchmark star, Arcturus (alpha-Boo), is used as a testbed for the analysis. Methods. An IDL script is written to select Fe-lines for stellar parameter diagnostics, with constraints based on blend-percentage and line-strength. Empirical line-strengths (astrophysical gf-values) of evaluated Fe-lines are solved against the solar spectrum using 1D-LTE MARCS model atmospheres and VALD to define atomic data. These are used as input to generate synthetic spectra of Arcturus which are confronted with high-resolution observations. The final Fe-linelists are used to retrieve the fundamental stellar parameters of Arcturus using SME for spectrum synthesis. Results. The derived gf-values follow a one-to-one correspondence with the available laboratory measurements. Theoretically calculated values show a larger spread, indicative of configuration interaction and level-mixing. With an independently determined surface gravity, the effective temperature of Arcturus can be retrieved using approximately 20 lines to a precision of ±50 K. The Fe-linelists constructed for the J- and H-bands each span excitation potentials of the same range as those constructed for the optical spectral region used in other studies. Conclusions. The results indicate that the NIR has a large potential for the determination of effective temperature and metallicity. The absence of observed Fe II lines suggests that surface gravity must be determined by an independent method, or alternatively using pressure-sensitive wings of strong lines in other atomic species.