Abstract

On the systematic errors of quantitative spectroscopy of OB stars

Carolina Sabín-Sanjulián (Instituto de Astrofísica de Canarias)

S. Simón-Díaz(1,2), M. García(1,2), N. Castro(3), A. Herrero(1,2) and J. Puls(4)

(1)Instituto de Astrofísica de Canarias,Tenerife,Spain

(2)Universidad de La Laguna,Tenerife,Spain

(3)National Observatory of Athens, Greece

(4)Universitäts-Sternwarte, Scheinerstrasse 1, D-81679 München, Germany

Our most powerful tool to study blue massive stars is quantitative spectroscopy, the comparison of synthetic spectra (calculated with an adequate code) and observations, to produce the parameters of the star from those of the best-fitting model.

However, our coverage of the parameter space is not continuous but discreet, and we rely on the assumption that small variations of certain parameters do not significantly alter our results. For some other parameters, we do not even have sufficient information to constrain their values.

The massive star team at the IAC has developed a chi-squared algorithm for the automatic analysis of O- and B- type stars. A key ingredient for the program is the vast grid of FASTWIND models, covering a wide range of effective temperatures, gravities and metallicities. The grid also includes models of varying wind-strength Q-parameter, beta exponent for the velocity law, microturbulence and He abundance. While we cannot usually constrain the latter set of parameters from typical optical spectra (because of poor spectral resolution, for instance), variations of these parameters or the adopted value can alter the final results from the analysis.

We use our grid and automatic fitting algorithm to evaluate the systematic effects and errors this may introduce in the analysis of OB stars. The method will be applied to Tarantula-FLAMES and IACOB OB-stars.

Mode of presentation: poster