Programm Potsdam-Berlin-Kolloquium, 11.10.2019
Brian Chaboyer (Dartmouth College)
| Wann | Am 23.06.2016 von 14:30 bis 15:30 |
|---|---|
| Was |
|
| Wo | SH Lecture Hall |
| Termin übernehmen | vCal / iCal |
Stellar evolution models are widely used to interpret observations of starlight. There is an increasing interest in the first stars which were formed in the universe, and interpretation of these observations requires accurate models of metal-poor stars. The reliability of the derived properties of stars (such as their age or intrinsic luminosity) depends upon the reliability of stellar models to accurate reproduce the conditions which occur within stars. There are a wealth of observations which can used to test the validity of solar type stars, but until recently there were few tests of the reliability of metal-poor stellar models. In this talk, I will discuss two projects which demonstrate the limitations of current stellar evolution models of metal-poor stars.
We have obtained HST FGS parallaxes for 8 metal-poor ([Fe/H] < -1.5) main sequence stars which are relatively close to the Sun. These parallaxes, with a typically accuracy of 100 microarcseconds, are used to determine the absolute magnitude of these main sequence stars and compared to Dartmouth stellar evolution models. As a star evolves up the red giant branch, the convection zone rapidly deepens, and then recedes towards the surface (in mass coordinates). This receding convection zone is predicted to leave behind a discontinuity in the hydrogen abundance, which slows down the evolution of the star, leading to an increase in stellar number counts at the red giant branch bump. This increase in stellar number counts (the red giant branch 'bump') has been observed in HST observations of globular clusters, and these observations are compared to the predictions from the Dartmouth stellar evolution models. This comparison suggests that convective overshoot becomes increasingly important in low-metallicity stars.
