lundi 18 février 2019, par Martin Farnir (University of Liège - STAR Institute, Belgium)
Vendredi 22 février 2019 à 14h00 , Lieu : Salle de réunion du bâtiment 14
Most of the information we gather about the Universe emanates from the stellar light. A good understanding of stars is therefore needed to, for instance, trace back the evolution of our Galaxy or of exoplanetary systems. In the last decades, the launch of the space-borne missions CoRoT (2006-2014) and Kepler (2009-2018) enabled a revolution in stellar physics. They provided us with photometric information on thousands of stars with an unprecedented quality. The amount and precision of the data allowed a direct probing of the internal properties of stars through the study of their oscillations, or asteroseismology. This additional piece of information leads to stringent constraints on stellar evolution that were not accessible with ‘classical’ techniques.
In this talk, I will present my work about the study of acoustic glitches in main-sequence stars. Acoustic glitches represent the small signature in the oscillation frequency pattern resulting from sharp gradients in the stellar structure. Those glitches are essential to our understanding of stellar structure and evolution as, for example, they provide information about the surface helium content in low-mass stars or on the extent of the mixed regions, inaccessible by other means. In order to analyse such signatures and provide significant inferences, methods as precise and accurate as possible are necessary. However, previous works interested in the information carried by acoustic glitches focused only on those signatures and neglected information contained in the oscillation spectrum as a whole. The resulting parameters characterising the stellar structure were correlated with rather important error-bars. Thus, in this presentation, I will introduce a new stellar seismic probing method that we developed, called WhoSGlAd (Whole Spectrum and Glitches Adjustement) which analyses in a comprehensive way stellar oscillation spectra. This leads to smaller error-bars and thus stricter constraints. Our method is precise and fast, and may be used with any minimisation scheme to find best fit models representing seismic data. This provides precise inferences about their structure and access to the surface helium content. As the TESS mission has been launched and PLATO will follow, providing methods able to fully benefit from the wealth of exquisite quality data while reducing at most the uncertainties of the inferences is in order. Therefore, WhoSGlAd should reveal to be of great help in characterising the targets of both missions.