Since most of the information we receive from outer space is carried by
stellar light, it comes as a necessity to properly characterise them.
With the advent of space missions, it becomes possible to do so, thanks
to the gathering of data of unprecedented quality. The past CoRoT and
Kepler missions provided the stellar scientists with a wealth of data
such that it allowed for the asteroseismology, the study of stellar
pulsations and their link with the stellar structure, to thrive. This
enabled to provide a detailed characterisation of distant solar-like
stars and to pinpoint the shortcomings of current models.
However, to provide precise inferences from observed oscillation
spectra, it is necessary to have methods which are able to take the most
advantage of the exquisite precision of the data. Both of the methods I
develop are tailor-made for such needs.
The first method, WhoSGlAd, accounts for the oscillations of
main-sequence solar-like stars and the acoustic glitches they may
exhibit. Acoustic glitches are an oscillating feature in the spectrum
caused by a sharp feature in the stellar structure. The adjustment is
done in such a way that the fitting parameters are completely
independent and the computations are extremely fast. The parameters are
then combined to build seismic indicators relevant of the stellar
structure as little correlated as possible. Those are then used as
constraints to stellar models.
The second method, EGGMiMoSA, aims at providing a precise adjustment of
the complex behaviour displayed by the mixed-modes oscillation spectra
of sugbiant and red giant stars. Mixed-modes constitute a unique
opportunity to probe the stellar interior from the surface to the core
of the star. Again, the objective of the method is to define seismic
indicators relevant of the stellar structure in order to constrain
stellar models.
During the present seminar, I will introduce both techniques and several
results obtained via their use.
Non-linear theory of pulsations could explain special characteristics observed in many variable star light curves, however, non-linear models are far from being complete. Ultra-precise photometric data captures the footprints of non-linear processes happening within a pulsating star, which are translated in their power spectra as combination frequencies (additions and subtraction of the eigenfrequencies). In this talk, I will present a study of combination frequencies in δ Scuti star power spectra, aiming to empirically characterize their non-linear behavior. Based on the study of frequency, amplitude and phase relations given by the Volterra series formulation (a general non-linear model), a self-consistent method to identify combination frequencies with high precision, a possible method to unambiguously identify the non-linear process behind a combination frequency, as well as the possible first use of combination frequencies for mode identification in δ Sct stars are, among others, the most relevant results.
The Kepler Space Telescope initiated a revolution by
detecting solar-like oscillations in more than 500 main-sequence and
subgiant stars. However, most Kepler stars are faint and
therefore have limited constraints from independent methods such as
long-term ground-based activity monitoring. In addition, population
studies of exoplanets orbiting seismic Kepler stars
yielded unprecedented insights for planet formation and evolution,
including the discovery of the so-called sub-Neptune desert. However, a
large fraction of Kepler planet candidates are still
awaiting confirmation and therefore, the population of planets orbiting
asteroseismic stars is still incomplete. Fortunately TESS is now
targeting stars in the nearby solar neighborhood, enabling the unique
opportunity to precisely characterize bright systems for which long-term
radial velocity time series are already available. In this talk, I will
present the TESS discovery of solar-like oscillations for naked-eye
solar-analogue ⍺
Mensae and discuss why this is a benchmark system for both stellar
astrophysics and exoplanet science. I will also discuss preliminary
results of new asteroseismic hosts from Kepler, K2 and TESS based on a systemic analysis of all available data.
Most of the international conferences being cancelled, we’re facing an issue: How can we make our PhD students known for their work? How can they start new collaborations, independently of their supervisors? How can they advertise for their papers? etc… So here’s the idea: what if we create a series of seminars, given by young researchers, dedicated to Asteroseismology? The Good Vibrations Initiative has already gathered 18 groups around the world and should launch its first event on Jan 6th 2021. This meeting aims at presenting the initiative and coordinating the contributions. If you’d like to take part, one way or another, feel free to join! Send an e-mail at: goodvibrations@obspm.fr.
Space-borne missions such as Kepler allow us to achieve a spectral resolution which is now sufficiently high to resolve the shape of the p-mode line profiles of solar-like stars, and in particular to measure the asymmetry they feature in the temporal Fourier domain. However, the physical origin of these asymmetries is still not fully understood, even for the Sun. In particular, there is no clear consensus to explain the asymmetry reversal between the velocity and intensity observables.
In this context, I will present an approach designed to better understand the physical origin of solar-like p-mode asymmetries, both in velocity and intensity. To that end, we model the spectral power density by convolving 1) the Green’s function associated to the oscillating mode, which we computed numerically using a 1D evolutionary stellar model, and 2) a source term, obtained by coupling an analytical turbulence model with CO5BOLD 3D simulations of the stellar atmosphere. I will show that we successfully reproduce the observed asymmetries in the case of the Sun, both in velocity and intensity. These results allow us to better understand the physical mechanisms pertaining to solar-like p-modes asymmetry, and go a long way towards explaining the asymmetry reversal puzzle.
The number of confirmed exoplanets orbiting evolved hosts is considerably small, compared to their dwarf counterparts. Of that already small sample, fewer still are known gas giants at close-in orbits. Nonetheless, finding and characterizing more of these systems provides a unique opportunity to not only test existing theories of giant planet formation and dynamical evolution, which does predict their low occurrence rate, but also to constrain other phenomena such as the radius inflation observed in “hot jupiters”.
To this end, I have worked to both improve the characterization of these systems and increase the sample of confirmed planets. To improve their characterization, I developed a tool to model both the transit signal and the stellar signals of the granulation and oscillations in the time domain simultaneously, using Gaussian Processes. The tool reduces the uncertainties in the recovered transit parameters, which, coupled with an asteroseismic characterization of the host star improves the precision of the determined planetary properties, and is also capable of determining properties of the stellar signals, such as granulation timescale and amplitude, as well as the frequency of maximum oscillation, with precision similar to analyses in the frequency domain. To increase the sample of confirmed planets, I am performing a search for planetary candidates in all bright (TESS mag < 10) low luminosity red-giant branch stars (LLRGBs) in the south sectors of the TESS mission. As of now, the sample of potential candidates found has been lower than expected by simulations, and might challenge estimates of the occurrence rate of these systems.
