Science Summaries

This page features summaries of selected recent publications, written for non-experts. They will highlight some of the range of research done by the group.

First ensemble seismology study of red giant stars with TESS

We perform an exploratory asteroseismic analyses of bright red giant stars in the TESS Southern Continuous Viewing Zone (SCVZ), and detect solar-like oscillations in about 6,500 stars. Based on this analysis, we predict that seismology will be attainable for ∼ 300,000 giants across the whole sky!
We also show that TESS CVZ-data make possible studies of, for example, stars in the Asymptotic Giant Branch bump, and the detection of mixed l = 1 modes and rotational splitting in the 1-year datasets. Finally, by combining TESS-CVZ data with spectroscopic constraints and with Gaia astrometry, we demonstrate TESS' strong potential for Galactic archaeology studies, making this data ideal for studies of the local star formation history and evolution of the Galactic disc.

A treasure trove (and some caveats): combining astrometric, spectroscopic and asteroseismic constraints of red giant stars in the Kepler field

What can we learn about the evolution of stars and the Milky Way when we have available unprecedented observational constraints? This paper explores relations between mass, age, evolutionary state, kinematics and surface chemical composition of about 5,000 red giant stars observed by the Kepler mission. Here is a list of some of the (quite varied) questions we address in this work:

  • quantify the intrinsic age spread of one of the oldest populations in the Galaxy,
  • measure the (integrated) mass lost by stars during the red-giant-branch phase,
  • evidence for stars that are likely products of coalescence or mass transfer,
  • different age-velocity-dispersion relations for the chemically defined thin and thick disks,
  • discussion about some of the biases stemming from stellar models and data analyses, and how to potentially address them.

Left: the inferred age distibutions of about 400 stars in the so-called thick disk of the Milky Way . The red solid line shows the intrinsic age distribution of the main population, while the dashed blue line represents contaminants, likely products of mass exchange / coalescence. Right: the inferred vertical-velocity-dispersion to age relation for stars in the low- and high-[α/Fe] populations.

Towards a more realistic depiction of the near-surface layers of stars

Nowadays, the determination of stellar properties often relies on comparisons between observations and the predictions of detailed numerical simulations. However, some aspects of stellar structures are notoriously hard to model. One prominent example is the atmospheres of the Sun and of other stars with convective envelopes. For such stars, state-of-the-art stellar models fail to recover the correct structure because they build on a set of simplifying approximations. In this paper, we go beyond these approximations and show that a more physically accurate depiction of stellar convection changes the picture. Although we only improve upon the description of a small fraction of the stellar structure, our improvements have a demonstrable impact on the global properties of stellar models. In the modern era of high-quality astronomical data, accurate stellar models are key to a better understanding of stars, planets, and galactic dynamics. Our results thus emphasize the need to employ all relevant physics when attempting to model stars, and our model improvements constitute one step towards the next generation of stellar models.

The figure above shows a comparison between the global properties of standard stellar models and our improved models, by illustrating the impact of our model improvements on the predicted stellar age and metallicity. On average, our model improvements shift the age estimates by about 10 per cent. If we were to conclude on the properties of stellar populations, systematic errors of the order of 10 per cent could be detrimental.

TESS dates an ancient collision with our galaxy

A single bright star in the constellation of Indus, visible from the southern hemisphere, has revealed new insights on an ancient collision that our galaxy the Milky Way underwent with another smaller galaxy called Gaia-Enceladus early in its history. The star was aged using its natural oscillations (asteroseismology), detected in data collected by NASA's recently launched Transiting Exoplanet Survey Satellite (TESS). When combined with data from the European Space Agency (ESA) Gaia Mission, the detective story revealed that this ancient star was born early in the life of the Milky Way, 11.5 billion years ago. The star bears hallmarks consistent with having been kinematically heated by the Gaia–Enceladus collision. Its age implies that the earliest the merger could have begun was 11.6 and 13.2 billion years ago, at 68% and 95% confidence, respectively.

The animation here above shows a schematic of the orbits of three stars within our Galaxy the Milky Way – whose extent is marked by the dashed line -- as viewed from above (left) and side-on (right), projected over the next half a billion years. The three stars are: ν Indi, the subject of the paper; a star accreted from Gaia-Enceladus (after its collision with the Milky Way); and the Sun. The accreted star came from outside the Galaxy, as part of the Gaia-Enceladus collision, and so has a very elongated orbit. The collision affected the motion and orbit of ν Indi. Notice how its orbit is quite different to that of the Sun.

    Media / press coverage:
  • Read here the full press release from the University of Birmingham

The stellar mass of the Milky Way halo and its components

In this paper, we use data from the APOGEE survey and Gaia to measure the mass in stars with different element abundances and eccentricity. By doing this, we place an estimate on the mass of the so-called Gaia-Sausage, showing that it is slightly less massive than previous expectations, but that the halo still has a lot of extra (likely) accreted mass. The stellar mass of the halo is difficult to pin down, but because of the extent of the APOGEE survey data both spatially and in element abundances, we were able to make this new constraint - a first step toward a complete map of our Galaxy's history of assembly.

HAYDN: high-precision AsteroseismologY in DeNse stellar field

In response to the ESA Voyage 2050 long-term plan proposal, we have put forward the concept of a new small/medium class space mission, which will improve our understanding of stellar astrophysics and the evolution of clusters, and elucidate the origins of the Milky Way bulge and nearby dwarf galaxies.

The core principle of HAYDN rests on the necessity for long period, high cadence photometry in dense and distant stellar systems, such as globular cluster, local dwarf galaxies and the Milky Way bulge.

Have a look at the HAYDN webpages and at the talk given at the ESA Voyage 2050 worshop for more info!

The vertical structure of the disc: evidence for a bimodal age distribution

Using data from the K2 mission, we are able, for the first time, to use asteroseismology to probe the age distribution of stars at various heights from the Galactic plane, well beyond the distances reached by e.g. Kepler or CoRoT.
As the distance from the Galactic plane increases, the stellar population becomes dominated by older stars. The changing distributions not only lend support to the theories indicating the thick disc is older than the thin disc, but reaffirm the desire for precise ages to allow for the confirmation of any possible epoch of quiescence in star formation and age bimodality associated with chemistry. A strong bimodality was also observed in the age distribution for both Kepler and the K2 fields (see figure below), with distinct young and old population. Clear associations with height from the plane and [α/Fe] were attributed to each peak: 5 Gyr - low-α, |Z| ≤ 1.0 kpc (thin disc); 14 Gyr - high-α, |Z| > 1.0 kpc (thick disc). The chemical age dichotomy was also confirmed with the Kepler sample, where the peak at 12 Gyr is due to the α-rich population. Each sample presented contains a minimum, suggestive of a time delay between the formation of these populations.

Normalised age distributions for the APOKASC and K2 populations. Left: Nominal (grey), α-rich (red) and α-poor ([α/Fe]< 0.1; blue dashed) APOKASC sample age distributions. Middle: K2HQ complete (blue), |Z| < 1.0 kpc (green) and |Z| > 1.0 kpc (purple) distributions. Right: K2 Spec.HQ (orange) and K2-APOGEE α-rich (red) and α-poor ([α/Fe]< 0.1; blue dashed) distributions.

Hertzsprung-Russell Diagrams from astrometric+asteroseismic+spectroscopic data

One of the main challenges of Galactic archaeology is to reveal the Galaxy assembly and evolution history via the age, chemical composition, and kinematics of stars in a large fraction of the volume of the Milky Way. In particular, asteroseismology provides us with the crucial chronological information: solar-like oscillating giants are unique evolutionary clocks owing to the availability of seismic constraints on their mass and to the tight age-initial mass relation they adhere to. Even so, a key information for a better understanding of stellar structure within red giants is still missing: their luminosity. This is where Gaia DR2 comes into play, providing distances for over 1.3 billion sources - which includes red-giant stars observed by Kepler. The synergy between Kepler, Gaia DR2, and spectroscopic surveys offers unprecedented possibilities for stress-testing models of stellar evolution unravelling long-standing problems of stellar evolution and, looking at the bigger picture, for helping disentangle the multidimensional problem of Galaxy assembly.

Hertzsprung-Russell Diagrams of red giants observed by Kepler. The K-band absolute magnitude is inferred using Gaia DR2 parallaxes and the parallax zero-point measured in the Kepler field by Khan et al. (2019). Spectroscopic constraints from APOGEE are used to provide information on the effective temperature and chemical composition. Finally, asteroseismology offers precise and accurate ages for these stars (Rodrigues et al. 2017), leading to a very informative Hertzsprung-Russell Diagram and allowing us to create cluster-like sequences of field stars. In the leftmost panel, several red-giant structures are indicated: the red-giant branch bump (RGBb; e.g. see Cassisi 2012), the red clump (RC), the secondary red clump (SRC), and the vertical red clump (VRC; e.g. see Girardi 2016). The other panels first show the selection in age, and then the effect of varying the metallicity.

Mean density inversions for red giants and red clump stars

Accurately weighing red giants is paramount for Galactic archaeology, as these stars act as the standard clocks and rulers to unravel the history of the Milky Way. Currently, the excellent data of the CoRoT and Kepler missions allows asteroseismology to fulfill this role for thousands of stars. More high-quality data can be expected in the coming years thanks to the TESS and PLATO missions. In this study, we adapt and test advanced seismic analysis techniques which had been developed for the main-sequence phase (core hydrogen burning stars) on red-giant stars. We show that these techniques can provide very accurate values of the mean density of red giants, beyond what is achievable by current approaches. We demonstrate the accuracy of our modelling approach by applying it to a sample of eclipsing binaries. The enclosed figure demonstrates that we obtain masses in all cases in good agreement with those determined from eclipses, where other seismic approaches failed to do so. Moreover, the robustness of our approach, which couples seismic inversions to the Bayesian AIMS modelling software, is well-suited for automated pipelines and thus the analyses of large samples of red giants required for Galactic archaeology. Coupled with the recent Gaia data release, our method leads to mass determinations of red giants below 10% in accuracy, reducing significantly the uncertainties on stellar ages and paving the way for in-depth Galactic archaeology.

Masses and radii for a subsample of eclipsing binaries determined from the eclipses and from various seismic methods.