Search for: All records

Abstract Seismic structure inversions have been used to study the solar interior for decades. With the high-precision frequencies obtained using data from the Kepler mission, it has now become possible to study other solar-like oscillators using structure inversions, including both main-sequence and subgiant stars. Subgiant stars are particularly interesting because they exhibit modes of mixed acoustic-buoyancy nature, which provide the opportunity to probe the deeper region of stellar cores. This work examines whether the structure inversion techniques developed for the pure acoustic modes of the Sun and other main-sequence stars are still valid for mixed modes observed in subgiant stars. We construct two grids of models: one of main-sequence stars and one of early subgiant stars. Using these grids, we examine two different parts of the inversion procedure. First, we examine what we call the “kernel errors,” which measure how well the mode sensitivity functions can recover known frequency differences between two models. Second, we test how these kernel errors affect the ability of an inversion to infer known structure differences. On the main sequence, we find that reliable structure inversion results can be obtained across the entire range of masses and large frequency separations we consider. On the subgiant branch, however, the rapid evolution of mixed modes leads to large kernel errors and hence difficulty recovering known structure differences. Our tests show that using mixed modes to infer the structure of subgiant stars reliably will require improvements to current fitting approaches and modifications to the structure inversion techniques. 

Abstract Asteroseismic inferences of main-sequence solar-like oscillators often rely on best-fit models. However, these models cannot fully reproduce the observed mode frequencies, suggesting that the internal structure of the model does not fully match that of the star. Asteroseismic structure inversions provide a way to test the interior of our stellar models. Recently, structure inversion techniques were used to study 12 stars with radiative cores. In this work, we extend that analysis to 43 main-sequence stars with convective cores observed by Kepler to look for differences in the sound speed profiles in the inner 30% of the star by radius. For around half of our stars, the structure inversions show that our models reproduce the internal structure of the star, where the inversions are sensitive, within the observational uncertainties. For the stars where our inversions reveal significant differences, we find cases where our model sound speed is too high and cases where our model sound speed is too low. We use the star with the most significant differences to explore several changes to the physics of our model in an attempt to resolve the inferred differences. These changes include using a different overshoot prescription and including the effects of diffusion, gravitational settling, and radiative levitation. We find that the resulting changes to the model structure are too small to resolve the differences shown in our inversions. 

Abstract Understanding the ages of stars is crucial for unraveling the formation history and evolution of our Galaxy. Traditional methods for estimating stellar ages from spectroscopic data often struggle with providing appropriate uncertainty estimations and are severely constrained by the parameter space. In this work, we introduce a new approach using normalizing flows—a type of deep generative model—to estimate stellar ages for evolved stars with improved accuracy and robust uncertainty characterization. The model is trained on stellar masses for evolved stars derived from asteroseismology and predicts the relationship between the carbon and nitrogen abundances of a given star and its age. Unlike standard neural network techniques, normalizing flows enable the recovery of full likelihood distributions for individual stellar ages, offering a richer and more informative perspective on uncertainties. Our method yields age estimations for 378,720 evolved stars and achieves a typical absolute age uncertainty of approximately 2 Gyr. By intrinsically accounting for the coverage and density of the training data, our model ensures that the resulting uncertainties reflect both the inherent noise in the data and the completeness of the sampled parameter space. Applying this method to data from the fifth-generation Sloan Digital Sky Survey Milky Way Mapper, we have produced the largest stellar age catalog for evolved stars to date. 

The vast majority of Milky Way stellar halo stars were likely accreted from a small number (<~3) of relatively large dwarf galaxy accretion events. However, the timing of these events is poorly constrained and predominantly relies on indirect dynamical mixing arguments or imprecise age measurements of stars associated with debris structures. Here, we aim to infer robust stellar ages for stars associated with galactic substructures to more directly constrain the merger history of the Galaxy. By combining kinematic, asteroseismic, and spectroscopic data where available, we infer stellar ages for a sample of 10 red giant stars that were kinematically selected to be within the stellar halo, a subset of which are associated with the Gaia–Enceladus–Sausage halo substructure, and compare their ages to 3 red giant stars in the Galactic disk. Despite systematic differences in both absolute and relative ages determined here, age rankings of stars in this sample are robust. Passing the same observable inputs to multiple stellar age determination packages, we measure a weighted average age for the Gaia–Enceladus–Sausage stars in our sample of 8+/-3 (stat.)+/-1 (sys.) Gyr. We also determine hierarchical ages using isochrones for the populations of Gaia–Enceladus–Sausage, in situ halo and disk stars, finding a Gaia–Enceladus–Sausage population age of 8.0+2.3-3.2 Gyr. Although we cannot distinguish hierarchical population ages of halo or disk structures with our limited data and sample of stars, this framework should allow a distinct characterization of Galactic substructures using larger stellar samples and additional data available in the near future