An official website of the United States government
ABSTRACT Age is a stellar parameter that is both fundamental and difficult to determine. Among middle-aged M dwarfs, the most prolific hosts of close-in and detectable exoplanets, gyrochronology is the most promising method to assign ages, but requires calibration by rotation-temperature sequences (gyrochrones) in clusters of known ages. We curated a catalogue of 249 late K- and M-type (Teff = 3200–4200 K) exoplanet host stars with established rotation periods, and applied empirical, temperature-dependent rotation–age relations based on relevant published gyrochrones, including one derived from observations of the 4-Gyr-old open cluster M67. We estimated ages for 227 of these stars, and upper limits for eight others, excluding 14 which are too rapidly rotating or are otherwise outside the valid parameter range of our gyrochronology. We estimated uncertainties based on observed scatter in rotation periods in young clusters, error in the gyrochrones, and uncertainties in temperature and non-solar metallicity. For those stars with measured metallicities, we provide but do not incorporate a correction for the effects of deviation from solar-metallicity. The age distribution of our sample declines to near zero at 10 Gyr, the age of the Galactic disc, with the handful of outliers explainable by large uncertainties. Continued addition or extension of cluster rotation sequences to more thoroughly calibrate the gyrochronology in time and temperature space, more precise and robust measurement of rotation periods, and more accurate stellar parameter measurements will enable continued improvements in the age estimates of these important exoplanet host stars.
more »
« less
This content will become publicly available on July 3, 2026
StarFlow: Leveraging Normalizing Flows for Stellar Age Estimation in SDSS-V DR19
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.
more »
« less
- Award ID(s):
- 2202135
- PAR ID:
- 10648559
- Publisher / Repository:
- The Astronomical Journal, Volume 170, Issue 2, id.66, 20 pp.
- Date Published:
- Journal Name:
- The Astronomical Journal
- Volume:
- 170
- Issue:
- 2
- ISSN:
- 0004-6256
- Page Range / eLocation ID:
- 66
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract Globular clusters (GCs) provide statistically significant coeval populations of stars spanning various evolutionary stages, allowing robust constraints on stellar evolution model parameters and ages. We analyze eight old Milky Way GCs with metallicities between [Fe/H] = −2.31 and −0.77 by comparing theoretical isochrone sets from the Dartmouth Stellar Evolution Program to Hubble Space Telescope (HST) observations. The theoretical isochrones include uncertainties introduced by 21 stellar evolution parameters such as convective mixing, opacity, diffusion, and nuclear reactions, capturing much of the quantifiable physics used in our code. For each isochrone, we construct simulated color–magnitude diagrams (CMDs) near the main-sequence turnoff region and apply two full-CMD-fitting methods to fit HST Advanced Camera for Surveys data across a range of distances and reddening and measure the absolute age of each GC from the resulting posterior distribution, which accounts for uncertainties in the stellar models, observations, and fitting method. The resulting best-fitting absolute ages range from ≈11.5 to 13.5 Gyr, with a typical error of 0.5–0.75 Gyr; the data show a clear trend toward older ages at lower metallicities. Notably, distance and reddening account for over 50% of the uncertainty in age determination in each case, with metallicity,αabundance, mixing length, and helium diffusion being the most important stellar physics parameters for the error budget. We also provide an absolute age–metallicity relation for Milky Way GCs.more » « less
-
Abstract Gyrochronology, the field of age dating stars using mainly their rotation periods and masses, is ideal for inferring the ages of individual main-sequence stars. However, due to the lack of physical understanding of the complex magnetic fields in stars, gyrochronology relies heavily on empirical calibrations that require consistent and reliable stellar age measurements across a wide range of periods and masses. In this paper, we obtain a sample of consistent ages using the gyro-kinematic age-dating method, a technique to calculate the kinematics ages of stars. Using a Gaussian process model conditioned on ages from this sample (∼1–14 Gyr) and known clusters (0.67–3.8 Gyr), we calibrate the first empirical gyrochronology relation that is capable of inferring ages for single, main-sequence stars between 0.67 and 14 Gyr. Cross-validating and testing results suggest our model can infer cluster and asteroseismic ages with an average uncertainty of just over 1 Gyr, and the inferred ages for wide binaries agree within 0.83 Gyr. With this model, we obtain gyrochronology ages for ∼100,000 stars within 1.5 kpc of the Sun with period measurements from Kepler and Zwicky Transient Facility and 384 unique planet host stars. A simple code is provided to infer gyrochronology ages of stars with temperature and period measurements.more » « less
-
Abstract Age is the most difficult fundamental stellar parameter to infer for isolated stars. While isochrone-based ages are in general imprecise for both main-sequence dwarfs and red giants, precise isochrone-based ages can be obtained for stars on the subgiant branch transitioning from core to shell hydrogen burning. We synthesize Gaia Data Release 3–based distance inferences, multiwavelength photometry from the ultraviolet to the mid-infrared, and three-dimensional extinction maps to construct a sample of 289,759 solar-metallicity stars amenable to accurate, precise, and physically self-consistent age inferences. Using subgiants in the solar-metallicity open clusters NGC 2682 (i.e., M67) and NGC 188, we show that our approach yields accurate and physically self-consistent ages and metallicities with median statistical precisions of 8% and 0.06 dex, respectively. The inclusion of systematic uncertainties resulting from nonsingle or variable stars results in age and metallicity precisions of 9% and 0.12 dex, respectively. We supplement this solar-metallicity sample with an additional 112,062 metal-poor subgiants, including over 3000 stars with [Fe/H] ≲ −1.50, 7% age precisions, and apparent GaiaG-band magnitudesG< 14. We further demonstrate that our inferred metallicities agree with those produced by multiplexed spectroscopic surveys. As an example of the scientific potential of this catalog, we show that the solar neighborhood star formation history has three components at ([Fe/H],τ/Gyr) ≈ (+0.0, 4), (+0.2, 7), and a roughly linear sequence in age–metallicity space beginning at ([Fe/H],τ/Gyr) ≈ (+0.2, 7) and extending to (−0.5, 13). Our analyses indicate that the solar neighborhood includes stars on disk-like orbits even at the oldest ages and lowest metallicities accessible by our samples.more » « less
-
Abstract Accurate stellar ages are essential for our understanding of the star formation history of the Milky Way and Galactic chemical evolution, as well as to constrain exoplanet formation models. Gyrochronology, a relationship between stellar rotation and age, appears to offer a reliable age indicator for main-sequence (MS) stars over the mass range of approximately 0.6–1.3M⊙. Those stars lose their angular momentum due to magnetic braking and as a result their rotation speeds decrease with age. Although current gyrochronology relations have been fairly well tested for young MS stars with masses greater than 1M⊙, primarily in young open clusters, insufficient tests exist for older and lower mass MS stars. Binary stars offer the potential to expand and fill in the range of ages and metallicity over which gyrochronology can be empirically tested. In this paper, we demonstrate a Monte Carlo approach to evaluate gyrochronology models using binary stars. As examples, we used five previously published wide binary pairs. We also demonstrate a Monte Carlo approach to assess the precision and accuracy of ages derived from each gyrochronology model. For the traditional Skumanich models, the age uncertainties areσage/age = 15%–20% for stars withB−V= 0.65 andσage/age = 5%–10% for stars withB−V= 1.5 and rotation periodP≤ 20 days.more » « less
