Researchers derive ages for 2,204 exoplanet-host stars using stellar rotation

Researchers derive ages for 2,204 confirmed exoplanet-host stars using stellar rotation in a new astro-ph.SR paper released on Oct. 31, 2026 as the first installment of the series Tracking the evolution of sub-Neptunes.

Stellar age underpins the reconstruction of planetary-system evolution because observed planetary-system properties emerge from mechanisms acting across timescales from protoplanetary-disc dispersal in the first few Myr to secular evolution over several Gyr. Yet age remains difficult to measure for unevolved main-sequence stars. Isochrone fitting works well for evolved stars but gives relatively weak limits for unevolved low-mass main-sequence stars because their luminosities and effective temperatures change slowly with age, a weakness the introduction contrasts with rotation-based age discrimination for the same class of stars. Low-mass main-sequence stars host the majority of known planetary systems, yet asteroseismology provides accurate ages for only a comparatively restricted sample of stars with suitable observations.

Rotation works as an age clock because stars with convective envelopes shed angular momentum through magnetized stellar winds, making surface rotation evolve systematically over time. The empirical link between stellar rotation and age comes from Sun and open-cluster observations and underlies gyrochronology. Skumanich’s 1972 observations of the Sun and open clusters established this connection, and work by Barnes in 2003 and 2007 formed the basis of gyrochronology.

Observations of open clusters show slow-rotator sequences that depend on mass, and the two-zone rotational models reproduce them. Higher-mass cool stars in the roughly 0.1 Gyr-old Pleiades follow this pattern, and progressively lower-mass stars converge onto the sequence at later ages, making rotation especially useful where conventional isochrone ages are poorly constrained.

The new study applies the two-zone rotational evolution model of Spada and Lanzafame (2026), which divides a star into a radiative core and a convective envelope, comparing observed rotation periods and de-reddened Gaia colors with model predictions while treating literature stellar masses as Gaussian priors. This Bayesian approach propagates observational uncertainties, the intrinsic width of the rotational sequence, and uncertainties in stellar mass into the age posterior. Each star enters the analysis with three observables: surface rotation period, de-reddened Gaia color, and an independent stellar-mass prior, while the team compiles effective temperature and metallicity as auxiliary parameters that do not enter the likelihood, and Section 2 of the paper describes the data compilation and specifies which quantities feed the statistical inference and which the team retains only for context. The framework also generates model predictions for stellar properties that researchers can check against independent literature.

Researchers compiled rotation periods and stellar parameters for 4,639 confirmed exoplanet hosts from an initial compilation of 4,716 systems in the Planetary Systems table of the NASA Exoplanet Archive, which they retrieved on June 19, 2026. Section 2.1 states that the team collected all stellar quantities from public literature and survey archives for each confirmed planet-hosting star in that archive compilation. Rotation measurements came from the Kepler, K2, and TESS photometric surveys, ground-based photometric monitoring, and spectroscopic chromospheric-activity time series, drawn from about two dozen dedicated catalogs and many individual system studies, spanning many photometric and spectroscopic techniques. Period-finding ran through auto-correlation, wavelet analysis, gradient-of-power-spectrum methods, Gaussian-process regression, deep learning, Lomb–Scargle periodograms, and generalized Lomb–Scargle periodograms.

Of the 2,874 stars with at least one published rotation measurement, 158 lost all their measurements during cleaning, leaving 2,716 stars with at least one usable period. Discordant periods often form harmonic families such as P, P/2, 2P, P/3, and higher harmonics produced by different techniques working on the same light curve, and most disagreements are not genuine. Section 2.2 explains that these harmonic aliases are sometimes contaminated by non-rotational signals such as asteroseismic oscillations or delta Scuti pulsations. The team throws out any period below the critical break-up period as physically impossible for rotation.

A star drops out when every one of its measured periods falls under that limit, since that pattern points to a non-rotational signal.

The team assigns each star a reliability tier, with tier 1 holding the most reliable periods and lower tiers progressively admitting periods that carry a quality or physical caveat. The recommended rotation period is the median of the retained measurements, and the authors deliberately avoid an inverse-variance weighted mean because a single spuriously precise quoted error could dominate the result. The selection removes impossible break-up periods, applies a v sin i upper limit, and reduces discordant period values to a single recommended median period per star. Once cleaning and selection were complete, 2,706 stars retained a recommended rotation period, while the team excluded or left unresolved ten of the 2,716 stars.

The likelihood uses the de-reddened Gaia DR3 G_BP-G_RP index, with reddening taken from the 3D extinction map of Vergely et al. (2022). Researchers integrate reddening to each star’s adopted distance using the most precise estimate among Gaia DR3, Bailer-Jones et al. (2021), TICv8 (Stassun and others, 2019), Berger et al.

(2020), and Berger et al. (2026). The reddening uncertainty is propagated into the colour error.

For stars with well-constrained posterior distributions, typical formal age uncertainties run to 10–15%, offering greater age resolution than isochrone fitting generally achieves for unevolved main-sequence stars. For unevolved stars, the rotation period can provide much greater age discrimination than the star’s position in the Hertzsprung–Russell diagram. The authors trace this precision to applying a single rotational evolution model across the full sample, which creates a common relative age scale and avoids methodological heterogeneity in literature age compilations. The introduction lists this homogeneity as the second attractive property of gyrochronology for exoplanet population studies, alongside the greater age discrimination of the rotation period.

The team calibrates the rotational evolution model against open-cluster rotation sequences spanning 0.4–1.25 solar masses. The calibration ceiling reflects the limited span of cluster ages with dependable rotation data, while older field stars show possible weakened magnetic braking at sufficiently large Rossby number. The team extended rotational tracks to the terminal-age main sequence, beyond the empirical calibration range, to characterize stars outside the domain directly constrained by open-cluster observations. Age estimates beyond that range are model-dependent rather than empirically anchored, and the work flags estimates falling outside that empirically calibrated domain.

Close-in sub-Neptunes are planets with sizes between Earth and Neptune and represent one of the most common outcomes of planet formation despite lacking a direct Solar System analog. Their radius distribution shows a pronounced deficit separating rocky super-Earths from larger sub-Neptunes, a feature known as the radius valley. Researchers generally interpret this gap as evidence that evolution after formation strongly affects the present-day radii of close-in planets. Two leading explanations for the radius valley are atmospheric mass loss driven by high-energy stellar irradiation and cooling-powered loss from planetary core and envelope, though other processes tied to planet formation and primordial atmosphere properties may also contribute.

The catalogue will support investigation of how the radius distribution and radius valley change across stellar ages, offering direct observational constraints on mechanisms that shape close-in planetary systems. The stellar parameters, their provenance, and the gyrochronological results are publicly available through the CDS, and the catalogue serves as the temporal basis for later papers in the series. Those later papers will study changes in the radius distribution and radius valley across stellar ages.

The paper’s authors are Alessandro C. Lanzafame, Vincenzo Drago, Nino Greco, Isabella Pagano, Gaetano Scandariato, and Federico Spada. The manuscript runs 17 pages with five figures and was submitted to Astronomy and Astrophysics under arXiv ID 2611.00856. Version 1 was submitted on Saturday, 31 October 2026 at 12:02:19 UTC, and the arXiv submission history lists a file size of 4,631 KB. A companion paper by Greco and others (2026, Paper II) presents the complete per-source compilation of exoplanet and host-star data, with the reduction of multiple values into single recommended estimates described in Appendix A.