Gaia DR3 reanalysis of Kepler stars: planet hosts with periods under 20 d rotate ~15% slower
Authors of an arXiv preprint report that Kepler stars with detected planets have a median rotation-period ratio about 15% larger than stars without them, for periods under 20 days.
A preprint on arXiv (arXiv · preprint, not yet peer reviewed) revisits whether stars with detected planets rotate more slowly than stars without them. The authors combined the Kepler rotation-period catalogue with updated Gaia DR3 stellar parameters and a current exoplanet catalogue. They report that for periods shorter than 20 days, the median ratio of observed to temperature-expected period is about 15% larger for planet hosts than for non-hosts (KS test p = 2.4 × 10^-4) arxiv.org
arXiv · препринтarxiv.orgOpen source ↗.
Data and method
The study keeps the rotation periods of McQuillan, Mazeh and Aigrain, measured with the autocorrelation-function method. This is the same catalogue used by Sibony et al. (2022), who found that planet hosts rotate slower than matched controls, so the two results can be compared directly. Effective temperatures, parallaxes, photometry, extinction estimates and metallicities now come from Gaia DR3 (GSP-Phot) instead of the Berger et al. (2020) catalogue. Planet hosts are stars associated with planets in the NASA Exoplanet Archive. All other selected stars count as non-hosts, meaning stars with no detected planets arxiv.org
arXiv · препринтarxiv.orgOpen source ↗.
Evolved stars were removed using an extinction-corrected colour–magnitude diagram, keeping stars within 1 mag of the fitted main-sequence region. The sample was also limited in distance and metallicity to the region where detected hosts are well represented. Before any period cut, it contains 400 detected hosts among 12 525 stars, or 3.2%. By comparison, 9 of 902 otherwise eligible metal-rich stars beyond 1.3 kpc and 18 of 1 398 metal-poor stars within 1.3 kpc are hosts, or 1.0 and 1.3% arxiv.org
arXiv · препринтarxiv.orgOpen source ↗.
The median non-host rotation period was fitted as a third-degree polynomial in effective temperature. For each star the authors then computed the ratio of its observed period to the period expected for its temperature, and compared the ratio distributions of hosts and non-hosts. A fractional measure was chosen because the same difference in days represents a different effect for fast and slow rotators arxiv.org
arXiv · препринтarxiv.orgOpen source ↗.
TESS check
The authors also examined the TESS All-Sky Rotation Survey (TARS) and found a similar trend. However, the host–non-host difference there is sensitive to sample selection, and the strongly structured period distribution precludes a robust comparison between hosts and non-hosts arxiv.org
arXiv · препринтarxiv.orgOpen source ↗.
The authors conclude that the updated analysis, with more robust stellar and planetary parameters, recovers the previously reported association between slower rotation and the presence of detected planets. They say this strengthens the case for a genuine connection between stellar rotation and planet occurrence arxiv.org
arXiv · препринтarxiv.orgOpen source ↗.
