Phygas stage 1: raw results of screening 33 million Gaia DR3 stars against 4,087 young clusters

What this is. The complete, unfiltered result of the first stage (a fast screening) of the Phygas search for the birth clusters of runaway stars: every Gaia DR3 star with a full 3-D velocity (33,162,595 stars) was traced backwards through the Milky Way and screened against 4,087 young open clusters and moving groups (Hunt & Reffert 2024). Phygas ("fugitive") runs on the volunteer-computing project bitboinc (https://bitboinc.athena.org.tr); every task was computed by two different volunteers and accepted only when both results agreed. Progress and an explanation are at https://bitboinc.athena.org.tr/phygas.php.

How the numbers were obtained

1. Stars. Position, parallax (zero-point corrected, Lindegren et al. 2021), proper motion and radial velocity of each star with their uncertainties (Gaia DR3).
2. Clusters. 4,087 clusters and moving groups with ages up to 100 Myr (Hunt & Reffert 2024), with position, distance, motion, radial velocity, half-mass radius, total radius and age.
3. Screening. Each star and each cluster is integrated backwards through the Allen & Santillan (1991) Milky Way potential; a linearised propagation of the uncertainties (a state-transition matrix from six neighbouring orbits) gives, for each pair, a chi2 of the approach, the probability p_r50 that the star was inside the cluster's half-mass radius at the closest approach, the corresponding p_rtot for the total radius, the time of closest approach and the relative speed. Pairs with p_r50 ≥ 0.001 are kept.

Files

The eight stage1_screen_*.txt.gz files hold the raw screening output of the volunteers' tasks (about 14 million pairs with p_r50 ≥ 0.01); young_clusters.txt.gz is the cluster list; README.txt documents every column.

How to read these data

These are screening probabilities, not birth-site probabilities. They include chance alignments and the known members of each cluster (stars currently inside their cluster have a closest approach at time zero); no control has been subtracted from this record. Chance-alignment control runs (20 controls and 20 measurement-error repeats of the whole catalogue) are being computed by the volunteers with an analysis plan that was registered in advance (doi:10.5281/zenodo.23680281); their results will follow in a new version of this record and in the paper. No claim about any individual star is made here.

Volunteers

45 volunteers returned 1,314 validated results for this part of the run. Volunteers who asked to be named: Alan W Nichols (United States), Alexander Piskun (Canada), Anand Bhat (Australia), Anthony Coleman (United States), Dejan Ivanek (Croatia), Detlef Lexut (Germany), Dmytro Taukchi (Ukraine), Erik Veit (United States), Franz Harvanek (Austria), Frolov Kirill (Russia), Glenn Goninan (Australia), Hans Sveen (Norway), Ian Dickinson (United Kingdom), Johannes Rensen (Netherlands), Jörg Steinmetz (Germany), José Miguel (Spain), Kyle Weber (International), Michael Doneske (United States), Michael Dörscheln (Germany), Mirko Tomaschek (Libya), Rea Rosa (Austria), Thomas Kledzik (Germany); and 23 other volunteers who preferred to stay anonymous. Thank you.

References

• Gaia Collaboration et al. 2023, A&A 674, A1 (Gaia DR3); Lindegren, L. et al. 2021, A&A 649, A4 (parallax zero point).
• Hunt, E. L. & Reffert, S. 2024, A&A 686, A42 (cluster catalogue, VizieR J/A+A/686/A42).
• Allen, C. & Santillan, A. 1991, RMxAA 22, 255 (Milky Way potential).
• Code: https://github.com/alplix/phygas.