A LightGBM classifier trained on NWAY Bayesian matches identifies true Chandra-Gaia counterparts for 113k X-ray sources, flags 7k ambiguous cases, and attributes half of 20k separation-only matches to chance coincidences, validated at 95% on COUP without positional features.
Membership of the Orion Nebula Population from the Chandra Orion Ultradeep Project
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
The Chandra Orion Ultradeep project (COUP) observation described in a companion paper by Getman et al. provides an exceptionally deep X-ray survey of the Orion Nebula Cluster and associated embedded young stellar objects. Membership of the region is important for studies of the stellar IMF, cluster dynamics, and star formation. The COUP study detected 1616 X-ray sources. In this study we confirm cloud membership for 1315 stars, identify 16 probable foreground field stars having optical counterparts with discrepant proper motions, and classify the remaining 285 X-ray sources, of which 51 are lightly and 234 heavily obscured. The 51 lightly obscured sources without known counterparts fall into three groups. (i) Sixteen are likely new members of the Orion Nebula Cluster. (ii) Two with unusually soft and non-flaring X-ray emission appear to be associated with nebular shocks, and may be new examples of X-rays produced at the bow shocks of Herbig-Haro outflows. (iii) The remaining thirty three are very weak uncertain sources, possibly spurious. Out of 234 heavily absorbed sources without optical or near-infrared counterparts 75 COUP sources are likely new embedded cloud members (with membership for 42 confirmed by powerful X-ray flares), and the remaining 159 are likely extragalactic AGN seen through the molecular cloud, as argued by a careful simulation of the extragalactic background population. Finally, a few new binary companions to Orion stars may have been found, but most cases of proximate COUP sources can be attributed to chance superpositions in this crowded field.
fields
astro-ph.IM 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
The Chandra-Gaia Catalog of Counterparts: Resolving ambiguous Gaia matches to X-ray sources in the Chandra Source Catalog using Machine Learning
A LightGBM classifier trained on NWAY Bayesian matches identifies true Chandra-Gaia counterparts for 113k X-ray sources, flags 7k ambiguous cases, and attributes half of 20k separation-only matches to chance coincidences, validated at 95% on COUP without positional features.