/ csc / csc_crossmatches.html

CSC Crossmatches

CSC 2.0 Preliminary Crossmatches | CSC 2.0 Crossmatches

Introduction

This page contains brief descriptions of the cross match tables between the Chandra Source Catalog and other catalogs across the electromagnetic spectrum. The CSC sources are extracted from the Master Sources Table. For each counterpart, the match probability (between 0.0 and 1.0) and separation (in arcsec) are given so that users may apply their own selection criteria appropriate for their goals. Starting with the cross matches for CSC version 2.1.1 the tables contain two additional items to aid users in applying more finely tuned selection criteria: normalized separation (separation divided by match σ) and match class (d, l, c, k, r; see below).

CSC 2.1.1 Crossmatches

All cross matches have been performed with Xmatch employing the Bayesian method developed by Budavari & Szalay (2008, ApJ 679, 301) as implemented and extended by A. Rots (2025). The Xmatch code takes into account local source density as well as both error ellipses and raw-size ellipses of the sources. For each cross-matched source pair the match probability (between 0.0 and 1.0), separation (in arcsec), normalized separation (separation divided by match σ uncertainty), and match class (see below) are provided. The details on these columns are provided in the Xmatch quick summary. The main match tables are provided in four different formats: TSV,

[Extended] CSV,

The proposed Enhanced CSV (ECSV) format has the following overall structure:

A header section which consists of lines that start with the # character and provide the table definition and data format via a YAML-encoded data structure.

An initial line in the header section which identifies the file as ECSV and provides a version number.

A CSV-formatted data section in which the first line contains the column names and subsequent lines contains the data values.

Version 1.0 of the ECSV format specification and the reference Python implementation assumes ASCII-encoded header and data sections. Support for unicode (in particular UTF-8) may be added in subsequent versions.

–Astropy Enhanced CSV file format specifications

FITS, and VOTABLE. In addition, ambiguous matches are provided in four separate tables: ambiguous pairs based on error ellipse matches, ambiguous pairs based on raw size matches, CSC sources with ambiguous matches, and counterpart sources with ambiguous matches. See the Primary Crossmatch Column Decriptions table for a description of the columns in the main crossmatch tables.

The cross matches are performed separately for each CSC Field: these are areas defined by the fields-of-view of observations covering a connected contiguous region on the sky. For CSC 2.1 there are 6019 such Fields, of which 5997 contain one or more master sources. It should be noted that match probabilities within each Field are (obviously) comparable, but their distribution may differ between Fields due to differences in source density and Bayes Factor distribution. This effect has been taken care of by applying a self-consistent probability threshold for each Field.

To aid users in selecting cross matches for their use, we have introduced five cross match classes. Users are advised to carefully consider which of the five cross match classes they want to include in order to obtain their desired level of robustness:

* 1.7σ corresponds to 76% confidence, 3.4σ to 99.7% confidence level; i.e., the likelihood of a match beyond 3.4σ is 0.3%.

** This means that the match has a high probability, but may be contaminated by another match with a distinctly lower but reasonable probability.
Cross Match Class
d	definite match, with a separation less than 1.7σ ^*
l	likely match, with a separation between 1.7σ and 3.4σ ^*
c	definite match, but with a potential ambiguity ^**
k	likely match, but with a potential ambiguity ^**
r	potential match based on raw source size; not robust

For more information, see the Xmatch quick summary or full documentation.