Research Note

Astrometric comparisons of quasar catalogues

¹ Observatoire de Paris, Systèmes de Référence Temps Espace (SYRTE), CNRS/UMR8630, Paris, France e-mail: Jean.Souchay@obspm.fr
² Observatório Nacional/MCT and Observatório do Valongo/UFRJ, Rio de Janeiro, Brasil
³ Technische Universität Dresden, Institut für Planetare Geodäsie, Lohrmann Observatorium, Dresden, Germany

Received: 7 August 2007
Accepted: 1 April 2008

Abstract

Context. Large surveys, such as the Sloan Digital Sky Survey (SDSS) and the 2-degree Field (2QZ) considerably increased the number of recorded quasars at optical wavelengths. At the same time, the densification of very long baseline interferometry observations of extragalactic radio sources enabled an increase in the accuracy of the the quasi-inertial radio reference frame that realizes the International Celestial Reference System (ICRS).

Aims. In the perspective of the realization of an accurate optical counterpart to the ICRF, we investigate the astrometric properties of quasars recorded by several surveys, both at radio and optical wavelengths.

Methods. Cross-identifications of quasars are made between catalogues, both at optical (SDSS vs. 2QZ) and radio wavelengths (VLBA Calibrator Survey vs. SDSS). Equatorial coordinates of the quasars are compared to evaluate the astrometric quality of the catalogues.

Results. We find more than 2000 objects in common between SDSS and 2QZ. We show that differences in position never exceed 1´´, with a root mean square of 0.2´´. The agreement in redshift determinations is good with a typical dispersion of 0.05 root mean square. Similar comparisons between the SDSS and the VLBA Calibrator Survey find 831 cross-correlated sources within a 0.23´´ search radius. Systematic errors, likely due to the partial sky coverage of the SDSS, are detected in the coordinate differences, which produces to a rotation of the optical frame, with respect to the radio counterpart, at the level of 13 mas. Thus both the SDSS and the 2QZ can provide invaluable information for the accurate determination of the ICRS at the fainter end of the optical domain. We finally point out that the highest probability of finding radio loud quasars involves searching in the redder regions of the color-color space.