The Gaia reference frame for bright sources examined using VLBI observations of radio stars

Lund Observatory, Department of Astronomy and Theoretical Physics, Lund University, Box 43, 22100 Lund, Sweden
e-mail: lennart@astro.lu.se

Received: 24 June 2019
Accepted: 22 October 2019

Abstract

Context. Positions and proper motions of Gaia sources are expressed in a reference frame that ideally should be non-rotating relative to distant extragalactic objects, coincident with the International Celestial Reference System (ICRS), and consistent across all magnitudes. For sources fainter than 16th magnitude, this is achieved through Gaia’s direct observations of quasars. At brighter magnitudes, it is difficult to validate the quality of the reference frame because comparison data are scarce.

Aims. The aim of this paper is to examine the use of very long baseline interferometry (VLBI) observations of radio stars to determine the spin and orientation of the bright reference frame of current and future Gaia data releases.

Methods. Simultaneous estimation of the six spin and orientation parameters makes optimal use of VLBI data and makes it possible to include even single-epoch VLBI observations in the solution. The method is applied to Gaia Data Release 2 (DR2) using published VLBI data for 41 radio stars.

Results. The VLBI data for the best-fitting 26 sources indicate that the bright reference frame of Gaia DR2 rotates relative to the faint quasars at a rate of about 0.1 mas yr⁻¹, which is significant at the 2σ level. This supports a similar conclusion based on a comparison with stellar positions in the HIPPARCOS frame. The accuracy is currently limited because only a few radio sources are included in the solution, by uncertainties in the Gaia DR2 proper motions, and by issues related to the astrophysical nature of the radio stars.

Conclusions. While the origin of the indicated rotation is understood and can be avoided in future data releases, it remains important to validate the bright reference frame of Gaia by independent observations. This can be achieved using VLBI astrometry, which may require re-observing the old sample of radio stars as well as measuring new objects. The unique historical value of positional measurements is stressed and VLBI observers are urged to ensure that relevant positional information is preserved for the future.