3.5.3 Source verification

The ranges and distributions of the astrometric parameters and various statistics were monitored as part of the internal validation performed by the AGIS team.

Residuals of the Gaia DR2 solution have been monitored during the iterative solution and extensively studied as part of the internal validation performed by the AGIS team. In particular the internal consistency checks were done to assess the level of systematic errors in the solution. See Section 5.3 of the Gaia DR2 astrometry paper (Lindegren et al. 2018) for a summary of results.

The internal consistency of the astrometric solution can be examined by comparing solutions based on complementary subsets of the observations. The observations can for example be divided depending on the CCD strip in the astrometric field (AF). Normally a source is observed in nine consecutive CCD strips, denoted AF1–AF9. Separate solutions have been obtained using observations on AF2–AF5 and AF6–AF9, respectively; these are called the “early” and “late” solutions. The results are compared in Section 5.5 (and Figs. 16-18) of the Gaia DR2 astrometry paper (Lindegren et al. 2018). The split-field solutions show the presence of a magnitude-dependent systematic error, probably affecting mainly the bright ($G$ $\lesssim 13$) sources, and spatial variations of a few tens of $\mu$as on a scale of several degrees.

The reference frame of Gaia DR2 is nominally aligned with ICRS and non-rotating with respect to the distant universe. This was achieved by means of a subset of 492 009 primary sources assumed to be quasars. These included 2844 sources provisionally identified as the optical counterparts of VLBI sources in a prototype version of ICRF3, and 489 163 sources found by cross-matching AGIS02.1 with the AllWISE AGN catalogue (Secrest et al. 2015, 2016).

The spin of the reference frame in Gaia DR2 is globally non-rotating to within $\pm 0.02$ mas yr${}^{-1}$ in all three axes. Particular attention was given to a possible dependence of the spin parameters on colour (using the effective wavenumber ${\nu }_{\text{eff}}$) and magnitude ($G$). The results suggest a small systematic dependence on colour, e.g. by $\pm 0.02$ mas yr${}^{-1}$ over the range $1.4\lesssim {\nu }_{\text{eff}}\lesssim 1.8\mu {\text{m}}^{-1}$ corresponding to roughly $G_{\mathrm{BP}}-G_{\mathrm{RP}}$ =0 to 2 mag. As this result was derived for quasars that are typically fainter than 15th magnitude, it does not necessarily represent the quality of the Gaia DR2 reference frame for much brighter objects.

Indeed, Fig. 4 in (Lindegren et al. 2018) suggests that the bright ($G\lesssim 12$) reference frame of Gaia DR2 has a significant ($\sim$0.15 mas yr${}^{-1}$) spin relative to the fainter quasars. The points in the left part of the diagram were calculated from stellar proper motion differences between the current solution and Gaia DR1 (TGAS). Although based on a much shorter stretch of observations than the present solution, TGAS provides a valuable comparison for the proper motions thanks to its $\sim$24 yr time difference from the Hipparcos epoch.

The most reasonable explanation are systematics in the Gaia DR2 proper motions of the bright sources. The gradual change between magnitudes 12 and 10 suggests an origin in the gated observations, which dominate for $G$ $\lesssim 12$, or possibly in observations of window class 0, which dominate for $G$ $\lesssim 13$. A more comprehensive analysis of the faint Gaia DR2 reference frame and the optical properties of the VLBI sources is given by Mignard et al. (2018).

The secondary solution was checked in a few selected areas using images obtained with the ESO VLT Survey Telescope (VST) for the GBOT project Section 3.2.2 and, for some very high-density areas in the Baade’s window region, with the HST Advanced Camera for Surveys (ACS/WFC). These did not check the astrometric precision of the secondary solution but rather the reality of the stars selected based on the astrometric quality indicators (number of matched observations and excess source noise).

Global astrometric satellites like Hipparcos and Gaia are able to measure absolute parallaxes, i.e. without zero-point error, but this capability is susceptible to various instrumental effects, in particular to a certain kind of basic-angle variations. As discussed by Butkevich et al. (2017), periodic variations of the basic angle ($\mathrm{\Gamma }$) are observationally almost indistinguishable from a global parallax shift.

It is believed that the basic-angle corrector derived from BAM data (Section 2.4.4) does a very good job of eliminating basic-angle variations, but a remaining small variations cannot be excluded. This would then show up as a small offset in the parallaxes. For this reason it is extremely important to investigate the parallax zero point by external means, i.e. using astrophysical sources with known parallaxes. It is also important to check possible dependences of the zero point on other factors such as position, magnitude, and colour, which could be created by errors in the calibration model.

The quasars are almost ideal for checking the parallax zero point thanks to their extremely small parallaxes (as for redshift $z>0.1$), large number, availability over most of the celestial sphere, and, in most cases, nearly point-like appearance. Main drawbacks are their faintness and peculiar colours. In order to create the largest possible quasar sample for validation purposes, a new cross-match of the final Gaia DR2 data with the AllWISE AGN catalogue Secrest et al. (2015) was made, choosing in each case the nearest positional match. The results and discussion of this comparison are outlined in Section 5.2 of Lindegren et al. (2018).