Testing stellar proper motions of the TGAS using data of the HSOY, UCAC5 and PMA catalogues

TL;DR

This study compares stellar proper motions from TGAS with ground-based catalogs, revealing magnitude-dependent systematics linked to Tycho-2 stars and differences between AGIS and classical methods, with implications for astrometric accuracy.

Contribution

It identifies magnitude-dependent systematic errors in proper motions derived from TGAS using AGIS, and compares these with ground-based catalogues to understand their origins.

Findings

Magnitude dependence of the $oldsymbol{ extomega_y}$ component reaches 1.5 mas/yr.

Differences between AGIS and classical methods affect proper motion accuracy.

Systematic errors are linked to Tycho-2 stars in TGAS proper motions.

Abstract

We present an investigation of stellar proper motions of the TGAS catalogue and ground-based HSOY, UCAC5 and PMA catalogues derived by combining with the Gaia DR1 space data. This investigation concerns with only those stars of ground-based catalogues that are contained in the TGAS. We analyze components of the mutual rotation vector between these systems. It has been found that in all three cases of comparison of the HSOY, UCAC-5 and PMA catalogues with TGAS, the $\omega_{y}$ component of the mutual rotation vector depends on magnitude nonlinearly within the range $9.5<m<11.5$, and has an amplitude reaching 1.5 mas/yr. The analysis has shown that the reason causing this effect is presence in proper motions of the Tycho-2-stars containing in the TGAS, of some inexplicable dependency on magnitude. It has been shown that proper motions of the TGAS stars derived using AGIS differ from those derived as a result of application the conventional (classical) method. At the same time, such a difference in proper motions of Hipparcos-stars from TGAS derived by both methods was not found. Investigation of systematic differences between proper motions of the TGAS stars derived by the classical method and proper motions of the HSOY, UCAC5 and PMA stars has shown that the values of the $\omega_{y}$ component in this case do not undergo any jumps and do not depend on magnitude. That fact unambiguously indicates that some magnitude error is contained in proper motions of Tycho2-stars derived by the use of AGIS. In the framework of the solid-body rotation model the coordinate systems set by the TGAS catalogue with classical proper motions (on the one hand) and by the HSOY, UCAC5 and PMA catalogues (on the other hand) have mutual rotation with the components along axis less than 0.2-0.3mas/yr.