The XPM Catalogue: absolute proper motions of 280 million stars

P. N. Fedorov 0 A. A. Myznikov 0 V. S. Akhmetov 0 0 Institute of Astronomy of Kharkiv National University , Sums'ka 35, 61022 Kharkiv , Ukraine A B S T R A C T We combined data from the Two-Micron All Sky Survey (2MASS) and USNO-A2.0 catalogues in order to derive the absolute proper motions of about 280 million stars distributed all over the sky excluding a small region near the Galactic Centre, in the magnitude range 12 < B < 19 mag. The proper motions were derived from the 2MASS Point Sources and USNO-A2.0 catalogue positions with a mean epoch difference of about 45 years for the Northern hemisphere and about 17 years for the Southern one. The zero-point of the absolute proper motion frame (the 'absolute calibration') was specified with the use of about 1.45 million galaxies from 2MASS. Most of the systematic zonal errors inherent in the USNO-A2.0 catalogue were eliminated before the calculation of proper motions. The mean formal error of absolute calibration is less than 1 mas yr1. The XPM Catalogue will be available via CDS in Strasbourg during 2010. The generated catalogue contains the International Celestial Reference System positions of stars for the J2000 epoch, original absolute proper motions, as well as B, R, J, H and K magnitudes. A comparison of the proper motions obtained in this work with the data of other recent catalogues of quasars was fulfilled. 1 I N T R O D U C T I O N The main goal of this work is to create the most comprehensive catalogue of absolute proper motions of stars using the extragalactic reference frame defined by faint galaxies. The concept of using galaxies as an inertial proper motion reference frame was initiated by Dneprovsky & Gerasimovic (1932) in Pulkovo. The results of most well-known absolute proper motion programmes using galaxies as a reference frame are presented by the following catalogues: General Compiled Catalogue of Absolute Proper Motions (GPM) (Rybka & Yatsenko 1997a), GPM1 (Rybka & Yatsenko 1997b), PUL2 (Bobylev, Bronnikova & Schakht 2004) for the faint stars programme (KSZ); NPM1 (Klemola, Jones & Hanson 1987) and NPM2 (Hanson et al. 2004) for the Lick Northern Proper Motion; and SPM2 (Platais et al. 1998), SPM3 (Girard et al. 2004) for the Yale Southern Proper Motion. We use the term absolute proper motions to describe about 280 million proper motions of stars with a zero-point derived using positions of about 1.45 million galaxies as the reference frame. As is well known, tangential velocities of galaxies (Chernin 2001) as compared to the Hubble flow are vanishingly small at distances from several Mpc. Even if their tangential motions Vt were equal in magnitude to the Hubble flow Vt = H R, the resulting proper E-mail: (PNF); (AAM); (VSA) motions should be as small as 0 = 1.5 105 arcsec yr1 for H = 70 km s1 Mpc1 (Klemola et al. 1987). It is evident that any rotation of the system of galaxies caused by their peculiar velocities is much less than 0. Consequently, the positions of galaxies over the time period of 100 yr may be considered to be time independent. Thus, the absolute proper motions are tangential components of the stars spatial velocities with respect to a quasi-inertial coordinate system, i.e. such a system that moves without rotation while its origin may have acceleration. Such coordinate systems are admissible in classical mechanics. In general relativity, such coordinate systems are admissible too, but they require some relativistic corrections (Einstein 1956; Weinberg 1972). A system of proper motions specified by any catalogue of absolute proper motions makes it possible to reproduce a quasi-inertial system of coordinates at any given time moment with an accuracy of up to the catalogue systematic errors. Since there are large numbers of faint galaxies that look like stars in the initial images and thus can be used as astrometric reference objects, the effect of the magnitude equation for stars fainter than 15 mag can be expected to be insignificant. Unfortunately, the position data for extragalactic point sources are very scanty. For example, in the Sloan Digital Sky Survey (SDSS) data release 5 (DR5) Quasar Catalogue (available at http://www.sdss.org/dr5/products/), there are only about 78 000 quasars, and 94 000 quasars are contained in the Lyon Extragalactic Data base (http://leda.univlyon1.fr/), and their distribution over the sky is very inhomogeneous. Though the magnitude equation may affect extended and point sources differently, the use of galaxies positions for absolute calibration seems to be reasonable from the viewpoint of minimization of the systematic errors. Therefore, the catalogue presented in this paper is an independent realization of the extragalactic reference system in the optical range, whose rate of rotation with respect to distant extragalactic objects is less than 1 mas yr1. This paper is the first one in a series representing a catalogue of the new absolute proper motions containing 280 million objects, which we called XPM. We hope that this catalogue will be available via CDS in Strasbourg during 2010 when we will complete an investigation of the obtained proper motions and compare the proper motions with those contained in the most recent catalogues. Here, we describe the initial considerations, procedures of cross-identification, error correction, linking to extragalactic objects and deriving the absolute proper motions. Also, we briefly discuss the results of external comparison that gives the estimate of errors of the proper motions. 2 T H E D ATA The Two-Micron All Sky Survey (2MASS) (Skrutskie et al. 2006) and USNO-A2.0 (Monet 1998) catalogues contain the most comprehensive data on the astrometric positions of stars. The positions of both the catalogues are nominally on the International Celestial Reference System (ICRS) (Arias et al. 1995). The mean difference of epochs between 2MASS and USNO-A2.0 is about 45 years for the Northern hemisphere and about 17 years for the Southern one. The 2MASS data contain two large data sets: the Point Source Catalogue (PSC; 470 992 970 point objects) and the Extended Source Catalogue (XSC; 1650 000 extended objects). Most of the extended objects in XSC are galaxies. Therefore, combining the 2MASS data with the earlier highly dense data sets for deriving the absolute proper motions of stars and providing the absolute zero-point of proper motion seems to be reasonable. The USNO-A2.0 catalogue is the densest data set suitable for solving this task. It contains about 526 million positions taken from 825 POSS I fields and from 606 SRC-J and ESO-R fields, but their combination with 2MASS to obtain precise proper motions is rather problematic due to the presence of the magnitude-dependent and zone-dependent systematic errors (http://vizier.u-strasbg.fr/vizbin/qcat?). In this paper, we use the term field in the sense as it has been considered by D. Monet in READUSE.V20 for USNO-A2.0. Another great problem in using these (...truncated)