An investigation of the absolute proper motions of the XPM catalogue

P. N. Fedorov 1 V. S. Akhmetov 1 V. V. Bobylev 0 A. T. Bajkova 0 0 Central Astronomical Observatory at Pulkovo of RAS , Pulkovskoye Chaussee 65/1, 196140, St Petersburg, Russia 1 Institute of Astronomy of Kharkiv National University , Sums'ka 35, 61022 Kharkiv, Ukraine A B S T R A C T XPM-1.0 is the regular version of the XPM catalogue. In comparison with XPM, this astrometric catalogue of about 280 millions stars covering the entire sky from 90 to +90 in declination and in the magnitude range 10 < B < 22 mag is somewhat improved. The general procedural steps were followed as for XPM, but some of these were performed on a more sophisticated level. The XPM-1.0 catalogue contains the star positions, the proper motions and the Two-Micron All-Sky Survey (2MASS) and United States Naval Observatory (USNO) photometry of about 280 million sources. We present some investigations of the absolute proper motions of the XPM-1.0 catalogue and also important information for users of the catalogue. Unlike the previous version, XPM-1.0 contains proper motions over the whole sky without gaps. In fields covering the zone of avoidance or those that contain fewer than 25 galaxies, quasi-absolute calibration was performed. The proper motion errors vary from 3 to 10 mas yr1, depending on a specific field. The zero-point of the absolute proper motion frame (the absolute calibration) was specified with more than one million galaxies from 2MASS and USNO-A2.0. The mean formal error of absolute calibration is less than 1 mas yr1. 1 I N T R O D U C T I O N In this paper, we describe some steps still to be taken towards the main goal, which is to create the most comprehensive catalogue of the absolute proper motions of stars, XPM (Fedorov, Myznikov & Akhmetov 2009, hereafter Paper I), using the extragalactic reference frame defined by faint galaxies. As is well known, there are few catalogues of the absolute proper motions of stars, while there are no catalogues that cover the whole celestial sphere. The Southern hemisphere is especially poor of data, as there is only one catalogue of absolute proper motions for the region south of 45, the Southern Proper Motion 1 catalogue (SPM1; Platais et al. 1998). This covers an area of approximately 720 deg2 near the South Pole. The limiting apparent stellar magnitude does not exceed 18 mag in all the catalogues. The catalogues are all based on photographic observations made in the 20th century. The most well-known of these are the GPM (Rybka & Yatsenko 1997, I/285 CDS), the PUL2 (Bobylev, Bronnikova & Shakht 2004, I/285 CDS) for the faint star programme (KSZ), the NPM1 (Klemola, Jones & Hanson 1987, III/199 CDS) for the Lick Northern Proper Motion programme and the SPM2 (Platais et al. 1998, III/277 CDS) for the Yale Southern Proper Motion programme. The maximal number of stars (287 000) is contained in the SPM2 catalogue, while the maximal number of galaxies (70 000) is in the NPM1 catalogue. The GPM, PUL2 and NPM1 catalogues cover the northern sky and partially the southern sky, and the SPM2 catalogue covers about one-third of the southern sky. The random error of proper motions in these catalogues depends on stellar magnitude and varies from 3 to 10 mas yr1, while the accuracy of the absolute calibration is 25 mas yr1. The above-mentioned catalogues of absolute proper motions are very important for astrometry, as they allow the local coordinate system to be implemented, which does not rotate with respect to galaxies. The global quasi-inertial coordinate system can be established through the catalogue of absolute proper motions of stars covering the whole sky. The data of these catalogues play a principal role in determining the kinematic parameters of the Galaxy, for example, in the framework of the OgorodnikovMiln model. It is worth noting that this model provides the most adequate parameters, on the condition that the proper motions representing the whole celestial sphere are used. As mentioned in Paper I, the XPM catalogue contains approximately 280 million absolute proper motions of stars and covers the whole celestial sphere, excluding a narrow zone near the galactic equator within the stellar magnitude range from 11 < B < 20 mag. The random error of its proper motions depends on stellar magnitude and lies within 310 mas yr1. The error of absolute calibration in the Northern hemisphere is approximately 0.3 mas yr1, and of the order of 1 mas yr1 in the Southern hemisphere. The creation of this catalogue is based mainly on the following three most important procedures: (i) cross-identification, which allows us to identify and compare objects in the United States Naval Observatory (USNO)-A2.0 and Two-Micron All-Sky Survey (2MASS) catalogues; (ii) elimination of systematic errors in the positions of USNOA2.0 objects with the use of the median filter; (iii) derivation of the absolute proper motions of stars. Evidently, the cross-identification procedure is crucial in the procedures listed above, as it determines all other procedures and the resulting accuracy of the absolute proper motions. It has been noted in Paper I that the cross-identification procedure mentioned above is not, strictly speaking, actual cross-identification, but rather it is an association that can result in false identifications. This leads in turn to forming false position differences for stars and galaxies. Thus, the values of the function F(, ) obtained with the median filter (see Paper I) will be burdened with errors, which will inevitably result in erroneous proper motions. Therefore, most of our attention must be given to the cross-identification procedure. In the XPM-1.0 version, we used an improved version of the cross-identification procedure, compared to the previous version of XPM described in Paper I. It was only for this procedure that proper motions from the USNO-B1 catalogue (Monet et al. 2003) were involved. This has made it possible combine three catalogues, USNO1, USNO-2.0 (Monet 1998) and 2MASS (Skrutskie et al. 2006), using a circular search window 1.5 arcsec in dimension. Moreover, the high-precision photometric data of 2MASS were used to calculate the USNO-2.0 magnitudes, which were compared to their original values when selecting the objects within the circular 1.5arcsec search window. This is described in detail in Section 2. There is no simple test at this stage that would allow us to quantitatively estimate the improvement in the properties of the catalogue. This is, first of all because of the absence of accurate estimates for the individual positions of stars in the initial catalogues. Nevertheless, we believe that using the improved version of the cross-identification procedure results in a decrease of random errors in the position differences, in some broadening of the stellar magnitude range and also in an improvement in linking to extragalactic objects. Consistent with the idea of creating the most comprehensive catalogue, we derive the pro (...truncated)