SpringerLink
Forum Springer Astron. Astrophys.
Forum Whats New Search Orders


Astron. Astrophys. 327, 577-586 (1997)

Previous Section Next Section Title Page Table of Contents

2. The data

For our study we base ourselves on new and published proper motions, radial velocities, and distances for subdwarf stars (Table 1). Most of the stars are part of a programme to investigate the nature and distance of horizontal-branch type stars (see Moehler et al.  1990, Theissen et al.  1993, Schmidt et al.  1997, Aguilar Sanchéz et al.  in prep.) from the Palomar-Green catalogue (Green et al.  1986) and from the Hamburg Survey (Hagen et al.  1995). Other stars have been selected purely because proper motions are available.

The proper motions used in this work have in part been determined by us (see below). Further data were taken from the catalogue of the Lick Northern Proper Motion (NPM) programme (Klemola et al.  1987), the already published proper motions for subdwarf stars from Colin et al.  (1994) and from Thejll et al.  (1997).

2.1. New absolute proper motions

We have determined absolute proper motions for twelve stars from recent CCD observations in combination with the Palomar Sky Survey (POSS). From the sample of stars with known distances we selected several which are located in fields with a sufficient number of background galaxies. New CCD observations were taken with the 'Weitwinkel Flächen Photometer and Polarimeter' (WWFPP, see Reif et al.  1995) at the 1.2m telescope at the Calar Alto in 1995 and with the Hoher List Camera (HOLICAM, see Sanner et al.  1997) on the 1m telescope at our Hoher List observatory in 1996. The positions of the star and the galaxies were used in combination with the published APM scans of the POSS as first epoch data.

The CCD frames were reduced using DAOPHOT to determine the rectangular coordinates x and y. The coordinates of the APM scans were transformed to rectangular coordinates x and y and a classical astrometric reduction was performed in a local astrometric system. By subtraction of the mean apparent proper motion of the galaxies from those of the stars we obtained absolute stellar proper motions. The results for our twelve stars are given in Table 2. This method of getting absolute proper motions was first tested in the field of the quasar 3C 273 (Geffert et al.  1994).


[TABLE]

Table 2. Absolute proper motions of stars using background galaxies (POSS & new CDD data)


Since these proper motions are based on only one first epoch position, we are not able to calculate proper motion errors. An indication of the proper motion uncertainty may be the uncertainty of the zero point shift, representing a lower limit to our errors. In Table 2, [FORMULA] and [FORMULA] designate the mean uncertainty of the zero point shift represented by the r.m.s.-values of the apparent proper motions of the galaxies.

As best data from the literature we have taken the proper motions from the NPM catalogue (Klemola et al.  1987; K+87), because these proper motions were calibrated with respect to extragalactic objects too. In fact, Klemola et al.  were among the first to use background galaxies to arrive at proper motions in a true inertial system.

In all we have for 21 stars absolute proper motions based on extragalactic calibration.

As second priority data we have taken the proper motions of Colin et al.  (1994; C+94). The proper motions of this set have the highest internal accuracy (1-2 mas/yr). Since these data are based on the PPM catalogue (Röser & Bastian 1991) we may expect additional systematic errors of our proper motions due to the local inhomogeneties of the PPM catalogue. A full use of the internal accuracy of these data will eventually be possible through a rereduction with Hipparcos reference stars.

Finally, proper motions were determined by Thejll et al.  (1997; T+97) from recent accurate meridian observations and the published positions of the Astrographic Catalogue. Since they did not sufficiently describe from which reduction the old position was taken, and since these proper motions are based on only one or two first epoch plates, we used these proper motions with least priority.

For the stars common to these samples we have compared the proper motions. There are 11 stars in the list of T+97 common with the NPM (K+87). The mean of the differences with their r.m.s. deviations are [FORMULA] 6.0 mas/yr in [FORMULA] and [FORMULA] 8.3 mas/yr in [FORMULA] (in the sense NPM minus T+97). For the five stars common to the catalogues by Colin et al.  (1994) and Thejll et al.  (1997) we found mean differences (in the sense T+97 minus C+94) of [FORMULA] 3.8 mas/yr in [FORMULA] and [FORMULA] 2.3 in [FORMULA]. Since the error of a single proper motion in the NPM is of the order of 5 mas/yr, we expect from our comparison nearly the same accuracy for the proper motions of T+97. The better agreement between the catalogues of C+94 and T+97 may be explained by the fact that the local inhomogenities of the PPM catalogue will affect both methods in a similar way.

2.2. Radial velocities

Radial velocities are available for the stars from the spectra in the Bonn data base (see papers cited with Table 1). Several of them have already been published, in some cases they have been determined for this paper. For stars from S+94 radial velocities have been taken from Saffer (1994). The radial velocities have accuracies of the order of 30 km s-1.

For some stars it is known (see Table 1) that the radial velocities are variable. This may be a sign for binary nature of the objects (see, e.g., Theissen et al. 1995, Paper V). However, the physical parameters derived for the selected stars were beyond doubt, so that the distance is reliable. We have investigated the effect of changes in the radial velocities on the orbits, as described in Sect. 5.2, and found them only of minor importance in a statistical sense.

2.3. Distances

The distances of the stars have been taken from the literature, as indicated with Table 1. Distances are based on the determination of [FORMULA] and log g, the reddening corrected visual magnitude, and the assumption that the star has a mass of 0.5 M [FORMULA]. The distances are accurate to about 30%. For a discussion of the distance determination method see the papers cited with Table 1.

For 2 stars distances are based on Hipparcos parallaxes from de Boer et al. (1997; B+97).

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1997

Online publication: April 6, 1998
helpdesk.link@springer.de