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


Astron. Astrophys. 332, 93-101 (1998)

Previous Section Next Section Title Page Table of Contents

4. Positioning the radio source within the centre of M 3

Within the precise VLA coordinates system, the position of the radio source (Kulkarni et al. 1990) is:

[EQUATION]

[EQUATION]

with a quoted error of [FORMULA] (on each axis).

At optical wavelengths, in the von Zeipel (1908) reference frame, the coordinates of AC999 (a bright star close to the adopted cluster centre) can be derived from Aurière and Cordoni (1983):

[EQUATION]

[EQUATION]

.

The relative position of the radio source with respect to AC999 (labelled "a" in Fig. 1) is then derived at:

[EQUATION]

[EQUATION]

However, Aurière & Cordoni (1983) have pointed out that the coordinates system adopted by von Zeipel (1908) differs by a few arcsec from that of Sawyer Hogg (1973). Gunn & Griffin (1979) have also noticed that for the star vZ807, the system used by von Zeipel differs from that of Barnard (1931) by 1.8 and [FORMULA] in R.A. and DEC. respectively, suggesting that the Right Ascension derived in the von Zeipel reference may be uncertain by an appreciable amount.

None of the recent reference catalogues of positions like the Automatic Plate Measuring System (APM) of the Institute of Astronomy at Cambridge or the Positions and Proper Motions catalogues (PPM1) (Roeser & Bastian, 1988) has any star within the field of the PC1 on which we can rely to set up a new safe coordinates system. The nearest PPM star is actually at [FORMULA] from the centre and no star closer than [FORMULA] from the cluster centre has been measured in the APM.

Using a different approach, we made an independent re-determination of the coordinates of AC999 in the APM system. To do this, we used the catalogues of stars measured in M 3 with known counterparts in the APM. First, the ground-based photometric catalogue of the central part of M 3 (Ferraro et al. 1997e) and the photometric catalogue of Buonanno et al. (1994) of the cluster outskirts were merged into one single catalogue. Besides magnitudes and colors, the two catalogues provide also the relative positions of stars with respect to a common origin within better than one tenth of an arcsec (the origin does not actually correspond to any star but falls in the immediate vicinity, [FORMULA] north of AC972). The merged catalogue lists the star AC999 and extends sufficiently far from the central part [FORMULA] to include a significant number of APM stars. For 309 matched stars, the relative positions of the merged catalogue were directly listed with their corresponding APM coordinates. The coordinates of the common origin were then re-determined as a free parameter from the set of equations, in the APM system of coordinates. The resulting gaussian distributions of the coordinates for the origin have the following mean values:

[EQUATION]

[EQUATION]

with a standard deviation [FORMULA] in [FORMULA] and [FORMULA] in [FORMULA] respectively. Since the star AC999 (#10852 in Ferraro et al. 1997e) is [FORMULA] and [FORMULA], hence the APM (B1950) coordinates of AC999 would be:

[EQUATION]

[EQUATION]

and the difference between the radio source and AC999 would be:

[EQUATION]

[EQUATION]

the corresponding position is labelled "b" in Fig. 1 and in Fig. 5.

The APM plates (Irwin & Trimble 1984) were originally aligned with respect to PPM astrometric standards. In the region of M 3, for 11 APM stars which have their counterpart in the PPM1, we have measured a mean difference of:

[EQUATION]

[EQUATION]

Although the dispersion is large and the number of stars in common is limited, the two catalogues are in reasonable agreement. Hence, the determination of the coordinates of AC999 and the position of the radio source are consistent with the PPM1.

Finally, an additional and independent determination of the location of the radio source within the PC chip was performed by using the IRAF task INVMETRIC which transforms the celestial coordinates into WFPC2 pixel coordinates making allowance of the geometrical distortions. In the HST image, the radio-source ([FORMULA] [FORMULA]) is positioned at pixel [FORMULA], which is nearly coincident with the position found if one uses the APM reference coordinates (position "b" in Fig. 1).

In conclusion, also given this remarkable coincidence we will adopt in the following, position "b" as the most probable location of the radio source in the PC1 frame.


[FIGURE] Fig. 5. The enlarged region of the best estimate of the radio source position ("b") is displayed at all wavelengths. The nearest faint feature (arrow) at UV wavelengths is actually a blend of faint regular main sequence stars, best seen in the visible filters; no variability over the period of observation nor an unusual color is found for these two stars.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1998

Online publication: March 10, 1998
helpdesk.link@springer.de