Astron. Astrophys. 335, 277-280 (1998)

3. Data-analysis

Cosmic rays were removed by comparing the multiple images per filter before we constructed the image shown as Plate 1. However, owing to the short exposure times, particle hits did not significantly disturb any of the stellar images of interest to us. Consequently, we applied the standard photometric and astrometric iraf tasks tasks to each of the four available images individually. This provided us with valuable information on consistency and repeatability. The accuracy of the individual positions in the HST celestial reference frame, as shown in Table 1, is estimated at 0.1 px (or 0.005 arcsec) for the central star of NGC 650 (No.1), and double that value for the companions (Nos. 2 and 3). The internal photometric errors are very small at a few thousands of a magnitude, but several additional factors need to be taken into account. In the first place, we had to use a very small extraction aperture (of 3 px, or 0.14 arcsec ) to separate the components 2 and 3. In contrast, WFPC2 photometric calibrations (Holtzman et al. 1995) are based on 1 arcsec images. We estimate the additional uncertainty introduced by this procedure, based on the PSF of the sufficiently isolated component 1, at no more than 0.03 mag. Some overlap of the PSF's occur for the stars 2 and 3, but the contribution to our extraction aperture, for which we corrected, is never more than 0.15 mag (i.e. to the I magnitude of No. 3). Finally, we converted the natural WFPC2 magnitudes to standard V and I values using the (2nd order) colour-transformation scheme of Holtzman et al. 1995. This also adds a small contribution to the error budget. We conclude that the zero-points of the magnitudes shown in Table 2 are probably correct to 0.05 mag, while the colours are a little better.

Table 1. HST/WFPC2 observations of the three resolved stellar components

Table 2. Intrinsic stellar colours for the three resolved stellar components of NGC650 assuming = -0.47 for the central star corresponding to E = 0.16 and A = 0.5.

The photometry in Table 1 may be compared to values given in the literature. Cudworth (1973) gives a photovisual magnitude of 17.5 for the central star and 17.7 for the companion. Shaw & Kaler (1985) give a value of V = 15.87 0.36 for the combination of stars, which is too bright. But they are measuring with a 40 diaphragm which not only allows a large amount of nebular light to enter the photometer (which they had to correct for) but measures the light of other stars in the field as well. As can be seen in Plate 1, there are several bright stars in a 40 field of view. Napiwotzki & Schönberner (1995) give V = 17.7 0.3 by measuring the continuum of spectrum of the central star. This value is only marginally fainter than our measurement, which might be caused by a small tracking error during the exposure through the slit of their spectrograph.

© European Southern Observatory (ESO) 1998

Online publication: June 12, 1998

helpdesk.link@springer.de