Astron. Astrophys. 317, 25-35 (1997)

2. Observations and data analysis

The X-ray source RX J0057.3-2222 was repeatedly observed with the XRT/PSPC telescope (Aschenbach, 1988; Pfeffermann et al., 1987) aboard the ROSAT satellite (Trümper 1983): once during the all-sky survey (RASS) in 1990 (Thomas et al., 1991) and several times during pointed observations at four occasions in 1992 and 1993. The detailed observation log is given in Table 1 together with the exposure and the registered source strength for each individual observation. The listed X-ray data refer to ROSAT's energy band (0.1 - 2.4)keV. The total exposure, spread over two and a half years, is 19648 s, during which more than 80.000 counts were accumulated.

Fig. 1. Sky field around Ton S 180 taken from the Digitized UK Schmidt Sky Survey. In addition, the 90% confidence error circles of the X-ray positions from the RASS (large circle) and from the merged ROSAT pointed observations (small circle) are shown. (Copyright 1993, 1994, Association of Universities for Research in Astronomy Inc.)

Fig. 2. X-ray image of RX J0057.3-2222 and its surroundings and the background field. a central part of the field of view with the target source near the center. b geometry of the sky field used to determine the X-ray background in the vinicity of the target source.

The celestial positions of the X-ray source as derived from the centroids of its X-ray images are for the RASS observation and for the merged pointed observations. The most precise measurement of the optical position of Ton S 180 found in literature are the Kiso coordinates (Kondo, Noguchi, and Maehara, 1984). The deviation of between the optical and the RASS X-ray position is well within the 90% confidence error circle of 32" radius (Fig. 1), which was determined for the positions of survey sources by comparison with optical star catalogues (Voges, 1992). The 90% confidence error circle for the merged pointed observations, also given in Fig. 1, comprises a statistical error of resulting from a maximum likelihood source detection algorithm and an intrinsic uncertainty of 6" of the attitude solution. In order to establish the X-ray data sets for the temporal and spectral analysis, the image of the source was extracted from the X-ray image of the sky field scanned by the telescope during the survey or viewed in pointed observations. The image reconstruction of the X-ray sky was achieved using the attitude solution. Each registered X-ray count was corrected for vignetting according to the off-axis angle of its impact point onto the PSPC. The extraction radius was chosen according to radial profiles of the point source's image. It is 2' in the RASS case and 3' for the brighter images of the pointed observations. For the RASS observation the background level was determined from two circular fields, free of sources and located on the scan path near the target object. The survey background surface brightness amounted to .

The X-ray data of the pointed observations consist of 15 subsets (Table 1) corresponding to 15 individual ROSAT orbits with a typical target exposure of about 1500 s. To obtain a reliable background field, we merged the sub sets to a total X-ray image. Its central part is reproduced in Fig. 2a. As can be seen from the picture, the target source RX J0057.3-2222 (Ton S180), located near the center of the field of view (FOV), is surrounded by numerous other X-ray sources. In addition, the shadow of the central annular supporting structure of the PSPC's entrance window is faintly visible in the X-ray image. To avoid this affecting of the image analysis, a circular sky field with a radius of 19' was cut out from the central FOV. Subsequently, this image field was subjected to a maximum likelihood source detection algorithm to identify all X-ray sources which were detected with a likelihood of more than 10. In total, 27 X-ray sources were detected and removed from this image section. After extracting the target source, the geometry of a background field was obtained as shown in Fig. 2b. In analyzing the individual data sets of each orbit this background geometry was always applied. The mean background surface brightness of the pointed observations as derived from the merged data set is . With reference to the source extraction area the mean background rate is therefore , about a factor of hundred lower than the mean source count rate corrected for instrumental effects (vignetting and dead time corrections).

© European Southern Observatory (ESO) 1997

helpdesk.link@springer.de