Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 320, 378-394 (1997)

Previous Section Next Section Title Page Table of Contents

2. Data reduction

M 82 was observed three times early in the ROSAT mission (Trümper 1984) by both the PSPC and the HRI (Table 1). Only the PSPC and the longer of the two HRI observations are used in the present analysis. The use of the HRI's good spatial resolution ([FORMULA] FWHM) to complement the spectral information available at moderate resolution ([FORMULA] at 1 keV) from the PSPC, is advantageous, especially with regard to clarifying source confusion. The data sets were obtained from the Leicester Data Archive (LEDAS) and were analysed using the Starlink ASTERIX X-ray analysis system.


Table 1. ROSAT observations of M 82. Although the first HRI observation rh600021 was ostensibly taken within a couple of days of the PSPC observation, only [FORMULA] seconds were taken in March 1991, the rest being taken in early May 1991.

2.1. Background subtraction of PSPC data

The data were cleaned of periods of high background (both particle and Solar) and poor pointing stability, leaving 21194 seconds of good data. A spectral image (or data cube) was formed over a [FORMULA] region centred on the [FORMULA] nucleus, with a pixel size of [FORMULA] and 22 energy bins between channel numbers 11 and 230 (corresponding to roughly 0.11-2.3 keV). A model of the background was constructed using data from an annulus [FORMULA], centred on M 82, with contaminating point sources removed. The particle contribution to the background was estimated using the master veto rate (Snowden et al. 1992) and the remainder was corrected for energy-dependent vignetting, to give a spatial-spectral model of the background covering the entire field. This model was then adjusted (by 3%) as a result of an iterative process of further source searching using the PSS programme (Allan 1995), removal of sources from the background annulus, and rescaling of the background to achieve a drop to zero surface brightness away from M 82. The resulting background subtraction should be accurate to 2%.

2.2. HRI reduction

The HRI's resolution and relative insensitivity to diffuse emission, make it an ideal instrument for investigating point sources in the field, clarifying the PSPC analysis. The data were binned into an image of size [FORMULA] centred on the [FORMULA] nucleus, with a [FORMULA] pixel size to exploit the HRI's superior spatial resolution. Given the flatness of the HRI vignetting function within the region of interest, no vignetting correction was applied. Source searching is described below.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1997

Online publication: June 30, 1998