Astron. Astrophys. 331, 193-210 (1998)

3. The ROSAT images

3.1. Image analysis

The ROSAT PSPC images each have exposures 10 ksec. Table 1 gives the (J2000) sky coordinates of the field centers, exposure times, galactic coordinates of the clouds and their adopted distances.

Table 1. Description of the fields observed by ROSAT

The energy range optimizing the signal-to-noise ratio (S/N) of the sources was established after examination of the detector background spectrum (Fig. 1). The "soft" band (0.1-0.4 keV) spectrum obtained from sourceless regions of the images cannot be distinguished from the spectra of the whole image including the sources. This probably has two causes: the background is high in this range, and the source soft emission is reduced by extinction. We have thus kept for analysis only the "hard" spectral band, 0.4 - 2.4 keV.

Fig. 1. ROSAT spectrum of the Mon R2 PSPC image, with (black squares), and without (open circles) the integrated source contribution. The sources are buried in the background below 0.4 keV, restricting the useful energy range to 0.4 - 2.4 keV. The figure would be almost exactly the same for the Rosette PSPC image.

Figs. 2a and 2b show the PSPC images of our target clouds. Sourceless regions in each image were selected, avoiding both the underexposed parts corresponding to the shadow of the mirror structure and bright diffuse areas (Sect. 3.2), to evaluate the mean background. In units of counts arcmin^-2 ksec^-1, the background level is 4.4 1.0 10^-2 for Monoceros and 4.2 0.9 10^-2 for Rosette. Figs. 2a and 2b also show enlargements of the central region, the inner 35' within the mirror support structure, which provides the best angular resolution and sensitivity. The PSPC images are displayed in the form of contours in Figs. 3a and 3b, superimposed on a digitized optical image (Sect. 4.1.1). These figures also show an outline of the PSPC field; the deviations from overall circular shapes are due to the wobbling of the satellite.

Fig. 2a. ROSAT observations of the Monoceros cloud in the 0.4-2.4 keV range. The entire image is obtained by smoothing the photon distribution with a Gaussian of =2 pixels (15"/pix). The 35' inner region is shown enlarged with a Gaussian smoothing of =1 pixel. The 3' 6' central region, indicated by an arrow, shows part of the JHK survey area of Carpenter et al. (1997).

Fig. 2b. Same as Fig. 2a, for the Rosette cloud. The delineated area corresponds to the near-IR survey of Phelps & Lada (1997).

Fig. 3a. ROSAT X-ray contours from the Monoceros field image of Fig. 2a superimposed on a digitized POSS(R) plate. The inner dashed lines show the ROSAT center field within the window support structure. The outer dashed lines show the effective field of view with the window support shadow.

Fig. 3b. Same as Fig. 3a. for the Rosette cloud area. The optical emission nebula is visible in the NW corner.

3.2. Source detection and extended features

Several tens of X-ray sources were detected in each image by using the ROSAT Standard Analysis Software System (SASS). Some have low ratios and could be spurious. Following FCMG and CMFA, we have measured the local of each source, and other weak localized excesses that were missed by the SASS process. Adopting a criterion that reliable X-ray sources have , we found 41 sources for Monoceros and 21 for Rosette. Seven (Monoceros) and eight (Rosette) excesses have , and some are positionally well-correlated with IR and/or visible sources (Sect. 4.2). In view of the Poissonian nature of the PSPC noise, a few of these low S/N sources must be real. In order not to introduce uncertainties in the statistics, however, the low S/N sources will not be used in our present analysis, but we give their list for possible future reference.

The resulting source lists are given in Table 2 (Monoceros) and Table 3 (Rosette). In these tables, col. [1] gives the running number of the ROSAT sources in the form "Mon X-n " and "Rosette X-n "; col. [2] the RXJ ROSAT source designation; cols. [3] and [4] give their equatorial coordinates (in J2000), corrected for boresight error in the ROSAT aspect solution (Sect. 4); col. [5] gives the positional errors in arcsec provided by the SASS; col. [6] gives the off-axis distance in arcmin; col.[7] gives the count rates measured in the 0.4 - 2.4 keV band and their statistical errors; col.[8] gives the obtained as described in FCMG; col.[9] gives the logarithm of the approximate X-ray luminosity and their statistical errors, derived assuming a typical average extinction (Sect. 5.1). The other columns correspond to optical stellar identifications described in Sect. 4 below.

Table 3. a. ROSAT sources detected in Rosette field

In addition to the point sources, two types of extended features are present. First, a large diffuse feature is seen near the NNW edge of the Rosette image. The point spread function is quite extended here, so it is a priori not possible to distinguish between unresolved emission of faint point sources and a truly diffuse emission. The problem is particularly relevant because the X-ray structure coincides with the edge of the optically bright HII region excited by the O star cluster. Second, bright "hot spots" are seen closer to the axis in both fields, corresponding to tight clusters of partially resolved X-ray sources. The nature of these hot spots will be discussed in Sect. 6.

© European Southern Observatory (ESO) 1998

Online publication: February 4, 1998
helpdesk.link@springer.de