A ROSAT medium-sensitivity Galactic plane survey at 180° < l < 280°

0 Osservatorio Astronomico di Palermo, Piazza del Parlamento 1, 90134 Palermo, Italy 1 Astrophysics Division, Space Science Department of ESA , ESTEC, Postbus 299, 2200 AG Noordwijk, the Netherlands 2 X-Ray Astronomy Group, Department of Physics and Astronomy, University of Leicester , Leicester LE1 7RH A B S T R A C T We have performed a moderately deep soft X-ray 0:1-2 keV survey of the Galactic plane using pointed observations with the ROSAT Position Sensitive Proportional Counter (PSPC). The survey is more than an order of magnitude more sensitive than previous X-ray surveys near the Galactic plane. The data consist of nine fields each of <10 ks exposure, pointed at positions on or very close to the Galactic plane bjj , 08:3 in the longitude range 1808 & l & 2808. This region has relatively low X-ray absorbing material out to distances of several hundred pc and presents fewer source-confusion problems than at other longitudes. The total sky area surveyed was 2.5 deg2; this yielded 93 sources, 89 of which were detected in the 'hard' 0:4-2:0 keV band. Nine sources were detected in both 'soft' 0:1-0:4 keV and hard bands. In the hard band, the survey coverage is *90 per cent for sources brighter than 0.002 count s21 , 2 10214 erg cm22 s21 , but falls steeply below this value, with the weakest sources being , 0.001 count s21. The median limiting flux is <0.0013 count s21 , 1:3 10214 erg cm22 s21 . There are 64 sources with hard-band count rates . 0.002 count s21. We present the catalogue of X-ray sources and the number - flux relations log N-log S . Eighteen sources have possible identifications from the SIMBAD data base. We have searched the Tycho-2 and USNO-A2.0 catalogues to find all possible optical counterparts brighter than , 19th magnitude, and attempt to classify these on the basis of log f X/ f opt versus optical colour diagrams and near-infrared photometry from the 2MASS Second Incremental Data Release. Hence, we have found the majority of these sources to be consistent with being late-type main-sequence stars, as previous studies have proposed from incompletely identified surveys. Comparison of the measured number - flux relations with predictions of Galactic (stellar) and extragalactic populations supports the view that the population of young stars in the plane is denser than previously thought. I N T R O D U C T I O N 900627 200084 900426 900628 900380 900428 200049 900427 400329 5 45 37.6 6 15 15.0 6 42 25.8 7 02 32.3 7 18 29.0 7 23 18.9 7 24 40.1 8 10 02.8 10 24 58.5 28 56 10.0 17 43 41.0 04 23 17.0 205 23 54.0 213 13 03.0 215 08 33.0 216 12 03.6 233 16 56.0 257 47 00.5 Texp Gal. plane 1 Target=G 104-27 Gal. plane 2 Gal. plane 6 Target=NGC 2359 Gal. plane 3 Target=HD 58343 Gal. plane 4 Target=1E 102457 to , 10214 erg cm22 s21, and has yielded 93 sources. Hence it complements well not only the EGPS and RGPS, but also ROSAT surveys at similar depths at high Galactic latitudes (e.g. Hasinger et al. 1993; Branduardi-Raymont et al. 1994; Boyle et al. 1995). The longitude range selected for our survey has relatively low X-ray absorbing material out to distances of several hundred parsecs, and presents less confusion problems than at other longitudes. The restriction in latitude essentially removes scaleheight effects when attempting to interpret the spatial distribution of sources, since jbj 08:3 corresponds to a height of only 5 pc off the plane at a distance from the Sun of 1 kpc. Previous surveys have typically found , 8090 per cent of identified sources to be active coronae, but left , 40 per cent of detected sources unidentified [although Motch et al. (1997) were able to identify 98 per cent of sources at a threshold of 0.03 PSPC count s21, < 3 10213 erg cm22 s21 . We try to classify the sources in our (deeper) survey on the basis of X-ray to optical flux ratios and optical and near-infrared colours, to find the percentage that are consistent with being due to active coronae. The paper is organized as follows. In Section 2 we describe the data selection and the source-detection method. The determination of the number flux relation is described in Section 3. Searches for optical counterparts are described in Section 4, and the methods used to classify these are detailed in Section 5. We compare our results to model predictions in Section 6, and summarize them in Section 7. O B S E R VAT I O N S Data selection The survey used nine ROSAT (Trumper et al. 1991) PSPC (Pfeffermann et al. 1986) fields, each of <10 ks exposure, to sample the Galactic plane at longitudes from roughly 1808 to 2708 (the third quadrant).1 A log of the observations is given in Table 1, while Fig. 1 shows the distribution of the fields on the sky. Five of the pointings (numbers 1, 3, 4, 6 and 8) were specifically performed for this programme, while the remaining four data sets were obtained from the archive.2 The target objects for these latter pointings are noted in Table 1. The field positions were selected to avoid molecular clouds, strong (*0.2 PSPC count s21) known X-ray sources, bright stars and supernova remnants (SNRs), all of which might distort the survey. Each observation was required to have at least 8 ks exposure and to 1 Some workers, e.g. Paresce (1984), refer to this as quadrant 4 or the Source detection Our analysis was restricted to the inner 18-arcmin radius of each field. This utilized the portion of the PSPC field of view (FOV) with the highest sensitivity and spatial resolution, and avoided potential problems due to shadowing by the window support structures. Those fields chosen from the archive also had the central 3.5-arcmin radius excluded from the analysis to prevent bias due to the central objects in these fields having been selected for their interesting nature.3 Data intervals with poor aspect solutions (error *1 arcsec) or high anticoincidence rates (Master Veto Rate . 170) were also rejected. Images were made in three photon energy bands: total 0:12:0 keV, PI channels 11202 , soft 0:10:4 keV, PI channels 1142 and hard 0:42:0 keV, PI channels 43202 . Each image had dimensions of 200 200 pixel2, with each pixel being 15 15 arcsec2. Sources were detected using the point-source search program PSS.4 PSS required as input a data image and a corresponding background map. The background map (one for each field and for each energy band) was constructed as follows. The average count rate was measured within the central 18-arcmin radius of an image. This was used for a preliminary PSS run. Detections made in this first run with a significance . 4.0s were then subtracted, and the resulting image smoothed using a 20 20 pixel2 top-hat function. This was used as the background model for a second (and final) PSS run where detections were accepted as X-ray sources if they exceeded a significance of 4.5s. A total of 93 sources was detected; however, most of these (89) were also detected in the hard band and only 12 were detected in the soft band (three (...truncated)