4 The catalogue and initial results

   

Table 2: Format of the object catalogue made available at CDS.

Label	Explanation
Seq	Sequential number
RA	Right ascension (J2000), internal accuracy $\pm0\hbox{$.\!\!^{\prime\prime}$ }15$
DE	Declination (J2000), internal accuracy $\pm0\hbox{$.\!\!^{\prime\prime}$ }15$
$B_{\rm mag}$	Aperture magnitude, scaled to total flux for stars
$e-B_{\rm mag}$	Mean error (sigma) of $B_{\rm mag}$
$V_{\rm mag}$	Aperture magnitude, scaled to total flux for stars
$e-V_{\rm mag}$	Mean error (sigma) of $V_{\rm mag}$
$R_{\rm mag}$	Aperture magnitude, scaled to total flux for stars
$e-R_{\rm mag}$	Mean error (sigma) of $R_{\rm mag}$
$ApC_{\rm mag}$	Estimated correction from aperture to total mag.

The object catalogue derived from the observations presented contains positions and BVR photometry of 63501 objects selected in R within a field of  $31\hbox{$.\mkern-4mu^\prime$ }5 \times 30\hbox{$^\prime$ }$ size. The depth and the seeing quality of our R-band imaging makes this catalogue potentially very useful for the scientific community. Therefore, the catalogue (format see Table 2) is available to the public at Centre de Données astronomiques de Strasbourg (CDS, http://cdsweb.u-strasbg.fr/ cgi-bin/qcat?J/A+A/377/442) and on the COMBO-17 survey homepage at MPIA (http://www.mpia.de/ COMBO/). In the following we discuss data quality issues and present a first sample of variable objects identified from the two epochs of R-band observations.

The quality of our photometry differs for point sources and extended sources. Essentially, it is a seeing-adaptive central surface brightness measurement giving accurate fluxes for point sources while underestimating the total flux of extended sources. But since it is performed on the individual frames in an optimal seeing-adaptive fashion, it yields more accurate colors and estimates the errors more realistically than measurements on a single co-added frame.

Figure 4: R-band histogram of objects with errors of $\sim$10% from combined photometry of both observing runs. From this distribution we read the 10$\sigma$ limiting magnitude as the median at $R=25\hbox{$.\!\!^{\rm m}$ }25$.

Figure 3 shows the photometric errors versus magnitudes of all objects we measured at less than  $0\hbox{$.\!\!^{\rm m}$ }1$ error. The photon noise limit can be seen as a sharp parabolic edge to the right of the object clouds. We use magnitude histograms of objects with errors of $\sim$10% to assess a representative 10-$\sigma$ magnitude limit for point-source photometry. In Fig. 4 we can see that all R-band images combined reach $R_{{\rm lim, 10}\sigma} = 25.25$ which is $R_{{\rm lim, 5}\sigma} = 26.00$.

Figure 5: Distribution of aperture correction magnitudes versus R magnitude. Stars are at zero level while extended objects reach down to negative values.

Figure 6: Number counts histogram of all objects excluding stellar objects with R<23. The lower staircase line is based on the aperture magnitude which is accurately calibrated for point sources. The upper staircase line is based on total magnitudes corrected for aperture effects, while the straight black line with a slope of 0.39 has been fitted to the latter staircase distribution. Note, that this plot contains $\sim$60000 objects and shows their counts in bins of  $0\hbox{$.\!\!^{\rm m}$ }1$.

Figure 7: B-V and V-R colors of observed stars compared to template colors. Shown are bright point sources from the object catalogue ( $R=17\ldots 21$, grey dots) and colors for the Pickles (1998) spectral library (black dots). Most stars in our field belong to the halo population, while most of the Pickles library are nearby stars from the disk population. The populations form two arms that are clearly separated for G and K stars ( $B-V \approx 0.3 \ldots 0.6$). We note a number of blue objects off the main sequence, which are probably quasars passing this purely morphological selection. The units of the colors are photon count color indices ( $CD_{\rm mag}$) as in Wolf et al. (2001), see also text.

However, the fluxes of extended objects are approximated by an aperture correction. The median correction among extended objects amounts to $\sim$ $-0\hbox{$.\!\!^{\rm m}$ }5$ (see Fig. 5). Figure 6 shows a number counts histogram for the aperture magnitude as well as for the corrected total magnitude excluding point sources at R<23. The counts in aperture magnitude suggest that our object list is at least complete to R>25 in terms of point-source photometry. The counts of total magnitude suggest completeness among galaxies provided to at least $R\approx 24.5$. With a slope of $\sim$0.39 the counts are consistent with galaxy counts from the literature. However, a detailed discussion of the counts is beyond the scope of this paper.

We checked our flux calibration by comparing measured colors of stellar objects with those predicted by synthetic photometry. We convolved the Pickles (1998) library of stellar spectra with the total efficiency curves of our filters and plotted their B-V and V-R colors as black dots in Fig. 7. Our own point sources are overplotted as grey dots and agree with the expected colors without any further correction. Shown are photon count color indices in units of $CD_{\rm mag}$ defined by Wolf et al. (2001b). As a physical magnitude definition besides $AB_{\rm mag}$ and $ST_{\rm mag}$ the $CD_{\rm mag}$ is

$$CD_{\rm mag} = -2.5\log {F_{{\rm phot}}} + 20\hbox{$.\!\!^{\rm m}$ }01$$ (1)

$$\quad\mbox{with} ~ F_{{\rm phot}} ~ \mbox{in $\gamma$\,m$^{-2}$\,s$^{-1}$\,nm$^{-1}$ } ,$$

and matches the common magnitude zeropoint of (astronomical) Vega-mag, $AB_{\rm mag}$ and $ST_{\rm mag}$ at $\lambda_0 = 548$ nm.

We note that most stars observed in our field belong to the halo population, while most stars in the Pickles library are nearby stars from the disk population. The populations form two arms that are separated for G/K stars ( $B-V_{\rm CD} \approx 0.3 \ldots 0.6$) and have quite different relative population densities in the data and the library. We also note a few blue objects off the main sequence, which are most likely quasars passing the purely morphological selection used here.