2. Observations and data reduction
Originally, four EIS cluster candidates were selected for the VLT-UT1 SV program, after visual inspection of all candidates found in the EIS-wide Patch B. The four targets were selected to cover a range in redshift and richness among the EIS candidates. However, due to time and weather constraints only two fields were actually observed.
The optical observations were conducted on the nights of August 18 and 23, 1998 with the Test Camera of the VLT-UT1, as part of the ESO VLT-UT1 Science Verification (1998). The VLT-TC was equipped with an engineering grade Tektronix CCD, covering a field of view of about 93 93 arcsec with an effective pixel size of 0.09 arcsec (after a rebinning). One of the cluster candidates (EIS 0046-2930) was observed in VRI, and the other (EIS 0046-2951) only in the V and I passbands. In Table 1 we summarize the available data, giving the passband, the corresponding total integration time and the median seeing of the combined images. During the exposure of EIS 0046-2930 the transparency was poor and variable, leading to fairly bright limiting magnitudes. Single exposures have been reduced by the ESO Science Verification Team using standard IRAF procedures, and then publicly released. These reduced images were then processed using the EIS pipeline which performed the astrometric and photometric calibration, and coaddition for each band (see Nonino et al. 1998). The VLT-TC optical data were calibrated against the EIS data, for which the uncertainty in the photometric zero-point was estimated to be 0.1 mag in V and 0.02 mag in I. The VLT-TC versus EIS comparison yields an additional uncertainty of about 0.1 mag. Therefore, we estimate that the overall uncertainty in the zero-points is mag in V and 0.12 in I.
Table 1. Summary of VLT-TC and SOFI observations
The IR J and Ks band images were obtained on October 8 and 9, 1998 using the SOFI infrared spectrograph and imaging camera (Moorwood, Cuby & Lidman 1998) at the NTT. SOFI is equipped with a Rockwell 10242 detector that, when used together with the large field objective, provides images with a pixel scale of 0.29 arcsec, and a field of view of about arcmin. The full set of SOFI observations will be described elsewhere (Jorgensen et al. 1999); here we describe only those for the fields including the two cluster candidates. Total integration times and the seeing measured on the combined images are given in Table 1. The data were reduced using the Eclipse data analysis software package (Devillard 1998), developed to combine jittered images. The resulting combined images were then input to the EIS pipeline for astrometric and photometric calibrations using observations of standard stars given by Persson (1997). From the scatter of the photometric solution we estimate the zero-point uncertainty in the J and Ks bands to be mag.
In order to facilitate the analysis of the whole dataset, the images from the EIS-wide survey were resampled to a common reference frame, centered on the initial estimate of the two candidate cluster positions, using the Drizzle routine of the EIS pipeline. The resampled images have the same pixel size as the SOFI images. To improve the sensitivity to faint objects the resampled EIS-wide and SOFI images were combined to produce one very deep image for each field. This image has a sufficiently large field of view ( arcmin) to allow a reliable estimate of the background source density to be obtained (see Sect. 3). The source extraction software SExtractor (Bertin & Arnouts 1996) was subsequently used to detect sources in these deep images, while measuring the flux parameters for each individual passband in the separate images. Magnitudes and colors were measured using a 4 arcsec diameter aperture.
Also all available VLT images were coadded to produce the and images shown in Fig. 2. The resulting images are considerably deeper than those from EIS, and also have much better resolution. This procedure has allowed us to reach approximately the same limiting magnitude in both fields ( at about ). Even though the transparency during the observations of the EIS 0046-2930 field was poor, this was compensated by the addition of the band exposure. We have also resampled and combined the VLT-TC images with those from SOFI, using the same method as above.
Using the available multicolor data from EIS-wide plus SOFI, we derived CM-diagrams for all galaxies within 1 arcmin of the nominal candidate cluster centers. From these diagrams, a tentative color-based selection was made, dividing galaxies into cluster candidate members and foreground/background objects. Based on this selection, we computed for each cluster candidate a new position, obtained as the flux-weighted center of mass of the candidate members. An identical procedure was carried out using the SOFI data only, leading to very similar results. In both cases the new positions were found to be within 0.4 arcmin of the position given in the EIS catalog (Olsen et al. 1998b). Note that this corresponds to the pixel size (0.45 arcmin) of the maximum-likelihood map used in the EIS cluster finding procedure.
© European Southern Observatory (ESO) 1999
Online publication: March 1, 1999