Astron. Astrophys. 358, 88-94 (2000)

2. Observations and data reduction

The observations were performed during two observing runs in September and October 1998 with HiRAC (High Resolution Adaptive Camera) on the 2.56 m Nordic Optical Telescope. The CCD used was a 2048² back-side illuminated thinned Loral with a pixel size of 0.1082 arcsec.

All 6 nights during which the observations were performed were photometric and with good seeing (median 0.7 arcsec FWHM in the I-band). Integration times in U were 1000-1500 sec in order to avoid the total noise to be dominated by readout noise. In I and B the integration times were 250-300 sec and 300-500 sec respectively. Between exposures the telescope was moved 2-4 arcsec to minimize the effects of bad pixels and fringing. The total integration times in each filter are given in Table 1.

Table 1. Observations of Q0151+048, Sept 17-20 and Oct 17 - 18, 1998

Also observed were several standard star sequences from Landolt (1992) and photometric solutions were obtained for each filter. For counts given as electrons per second we derive zero-points in the Landolt system of 22.77, 25.13 and 24.48 for u, B and I respectively. All magnitudes subsequently quoted in this paper are on the AB system. For the U-band we determined the colour equation relating the instrumental magnitude u to the standard Johnson U. There was substantial scatter around the fit near of mag., which is typical for this band (e.g. Bessell 1990). The instrumental magnitudes u were converted to AB magnitudes using the equation , determined by integrating the spectrum of the star GD71 over the passband. Here we have retained the lower case u for the AB magnitude indicating that the effective wavelength of the filter lies significantly away from the standard value. The colour term for the I and B filters are consistent with zero i.e. and , and we used the equations and (Fukugita et al. 1995) to put the I and B magnitudes onto the AB system.

The data were bias-subtracted, and twilight sky frames were used to flatten the frames in the standard way.

For the deconvolution, the individual bias subtracted and flattened I-band images were divided into six groups (chronologically) and the frames of each group were then combined using the optimal combination code described by Moller & Warren (1993), which maximizes the signal-to-noise ratio for faint sources. These six combined images were used in the simultaneous deconvolution process (see Sect. 3.1 below). Furthermore, all the individual I-band images were combined into one combined image, which was used in the PSF subtraction described in Sect. 3.2. In the same way we divided the individual bias subtracted and flattened B-band frames in four groups and calculated combined images for each group and for all images. Finally we combined the ten individual bias subtracted and flattened U-band frames into one combined image. The details of the sky noise in the combined images are provided in Table 2.

Table 2. Measured sky level and rms of sky surface brightness.

© European Southern Observatory (ESO) 2000

Online publication: June 26, 2000
helpdesk.link@springer.de