During the past decade, the photometric and spectroscopic surveys have allowed
to improve our knowledge of galaxy formation and evolution.
Since the 1980's, the new technology based on CCD detectors has improved the
photometric efficiency (sensitivity, linearity, high dynamic range,...) as compared
to photographic plates. Although the photographic plates are un-rivaled in their
ability to cover large areas of the sky,
the large improvements brought by digital surveys, both in the photometric accuracy
and in the faint limiting magnitude which can be reached, have allowed
important new
insight into the properties and the evolution of the different galaxy
populations.
Several deep digital surveys (Tyson 1988;
Lilly et al. 1991; Metcalfe 1991, 1995;
Driver et al. 1994; Smail et al.
1995) were performed in different regions of the sky (with a typical field size
0.012 deg
) at faint
or very faint magnitudes 
.
First, these surveys show that the number-counts of galaxies at
are in excess with respect to a non-evolving model and the
disagreement increases with the apparent magnitude. Second, the galaxy colour
distributions become significantly bluer at fainter magnitudes.
Several models have been elaborated based on the
cosmological parameters (
, 
) and the luminosity function parameters
(
) allowing a good fit to the photometric data
(see Koof Kron 1992 for a review).
Models with pure luminosity evolution where 
evolves with look-back time (Bruzual 1983;
Guiderdoni & Rocca-Volmerange
1990; Yoshii & Takahara 1988) predict
that the tail of the redshift distribution of very faint galaxies should be
extended towards high redshifts.
This is not observed in the recent redshift
surveys to 
24 (Colless et al. 1990, 1993;
Cowie et al. 1991;
Lilly et al. 1991; Tresse et al. 1994;
Glazebrook et al. 1995a),
which are in good agreement with the redshift
distribution expected for a non-evolving model.
Models with number-density evolution (Rocca-Volmerange &
Guiderdoni 1990; Broadhurst et al. 1992)
are based on a population of dwarf
galaxies at 
which would have merged into brighter galaxies by
.
However, these models are difficult to reconcile
with both the recent observations of weak clustering in the correlation function
of faint galaxies
(Efstathiou et al. 1991 & Roche et al. 1993)
and with the physical mechanisms for merging
(Ostriker 1990;
Dalcanton
1993). The most
recent models use a new estimation of the slope 
of the local
luminosity function
(Koo et al. 1993; Driver et al. 1994) which is assumed
to increase from 
to 
by the presence
of a large number of dwarf galaxies 
. This model is supported by the recent
observations of the Medium Deep Survey with the HST to I=22
(Glazebrook et
al. 1994b), where the counts of morphologically normal galaxies are well fitted
by a non-evolving model
and where a large excess of Irregular and Peculiar galaxies is detected which could
contribute to the excess of blue galaxies.
Faint photometric and spectroscopic surveys also provide maps of the
distribution of galaxies in three dimensions. The nearby surveys show that galaxies
are distributed within sharp walls delineating voids with diameters between 20
and 
Mpc (de Lapparent et al. 1986;
Geller & Huchra 1989;
da Costa et al. 1994) (where 
).
Very deep pencil-beam surveys were obtained in particular directions of the sky
(Broadhurst et al. 1988;
Colless et al. 1990, 1993; Cowie et al. 1991;
Lilly et al. 1991; Tresse et al. 1994
& Glazebrook et al. 1995a), and in
some of these an apparent periodicity on scales of 
Mpc
has been detected (Broadhurst et al. 1990).
At these depths (
, the large amount of time required
to obtain the redshift distribution for a complete magnitude-limited sample
constrains observations to
narrow solid angles. Although the existing deep pencil-beam probes are adequate for
establishing the evolutionary history of galaxies, biases caused by sparse
sampling may affect the data when
used to study the large-scale structures (de Lapparent et al. 1991;
Ramella et al.
1992).
In this context, a deep redshift survey near the southern galactic
pole was started with the main goal to characterize
the large-scale structure at large distances (de Lapparent et al. 1993).
The spectroscopic survey covers a continuous solid angle of 0.3 deg
and
contains 
galaxies with 
(i.e. 
.
The median redshift is at 
.
The entry photometric catalogue for the redshift survey was obtained by
observing in the B, V, R photometric bands up to
24.5, 24.0, 23.5 respectively in a longer strip of 
(this area was
not fully covered by spectroscopic observations).
The photometric data provide an adequate sample for measuring
with a high confidence level the galaxy number counts and
the distribution of galaxy colours.
The description of the spectroscopic sample of the ESO-Sculptor survey is given
in Bellanger et al. (1995a) and the first results about the properties of the
large-scale structure are shown in Bellanger & de Lapparent (1995b).
Here we describe in detail the procedures used in the reduction and analysis of the photometric sample. The paper is organized as follows. In Sect. 2, we describe the photometric observations. Sections 3 and 4 outline the data reduction procedures and analyses. In Sects. 5, 6 and 7 we discuss the transformation into astronomical coordinates, the photometric calibration of fields and the magnitude transformations into the Johnson-Cousins standard system. In Sect. 8 we present the method used to match the photometry over the whole survey in each band. In Sect. 9 we show the first results on the star colour distributions, the galaxy colours and the galaxy number-counts. Finally, in Sect. 10 we summarize the major steps of our photometry and we present the scientific perspectives for the near future.