1. Introduction

Large sky surveys are of fundamental importance to astronomy. They provide an overall description of the universe and the generic properties of its many different constituents. Furthermore, complete and unbiased surveys are an indispensable tool in creating well defined samples of particular objects.

In the radio regime there is a long history of ever more sensitive and accurate large sky surveys over a wide range of frequencies. Both the Westerbork Synthesis Radio Telescope (WSRT) and the Very Large Array (VLA) are currently dedicating substantial parts of their observing time to conducting new large-scale radio surveys. During the next few years, exploitation of these surveys should produce an enormous amount of new information about the radio universe.

The Westerbork Northern Sky Survey (WENSS) is a new low-frequency radio survey, designed to cover the whole sky north of declination at a wavelength of 92 cm (325 MHz), and about a quarter of this region, concentrated at high galactic latitudes, at a wavelength of 49 cm (609 MHz), to a limiting flux density of approximately 18 mJy () and 15 mJy respectively. The products from WENSS are maps and source lists for all four Stokes parameters (I, Q, U, V). Maps will be produced at a resolution (FWHM of the restoring beam) of at 92 cm and at 49 cm. The positional accuracy for strong sources is 1.5'' at both 92 cm and 49 cm. WENSS will distribute its maps in a standard format. These maps we call frames.

To carry out this survey in a reasonable amount of time, WENSS utilizes the mosaicing capability of the WSRT. Exploiting this technique, a pattern of 80 evenly spaced fields is observed at regular intervals over several 12 hour syntheses with different array configurations. In this way it is possible to sufficiently sample the visibilities for all 80 fields and thus reconstruct a map of the sky that is many times larger than the field of view of the WSRT ( HPBW at 92 cm). These maps are referred to as mosaics.

   

WENSS 87GB NVSS FIRST 6C 7C 8C

Sky Area (sr) 0.7 3.1 6.1 10.3 3.1 2.8 4 0.8

Lim. Flux density 18 15 18 2.5 1 300 150 1000

Resolution 45'' 5''

Positional uncertainty 1.5'' 10'' 1'' 0.1'' 5''-10'' 1'' 30'' (strong sources)

Polarization I, Q, U, V I I, Q, U I I I I

1 2 3 4 5 6 7

Notes: a) The 6C catalogue does not completely cover this area. b) The 49 cm survey will only cover about 0.7 sr of this area. c) References: 1) This paper, 2) Gregory et al. (1996), 3) Condon et al. (1993), 4) Becker et al. (1995), 5) Hales et al. (1993), and references therein 6) Visser et al. (1996), and references therein, 7) Rees (1990).

WENSS is complemented by two other radio surveys with comparable beam-sizes, that are currently underway; the NRAO VLA Sky Survey (NVSS) at 1.4 GHz (Condon et al. 1993) and the 151 MHz 7C survey (McGilchrist et al. 1990; Lacy et al. 1994; Visser et al. 1995). These surveys will, when combined, provide well defined spectral information on an estimated radio-sources. The 4.85 GHz NRAO-Greenbank survey (Condon et al.\ 1989; Becker et al. 1991; Gregory & Condon 1991), recently updated (Gregory et al. 1995), the 6C survey at 151 MHz (Hales et al. 1993), and the 8C survey at 38 MHz (Rees 1990; Lacy et al. 1992) provide additional information for a subset of these sources. The high resolution VLA survey at 21 cm of the north galactic cap, the Faint Images of the Radio Sky at Twenty cm (FIRST, Becker et al. 1995), will provide accurate positions and morphological information for a large number of WENSS sources in this region. Table 1 (click here) summarizes the characteristics of WENSS and several other major radio surveys.

The most important additional information that WENSS provides compared to previous radio surveys are:

• Radio spectral information on an unprecedented number of sources over a substantial fraction of the sky. WENSS will provide spectral information, both internally (327/610 MHz), and by comparison with radio surveys at other frequencies. Thus WENSS will permit the study of very large numbers of sources with extreme spectra. These include: ultra steep spectrum (USS) sources (e.g. high-redshift radio galaxies, cluster halos, head-tail galaxies), peaked spectrum sources (e.g. GPS sources), and flat spectrum sources (e.g. high-redshift quasars)
• WENSS yields good positional information (from 5-10'' for the faintest sources to 1.5'' for the brighter ones). For a large number of sources this is sufficient to obtain an optical identification.
• The sensitive polarization information coupled with the large number of sources give WENSS unique capabilities in searching for radio sources having (anomalously) high linear polarizations at low frequencies. These include pulsars as well as interesting variable extra galactic radio sources. The sensitivity to extended structure has made it possible to study the large scale distribution of diffuse polarized galactic foreground emission.
• Because of the large dynamic range of the WSRT, the good coverage of short baselines, and the relatively low resolution, WENSS will provide data on faint extended structures, ranging in sizes from 30'' to , over a large region of the sky. This will be particularly valuable for detecting large galactic and extra galactic radio sources.
• The mosaicing technique, used in constructing WENSS, requires several observations per field, and provides limited data on the low-frequency variability of a large number of sources over time-scales from hours to years.

1.1. Outline

In Sect. 2 of this paper we start with an introduction to the observational techniques used in producing WENSS. Included in this section are a discussion of the mosaicing concept (2.1), the observational set-up (2.2), the data reduction (2.3), the frame production (2.4), and the source extraction (2.5).

The first WENSS fields to be observed (in the spring of 1991) covered a 570 square degree area centered on the north ecliptic pole: the mini-survey. We present results for the total intensity data from this region. The mini-survey is described in Sect. 3. An overview of the region is presented (3.1), followed by a description of the source list (3.2). This source list is only available in electronic form. A correction to the measured flux densities of faint sources is discussed (3.3, 4), followed by a description of the errors (3.5) and the completeness (3.6).

Plots for a selection of sources characterized by an exceptional morphology (Appendix A) are given at the end of the paper.