(Paper I) We provide details of a multicolor ( u, b_j_, v, or, r, i), wide-field, faint magnitude survey for high-redshift (z >= 2.2 ) quasars. The survey extends over the magnitude range 16<=m_or_<=20 and covers a total area of 58.6 square degrees. Sources of incompleteness in the photometric catalog are quantified, giving the survey an effective area of 45.7 square degrees. Particular attention is paid to the details of the plate-matching and image-classification procedures used in the generation of the photometric catalog from which quasars are selected, in order that the number of spurious quasar candidates be kept to an absolute minimum. The selection of candidates incorporates a number of novel features, including treating the multicolor information as low-resolution spectra, adopting a quantitative candidate identification algorithm that employs all the available information, and requiring only that objects be detected in the or passband. We describe how model quasar spectra are used to calculate the selection function for the survey, and we present the computed sample completeness, as a function of redshift and magnitude, for a range of representative quasar types. The spectroscopic results of the survey are reported in a companion paper and the derivation of the quasar luminosity function at high redshift will be reported in a third paper. (Paper II; see also documentation files "doc.tex" or "doc.txt") In a wide-field multicolor survey (45.7 deg^2^, 16.0<=m_or_=2.2. There are 49 new quasars of redshift z>=3.0 including three of z>=4.0. We provide spectra, coordinates, redshifts, broad-band magnitudes (u, b_j_, v, or, r, i), line-equivalent widths for Lyman-{alpha}/NV and CIV, FWHM CIV, and continuum spectral indices for all the new quasars. The sample includes 96 quasars selected according to the rigid criteria detailed in the companion paper by Warren, Hewett, Irwin, and Osmer. These are combined with 14 previously known quasars in one of our fields, which also meet these selection criteria, to form a complete sample. The median equivalent width for Lyman-{alpha}/NV for the complete sample is 67A, and for CIV is 31{AA}. The median FWHM CIV is 35{AA} and the median spectral index is {alpha}=-0.60. The complete sample contains at least five broad absorption-line quasars. We have obtained spectra of a total of 473 multicolor selected candidates. The proportion of quasars found, including previously known quasars, of all redshifts, is 30%. The proportion of quasars of redshift z>=2.2 in the list of candidates that satisfy the selection criteria of the complete sample is 43%. Finally we limit the sample to the 85 objects of redshift z>=2.2 in the complete sample, and summarize the relevant observational data input to the calculation of the luminosity function, comprising the redshift versus magnitude distribution and the distribution of the spectral properties. The analysis of the luminosity function will be presented in a forthcoming paper.

Cone search capability for table VII/182/table4 (Known non-UV-excess QSOs in the SGP (paper I))

Cone search capability for table VII/182/tables (High-z QSOs in SGP and F401 (paper II, tables 1-4))

Cone search capability for table VII/182/table5 (previously-known high-z QSOs in SGP (paper II))

Cone search capability for table VII/182/table6 (new UV excess quasars (paper II))