We present the VIMOS Ultra Deep Survey (VUDS), a spectroscopic redshift survey of ~10.000 very faint galaxies to study the major phase of galaxy assembly 2<z<~6. The survey covers 1 deg^2 in 3 separate fields: COSMOS, ECDFS and VVDS-02h, with targets
selection based on an inclusive combination of photometric redshifts and color properties. Spectra covering 3650<lambda<9350 A are obtained with VIMOS on the ESO-VLT with integration times of 14h. Here we present the survey strategy, the target selection, the data processing, as well as the redshift measurement process, emphasizing the specific methods adapted to this high redshift range. The spectra quality and redshift reliability are discussed, and we derive a completeness in redshift measurement of 91%, or 74% for the most reliable measurements, down to i_AB=25, and measurements are performed all the way down to i_AB=27. The redshift distribution of the main sample peaks at z=3-4 and extends over a large redshift range mainly in 2 < z < 6. At 3<z<5, the galaxies cover a large range of luminosities -23< M_U < -20.5, stellar mass 10^9 M_sun< M_star < 10^{11} M_sun, and star formation rates 1 M_sun/yr< SFR < 10^3 M_sun/yr. We discuss the spectral properties of galaxies using individual as well as stacked spectra. The comparison between spectroscopic and photometric redshifts as well as color selection demonstrate the effectiveness of our selection scheme. With ~6000 galaxies with reliable spectroscopic redshifts in 2<z<6 expected when complete, this survey is the largest at these redshifts and offers the opportunity for unprecedented studies of the star-forming galaxy population and its distribution in large scale structures during the major phase of galaxy assembly.
We measure and analyse the redshift distribution N(z) of magnitude-selected samples using spectroscopic redshift measurement from the magnitude-selected VIMOS VLT Deep Survey (VVDS) with 17<iAB<24.75. We compute the N(z) and provide reference paramet
ric fits for i band, J, H and Ks band magnitud limited samples. The N(z) of a sample with iAB<24 has a mean redshift z=0.92, with 8.2% of the galaxies with z>2. Down to iAB<24.75 the sample has a mean redshift z=1.15 and 17.1% of the galaxies are beyond z=2. The projected sky density is 2.07+/-0.12 gal/arcmin2 at 1.4<z<2.5 and KsAB<22.5, 1.72+/-0.15 gal/arcmin2 at 2.7<z<3.4 and 0.59+/-0.09 gal/arcmin2 at 3.4<z<4.5 brighter than iAB=24.75. Galaxies at z~3 identified from magnitude-selected samples are 1.5 to 3 times more numerous than when they are colour-colour selected. We demonstrate that colour-colour selected samples over 1.4<z<4.5 are strongly contaminated by galaxies at other redshifts. Semi-analytic models on the Millennium simulations under-predict the number of luminous star-forming galaxies at zsim1.8-2, as well as over-predict the number of low-luminosity galaxies at z<0.8. Our study provides comprehensive galaxy number counts N(z) from galaxies with spectroscopic redshifts over a large redshift domain 0<z<5, a solid basis for the measurement of volume-complete quantities. Magnitude-selected surveys identify a higher number of galaxies at z>2 than in colour-colour selected samples, and we use the magnitude-selected VVDS to emphasize the large uncertainties associated to other surveys using colour or colour-colour selected samples. Our results further demonstrate that semi-analytical models on dark matter simulations have yet to find the right balance of physical processes and time-scales to properly reproduce a fundamental galaxy population property like the observed N(z).
We describe the completed VIMOS VLT Deep Survey, and the final data release of 35016 galaxies and type-I AGN with measured spectroscopic redshifts up to redshift z~6.7, in areas 0.142 to 8.7 square degrees, and volumes from 0.5x10^6 to 2x10^7h^-3Mpc^
3. We have selected samples of galaxies based solely on their i-band magnitude reaching i_{AB}=24.75. Spectra have been obtained with VIMOS on the ESO-VLT, integrating 0.75h, 4.5h and 18h for the Wide, Deep, and Ultra-Deep nested surveys. A total of 1263 galaxies have been re-observed independently within the VVDS, and from the VIPERS and MASSIV surveys. They are used to establish the redshift measurements reliability, to assess completeness, and to provide a weighting scheme taking into account the survey selection function. We describe the main properties of the VVDS samples, and the VVDS is compared to other spectroscopic surveys. In total we have obtained spectroscopic redshifts for 34594 galaxies, 422 type-I AGN, and 12430 Galactic stars. The survey has enabled to identify galaxies up to very high redshifts with 4669 redshifts in 1<=z_{spec}<=2, 561 in 2<=z_{spec}<=3 and 468 with z_{spec}>3, and specific populations like LAE have been identified out to z=6.62. We show that the VVDS occupies a unique place in the parameter space defined by area, depth, redshift coverage, and number of spectra. The VVDS provides a comprehensive survey of the distant universe, covering all epochs since z, or more than 12 Gyr of cosmic time, with a uniform selection, the largest such sample to date. A wealth of science results derived from the VVDS have shed new light on the evolution of galaxies and AGN, and their distribution in space, over this large cosmic time. A final public release of the complete VVDS spectroscopic redshift sample is available at http://cesam.lam.fr/vvds.
