Paper detail

MASSIV: Mass Assemby Survey with SINFONI in VVDS. I. Survey description and global properties of the 0.9 < z < 1.8 galaxy sample

Understanding how galaxies evolve and assemble their mass across cosmic time is still a fundamental unsolved issue. To get insight into the various processes of galaxy mass assembly, the Mass Assembly Survey with SINFONI in VVDS (MASSIV) aims at probing the kinematical and chemical properties of a significant and representative sample of high-redshift (0.9 < z < 1.8) star-forming galaxies. This paper presents the selection function, the observing strategy and the global properties of the MASSIV sample. This sample contains 84 star-forming galaxies, selected from the VIMOS VLT Deep Survey (VVDS) and observed with the SINFONI IFU at the VLT. We present the redshift distribution, and derive the stellar masses and SED-based star formation rates (SFR). Integrated metallicities and the presence of type-2 AGNs are investigated using composite 1D spectra built from VIMOS and SINFONI observations. The MASSIV selection function, based on star formation criteria ([OII] emission-line strength up to z~1.5 and colors/UV absorption lines at higher redshifts), provides a good representation of "normal" star-forming galaxies with SED-based SFRs between 5 and 400 Msun/yr in the stellar mass regime 10^9 - 10^11 Msun. Analysis of typical emission-line ratios performed on composite spectra reveals that the contamination by type-2 AGNs is very low and that the integrated metallicity of the galaxies follows the well-known mass-metallicity relation. The MASSIV sample has been built upon a simple selection function, fully representative of the star-forming galaxy population at 0.9 < z < 1.8 for SFR > 5 Msun/yr. Together with the size of the sample, the spatially-resolved SINFONI data therefore enables us to discuss global, volume averaged, galaxy kinematic and chemical properties all accross the mass and SFR range of the survey to derive robust conclusions on galaxy mass assembly over cosmological timescales.