No abstract

We reproduce the blue and red sequences in the observed joint distribution of colour and magnitude for galaxies at low and high redshifts using hybrid N‐body/semi‐analytic simulations of galaxy formation. The match of model and data is achieved by mimicking the effects of cold flows versus shock heating coupled to feedback from active galactic nuclei (AGNs), as predicted by Dekel and Birnboim. After a critical epoch z∼ 3, only haloes below a critical shock‐heating mass Mshock∼ 1012 M⊙ enjoy gas supply by cold flows and form stars, while cooling and star formation are shut down abruptly above this mass. The shock‐heated gas is kept hot because being dilute it is vulnerable to feedback from energetic sources such as AGNs in their self‐regulated mode. The shutdown explains in detail the bright‐end truncation of the blue sequence at ∼L*, the appearance of luminous red‐and‐dead galaxies on the red sequence starting already at z∼ 2, the colour bimodality, its strong dependence on environment density and its correlations with morphology and other galaxy properties. Before z∼ 2–3, even haloes above the shock‐heating mass form stars by cold streams penetrating through the hot gas. This explains the bright star forming galaxies at z∼ 3–4, the early appearance of massive galaxies on the red sequence, the high cosmological star formation rate at high redshifts and the subsequent low rate at low redshifts.

This is the first paper of a series that describes the methods and basic results of the GALICS model (Galaxies In Cosmological Simulations). GALICS is a hybrid model for hierarchical galaxy formation studies, combining the outputs of large cosmological N-body simulations with simple, semi-analytic recipes to describe the fate of the baryons within dark matter haloes. The simulations produce a detailed merging tree for the dark matter haloes, including complete knowledge of the statistical properties arising from the gravitational forces. We intend to predict the overall statistical properties of galaxies, with special emphasis on the panchromatic spectral energy distribution emitted by galaxies in the ultraviolet/optical and infrared/submillimetre wavelength ranges.In this paper, we outline the physically motivated assumptions and key free parameters that go into the model, comparing and contrasting with other parallel efforts. We specifically illustrate the success of the model in comparison with several data sets, showing how it is able to predict the galaxy disc sizes, colours, luminosity functions from the ultraviolet to far infrared, the Tully-Fisher and Faber-Jackson relations, and the fundamental plane in the local Universe. We also identify certain areas where the model fails, or where the assumptions needed to succeed are at odds with observations, and pay special attention to understanding the effects of the finite resolution of the simulations on the predictions made. Other papers in this series will take advantage of different data sets available in the literature to extend the study of the limitations and predictive power of GALICS, with particular emphasis put on high-redshift galaxies.

We present the MUSE Hubble Ultra Deep Survey, a mosaic of nine MUSE fields covering 90% of the entire HUDF region with a 10-h deep exposure time, plus a deeper 31-h exposure in a single 1.15 arcmin 2 field. The improved observing strategy and advanced data reduction results in datacubes with sub-arcsecond spatial resolution (0 . 65 at 7000 Å) and accurate astrometry (0 . 07 rms). We compare the broadband photometric properties of the datacubes to HST photometry, finding a good agreement in zeropoint up to m AB = 28 but with an increasing scatter for faint objects. We have investigated the noise properties and developed an empirical way to account for the impact of the correlation introduced by the 3D drizzle interpolation. The achieved 3σ emission line detection limit for a point source is 1.5 and 3.1 × 10 −19 erg s −1 cm −2 for the single ultra-deep datacube and the mosaic, respectively. We extracted 6288 sources using an optimal extraction scheme that takes the published HST source locations as prior. In parallel, we performed a blind search of emission line galaxies using an original method based on advanced test statistics and filter matching. The blind search results in 1251 emission line galaxy candidates in the mosaic and 306 in the ultradeep datacube, including 72 sources without HST counterparts (m AB > 31). In addition 88 sources missed in the HST catalog but with clear HST counterparts were identified. This data set is the deepest spectroscopic survey ever performed. In just over 100 h of integration time, it provides nearly an order of magnitude more spectroscopic redshifts compared to the data that has been accumulated on the UDF over the past decade. The depth and high quality of these datacubes enables new and detailed studies of the physical properties of the galaxy population and their environments over a large redshift range.