KMOS is a cryogenic infrared spectrograph fed by twentyfour deployable integral field units that patrol a 7.2 arcminute diameter field of view at the Nasmyth focus of the ESO VLT. It is well suited to the study of galaxy clusters at 1 < z < 2 where the well understood features in the restframe V-band are shifted into the KMOS spectral bands. Coupled with HST imagining, KMOS offers a window on the critical epoch for galaxy evolution, 7-10 Gyrs ago, when the key properties of cluster galaxies were established. We aim to investigate the size, mass, morphology and star formation history of galaxies in the clusters. Here we describe the instrument, discuss the status of the observations and report some preliminary results.

The total density profiles of elliptical galaxies can be fit by a single power law, i.e., ρtot ∝ rγ with γ ≈ −2. While strong lensing observations show a tendency for the slopes to become flatter with increasing redshift, simulations indicate an opposite trend. To understand this discrepancy, we study a set of simulated spheroids formed within the cosmological framework. From our simulations we find that the steepness of the total density slope correlates with the compactness of the stellar component within the half-mass radius, and that spheroidal galaxies tend to be more compact at high redshifts than their present-day counterparts. While both these results are in agreement with observations, the observed trend of the total density slope with redshift remains in contradiction to the results from simulations.

Various programs aimed at exploring the still largely unknown low surface brightness Universe with deep imaging optical surveys have recently started. They open a new window for studies of galaxy evolution, pushing the technique of galactic archeology outside the Local Group (LG). The method, based on the detection and analysis of the diffuse light emitted by collisional debris or extended stellar halos (rather than on stellar counts as done for LG systems), faces however a number of technical difficulties, like the contamination of the images by reflection halos and Galactic cirrus. I review here the on-going efforts to address them and highlight the preliminary promising results obtained with a systematic survey with MegaCam on the CFHT of nearby massive early-type galaxies done as part of the ATLAS3D, NGVS and MATLAS collaborations.

We present results on the stellar population properties of massive galaxies at z = 0.7 based on deep, medium-resolution IMACS spectra for a sample of ~ 70 galaxies in the ECDFS with M* > 1010M⊙. The age–mass and stellar metallicity–mass relations for the population as a whole have a similar shape as the local relations over the probed mass range, but offset to ages younger by ~ 4 Gyr and metallicities lower by ~ 0.13 dex. Quiescent galaxies alone have stellar ages and metallicities consistent with passive evolution onto the local quiescent galaxies relations. The evolution in metallicity is driven by star-forming galaxies. However a significant fraction of massive star-forming galaxies have metallicities comparable to those of local quiescent galaxies. If quenched at z < 0.7 they can provide the necessary population to reproduce the scatter in age and metallicity of local quiescent galaxies.

There is growing evidence suggesting that massive galaxies have growth both in mass and size mainly by the accretion of smaller satellites. This evolutionary path qualitative explains many different observations. However, there is still much work to do on trying to quantify whether this merging scenario is enough to explain the significant growth of massive galaxies since z~2. In this contribution, we show that both the number of satellite galaxies around massive galaxies at all redshifts, and the existence of massive relic galaxies in the nearby Universe are in quantitative agreement with this merging picture. Nonetheless, there is still open questions related to the properties of the stellar halos of present-day massive galaxies and the number density of massive relics at all redshifts that remain open.

We present the results on the stellar Initial Mass Function (IMF) normalisation of 68 massive (M* = 1011 - 1012M⋖) Early-Type Galaxies (ETGs) at redshift of ~1. This was achieved by deriving the stellar Mass-to-Light ratio (M/L) of the galaxies through axis-symmetric dynamical modelling and comparing it to the same derived via stellar population modelling through full spectrum fitting. The study also employs an Abundance Matching technique to account for the dark matter within the galaxies. The results demonstrate that massive ETGs at high redshifts on average have a Salpeter-like IMF normalisation, while providing observational evidence supporting previous predictions of low dark matter fraction in the inner regions (<1Re) of galaxies at higher redshift.

We have learned much about galaxy evolution since z = 2, and something to even higher redshifts. How can it be that we know so little about! the star formation histories (SFHs) of individual galaxies? Although great progress has been made accumulating huge samples with only rudimentary properties, progress in galaxy evolution means connecting what we've learned to detailed measurements of the life-histories of specific –- not just representative –- systems.

We investigate the evolution of the Tully–Fisher relation out to z = 1 with 137 emission-line galaxies in the field that display a regular rotation curve. They follow a linear trend with lookback time being on average brighter by 1.1 Bmag and 60% smaller at z = 1. For a subsample of 48 objects with very regular gas kinematics and stellar structure we derive a TF scatter of 1.15mag, which is two times larger than local samples exhibit. This is probably due to modest variations in their star formation history and chemical enrichment.

In another study of 96 members of Abell 901/902 at z = 0.17 and 86 field galaxies with similar redshifts we find a difference in the TFR of 0.42mag in the B-band but no significant difference in stellar mass. Comparing specifically red spirals with blue ones in the cluster, the former are fainter on average by 0.35 Bmag and have 15% lower stellar masses. This is probably due to star formation quenching caused by ram-pressure in the cluster environment. Evidence for this scenario comes from strong distortions of the gas disk of red spirals that have at the same time a very regular stellar disk structure.

Do the kinematics and mass profiles of dwarf galaxies present a fundamental challenge to standard cold dark matter (CDM) models? New, deep spectroscopy using DEIMOS on Keck for hundreds of low stellar mass (107-109 M⊙ star-forming galaxies at intermediate redshift (0.2 < z < 1) addresses this inquiry in a way that is less subject to cosmic variance and environmental bias than previous, more local work. Half of this sample reveals resolved, doppler-shifted nebular emission, used to constrain rotation curves. From these we can construct the stellar mass Tully-Fisher relation to masses as low as ~107 M⊙, and a persistent discrepancy is found between predictions from simulations and models compared to our observations. We suggest on-going and future tests that will be more effective in distinguishing between the effects of baryonic feedback and alternative models of dark matter in this remarkable regime.