We present the KMOS 3D survey, a new integral field survey of over 600 galaxies at 0.7 < z < 2.7 using KMOS at the Very Large Telescope. The KMOS 3D survey utilizes synergies with multi-wavelength ground-and spacebased surveys to trace the evolution of spatially resolved kinematics and star formation from a homogeneous sample over 5 Gyr of cosmic history. Targets, drawn from a mass-selected parent sample from the 3D-HST survey, cover the star formation-stellar mass (M * ) and rest-frame (U − V ) − M * planes uniformly. We describe the selection of targets, the observations, and the data reduction. In the first-year of data we detect Hα emission in 191 M * = 3 × 10 9 -7 × 10 11 M galaxies at z = 0.7-1.1 and z = 1.9-2.7. In the current sample 83% of the resolved galaxies are rotation dominated, determined from a continuous velocity gradient and v rot /σ 0 > 1, implying that the star-forming "main sequence" is primarily composed of rotating galaxies at both redshift regimes. When considering additional stricter criteria, the Hα kinematic maps indicate that at least ∼70% of the resolved galaxies are disk-like systems. Our high-quality KMOS data confirm the elevated velocity dispersions reported in previous integral field spectroscopy studies at z 0.7. For rotation-dominated disks, the average intrinsic velocity dispersion decreases by a factor of two from 50 km s −1 at z ∼ 2.3 to 25 km s −1 at z ∼ 0.9. Combined with existing results spanning z ∼ 0-3, we show that disk velocity dispersions follow an evolution that is consistent with the dependence of velocity dispersion on gas fractions predicted by marginally stable disk theory.

Exploiting the deep high-resolution imaging of all 5 CANDELS fields, and accurate redshift information provided by 3D-HST, we investigate the relation between structure and stellar populations for a mass-selected sample of 6764 galaxies above 10 10 M ⊙ , spanning the redshift range 0.5 < z < 2.5. For the first time, we fit 2-dimensional models comprising a single Sérsic fit and two-component (i.e., bulge + disk) decompositions not only to the H-band light distributions, but also to the stellar mass maps reconstructed from resolved stellar population modeling. We confirm that the increased bulge prominence among quiescent galaxies, as reported previously based on rest-optical observations, remains in place when considering the distributions of stellar mass. Moreover, we observe an increase of the typical Sérsic index and bulge-to-total ratio (with median B/T reaching 40-50%) among starforming galaxies above 10 11 M ⊙ . Given that quenching for these most massive systems is likely to be imminent, our findings suggest that significant bulge growth precedes a departure from the starforming main sequence. We demonstrate that the bulge mass (and ideally knowledge of the bulge and total mass) is a more reliable predictor of the star-forming versus quiescent state of a galaxy than the total stellar mass. The same trends are predicted by the state-of-the-art semi-analytic model by Somerville et al. In the latter, bulges and black holes grow hand in hand through merging and/or disk instabilities, and AGN-feedback shuts off star formation. Further observations will be required to pin down star formation quenching mechanisms, but our results imply they must be internal to the galaxies and closely associated with bulge growth.

It remains a major challenge to derive a theory of cloud-scale ( 100 pc) star formation and feedback, describing how galaxies convert gas into stars as a function of the galactic environment. Progress has been hampered by a lack of robust empirical constraints on the giant molecular cloud (GMC) lifecycle. We address this problem by systematically applying a new statistical method for measuring the evolutionary timeline of the GMC lifecycle, star formation, and feedback to a sample of nine nearby disc galaxies, observed as part of the PHANGS-ALMA survey. We measure the spatially-resolved (∼ 100 pc) CO-to-Hα flux ratio and find a universal de-correlation between molecular gas and young stars on GMC scales, allowing us to quantify the underlying evolutionary timeline. GMC lifetimes are short, typically 10−30 Myr, and exhibit environmental variation, between and within galaxies. At kpc-scale molecular gas surface densities Σ H 2 8 M pc −2 , the GMC lifetime correlates with time-scales for galactic dynamical processes, whereas at Σ H 2 8 M pc −2 GMCs decouple from galactic dynamics and live for an internal dynamical time-scale. After a long inert phase without massive star formation traced by Hα (75−90 per cent of the cloud lifetime), GMCs disperse within just 1−5 Myr once massive stars emerge. The dispersal is most likely due to early stellar feedback, causing GMCs to achieve integrated star formation efficiencies of 4−10 per cent. These results show that galactic star formation is governed by cloud-scale, environmentally-dependent, dynamical processes driving rapid evolutionary cycling. GMCs and HII regions are the fundamental units undergoing these lifecycles, with mean separations of 100−300 pc in star-forming discs. Future work should characterise the multi-scale physics and mass flows driving these lifecycles.