No abstract

DYNAMO is a multi-wavelength, spatially-resolved survey of local (z ∼ 0.1) star-forming galaxies designed to study evolution through comparison with samples at z 2. Half of the sample has integrated H α luminosities of > 10 42 erg s −1 , the typical lower limit for resolved spectroscopy at z 2. The sample covers a range in stellar mass (10 9 -10 11 M ) and star-formation rate (0.2-100 M yr −1 ). In this first paper of a series, we present integral-field spectroscopy of H α emission for the sample of 67 galaxies. We infer gas fractions in our sample as high as 0.8, higher than typical for local galaxies. Gas fraction correlates with stellar mass in galaxies with star-formation rates below 10 M yr −1 , as found by COLDGASS, but galaxies with higher star-formation rates have higher than expected gas fractions. There is only a weak correlation, if any, between gas fraction and gas velocity dispersion. Galaxies in the sample visually classified as disc-like are offset from the local stellar-mass Tully-Fisher relation to higher circular velocities, but this offset vanishes when both gas and stars are included in the baryonic Tully-Fisher relation. The mean gas velocity dispersion of the sample is 50 km s −1 , and V /σ ranges from 2 to 10 for most of the discs, similar to 'turbulent' galaxies at high redshift. Half of our sample show disc-like rotation, while ∼20 per cent show no signs of rotation. The division between rotating and non-rotating is approximately equal for the sub-samples with either star-formation rates > 10 M yr −1 , or specific-star-formation rates typical of the star-formation 'main sequence' at z 2. Across our whole sample, we find good correlation between the dominance of 'turbulence' in galaxy discs (as expressed by V /σ) and gas fraction as has been predicted for marginally stable Toomre discs. Comparing our sample with many others at low-and high-redshift reveals a correlation between gas velocity dispersion and star formation rate. These findings suggest the DYNAMO discs are excellent candidates for local galaxies similar to turbulent z 2 disc galaxies.

We present an analysis of the size growth seen in early-type galaxies over 10 Gyr of cosmic time. Our analysis is based on a homogeneous synthesis of published data from 17 spectroscopic surveys observed at similar spatial resolution, augmented by new measurements for galaxies in the Gemini Deep Deep Survey. In total, our sample contains structural data for 465 galaxies (mainly early-type) in the redshift range 0.2 < z < 2.7. The size evolution of passively-evolving galaxies over this redshift range is gradual and continuous, with no evidence for an end or change to the process around z ∼ 1, as has been hinted at by some surveys which analyze subsets of the data in isolation. The size growth appears to be independent of stellar mass, with the mass-normalized half-light radius scaling with redshift as R e ∝ (1 + z) −1. 62±0.34 . Surprisingly, this power law seems to be in good agreement with the recently reported continuous size evolution of UV-bright galaxies in the redshift range z ∼ 0.5 − 3.5. It is also in accordance with the predictions from recent theoretical models.

High spatial and spectral resolution observations of star formation and kinematics in early galaxies have shown that two-thirds are massive rotating disk galaxies [1][2][3][4][5] with the remainder being less massive non-rotating objects 2,4,[6][7][8] . The line of sight averaged velocity dispersions are typically five times higher than in today's disk galaxies. This has suggested that gravitationally-unstable, gas-rich disks in the early Universe are fuelled by cold, dense accreting gas flowing along cosmic filaments and penetrating hot galactic gas halos 9,10 . However these accreting flows have not been observed 11 , and cosmic accretion cannot power the observed level of turbulence 12 . Here we report on a new sample of rare high-velocity-dispersion disk galaxies we have discovered in the nearby Universe where cold accretion is unlikely to drive their high star-formation rates. We find that the velocity dispersion is most fundamentally correlated with their star-formation rates, and not their mass nor gas fraction, which leads to a new picture where star formation itself is the energetic driver of galaxy disk turbulence at all cosmic epochs.Understanding how these different kinematic states of star-forming galaxies fit together is complicated by selection, surface-brightness and angular-resolution effects. Particularly at high redshift, resolution is a major limitation. The resolution gain of adaptive optics has enabled kinematic observations of early disks but not all observations, even within a particular survey To quantify how these difficulties might affect previous results, we turned to the well-studied Sloan Digital Sky Survey (SDSS) to put kinematic galaxy states in the context of a large, uniformlyselected sample. We undertook the first IFS observations of 65 star-forming 13 galaxies at redshift z ∼ 0.1. As active galactic nuclei interfere with Hα emission as a star formation tracer, they have been excluded. (Further details of the selection criteria are in the Supplementary Information). Observed galaxies have Hα luminosities of 10 40.7 to 10 42.6 erg/s and median of 10 41.9 erg/s and stellar mass range and median of 10 9.1 to 10 10.9 and 10 10.3 solar masses, respectively 14 . Because only bright (which we shall define here as L Hα > 10 42 erg/s) objects are typically detected at z ∼ 2, such objects make up half our observations despite representing only 3.2% of SDSS galaxies which otherwise meet our criteria. A broad selection illustrates the impact of surface brightness and luminosity on our results and those reported for the high-redshift Universe.Chosen galaxies were observed using the integral-field spectrographs SPIRAL 15 on the 3.9mAnglo-Australian Telescope or WiFeS 16 on the ANU 2.3m telescope. Median seeing of 1.3" corresponds to a median spatial resolution of 2.3 kpc and the field of view is 10-40 kpc. This is closely matched to high-redshift samples observed with adaptive optics, but with better spectral resolution (λ/∆λ = 7000 to 11,500). Following standard methods, we fit a Gaussian p...

We present ∼100 pc resolution Hubble Space Telescope Hα images of 10 galaxies from the DYnamics of Newly-Assembled Massive Objects (DYNAMO) survey of lowz turbulent disk galaxies, and use these to undertake the first detailed systematic study of the effects of resolution and clump clustering on observations of clumps in turbulent disks. In the DYNAMO-HST sample we measure clump diameters spanning the range d clump ∼ 100 − 800 pc, and individual clump star formation rates as high as ∼ 5 M ⊙ yr −1 . DYNAMO clumps have very high SFR surface densities, Σ S FR ∼ 1−15 M ⊙ yr −1 kpc −2 , ∼ 100× higher than in Hii regions of nearby spirals. Indeed, SFR surface density provides a simple dividing line between massive star forming clumps and local star forming regions, where massive star forming clumps have Σ S FR > 0.5 M ⊙ yr −1 kpc −2 . When degraded to match the observations of galaxies in z ∼ 1 − 3 surveys, DYNAMO galaxies are similar in morphology and measured clump properties to clumpy galaxies observed in the high-z Universe. Emission peaks in the simulated high-redshift maps typically correspond to multiple clumps in full resolution images. This clustering of clumps systematically increases the apparent size and SFR of clumps in 1 kpc resolution maps, and decreases the measured SFR surface density of clumps by as much as a factor of 20×. From these results we can infer that clump clustering is likely to strongly effect the measured properties of clumps in high-z galaxies, which commonly have kiloparsec scale resolution.