ALBERT

Description: We have exploited the Hubble Space Telescope ( HST ) Cosmic Assembly Near-IR Deep Extragalactic Legacy Survey (CANDELS) J and H -band Wide Field Camera 3 (WFC3)/infrared (IR) imaging to study the properties of (sub-)millimetre galaxies within the Great Observatories Origins Deep Survey South (GOODS-South) field. After using the deep radio (Very Large Array 1.4 GHz) and Spitzer (Infrared Array Camera 8 μm) imaging to identify galaxy counterparts for the (sub-)millimetre sources, we have then utilized the new CANDELS WFC3/IR imaging in two ways. First, the addition of new deep near-IR photometry from both HST and (at K band) the VLT to the existing GOODS-South data base has enabled us to derive improved photometric redshifts and stellar masses, confirming that the (sub-)millimetre sources are massive (〈 M * 〉 = 2.2  x 10 11  ± 0.2 M ) galaxies at z ~= 1-3. Secondly, we have exploited the depth and resolution of the WFC3/IR imaging to determine the sizes and morphologies of the galaxies at rest-frame optical wavelengths rest  〉 4000 Å. Specifically, we have fitted two-dimensional axisymmetric galaxy models to the WFC3/IR images, varying luminosity, axial ratio, half-light radius r 1/2 and Sérsic index n . Crucially, the wavelength and depth of the WFC3/IR imaging enables modelling of the mass-dominant galaxy, rather than the blue high surface-brightness features which often dominate optical (rest-frame ultraviolet) images of (sub-)millimetre galaxies, and can confuse visual morphological classification. As a result of this analysis, we find that 〉95 per cent of the rest-frame optical light in almost all of the (sub-)millimetre galaxies is well described by either a single exponential disc ( n  ~= 1), or a multiple-component system in which the dominant constituent is disc like. We demonstrate that this conclusion is completely consistent with the results of recent high-quality ground-based K -band imaging sampling even longer rest-frame wavelengths, and explain why it is so. These massive disc galaxies are reasonably extended (〈 r 1/2 〉 = 4.5 ± 0.5 kpc; median r 1/2  = 4.0 kpc), consistent with the sizes of other massive star-forming discs at z  ~= 2. In many cases, we find evidence of blue clumps within the sources, with the mass-dominant disc component becoming more significant at longer wavelengths. Finally, only a minority of the sources show evidence for a major galaxy–galaxy interaction. Taken together, these results support the view that most (sub-)millimetre galaxies at z  ~= 2 are simply the most extreme examples of normal star-forming galaxies at that era. Interestingly, the only two bulge-dominated galaxies are also the two lowest redshift sources in the sample ( z  ~= 1), a result which may reflect the structural evolution of high-mass galaxies in general.

Description: We investigate the redshift evolution of the [O  iii] /H  β nebular emission line ratio for a sample of galaxies spanning the redshift range 0 〈  z  〈 4. We compare the observed evolution to a set of theoretical models which account for the independent evolution of chemical abundance, ionization parameter and interstellar medium (ISM) pressure in star-forming galaxies with redshift. Accounting for selection effects in the combined data sets, we show that the evolution to higher [O  iii] /H  β ratios with redshift is a real physical effect which is best accounted for by a model in which the ionization parameter is elevated from the average values typical of local star-forming galaxies, with a possible simultaneous increase in the ISM pressure. We rule out the possibility that the observed [O  iii] /H  β evolution is purely due to metallicity evolution. We discuss the implications of these results for using local empirical metallicity calibrations to measure metallicities at high redshift, and briefly discuss possible theoretical implications of our results.

Description: The KMOS Redshift One Spectroscopic Survey (KROSS) is an ESO-guaranteed time survey of 795 typical star-forming galaxies in the redshift range z = 0.8–1.0 with the KMOS instrument on the Very Large Telescope. In this paper, we present resolved kinematics and star formation rates for 584 z ~ 1 galaxies. This constitutes the largest near-infrared Integral Field Unit survey of galaxies at z ~ 1 to date. We demonstrate the success of our selection criteria with 90 per cent of our targets found to be H α emitters, of which 81 per cent are spatially resolved. The fraction of the resolved KROSS sample with dynamics dominated by ordered rotation is found to be 83 ± 5 per cent. However, when compared with local samples these are turbulent discs with high gas to baryonic mass fractions, ~35 per cent, and the majority are consistent with being marginally unstable (Toomre Q ~ 1). There is no strong correlation between galaxy averaged velocity dispersion and the total star formation rate, suggesting that feedback from star formation is not the origin of the elevated turbulence. We postulate that it is the ubiquity of high (likely molecular) gas fractions and the associated gravitational instabilities that drive the elevated star formation rates in these typical z ~ 1 galaxies, leading to the 10-fold enhanced star formation rate density. Finally, by comparing the gas masses obtained from inverting the star formation law with the dynamical and stellar masses, we infer an average dark matter to total mass fraction within 2.2 r e (9.5 kpc) of 65 ± 12 per cent, in agreement with the results from hydrodynamic simulations of galaxy formation.

Description: We present new accurate measurements of the mass, metallicity and star-formation rate of a statistically significant sample of 93 galaxies at $z$ 2 using near-infrared spectroscopy taken as part of the 3D- Hubble Space Telescope survey. We derive a mass–metallicity relation (MZR) for our sample with metallicities based on the oxygen and Hβ nebular emission lines. We find the MZR derived from our data to have the same trend as previous determinations in the range 0 〈 $z$ 〈 3 with metallicity decreasing with stellar mass. However, we find that our MZR is offset from a previous determination at $z$ 2 which used metallicities derived from the [N ii ]/Hα ratio. Incorporating star formation rate information, we find that our galaxies are also offset from the fundamental metallicity relation (FMR) by ~0.3 dex. Using the Baldwin–Phillips–Terlevich (BPT) diagram we argue that, if the physical conditions of star-forming regions evolve with redshift, metallicity indicators based on [N ii ] and Hα, calibrated in the local Universe, may not be consistent with the ones based on oxygen lines and Hβ. Our results thus suggest that the evolution of the FMR previously reported at $z$ ~ 2–3 may be an artefact of the differential evolution in metallicity indicators and caution against using locally calibrated empirical metallicity relations at high redshift, which do not account for evolution in the physical conditions of star-forming regions.

Description: We present a new method to classify the broad-band optical–near-infrared spectral energy distributions (SEDs) of galaxies using three shape parameters (super-colours) based on a principal component analysis of model SEDs. As well as providing a compact representation of the wide variety of SED shapes, the method allows for easy visualization of information loss and biases caused by the incomplete sampling of the rest-frame SED as a function of redshift. We apply the method to galaxies in the United Kingdom Infrared Telescope Infrared Deep Sky Survey Ultra Deep Survey with 0.9 〈  z  〈 1.2, and confirm our classifications by stacking rest-frame optical spectra for a fraction of objects in each class. As well as cleanly separating a tight red sequence from star-forming galaxies, three unusual populations are identifiable by their unique colours: very dusty star-forming galaxies with high metallicity and old mean stellar age; post-starburst galaxies which have formed 10 per cent of their mass in a recent unsustained starburst event; and metal-poor quiescent dwarf galaxies. We find that quiescent galaxies account for 45 per cent of galaxies with log M */M 〉 11, declining steadily to 13 per cent at log M */M = 10. The properties and mass function of the post-starburst galaxies are consistent with a scenario in which gas-rich mergers contribute to the growth of the low- and intermediate-mass range of the red sequence.

Description: We present the results of a new and improved study of the morphological and spectral evolution of massive galaxies over the redshift range 1 〈 z 〈 3. Our analysis is based on a bulge–disc decomposition of 396 galaxies with M * 〉 10 11 M uncovered from the Cosmic Assembly Near-infrared Deep Extragalactic Legacy Survey (CANDELS) Wide Field Camera 3 (WFC3)/IR imaging within the Cosmological Evolution Survey (COSMOS) and UKIRT Infrared Deep Sky Survey (UKIDSS) UDS survey fields. We find that, by modelling the H 160 image of each galaxy with a combination of a de Vaucouleurs bulge (Sérsic index n = 4) and an exponential disc ( n = 1), we can then lock all derived morphological parameters for the bulge and disc components, and successfully reproduce the shorter-wavelength J 125 , i 814 , v 606 HST images simply by floating the magnitudes of the two components. This then yields sub-divided four-band HST photometry for the bulge and disc components which, with no additional priors, is well described by spectrophotometric models of galaxy evolution. Armed with this information, we are able to properly determine the masses and star formation rates for the bulge and disc components, and find that: (i) from z = 3 to 1 the galaxies move from disc dominated to increasingly bulge dominated, but very few galaxies are pure bulges/ellipticals by z = 1; (ii) while most passive galaxies are bulge dominated, and most star-forming galaxies disc dominated, 18 ± 5 per cent of passive galaxies are disc dominated, and 11 ± 3 per cent of star-forming galaxies are bulge dominated, a result which needs to be explained by any model purporting to connect star formation quenching with morphological transformations; (iii) there exists a small but significant population of pure passive discs, which are generally flatter than their star-forming counterparts (whose axial ratio distribution peaks at b / a ~= 0.7); (iv) flatter/larger discs re-emerge at the highest star formation rates, consistent with recent studies of sub-mm galaxies, and with the concept of a maximum surface density for star formation activity.

Description: We have constructed a mass-selected sample of M *  〉 10 11 M galaxies at 1 〈  z  〈 3 in the CANDELS UKIDSS UDS and COSMOS fields and have decomposed these systems into their separate bulge and disc components according to their H 160 -band morphologies. By extending this analysis to multiple bands, we have been able to conduct individual bulge and disc component SED fitting which has provided us with stellar-mass and star formation rate estimates for the separate bulge and disc components. Having utilized the new decomposed stellar-mass estimates, we confirm that the bulge components display a stronger size evolution than the discs. The median sizes of the bulge components is 3.09 ± 0.20 times smaller than similarly massive local galaxies over the full 1 〈  z  〈 3 redshift range; for the discs, the corresponding factor is 1.77 ± 0.10. Moreover, by splitting our sample into the passive and star-forming bulge and disc sub-populations and examining their sizes as a fraction of their present-day counter-parts, we find that the star-forming and passive bulges are equally compact, star-forming discs are larger, while the passive discs have intermediate sizes. This trend is not evident when classifying galaxy morphology on the basis of single-Sérsic fits and adopting the overall star formation rates. Finally, by evolving the star formation histories of the passive discs back to the redshifts when the passive discs were last active, we show that the passive and star-forming discs have consistent sizes at the relevant epoch. These trends need to be reproduced by any mechanisms which attempt to explain the morphological evolution of galaxies.

Description: Motivated by the current controversy over the redshift distribution and physical properties of luminous (sub-)mm sources, we have undertaken a new study of the brightest sample of unlensed (sub-)mm sources with pre-Atacama Large Millimeter/submillimeter Array (ALMA) interferometric follow-up in the Cosmological Evolution Survey field. Exploiting the very latest multifrequency supporting data, we find that this sample displays a redshift distribution indistinguishable from that of the lensed sources uncovered with the South Pole Telescope, with z median ~= 3.5. We also find that, over the redshift range z ~= 2–6, the median stellar mass of the most luminous (sub-) mm sources is M * ~= 3 x 10 11 M , yielding a typical specific star formation rate sSFR ~= 3 Gyr – 1 . Consistent with recent ALMA and the Submillimeter Array studies, we confirm that source blending is not a serious issue in the study of luminous (sub-) mm sources uncovered by ground-based, single-dish surveys; only ~=10–15 per cent of bright ( S 850 ~= 5–10 mJy) (sub-) mm sources arise from significant (i.e. 〉20 per cent) blends, and so our conclusions are largely unaffected by whether we adopt the original single-dish mm/sub-mm flux densities/positions or the interferometric data. Our results suggest that apparent disagreements over the redshift distribution of (sub-)mm sources are a result of ‘down-sizing’ in dust-enshrouded star formation, consistent with existing knowledge of the star formation histories of massive galaxies. They also indicate that extreme star-forming galaxies at high redshift are, on average, subject to the same star formation rate-limiting processes as less luminous objects, and lie on the ‘main sequence’ of star-forming galaxies at z 〉 3.

Description: Despite decades of study, we still do not fully understand why some massive galaxies abruptly switch off their star formation in the early Universe, and what causes their rapid transition to the red sequence. Post-starburst galaxies provide a rare opportunity to study this transition phase, but few have currently been spectroscopically identified at high redshift ( z 〉 1). In this paper, we present the spectroscopic verification of a new photometric technique to identify post-starbursts in high-redshift surveys. The method classifies the broad-band optical–near-infrared spectral energy distributions (SEDs) of galaxies using three spectral shape parameters (supercolours), derived from a principal component analysis of model SEDs. When applied to the multiwavelength photometric data in the UKIDSS Ultra Deep Survey, this technique identified over 900 candidate post-starbursts at redshifts 0.5 〈 z 〈 2.0. In this study, we present deep optical spectroscopy for a subset of these galaxies, in order to confirm their post-starburst nature. Where a spectroscopic assessment was possible, we find the majority (19/24 galaxies; ~80 per cent) exhibit the strong Balmer absorption (H  equivalent width W 〉 5 Å) and Balmer break, characteristic of post-starburst galaxies. We conclude that photometric methods can be used to select large samples of recently-quenched galaxies in the distant Universe.

Description: We investigate the properties of the galaxies selected from the deepest 850-μm survey undertaken to date with (Submillimetre Common-User Bolometer Array 2) SCUBA-2 on the James Clerk Maxwell Telescope as part of the SCUBA-2 Cosmology Legacy Survey. A total of 106 sources (〉5) were uncovered at 850 μm from an area of ~=150 arcmin 2 in the centre of the COSMOS/UltraVISTA/Cosmic Assembly Near-infrared Deep Extragalactic Legacy Survey (CANDELS) field, imaged to a typical depth of 850 ~= 0.25 mJy. We utilize the available multifrequency data to identify galaxy counterparts for 80 of these sources (75 per cent), and to establish the complete redshift distribution for this sample, yielding $\bar{z} = 2.38\pm 0.09$ . We have also been able to determine the stellar masses of the majority of the galaxy identifications, enabling us to explore their location on the star formation rate:stellar mass (SFR: M *) plane. Crucially, our new deep 850-μm-selected sample reaches flux densities equivalent to SFR ~= 100 M yr –1 , enabling us to confirm that sub-mm galaxies form the high-mass end of the ‘main sequence’ (MS) of star-forming galaxies at z 〉 1.5 (with a mean specific SFR of sSFR = 2.25 ± 0.19 Gyr –1 at z ~= 2.5). Our results are consistent with no significant flattening of the MS towards high masses at these redshifts. However, our results add to the growing evidence that average sSFR rises only slowly at high redshift, resulting in log 10 sSFR being an apparently simple linear function of the age of the Universe.