EMU is a wide-field radio continuum survey planned for the new Australian Square Kilometre Array Pathfinder (ASKAP) telescope. The primary goal of EMU is to make a deep (rms ,10 mJy/beam) radio continuum survey of the entire Southern sky at 1.3 GHz, extending as far North as þ308 declination, with a resolution of 10 arcsec. EMU is expected to detect and catalogue about 70 million galaxies, including typical star-forming galaxies up to z , 1, powerful starbursts to even greater redshifts, and active galactic nuclei to the edge of the visible Universe. It will undoubtedly discover new classes of object. This paper defines the science goals and parameters of the survey, and describes the development of techniques necessary to maximise the science return from EMU.

Here, we describe the Compact Array Broad‐band Backend (CABB) and present first results obtained with the upgraded Australia Telescope Compact Array (ATCA). The 16‐fold increase in observing bandwidth, from 2 × 128 to 2 × 2048 MHz, high‐bit sampling and the addition of 16 zoom windows (each divided into further 2048 channels) provide major improvements for all ATCA observations. The benefits of the new system are: (1) hugely increased radio continuum and polarization sensitivity as well as image fidelity; (2) substantially improved capability to search for and map emission and absorption lines over large velocity ranges; (3) simultaneous multi‐line and continuum observations; (4) increased sensitivity, survey speed and dynamic range due to high‐bit sampling and (5) high‐velocity resolution, while maintaining full polarization output. The new CABB system encourages all observers to make use of both spectral line and continuum data to achieve their full potential. Given the dramatic increase of the ATCA capabilities in all bands (ranging from 1.1 to 105 GHz) CABB enables scientific projects that were not feasible before the upgrade, such as simultaneous observations of multiple spectral lines, on‐the‐fly mapping, fast follow‐up of radio transients (e.g. the radio afterglow of new supernovae) and maser observation at high‐velocity resolution and full polarization. The first science results presented here include wide‐band spectra, high dynamic‐range images and polarization measurements, highlighting the increased capability and discovery potential of the ATCA.

We present a CO(1-0) survey for cold molecular gas in a representative sample of 13 high-z radio galaxies (HzRGs) at 1.4 < z < 2.8, using the Australia Telescope Compact Array. We detect CO(1-0) emission associated with five sources: MRC 0114-211, MRC 0152-209, MRC 0156-252, MRC 1138-262 and MRC 2048-272. The CO(1-0) luminosities are in the range L CO ∼ (5 − 9) × 10 10 K km s −1 pc 2 . For MRC 0152-209 and MRC 1138-262 part of the CO(1-0) emission coincides with the radio galaxy, while part is spread on scales of tens of kpc and likely associated with galaxy mergers. The molecular gas mass derived for these two systems is M H2 ∼ 6 × 10 10 M (M H2 /L CO = 0.8). For the remaining three CO-detected sources, the CO(1-0) emission is located in the halo (∼50-kpc) environment. These three HzRGs are among the fainter far-IR emitters in our sample, suggesting that similar reservoirs of cold molecular halo gas may have been missed in earlier studies due to pre-selection of IR-bright sources. In all three cases the CO(1-0) is aligned along the radio axis and found beyond the brightest radio hot-spot, in a region devoid of 4.5µm emission in Spitzer imaging. The CO(1-0) profiles are broad, with velocity widths of ∼ 1000 -3600 km s −1 . We discuss several possible scenarios to explain these halo reservoirs of CO(1-0). Following these results, we complement our CO(1-0) study with detections of extended CO from the literature and find at marginal statistical significance (95% level) that CO in HzRGs is preferentially aligned towards the radio jet axis. For the eight sources in which we do not detect CO(1-0), we set realistic upper limits of L CO ∼ 3 − 4 × 10 10 K km s −1 pc 2 . Our survey reveals a CO(1-0) detection rate of 38%, allowing us to compare the CO(1-0) content of HzRGs with that of other types of high-z galaxies.

The largest galaxies in the Universe reside in galaxy clusters. Using sensitive observations of carbon-monoxide, we show that the Spiderweb Galaxy -a massive galaxy in a distant proto-cluster -is forming from a large reservoir of molecular gas. Most of this molecular gas lies between the proto-cluster galaxies and has low velocity dispersion, indicating that it is part of an enriched inter-galactic medium. This may constitute the reservoir of gas that fuels the widespread star formation seen in earlier ultraviolet observations of the Spiderweb Galaxy. Our results support the notion that giant galaxies in clusters formed from extended regions of recycled gas at high redshift.The formation of the largest galaxies in the Universe is thought to be a two-stage process. For the last 10 Gyr, these giant galaxies have grown mostly by cannibalizing smaller galaxies (1,2). However, computer simulations predict that in an earlier phase, lasting a few Gyr, their stars condensed directly out of large reservoirs of accreted gas (3,4).We present observational evidence for an extended gas reservoir fueling star formation in the massive Spiderweb Galaxy, MRC 1138-262, which is located in a proto-cluster at a redshift of z = 2. 161 (5-9). The Spiderweb Galaxy is not a single galaxy, but an aggregation of protocluster galaxies. They are embedded in a giant halo of atomic (neutral and ionized) hydrogen gas, which radiates Lyα emission across a region of ~200 kpc (6). The central proto-cluster galaxy has a super-massive black hole at its core, which emits jets of relativistic particles visible in radio observations (5). Observations suggest that the proto-cluster galaxies will eventually merge and evolve into a single, giant elliptical galaxy in the center of the cluster (10). We therefore refer to the Spiderweb Galaxy as the entire region encompassed by the Lyα halo, and to the gas between the proto-cluster galaxies as the inter-galactic medium (IGM).Earlier observations of line emission by carbon-monoxide revealed the presence of large amounts of molecular gas in the Spiderweb Galaxy (11). Molecular gas is the raw fuel for the formation of stars, so observations of molecular gas give us insight into the processes driving the evolution of the distant Spiderweb Galaxy. We have obtained new, sensitive observations of 12 CO (J=1→0) with the Australia Telescope Compact Array (ATCA, 90 hour exposure time) and the Karl G. Jansky Very Large Array (VLA, 8 hour exposure time) (12). The ATCA observations were optimized for detecting low-surface-brightness emission from broadly distributed CO, with a 4.8ʺ″×3.5ʺ″ resolution. The VLA observations complement the ATCA data with a higher 0.7ʺ″×0.6ʺ″ resolution, sensitive to small-scale features but not to large-scale ones. Sampling these different spatial scales allows us to obtain a complete picture of the CO distribution, from the gas in the individual proto-cluster galaxies to that across the IGM. Figure 1 shows that the CO emission in the ATCA data covers a region of ~70 kpc around the central...