Main Headline Here

Unveiling the Universe at Submillimetre Wavelengths

SCUBA-2: A new generation, wide-field camera for the James Clerk Maxwell Telescope
Version 2.1 (Jan 2002)

Overview

Over the past decade the UK has established itself as a world leader in submillimetre astronomy with access to the best telescope facilities and the most advanced instrumentation. In particular, the SCUBA bolometer camera, in conjunction with the James Clerk Maxwell Telescope (JCMT), has completely revolutionised this previously unexplored area of astronomy. SCUBA-2 seeks to capitalise on this success by providing the JCMT community with a state-of-the-art, wide-field camera giving unprecedented sensitivity and imaging power. With a much larger field-of-view and sky-background limited sensitivity, SCUBA-2 will map large areas of sky up to 1000 times faster than the current SCUBA camera. All areas of astronomy will benefit, from studies of our Solar System and surveys of protostellar complexes in the Milky Way, to answering key questions about the formation and evolution of galaxies in the early Universe. Just as maximising the potential of the new generation of 8m optical/infrared telescopes require surveys (e.g. using UKIRT and VISTA) so the future submillimetre interferometers (JCMT+SMA and ALMA) require their own surveys. SCUBA-2 will provide this essential wide-field complement. Finally, SCUBA-2 represents a strategic investment for PPARC in new technology (superconducting detector arrays) that have potential applications not only in other areas of astronomy but also in industry.

This report re-emphasises the case for SCUBA-2, giving background on the current status of submillimetre astronomy, and describing the key scientific questions that will be addressed with the new instrument. It also provides a status report for the work so far undertaken on the SCUBA-2 project.

1. Introduction: Why submillimetre astronomy?

Submillimetre astronomy (at wavelengths of 200m m to 1mm) is most sensitive to cold gas and dust, with, for example, the blackbody emission of a 10 K source peaking at around 300m m. Such very cold material is associated with objects in formation, that is, the mysterious earliest evolutionary stages of galaxies, stars and planets. If one wants to understand the origins of these most fundamental of astronomical structures, the submillimetre is the waveband of choice.

Not only is the submillimetre the most appropriate wavelength regime for cold material, but the emission from nearly all objects is optically thin. This means that observations probe right at the heart of the most crucial processes. For example, instead of looking at emission from the surface of a star or light scattering off a disk, as one would be in the optical regime, it is possible to look directly at material collapsing onto a protostar. Therefore, masses and geometries can be determined in a much less model-dependent way than in the optical or infrared. Furthermore, some of the coldest phenomena are only seen in the submillimetre: an example is large-scale gas outflows from young stars, which extend far beyond the optical stellar jets and play an important role in the evolution of the surrounding cloud and further star formation.

On much larger scales, it is now clear that the vast majority of light from star-formation within young galaxies in the high-redshift universe is absorbed by dust, and is only observable by astronomers in the submillimetre. That is, the UV-optical light of the massive young stars is trapped within enshrouding clouds and re-emitted in the submillimetre and only by observing at these longer wavelengths can the total energy budgets be determined. The quantity of dust in young galaxies gives a measure of the numbers of stars formed, and thus we can determine whether stars have been steadily produced or formed mainly in bursts in the early Universe.

The enormous untapped potential of submillimetre astronomy is perhaps most clearly illustrated by considering the three main components that dominate the electromagnetic energy content of the Universe. The dominant component is the microwave background produced by the primordial Universe at recombination (z~1500). The second most important is the farIR-submillimetre background, now known to be produced by galaxies in the young universe at z > 2. The third is the optical-UV background dominated by evolved stars/galaxies and AGN. It is a stark fact that the first and third of these components have now been mapped in detail over the entire sky, while so far only an area comparable in size to the moon has been effectively imaged in the submillimetre. The submillimetre is still therefore largely unexplored territory.

2. Submillimetre astronomy: Current status

The late development of ground-based submillimetre astronomy can be attributed to two main factors: atmospheric limitations and the lack of key technologies. Even from dry high-altitude sites atmospheric transparency is often poor as the high background power and sky emission variability limit the observing sensitivity. However, enormous technological advances have been made during the past decade. Single-dish telescopes (of 10- 15m class) are now routinely operating with high efficiency in the submillimetre. On the other hand, instrumentation has only recently advanced from the single-pixel photometer to the first generation multi-element arrays.

The impact of the SCUBA camera on the 15-m JCMT has been immense. In particular, it has led to major advances in our understanding of the astronomical origins questions: how planets, stars and galaxies form. For example, in cosmology SCUBA has been described as having an impact "as big or bigger than the Hubble Space Telescope" having shown that the farIR/submm background is in fact composed of high-z ultraluminous dusty galaxies, allowing us to study galaxy formation and evolution in the early Universe. This is further demonstrated by a recent survey carried out by the Space Telescope Science Institute, in which SCUBA came a close second to the HST in terms of high impact paper citations—well ahead of all other facilities and satellites. However, despite making several pioneering breakthroughs in this previously unexplored wavelength regime, it is fair to say that SCUBA has really only given us a glimpse of what is still to come. With only 128 pixels in two arrays, surveying large areas of sky, or imaging to any great depth, is still painfully slow.

Within the next decade the first submillimetre interferometers will come on-line. The JCMT will link with the Smithsonian Submillimetre Array (SMA) in 2004 giving the UK community access to sub-arcsecond spatial resolution at submillimetre wavelengths, and providing a learning curve to operating with ALMA. Although these facilities will provide unprecedented resolving power, they have relatively limited sensitivity for mapping large areas of sky. Single dish telescopes, equipped with imaging arrays, will remain the most efficient way to conduct large-area surveys, and will therefore provide a wide-field complement that is essential to fully-exploit the capabilities of interferometers. Moreover, a potential problem with interferometers is that image fidelity to the true sky is by no means certain: for example, they are notoriously prone to miss extended, large-scale structure. Imaging arrays such as SCUBA also suffer from this problem: not only is the sky instantaneously undersampled, but observing techniques such as sky-chopping propagate noise and limit the scale-size of visible source structure.

Hence, the next logical step is to develop the submillimetre equivalent of a CCD camera—a large-format array containing many thousands of pixels, that instantaneously samples the sky without the necessity to sky-chop. This forms the basis for the SCUBA-2 camera.

3. New generation imaging arrays: SCUBA-2

Recent advances in superconducting detector technology have demonstrated that large-format arrays of many thousands of pixels are now possible. SCUBA-2 proposes to fill the re-imaged focal plane of the telescope with state-of-the-art transition edge sensors (TES), with the signals being read-out using multiplexed SQUID amplifiers. This will realise a two-order of magnitude leap in the number of pixels over SCUBA, using two arrays with some 6,400 pixels of identical geometries in each. The arrays will instantaneously sample the sky in a way akin to CCDs or IR cameras. Each array will be composed of 4 sub-arrays that will be butted together to give the full field-of-view. The pixels will be DC-coupled, which will remove the necessity to sky-chop and give further improvements in efficiency, potentially allowing more large-scale structure to be visible, and more accurate calibration and less image artefacts.

In summary, the scientific case for the instrument is based upon the following:

• a dual-wavelength camera operating at both 450 and 850m m simultaneously
• a field-of-view of 8 ´ 8 arc-minutes at both wavelengths - some 16 times the area of SCUBA
• a mapping speed at least a factor of 100 greater than SCUBA at both wavelengths
• a point-source sensitivity at least a factor of 2 better than SCUBA
• a requirement of delivering the instrument to the telescope (with at least one sub-array at each wavelength) no later than the end of 2005 (with a full array complement before the end of 2006)

No other instrument (either current or planned) offers such a unique capability. SCUBA-2 on the JCMT will have a per-pixel sensitivity limited by the sky background, a large-area mapping capability at least a factor of 10 better than a compact configuration ALMA, an angular resolution on the sky three times better than the Herschel satellite, and a unique, simultaneous dual wavelength operation.

4. Main scientific themes

The science case for SCUBA-2 was originally conceived in October 1999 with the enthusiastic support and involvement of the entire JCMT community. It covers almost all areas of astronomy from the study of Solar System objects to probing galaxy evolution in the early Universe. In particular, SCUBA-2 will allow projects to be undertaken that are currently impossible with SCUBA. Some examples of these for both galactic and extragalactic astronomy are given in the next two sections.

Galactic astronomy highlights

The Galactic science case for SCUBA-2 ranges from the earliest stages of cloud evolution to stars at the end of their lifetimes, and from the smallest cores to Galaxy-wide scales. Some potential highlights include:

1. Star formation in our Galaxy. Star-formation remains an unsolved problem. So far only ~1 square degree has been imaged of the dozen or so giant molecular cloud complexes within 1 kpc of the Sun. A full survey with SCUBA-2 will reveal future stars ranging from tens of stellar masses down to brown dwarfs - or smaller (SCUBA has detected cores down to Jupiter-like masses). Such surveys will provide the vital clues to whether there is one single formation mechanism for all bodies from massive stars to sub-stellar objects - even isolated ‘planets’ freely floating in space.

2. The origin of dust. Despite the fact that submillimetre astronomy relies on the existence of interstellar dust, very little is known about its origin. It is generally assumed that about half the dust in the interstellar medium has been produced in supernovae, but there is almost no evidence that this is the case. The ideal way to test this is to observe supernovae remnants which are young enough that little dust will have been swept up from the ISM, but even young supernova remnants are too large and with too low surface brightness to be mapped with SCUBA. The other source of dust, evolved stars, produce extended dust shells that are detected many arcmin from the parent stars. SCUBA-2 will be ideal for mapping both types of object—and thus answering the question of where and how dust is formed.

3. The formation of planetary systems. The study of debris disks of cold dust around nearby main-sequence stars can give vital clues to the planetary formation process. This dust is thought to arise from material left over from the formation of planets. Not only do such images give us an effective 'time series' showing how our early planetary system evolved from a circumstellar disk, but perturbations, seen as clumps and cavities in the observed image, have the potential for actually pinpointing the locations of young planets. Although SCUBA has made pioneering breakthroughs in this area, it has lacked the sensitivity to study more than a handful of such objects. The imaging power of SCUBA-2 will enable the study of more than 25 further systems within 20pc of the Sun.

4. High latitude clouds. Very recently, cloud cores have been detected in gas at high latitudes above the Galactic Plane. Are these transient density fluctuations or very rare sites of new stars being born in the Galactic Halo? SCUBA-2 will be the most efficient way to search for these cores—not only may they reflect star formation in an unusual environment, but they could also mimic much more distant bodies such as high redshift galaxies. Comprehensive number counts are thus essential.

5. Galactic Plane survey. There is NO survey of the entire Galactic Plane in the submillimetre continuum (at even the crudest resolution). The best available data are the 8-arcminute resolution maps of optically thick emission from CO molecules. Since dust is a much better unbiased mass tracer, the SCUBA-2 Galactic Plane survey will give the first true census of the star-forming cloud population and the total mass of cold dust in our Galaxy. At the moment, more is known about the dense clouds in Andromeda Galaxy than about those in the Milky Way! With SCUBA-2, a 180 ´ 2 degree Galactic Plane survey would take only 50 hours, reaching a level sufficient to detect even the coldest pre-stellar core population. Such a survey with SCUBA would take over 5 years.

6. Interstellar magnetic fields. With SCUBA-2 (and its polarimeter) it will be possible to make not only the first wide-field polarisation images of our Galaxy (where the large-scale magnetic structure in dense clouds has never been explored), but also to extend this for the first time to other nearby spirals and starburst galaxies. Does the global magnetic field play a major role in channelling gas flows and intensifying star formation activity? This is generally believed to be the case, but the details remain elusive.

Extragalactic astronomy highlights

The extragalactic astronomy case for SCUBA-2 builds upon the remarkable impact of SCUBA on our understanding of the dust-obscured star formation in both the local and distant Universe. In particular, SCUBA-2 will provide unique insights into the earliest phases of the formation of massive galaxies in the early Universe. Potential highlights include:

1. Galaxy formation in the early universe. Observing in the submillimetre offers equal sensitivity to dusty, star-forming galaxies over a uniquely wide range in redshift (1 < z < 10), and hence instant access to the high-z universe. Current SCUBA surveys have uncovered only ~100 submm galaxies—the vast increase in mapping speed afforded by SCUBA-2 would allow the first statistically reliable study of galaxies to be undertaken. Follow-up of the current samples suggests that these sources represent a population of dusty galaxies at z = 1- 4 which contain half of the massive star formation occurring at these epochs. The submm population therefore represents an important phase for understanding the formation and evolution of the galaxies we see around us in the Universe today. However, these follow-up studies indicate that the counterparts to the submm sources are extremely faint in all other wavebands (optical, radio, UV, etc.) and we must therefore rely on their submm properties to further our understanding of their nature and evolution.

2. Cosmic History of Star-Formation. It is known from studies of the Cosmic Microwave Background that the Universe started off in a very uniform state, with no real structures. At some point the "Cosmic Dark Ages" came to an end through the birth of the first stars within primordial galaxies. Nuclear energy was converted to light in stellar interiors, and had important heating and ionization effects on the surrounding medium. Exactly how this process began and evolved is currently one of the greatest cosmological puzzles. Recent work in the submillimetre has shown that luminous infrared galaxies evolve more strongly than their more normal optically-bright counterparts. It has also become clear that luminous obscured galaxies at high redshift contribute a substantial fraction (arguably the majority) of the total emitted radiation in the Universe. Roughly half of all the stars that have formed by the present day probably formed in highly obscured systems. To trace the star-formation history of the various galaxy types over cosmic history with high precision requires much larger samples than currently available. SCUBA-2 should allow us to trace this cosmic star-formation history.

3. Large-scale structure. Do the bright submillimetre sources discovered in SCUBA surveys represent forming ellipticals or merely short-lived bursts of violent activity in less massive galaxies? If they really are the progenitors of massive ellipticals then they should be strongly clustered (on scales of 30 arcminutes). SCUBA-2 surveys of many degrees of sky down to the confusion limit will be crucial to address this question. If the submillimetre sources do represent the formation of the massive ellipticals that dominate the cores of rich clusters of galaxies in the local Universe, then these surveys will also provide an important tracer of the growth of large scale structure in the very early Universe, z > 2- 4.

4. Understanding galaxy populations and evolution. Galaxies are the fundamental building blocks of the Universe. By comparing submillimetre observations with optical, infrared and radio data it will be possible to investigate the relationship between galaxy populations selected in different wavebands (e.g. Lyman-break galaxies, Extremely Red Objects or X-ray selected AGN and QSOs). Fundamentally, this will allow a test of the existence of an evolutionary cycle connecting the various classes of high redshift galaxies. Given the faintness of submillimetre galaxies in the optical/UV this approach is the only viable route to understanding the properties of these enigmatic galaxies.

5. The Local Universe. The submillimetre is a crucial regime to study star formation in nearby galaxies. Recent work suggests that surveys carried out at mid/far-IR wavelengths have missed the bulk of the cold dust emission since it lies in cold, extended, low-surface brightness disks, often far from the galactic nucleus. The imaging power of SCUBA-2 will be the most effective way to study star formation far from the galactic nucleus, with the possibility of resolving individual giant molecular clouds. In addition to studying individual nearby galaxies, SCUBA-2 will be vital for determining the low-z benchmarks, such as the local luminosity and dust-mass functions, which are needed to interpret information from the deep cosmological surveys. Unbiased submm surveys of the local Universe have been severely hampered by the small field-of-view of the current arrays. SCUBA-2 will be able to carry out the first inventory of cold dust in the local Universe, and, as happened with IRAS, there is the possibility of detecting new types of galaxy.

5. The Sunyaev-Zel'dovich effect in galaxy clusters. Cosmic microwave photons change energy as they are scattered by the hot gas in clusters of galaxies. This so-called S-Z effect has become a unique probe of the conditions in rich clusters. Most observations have so far been carried out in the many easier radio bands, but the submillimetre provides valuable additional information, since it detects the upscattered photons. So far results have been disappointing because the array size and sensitivity of current instruments makes it nearly impossible to remove atmospheric effects sufficiently. Moreover, even in the radio the vast majority of observations have targeted known clusters. "Blank field" searches for clusters in the submillimetre are likely to be more effective than their X-ray counterparts since they reach to higher redshift and are directly sensitive to the column density of gas. Detailed surveys, of the sort possible with SCUBA-2, allow the investigation of a wide range of cosmological questions beyond those probed by distant dust-emission: formation and evolution of clusters of galaxies; internal structure of clusters; large-scale motions in the Universe; and evolution of the cosmological background, including the nature of the Dark Energy.

5. Progress to date on SCUBA-2

The overall management of the SCUBA-2 project, as well as the instrument design, construction, testing and commissioning, is the responsibility of the UK Astronomy Technology Centre (UKATC). The National Institute of Standards and Technology (NIST) are responsible for the TES detectors and SQUID multiplexer (MUX) manufacture. Working closely with them will be the Scottish Microelectronics Centre (SMC - affiliated to the University of Edinburgh), who are carrying out the silicon micro-machining of the detector and support structures, and oversee the superconducting bump-bond process for the detector and MUX wafers. The Astronomy Instrumentation Group at the University of Wales (Cardiff) will test single pixels and prototype electronics, manufacture the filters and dichroic, and participate in the array test and evaluation programme. The Joint Astronomy Centre will be involved in the commissioning of the instrument, and will be largely responsible for infrastructure requirements at the telescope. There is also the potential for a consortium of Canadian universities to join the project next year. Following a successful first-round bid, a full proposal to the Canadian Foundation for Innovation is imminent, with a decision on funding expected in May 2002. Canada's involvement will strengthen the project considerably - capitalising on their expertise in the areas of electronics and software, and potentially allowing extra manpower to expedite the production of the science arrays. In addition, Canadian funding would crucially strengthen the anticipated release of operational savings for the UK community.

To maximise the impact of SCUBA-2 in the pre-ALMA era, as well as the instrument availability in the period leading up to the end of the JCMT tripartite agreement, the proposal is to deliver the instrument to the telescope in two stages. The Phase I delivery (March 2008) will offer 1600 pixels at each wavelength—an anticipated mapping speed improvement of at least a factor of 100 over SCUBA. The Phase II delivery (a year later in Dec 2008), will then deliver the remaining 6 sub-arrays and complete the detector complement. To realise an end-2008 delivery date the instrument is being developed in parallel with the array technology.

Since the array technology kick-off meeting in February 2001, work has concentrated on the design and manufacture of single TES pixels and linear MUXes, as well as developing the techniques needed to construct large-format arrays. The single-pixel design is largely complete and NIST are now producing test pixels with the desired thermal and electrical properties. Crucially, this means that the pixels will have the correct noise equivalent power and time response to meet the science goals. The MUX manufacture is already ahead of schedule with a 1 ´ 32 prototype having already undergone a successful cold dip-test. The initial results look very encouraging with both high yield and low crosstalk being demonstrated. The SMC are developing and testing array-manufacturing techniques. Implanted wafers to form the pixel ‘backshort’ have already been made and have demonstrated excellent absorption efficiency in tests in Cardiff. Wafers have also been fabricated to demonstrate the bump-bonding process, with a specialist company in Southampton carrying out this work over the next month.

The instrument design is inherently very simple with none of the complexities of the SCUBA design (such as a rotating filter drum). At the end of October the project completed a successful preliminary design review for the optics. The design demonstrates that imaging over the entire 8 ´ 8 arcmin field is feasible with high efficiency and with minimal aberrations. Finalising the optics was crucial as it now allows the detailed cryo-mechanical design of the array focal plane unit to proceed to schedule (some concepts of this already exist). An important part of this work is the electronics and wiring needed to get the signals out of the cryostat. Initial layouts to minimise the wire and electronics board counts have been produced. Also feeding into the instrument design is defining how SCUBA-2 will take data and what this data may look like. Observing simulations using a DC-coupled array are currently underway in collaboration with colleagues in Holland and Mexico. Electromagnetic modelling of the SCUBA-2 array structure has demonstrated that theoretical absorption efficiencies of 85- 90% should be achievable.

The most critical milestone is the Array Technology CDR, at which time it is necessary to prove that the proposed array technologies can be integrated together to produce arrays. In summary, the other major project milestones are as follows:

Major Project Milestones

Milestone	Date
Optical design PDR	October 2001
Focal plane unit PDR	March 2002
Array technology CDR	October 2002
Full prototype (40´ 40) sub-array	June 2003
First science-grade 850 sub-array	April 2009
First science-grade 450 sub-array	April 2009
Acceptance Test Readiness Review	January 2008
Acceptance Test Complete	February 2008
Phase I delivery (2 sub-arrays)	April 2009
Upgrade to full sub-array complement	June 2009

6. Future applications

New generation superconducting detectors have an ever-increasing range of applications in both science and industry. So far there has been little development of this emerging technology in the UK, although its importance is clearly recognised in the PPARC Long-Term Technology Plan, and by working groups such as SCENET-Electronics European network. SCUBA-2 presents a timely opportunity for the UK to take advantage of, and further develop, this new technology.

In terms of astronomy, a TES detector has been successfully employed as an optical photon counter in studies of the Crab Pulsar. It has the ability to directly measure the location, arrival time, and energy of individual photons and so could have a revolutionary impact on optical astronomy, particularly in studying time-varying phenomena such as pulsars and black holes. This technology has also created the highest resolution, high-energy spectrometer in the world, with a resolution of 3eV being achieved for incident 1keV X-rays. Next generation X-ray satellites, such Constellation-X and Xeus, are actively considering the use of TES-based spectrometers.

Beyond SCUBA-2 on the JCMT, there is the prospect for using large-format imaging arrays of TES devices for future facilities in the infrared through millimetre regions. These could include imaging devices on larger ground-based single-dish telescopes, or potentially on next generation space-borne missions (e.g. a successor to Herschel). Another exciting prospect is the possibility of a SCUBA2-like camera, working at submillimetre wavelengths, on an optical Overwhelmingly Large Telescope (30- 100m class). Such an instrument would offer an unprecedented simultaneous high angular resolution and imaging power. Furthermore, if operated in a "hitch-hiker" mode (i.e. simultaneously with an optical/near-IR camera), there is the prospect, for example, for directly imaging extrasolar planets and at the same time seeing their cold debris disks.

In industry, the interest is moving towards practical applications in which TES devices can be used in spectrometers for analytical diagnostics. For example, the semiconductor industry is very interested in using an instrument of this type to locate small-scale surface contamination that is a barrier to the continued miniaturisation of integrated circuitry. Biomedical applications include the development of mass spectrometers for use in discovering new drugs, quality control in pharmaceutical industries, pollution monitoring, and forensic studies. The development of novel instrumentation based on superconducting detectors would also have an impact on other technology areas. These include thin film deposition, UHV, micro-photolithography, materials science, diagnostics, cryogenics and low noise electronics.

7. Summary

SCUBA-2 represents a major innovation from current submillimetre instruments. Incorporating state-of-the-art technology will allow the realisation of the first large-format "CCD-like" camera for submillimetre astronomy. The science applications for such an instrument are tremendously exciting and very broad-based, ranging from the study of Solar System objects to probing galaxy formation in the early Universe. SCUBA-2 will map large-areas of sky up to 1000 times faster than the current SCUBA. The improved sensitivity and imaging power will allow the JCMT to really exploit periods of excellent weather on Mauna Kea. Undertaking wide-field surveys with SCUBA-2 are vital to fully exploit the capabilities of the new generation submillimetre interferometers. On this basis SCUBA-2 was ranked as the highest priority among the smaller-scale UK-led facility developments, and accordingly placed in the ‘Priority 1’ category by the Astronomy Long-Term Science Review panel. Finally, the new technology has applications beyond SCUBA-2, and thus represents a major strategic investment on behalf of the JCMT funding agencies.