Evolution of the SKA Science Case

Abstract

This chapter describes the evolving science case and how it informed the instrumental developments needed to build the SKA. It starts with an outline of the scientific context in 1990, and the tenth anniversary meeting for the Very Large Array in the USA (VLA). In the early 1990s a formal international working group was established through the International Union of Radio Science (URSI) to develop the science case for an SKA.

The dominant components of the science case involve the evolution of the universe as it is traced by neutral hydrogen, understanding the dark ages when the first stars are changing the state of the primordial hydrogen (the so-called Epoch of Reionisation) and the effect of the super massive black holes in the nuclei of galaxies. Radio observations are identified which provide unique and complementary information about the universe. These include magnetic fields, transient radio signals from pulsars and fast radio bursts (FRBs) and tests of general relativity.

The chapter includes a discussion of the way the case for building an SKA was adapted to meet the aspirations of the different stakeholders in a global collaboration and concludes with a description of the scientific advances already being made by the SKA pathfinders.

5.1 Introduction

By looking at the science proposed for the SKA over three decades, in various talks and reports, the way in which the emphases changed as the SKA concept was developed can be traced. Chapter 2 shows how the historical development of designs for a range of radio telescopes influenced the ideas behind the SKA. This chapter focuses on the role of the science case in shaping the final SKA design and discusses how it has evolved over time, reflecting the transition from the initial broad scientific opportunities through to more focussed key science projects and finally to the detailed and comprehensive science requirements that were needed to refine design options as the construction stage approached.

New scientific ideas emerged, and priorities have changed throughout this period. These changes will continue in the future, and it will be a challenge for the SKA project to adapt as new scientific opportunities arise. Unlike space projects which are fixed after launch, terrestrial telescopes are never finished.

The introduction chapter discussed the difference between astronomy which is an observational science and experimental sciences. These differences have influenced the ways the SKA developed into a global mega-science project. The tension between emphasis on a small number of key questions (experimental physics approach) and exploring a broader parameter space (observational astronomy approach) has affected the formulation of the science case.

5.2 The Context in 1990

The state of radio astronomy in the 1990s when the first ideas of a future SKA-scale radio telescope emerged shaped the scientific drivers for some of the specific pre-SKA projects discussed in Chap. 2.

By 1990, the Very Large Array (VLA) had been in operation for a decade. Radio astronomy was a very active area of research and had been in a privileged position from the 1950s through to the 1980s, making significant impact as the first astronomy outside the optical wavelength band. But then, quite dramatically, astronomy started to develop rapidly in many other wavebands (Infra-red, Ultra-Violet, X-ray, γ -ray) as observations from space became possible. Radio astronomy developments plateaued with relatively few new facilities being built for many years after completion of the VLA in 1980.

The possibility of measuring the epoch of reionisation in the early universe—now a major part of the SKA science case—was not yet being discussed. Detection of gravitational waves was only an aspirational dream. There was no accelerating universe to explain and no case yet for dark energy. At this time, observations of the cosmic microwave background and the big optical surveys measuring galaxy distances (redshifts) were making a significant impact on our understanding of the evolution and large-scale structure in the universe. The most distant objects in the universe had been pushed out to redshifts of five or six¹ and the hot topic was how galaxies formed and how they evolved. The case for dark matter had already been around for 30-plus years but the nature of dark matter was unknown then and is still unknown at the time of writing.

5.3 Pre-1990 Science Cases for a Large Collecting Area Radio Telescope

As recounted above, and in Sect. 2.4.2, the VLA meeting in Socorro in 1990 was the point at which a number of independent ideas and concepts for a very large telescope came together for the first time. Science goals that could only be achieved with a major increase in sensitivity were being discussed in different groups around the world. These initial science goals are described in more or less chronological order as an introduction to a discussion of the broader SKA science case and its evolution in the remainder of this chapter.

5.3.1 SETI and Project Cyclops: 1971

Cocconi and Morrison (1959) published an article in Nature suggesting the feasibility of interstellar signalling using radio, saying “The probability of success is difficult to estimate, but if we never search, the chance of success is zero.” As was described in Sect. 2.2.2.2, the 1971 NASA-AMES Cyclops design study led by Barney Oliver (Hewlett-Packard, USA) and John Billingham (NASA)² looked at the requirements for a radio telescope with the sensitivity needed to detect radio signals emanating from other civilisations spread throughout the galaxy, which would be extremely weak when they arrived at Earth. The final version of this inspirational report proposed an array of 1000 100-metre dishes. The Search for Extra-Terrestrial Intelligence (SETI) has been included as one of the SKA goals from the earliest days.

5.3.2 Govind Swarup’s Vision for the Next Generation Radio Telescope, 1978–1991

Govind Swarup (TIFR, India) had a very well-defined scientific project in mind in 1970 when he built the large cylindrical Ooty Radio Telescope discussed in Sect. 2.2.2.5. The lunar occultation observations of the radio source 3C273, with the Parkes radio telescope in 1963, led to the discovery of quasars³ by Maarten Schmidt (1963) and the first evidence for the existence of black holes. Swarup wanted to build a telescope with enough sensitivity to observe occultations of weaker radio sources to use as high redshift probes for cosmology studies. His innovative equatorially-mounted cylindrical telescope was a great success, providing high sensitivity at relatively low cost. As discussed in Chap. 2, Swarup went on to plan and build a succession of radio telescopes using innovative technology and always focussed on clear scientific objectives.

In 1988, he wrote a paper on the idea of building 1000 GMRT 45-m antennas (1 million m²) primarily for SETI observations (Swarup, 1988) and during GMRT construction, in 1991, he published another article on a concept for a 700,000 m² telescope comprising 160 × 75-m dishes (Swarup, 1991). A primary motivation was to understand how galaxies form, and what the gas was doing before it formed into galaxies. This had been the basic aim of the earlier Giant Equatorial Radio Telescope (GERT) , see Sects. 2.2.1.6 and 2.4.1.1. To do this there had to be enough sensitivity to see neutral hydrogen⁴ in the early universe before the first stars formed, something that would require a significantly scaled-up version of the GMRT. In the proposed science case for the 160 × 75-m telescope Swarup also mentioned observing pulsars⁵ and using the pulsar timing for gravitational wave detection, as well as for SETI. All these goals were prominent aspects of what became the SKA science case.

5.3.3 Soviet Union Square Kilometre Telescope, 1982–1990

As mentioned in Sect. 2.4.1.4, discussions of large collecting area radio telescopes go back to Semyon Khaikin and Yuri Pariiskii in the 1960s. By the 1980s the USSR Academy of Sciences had established a working group to explore the science case for a Square Kilometre Telescope (SKT) (Gurvits, 2019). The main science cases for the SKT included studies of the statistics of extragalactic sources (log N–log S⁶), pulsars, and extra-galactic radio recombination lines.⁷ Neutral hydrogen in our own galaxy and other galaxies, and the search for extra-terrestrial intelligence were also part of the science case. Pariiskii complemented this activity with his diagram (see Fig. 2.20) displaying the exponential increase in collecting area of radio telescopes over time and noting that the largest radio telescopes would reach 1 million square metres collecting area by 2000 if the trend from the first few five decades of radio astronomy continued.

5.3.4 UK, Hydrogen Array, Peter Wilkinson, 1985

Following a visit to the VLA in 1984, Peter Wilkinson.⁸ (U. Manchester, UK) made a proposal to the “Priorities for Astronomy in the United Kingdom” study group for a Large Radio Flux Collector. This proposal was based on the case for high angular resolution imaging and hence determining the velocity distribution and column density of atomic neutral hydrogen in nearby galaxies.⁹ This would require a collecting area 100 times larger than the VLA, and ten times larger than Arecibo (see Sect. 2.4.1.2). There was no UK support for this proposal and the idea lapsed until Wilkinson came to Socorro in October 1990 to attend IAU Colloquium 131 as discussed in Sect. 5.4.1.

5.3.5 The Netherlands Large Telescope: Robert Braun, Ger de Bruyn and Jan Noordam

After Robert Braun left the VLA in the late 1980s and went to the Netherlands Foundation for Research in Astronomy (NFRA) he was thinking about how to get sufficient sensitivity in a radio telescope to do extragalactic HI (see Sect. 2.4.1.3). Braun used Ger de Bruyn’s cosmological HI signal strength calculation code and it was apparent to him that a square kilometre collecting area would be needed to get galaxy HI emission detections out to cosmologically interesting distances.¹⁰ After extensive discussion with de Bruyn and Jan Noordam at NFRA on how to build such a telescope, they proposed an array of large collectors with a total collecting area of 1 km². This was later named Euro-16¹¹ It was similar in scale to the Swarup proposals and made a similar science case, but also emphasised the study of interstellar HI in extragalactic systems with sufficient sensitivity and resolution to observe the HI structure with a level of detail previously only possible in our own galaxy. Their science case also included: pulsars, transient radio sources and galaxy clusters, studying both thermal and non-thermal radio emission. They also envisaged this as the core of an incredibly sensitive very long baseline (VLBI) array.

5.3.6 Penticton Meeting “Radio Schmidt Telescope” 1989

In 1989 a meeting was held in Penticton to discuss an early initiative in Canada, led by Peter Dewdney (DRAO, Canada), to build the equivalent of a radio Schmidt telescope, a survey telescope but focusing on wide field-of-view science rather than higher sensitivity. This workshop was well attended by the international community and many of the technical topics raised were important for the SKA instrumental developments discussed in Chap. 6. For example, the Canadian proposal for an array of 100 small 12 m dishes (see Sect. 2.2.2.4) to maximise the field of view was the forerunner of the “large N—small D” design concept¹² that has played a major role throughout the development of the SKA. Small D increases the field of view and large N improves the imaging dynamic range and sensitivity.

Because the focus for this meeting was wide field of view and brightness sensitivity, galactic astronomy applications dominated the science case with both spectral line and continuum proposals to trace the large-scale structures in the galactic interstellar medium (ISM) and to measure nearby. The proposed array of small dishes was also well suited to radio imaging of the active sun. However, these goals did not require higher sensitivity and they did not get much emphasis in the future SKA science case. Extreme scattering events (ESEs) had just been discovered (Fiedler et al., 1987) and these made a strong case for surveys of transient radio sources, but even though this phenomenon has never been fully understood it did not re-appear in the SKA science case! Extragalactic astronomy requiring observations of large-scale structures such as the distribution and dynamics of HI in relatively nearby galaxies was included, as were the continuum observations of low brightness features in galaxy clusters. Observations of structure in the radio continuum emission from the cosmic microwave background (CMB) and the Sunyaev-Zeldovitch effect were considered. Multiwavelength astronomy had become popular so there was a general case for large radio surveys to match those being made at infrared and X-rays wavelengths.

5.3.7 Unexpected Discoveries

As discussed in Chap. 1, De Solla Price (1963) reached the conclusion that most scientific advances follow laboratory experiments and that the normal mode of growth of science is exponential. Subsequently, Harwit (1981) analysed discoveries in astronomy and concluded that most important astronomical discoveries were a result of technical innovation. De Solla Price (1984) pointed out that practitioners using new technology apply it to “everything in sight” often leading to the discovery of novel and surprising phenomena and this led to a recognition of the importance of exploration of the unknown when discussing the design of new facilities (Wilkinson et al., 2004). Some discoveries are predicted new phenomena which are either confirmed by an observation or are observed accidentally but still confirm an existing prediction. In radio astronomy there had been a very large number of serendipitous discoveries of the unexpected and this was the theme for a meeting on the 50th anniversary of Jansky’s discovery of radio emission coming from outside the earth (Kellermann & Sheets, 1984). The “Serendipity pattern” and its broader role in the nature of scientific discoveries is discussed by Merton and Barber (2004). It was well known in the radio astronomy community that while successful telescopes were built by visionaries, these telescopes often became best known for their unexpected discoveries, not for the reason they were built. There have now been a great many unexpected discoveries in radio astronomy; as summarised in a new book on the discoveries in radio astronomy (Kellermann & Bouton, 2023). Their Chap. 12 includes a dramatic discovery rate plot (Fig. 5.1)—with peak discovery rate between 1960 and 1980. These discoveries include many of the constituents of modern radio astronomy; radio galaxies, quasars, the cosmic microwave background, pulsars, masers, blackholes and many others.

Fig. 5.1

Cumulative number of discoveries at radio wavelengths vs time. (Credit: K. I. Kellermann, NRAO)

5.4 Convergence of Visions: VLA Tenth Anniversary, 8 Oct 1990

The IAU colloquium 131 celebrated the 10-year birthday of the VLA. A meeting was held in Socorro, New Mexico, 8–12 October 1990. The VLA had been operating for a decade and was having a huge impact in astronomy, as is discussed in Chap. 1.

This meeting is generally recognised by the radio astronomy community as where the birth of the global SKA concept took place. The ideas of a next generation radio astronomy facility with a square kilometre of collecting area at centimetre wavelengths, which were discussed at this meeting, had arisen spontaneously in four different countries (as described above) and led to the “born global” concept.

5.4.1 Hydrogen Array: Peter Wilkinson

During a break at the Socorro meeting, Jan Noordam talked to Peter Wilkinson about the Dutch proposal to build 16 100-m dishes. For a number of years Peter Wilkinson had also been thinking about the sensitivity requirements needed to observe HI in distant galaxies (see Sect. 2.4.1.2), so it was agreed that a talk on this topic be included in the meeting. Wilkinson agreed to give the talk and made the case for a square-kilometre collecting area array, which he called the Hydrogen Array (Wilkinson, 1991). Figure 5.2 shows an optical and an HI picture of the nearby grand design spiral galaxy, M81. These two pictures have some similarities but are also dramatically different. Wilkinson’s concept was that there is a universe of stars and a universe of gas. So, in addition to looking at the universe in the light of stars, there is huge value in looking at the universe in the light of hydrogen gas, the gas from which the stars and first galaxies formed. Wilkinson penned this often-quoted comment, “The encyclopaedia of the universe is written in very small typescript and to read it requires a very sensitive telescope.”

Fig. 5.2

Peter Wilkinson’s comparison of a Universe of Stars and of Hydrogen gas. The M81-M82-NGC 3077 group. (Left) Palomar Sky Survey Image. (Right) VLA image of the 21 cm line of HI. Yun et al. (1994). Credit: NRAO/AUI/NSF

5.4.2 Exponential Growth of Radio Telescope Sensitivity: Yuri Parijskii

Yuri Pariiskii (Pulkovo Observatory, Russia) also gave a talk at the Socorro meeting and showed the exponential increase in sensitivity of radio telescopes vs time (see Fig. 5.3). Ekers had shown a similar plot at the Prague URSI General Assembly in September 1990. The exponential growth curves have been discussed in Sect. 1.2.3 and the sensitivity plot for some of the major radio telescopes v time is plotted in Fig. 1.1. Kellermann (NRAO, USA) may have been the first to describe this exponential improvement in sensitivity with time and his plot is included in Sullivan (1984, Preface p. x). Pariiskii’s version of this plot (Fig. 5.3) is a measure of sensitivity (which improves as you go down in this plot) versus time. It is a log-linear plot so, again, there is the exponential trend, and the black dot in Pariiskii’s plot at the year 2000 would correspond to something like an SKA. When Pariiskii gave this talk in Socorro he added a fascinating twist to the story. In Fig. 5.3 the sensitivity is improving exponentially but the world’s man-made radio frequency interference (RFI) environment is getting exponentially worse as more RFI generating technology develops. His key message was that human ability to observe the universe at radio wavelengths is shrinking with time and therefore the next generation radio telescope must be built before these lines cross. Experience with radio observatories all around the world demonstrate that we are indeed running out of time! The impact of RFI, and the procedures needed to avoid this fate, are discussed in Sect. 5.8.2. By being clever, we may be able to suppress RFI by using adaptive RFI excision.¹³ We can also select radio quiet sites,¹⁴ however the increasing threat from low earth orbiting satellites (LEOs) will affect all sites on earth. Pariiskii did not publish his talk in the IAU Colloquium Series, but it was published a year later (Parijskij, 1992).

Fig. 5.3

Interference and sensitivity versus time. (Credit: Fig. 10 from Pariiskii, Y. (1992), “Radio astronomy of the next century”. Astronomical and Astrophysical Transactions, 1(2), 85–106. Reprinted with permission from the Eurasian Astronomical Society)

5.5 The Evolving Science Case

It is interesting to see how the science case evolved over the 30 years that have elapsed since the story started to be written in 1990. We have used four major sources of information:

• i. hard copies of presentations including the science case made at meetings involving the SKA in various countries over this period,

• ii. summaries of the science case included in the minutes of SKA meetings,

• iii. reports included in the SKA memo series and newsletter articles and

• iv. the SKA science books.

The first SKA science book was based on the Calgary meeting in July 1998 (Taylor & Braun, 1999), this was followed by the summary of the 7–9 April 1999 Amsterdam meeting “Perspectives on Radio Astronomy: Science with Large Antenna Arrays” (van Haarlem, 2000). Then a complete revision of the science case, edited by Chris Carilli (NRAO) and Steve Rawlings (Oxford, UK) was published in 2004. Finally, there is the massive two-volume edifice on the current science case published by SKAO in 2015.

The historical sequence of developments is given in the following sections and includes summaries of the actual presentation text in the boxes. We discuss the evolution of the different individual science areas in Sect. 5.10 and we have prepared a retrospective over-view of the evolving science case in the form of a matrix (Fig. 5.4). To simplify these comparisons, we have used a consistent terminology in the matrix but the text in the boxes is kept close to that used in the presentations at the time. This has been prepared from a much more detailed matrix with all the sub-categories included. Finally, we comment on how the science case impacted the development of the SKA concept in Sect. 5.11.

Fig. 5.4

An overview of the evolving science case. Blank regions are included in periods when a particular case was not emphasised

5.5.1 SKA Science Case in 1990

Box 5.1 is a composite science case from Swarup, Braun and Wilkinson presentations in the early-1990s. Note that the concept of the key science drivers did not materialise until well after 1990.

Box 5.1 1990 Science Case

• 21 cm HI from pancakes of gas before galaxy formation

  • Zel’dovich pancakes [the case for EoR had not yet been recognised]

• HI in high redshift galaxies

  • Surveys, large scale structure

  • Dynamical dark matter estimates

  • The “Great Attractor” debate

• Pulsars

  • Using pulsar timing for gravitational wave detection

  • Extragalactic pulsars detections

• Non-thermal radio continuum

  • The low luminosity emission from normal galaxies and weaker AGN at significant cosmological distances

  • New classes of stellar radio emission from stars in our galaxy

• Radio transients

  • The evolution of the radio emission from extra-galactic supernovae remnants

• Search for Extra-Terrestrial Intelligence (SETI)

The case for detecting the Epoch of Reionisation (EoR) had not yet been made but it was already clear that galaxies formed out of collapsing gas clouds. Ya Zel’dovich (Sternberg Institute, Russia) had proposed that to get rid of angular momentum, the gas would first collapse into flat pancakes and, after that, galaxies of stars would form (Zel’dovich, 1970). There were unsuccessful efforts at that time to try and detect these HI pancakes which would be the predecessors of the first galaxies with stars. Uson et al. (1991) claimed a VLA detection, but this was not confirmed by de Bruyn using the Westerbork Synthesis Radio Telescope (WSRT) or by Subramanyan and Swarup using the GMRT.

There was a case for measuring hydrogen over a large range of redshifts, to complement the big galaxy surveys of large-scale structure. HI observations to measure the velocity distribution could also provide mass estimates of galaxies including their dark matter. In 1990 the “Great Attractor” was a very hot topic. This was a large-scale gravitational anomaly traced by galaxy peculiar motions in the relatively nearby universe. Where was all the missing mass in the local universe causing these peculiar motions falling towards the Great Attractor?

From the very beginning, there was a clear science case for pulsars research which has become stronger over time. In 1993 the Nobel Prize in Physics¹⁵ was awarded jointly to Russell A. Hulse and Joseph H. Taylor Jr. (U. Massachusetts, USA) “for the discovery of a new type of pulsar, a discovery that has opened up new possibilities for the study of gravitation.” This was a pulsar in a binary neutron star system discovered using the Arecibo radio telescope (Hulse & Taylor, 1975). The circumstances leading up to this discovery are described by (Kellermann & Bouton, 2023) Chap. 7. The observation of the orbital decay of the binary pulsar confirmed the existence of gravitational waves (Taylor & Weisberg, 1982). Gravitational waves have now been detected directly [see Sect. 5.10.3] making this an even more active area of research. Already in 1990 it was also understood that pulsar timing could be used for detecting the predicted very long wavelength cosmological gravitational waves generated in the early universe. The possible detection of extragalactic pulsars was also considered at this time.

The ability to detect supernovae¹⁶ at great distance by observing non-thermal radio emission was considered important and radio continuum observations of the normal galaxies and weaker Active Galactic Nuclei (AGN) was included but with no specific additional goals.

There was some emphasis on finding new classes of stellar radio sources that might be detected, a topic which then disappeared out of the science case and has only re-appeared in the last few years as part of the search for radio transients and possibly transient radio emission from exoplanets.

The science cases nearly always included the evolution of life in the universe as one of the big questions. The emphasis on this varies between Searching for Extra-Terrestrial Intelligence (SETI) transmissions or other evidence for technology (“technosignatures”) and looking for markers for planetary formation and other evidence for living organisms (“biosignatures”).

5.5.2 Utrecht Meeting and the Euro-16 Proposal

On 27 September 1990 a “brainstorming” meeting was held in Utrecht¹⁷ to discuss future developments in Dutch radio astronomy following the huge success of the Westerbork Synthesis Radio Telescope which had been in operation since the 1970s. This brainstorming meeting led to the proposal¹⁸ for the “Euro-16” array with a total collecting area of 1 km² to provide enough sensitivity to detect neutral atomic hydrogen (HI) at cosmological distances, as already discussed in Sect. 5.3.5.

5.5.3 SKA Science Case 1994 and the URSI Large Telescope Working Group (LTWG)

One of the terms of reference for the URSI Large Telescope Working Group, the LTWG (Sect. 3.2.1), was to produce a concrete proposal supported by a well-defined scientific case. They did this at their first meeting,¹⁹ which was held at Jodrell Bank on March 21 and 22, 1994. They decided the major scientific drivers should be defined by assessing what science could be envisaged with a 100-fold improvement in telescope performance, and then seeing how this would influence the instrumental specifications. Box 5.2 is a summary of the science case which emerged from the Large Telescope Working Group study, together with items from hard copies of the overhead projector presentations made in this period.

Box 5.2 1994 Science Case

• Cosmology

  • HI detections of galaxies at cosmologically interesting distances

  • Statistics of emission from powerful radio galaxies and quasars

• Galaxy evolution, dark matter and large scale structure

  • Galaxy evolution between z = 0 and 5 using radio continuum and neutral hydrogen

  • Proto-galaxy and proto-cluster evolution at red-shifts of 1 to 10 via HI emission

  • Weak gravitational lensing to probe the distribution of dark matter

• Interstellar medium

  • Deuterium emission imaging in the Galaxy

  • Recombination line imaging in H, He, C and S

  • Magnetic fields

• Galaxies

  • Extend galaxy rotation curves to measure dark matter content

  • The power source for quasars and active galactic nuclei (AGN)

  • Extragalactic SN and SNR

  • Magnetic fields

• Pulsars

  • Using pulsars to find rare objects such as black-hole binaries

  • Detecting extragalactic pulsars

  • Pulsar timing to Test General Relativity

  • Pulsar timing to detect gravitational radiation

• Stars

  • Many research areas discussed; mass loss, planetary companions, proto-stellar discs and jets, solar flares.

• Solar system

  • Imaging planets

  • Planetary radar

  • Detecting asteroids

• SETI

  • Detecting extra solar system planets

  • ETI communication detection (SETI)

This report included some insightful additional suggestions. They asked how will the study of proto-galaxy evolution be impacted by the new competition with observations of the much stronger CO emission at mm wavelengths? They included the possibility of detecting the primordial recombination lines of H_n-alpha for n = 20 to 40 transitions from the cosmological recombination epoch in the very early universe (redshift z=1500) when the plasma from the big bang first combined. This is a very surprising inclusion which was not discussed further at the time but it would have had a big impact. It is now considered an exceptionally important opportunity for a specialised future telescope (Rao et al., 2015). The case for searching for Zel’dovich pancakes was fading away as the possible detections (see Sect. 5.5.1) were never confirmed. Magnetic field measurements using the Zeeman splitting of right and left circularly polarised HI emission or absorption were discussed, and the use of Faraday rotation was analysed, setting requirements on the frequency range and frequency resolutions. As a consequence of the recent discovery of exoplanets,²⁰ pulsar timing now includes searches for exoplanets as well as general relativity theory tests.

5.5.4 Pesek Lecture 1995

Ekers (1995) gave the Pesek Lecture, “SETI and the One Square Kilometre Radio Telescope” on 2 October 1995 at the 48th International Astronautical Congress in Oslo, Norway. The SKA telescope could be considered a significant step towards the realisation of project Cyclops for SETI (see Sect. 2.2.2.2). The SKA would provide a sensitivity 400 times that of the recently completed Project Phoenix survey using the Parkes radio telescope in Australia.²¹

In addition to summarising the SKA Science case and the exponential growth in sensitivity of radio telescopes, the lecture included the first discussion with the space community of SKA opportunities for solar system astronomy. Radar and thermal imaging would be possible for the more distant planets and for many of the planetary satellites. Radar observations of near-earth asteroids were crucial to make orbit determination possible at significantly greater range, an essential requirement for any asteroid collision avoidance program.

If the SKA were used as a communication ground station for deep space missions, high band width communication would be possible with extremely modest spacecraft communication antenna and power requirements. For example, the SKA would have had adequate sensitivity to achieve the original communication objectives of the Galileo spacecraft mission to Jupiter in 1995 even when using its low gain omni-directional antenna.²²

5.5.5 Oort Workshop 1997

On 2 June 1997 the topic of the Oort workshop in Leiden was Scientific Drivers for the Next Generation Radio Telescope.²³ This was a small workshop with 16 international experts from 7 different countries discussing the range of science that could be done with a next generation centimetre wavelength radio telescope. It was assumed that such a telescope should be capable of making high angular resolution high dynamic range images with at least a factor of 10 more sensitivity than existing arrays such as Westerbork and the VLA. The focus of the workshop was on the science, but Harvey Butcher (ASTRON, Netherlands) provided an insightful overview of the technical, political and social issues involved in such a major global project. Science topics included for discussion covered the full range of topics identified by the Large Telescope Working Group (Sect. 5.5.3). It also included, for the first time, a presentation by Piero Madau on the possibility of using the spectral signature of the 21 cm line to probe the epoch of reionisation (EoR) and heating in the early universe (EoR). This is the first presentation of a science case which over time became the most important driver for SKA-low (Sects. 5.5.20 and 5.10.2).

Following that meeting, George Miley (U. Leiden, Netherlands) made a proposal to ASTRON in July 1997 for a design study of a simpler high sensitivity low frequency array that could be built on a much shorter time scale than was projected for the SKA (see also Sect. 3.2.6.1). Miley was concerned that interest in radio astronomy would decline without such a focussed mission. There was an obvious need for different technologies at lower and higher frequencies (transition at about 300 MHz) and this led to the beginning of LOFAR (see Sects. 3.2.6.1 and 3.3.3.4.1) and the technology-based split of the SKA into separate lower and higher frequency solutions. However, the overlapping science cases were not split, and they continued to be discussed jointly under the one SKA umbrella.

5.5.6 Science with the SKA 1998

URSI had established the LTWG in September 1993. The six subsequent meetings of this working group were a forum for mobilising a broad scientific community to discuss the technical requirements and to cooperate in establishing the science objectives. In December 1997 the 1kT workshop²⁴ in Sydney included a half-day meeting of the LTWG to discuss the science drivers.²⁵ The May 1998 report developed from these discussions²⁶ represents the first international effort to document some of the many science goals that will be addressed by the SKA. They drew attention to a particularly noteworthy aspect. “This will be the world’s premier astronomical imaging instrument. No other existing or planned instrument in any wavelength regime can provide simultaneously: spatial resolution better than the Hubble Space Telescope (<0.1”), a field of view significantly larger than the full moon (1 square degree), the spectral coverage of more than 50% (γ/Δγ < 2) and a spectral resolution sufficient for kinematic studies (γ/dγ > 10⁴) and all at a sensitivity which is about 100 times that currently achievable.”

Box 5.3 1998 Science Case

• The Dark Ages

• Large-scale Structure and Galaxy Evolution

  • HI surveys out to z = 3

• Very Deep Fields

• Probing Dark Matter with Gravitational Lensing

• Circum-nuclear masers

  • H₂O megamasers

  • 0H megamasers

• Synchrotron Ageing and Evolutionary Studies of Radio Galaxies

• Fossil galaxies

• Halo Emission

• Parsec-scale radio structure in Active Galactic Nuclei

• Interstellar Processes

  • Cosmic Ray Origin

  • Supernova Remnants

  • HII Regions

  • Interstellar Propagation Effects

  • Carbon Recombination Lines

• Magnetic Fields

• Formation and Evolution of Stars

  • Detection of Stellar Radio Continuum Emission

  • Circumstellar Environments

  • Stellar Astrometry

• Transient Phenomena

  • Supernovae and Gamma-ray Bursts

  • Coherent Processes

  • Extra Solar System Planets (exoplanets)

  • Flare Stars

• Gravitational Radiation and General Relativity

  • Pulsar timing and searches

• Formation and Evolution of Life

  • Solar System Science

  • SETI

The summary of this very extensive science case is given in Box 5.3. It now included many topics not previously discussed and for each topic they included considerable detail on the value of the science and the technical requirements. They did not restrict the list to new science areas which will only be possible with the SKA. For the first time the “Dark Ages” were included as a major component of the science case. They explored the observable effects of the various possible sources of ionising radiation on the neutral hydrogen gas. The frequency range of interest now extended from 20 GHz down to 30 MHz but it was noted that different antenna technologies would be needed to cover this frequency range (see Sect. 5.7.3 and Chap. 6). Weak Gravitational Lensing (Kaiser & Squires, 1993) was another field which had opened up since the initial science discussions and SKA, with its well defined point spread function (PSF), as a very promising observational technique.

These topics were the basis for discussions at the International SKA Science meeting in Calgary, 17 July 1998. An interesting presentation²⁷ made at this meeting was “missing items” which identified new ideas that had emerged since the last URSI LTWG discussions. These included highly redshifted CO where the 3 mm lines were redshifted into the 1.5 cm SKA band, Sunyaev-Zel’dovich effects, gravitationally lensed HI, fast (msec) transients²⁸ and deep space communications.

5.5.7 Impact of a Multibeam Design on the Science Case: 1998

At the 1998 Calgary meeting the use of simultaneous multiple beams pointing in many different directions was being discussed. This was analogous to the multiple high energy projects sharing particle accelerator beams. This concept had quite an impact on the science case as is illustrated in Fig. 5.5. The multibeam concept emerged as an innovative new technology development, first as a natural option for an aperture array and later also as a driver for the Luneberg lens proposals—see Chap. 6. Note that the possibility of splitting the signal and forming multiple beams with no loss in S/N is only possible at radio wavelengths and is a fundamental quantum effect, see Radhakrishnan (1999), that dramatically changes design and observational strategies possible at radio wavelengths.²⁹

Fig. 5.5

SKA multibeam concept prepared for a brochure in 2003. Design: H. Sim, Credit: CSIRO Radio Astronomy Image Archive CRAIA-SKA006

In 2002 the International Science Advisory Committees (ISAC) Radio Transient Working Group³⁰ suggested at least 10 simultaneous beams were needed to monitor multiple sources simultaneously. They also pointed out that the concept of simultaneously doing many different observations at the same time opened up the opportunity for high-risk science which can be done commensally. For example, SETI was often included in the science case but rarely given much emphasis until the multibeam option opened more opportunities for such high-risk observations. The ISAC also pointed out that the multiple-beaming capability would make it possible to identify terrestrial radio-frequency interference in one beam in order to remove it from other beams.

5.5.8 SKA Science Book (Eds Taylor and Braun) 1999

As discussed in Sect. 3.2.4.2 a major international meeting focused on SKA science was held in Amsterdam in March 1999 (van Haarlem, 1999) and this raised greater awareness of the SKA in the wider astrophysics’ community. The detailed science case discussions which now included the broader astronomy community together with the science case developed by the LTWG led to the first publication of the SKA science case: Science with the Square Kilometre Array: a next generation world radio observatory (Taylor & Braun, 1999). This was published by the Netherlands Foundation for Radio Astronomy but not widely distributed outside the SKA community. Science with the SKA included material from 67 contributors. These were largely based on presentations made at the SKA Science Workshop held in Calgary, Canada, in July 1998 and in Amsterdam the following year. Planning for a next generation facility had led to the conclusion that a revolutionary new instrument at radio wavelengths was needed, one with an effective collecting area 30 times greater than the largest telescope ever built.

Taylor and Braun (1999) expressed the view that “With a spatial resolution better than the Hubble Space Telescope, a field-of-view (FoV) larger than the full Moon, and the ability to simultaneously image a wide range of red shifts (as many objects at high redshift in one long integration as the whole Las Campanas redshift survey of galaxies!), the SKA will be a discovery instrument to rival the NGST [Next Generation Space Telescope]³¹”. They summarised the main goals of SKA with a focus on the evolution of structure in the Universe on all scales.

• Probe the structure and kinematics of the Universe before the dawn of galaxies to understand the physics of the early Universe and how galaxies arose.

• To chart the formation and evolution of galaxies from the epoch of formation. To measure the evolution of the properties of galaxies, including dark matter halos, trace the star formation history of the Universe, and explore the origin of cosmic magnetic fields and their role in galaxy evolution.

• To understand key astrophysical processes relating to the process of star formation and the physical and chemical evolution of galaxies by studies of the local Universe.

• To trace the physical mechanisms that give rise to planetary systems, to understand the evolution of our own solar system, and to engage in definitive experiments to answer the question, “Are we alone?”

• To detect long-period gravitational waves, conduct exhaustive tests of general relativity, and explore the properties of nuclear matter within neutron stars.

Taylor and Braun optimistically assumed the SKA would need to be completed by 2010 to complement developments at other wavelengths. But they also noted that while the SKA had been born global, there was no international vehicle such as ESO or CERN to develop this concept.

The value of a very wide FoV had already been identified by the discussion of the Radio Schmidt telescope in Penticton in 1989 (see Sect. 2.2.2.4). Figure 5.6 compares the wide FoV of the SKA (Table 5.1) with that of the Hubble Deep Field and the MMA (Milli-Meter Array).³² By combining interferometry and phased-array receiver technology, the SKA will image a FoV of one degree at λ21 cm with angular resolution of 0.1”. Taylor and Braun again emphasised this advantage compared to any other existing or planned instrument in any wavelength regime.

Fig. 5.6

Comparison of the fields of view of the SKA (1 square degree at 20 cm) with those of the Hubble Deep Field and the MMA the proposed MilliMetre Array, which evolved into the Atacama Large Millimetre/sub-millimetre wave Array (ALMA). Credit: CSIRO Radio Astronomy Image Archive CRAIA-SKA002

This science case was used to develop the first quantitative set of SKA design goals which have not changed significantly since then.

The potential to study the epoch of reionisation (EoR) was clearly emerging as can be seen in the following quote from Taylor and Braun (1999, p. 23):

Prior to the epoch of full reionization, the intergalactic medium and gravitationally collapsed systems will be detectable in 21-cm radiation. Physical mechanisms that would produce a 21-cm signature are Lyα coupling of the hydrogen spin temperature to the kinetic temperature of the gas resulting from the radiation by an early generation of stars, preheating by soft x-rays from collapsing dark matter halos, and preheating by ambient Lyα photons. A patchwork of either 21-cm emission, or absorption against the Cosmic Microwave Background, will result. The Square Kilometre Array offers the prospect of measuring this signature, and so detecting the transitional epoch from a dark universe to one with light.

Table 5.1 SKA design goals^a

Dark Ages refer to a time before light existed in the Universe. As the first stars and galaxies formed, their light re-ionised the surrounding neutral intergalactic medium. This ended the Dark Ages and brought us the nearly completely ionised, light-filled Universe in which we live today. Figure 5.7 illustrates the Dark Ages and the reionisation era in the context of the cosmic history of the evolving universe.

At that time, the most distant quasar known had z = 5 (corresponding to redshifted HI at 200 MHz) so there was much speculation about the observability of this “epoch of first light” at low radio frequencies, a dream which has still not been realised more than 20 years later. The physics of the 21 cm HI emission and absorption and the experimental difficulties are discussed in more detail in Sects. 5.5.20 and 5.10.2.

Fig. 5.7

Cosmic History and the Dark Ages. Credit: S. Djorgovski et al., produced with the help of the Caltech Digital Media Centre

Hydrogen is the most abundant element in the Universe. With a sensitivity to the 21 cm hyperfine transition of H I, allowing detection out to z > 1, the SKA will follow the assembly of galaxies and can use their H I emission to measure large scale structure and early galaxy evolution. This provides a strong SKA science case and has been a key science driver since its inception. Details of narrow-deep and wider-shallow surveys were considered within the context of surveys proposed at other wavelengths. Highly redshifted CO was also considered a possibility.

The deep radio continuum surveys are an obvious application with the easy detectability and spectral resolution of regions of star forming activity added to the fainter Active Galactic Nuclei (AGN)s (e.g. Hopkins et al., 1998). Probing dark matter through weak gravitational lensing (Kaiser & Squires, 1993) of radio continuum sources was a new application taking advantage of the SKA’s very well-defined point spread function (PSF) and very large field of view. Another obvious extension of current radio galaxy research was to make high angular resolution observations of the radio structure surrounding the central black holes in active galaxies. This led to a science case to use the SKA core as the centre of a Very Long Baseline (VLBI) array spread over thousands of kilometres with milli-arcsecond angular resolution. The use of SKA stations in a future VLBI arrays extending over global baselines was also raised.

Other extragalactic topics covered were: OH and H₂O mega masers,³³ extragalactic supernovae remnants, scattering, and Faraday rotation. Taylor and Braun (1999) included a substantial discussion of the uses of high sensitivity radio observations to study many different stellar processes. This aspect of the SKA science did not receive much attention since then and has only recently re-emerged with the possible detection of stellar systems with exoplanets. The pulsar case included surveys to find rare binary systems which, they note, would become future gravitational laboratories. The case was well emphasised but not developed further in the SKA Science book (eds Taylor and Braun). Pulsar timing arrays and their potential to detect gravitational waves were noted. Solar system science, including radar, was another topic raised in this SKA Science book but not subsequently followed up (see Sect. 5.10.11).

SETI opportunities were quantified and the huge sensitivity advance over any existing surveys was tabulated. This may have partially triggered the inclusion of SKA advocates in discussions of future technologies which were taking place in the SETI community—e.g. (Ekers et al., 2002) “SETI 2020: A roadmap for the Search for Extraterrestrial Intelligence” which summarised a series of workshops held in Silicon Valley which included the SETI community, radio and optical astronomers and industry.

5.5.9 SKA Key Science Goals: 1999

This is a simplified version of the summary from Taylor and Braun (1999) extracting the key science goals.

• Probing the dark ages before the first stars

• Evolution of galaxies and large scale structure in the universe

• Origin and evolution of cosmic magnetism

• The cradle of life (terrestrial planets)

• Strong field tests of gravity via pulsars and black holes

• Exploration of the unknown

A key change is the inclusion of the potential detection of HI spectral line during the epoch of reionisation (EoR) which was triggered by the formation of the first stars and given the evocative name “the dark ages” . This replaces the old concept of searching for Zel’dovich pancakes (Zel’dovich, 1970) which would have very low-density contrast and would be difficult to observe compared to spectral changes caused by reionisation. This possibility opened up a new research area of astrophysical modelling and made a strong case for extending the frequency range down to below a few hundred MHz to look for the signature of the HI line at redshifts greater than 6.

By 1999 the other topics are still broadly similar except SETI has been turned into the “cradle of life” and now includes the search for extra-terrestrial planets. Whether or not SETI is explicitly included depends very much on the personal view of the presenter. Some astronomers have always questioned whether SETI is a legitimate area of scientific research.

Exploration of the unknown was often listed and from time to time was included specifically as a key science goal—see Wilkinson et al. (2004). In addition, recognising the long history of discovery at radio wavelengths (pulsars, cosmic microwave background, quasars, masers, the first extrasolar planets, etc.), the international science community also recommended that the design and development of the SKA include “exploration of the unknown” as a philosophy. Wherever possible, the design of the telescope should be developed in a manner to allow maximum flexibility and evolution of its capabilities in new directions and to probe new parameter space (e.g., time variable phenomena). This philosophy is essential given that many of the outstanding questions when the SKA will be in its most productive years,—are not even known today. This philosophy generated pressure for flexibility in the instrumental design as discussed in Chap. 6. However, some astronomers felt that it was difficult to easily include this in the science case because it does not generate clear-cut specifications and because there is also a perception that it indicates that astronomers do not know what they are looking for.

5.5.10 Australian Mid-Term Review: 2001

The Australian mid-term review “Beyond 2000”³⁴ included an assessment that the SKA would be an extremely versatile instrument that will be able to make major contributions to a broad range of astronomical topics. Box 5.4 lists the topics that were included in the science case.

Box 5.4 2001 Science Case

• The first stars and galaxies—The SKA will detect the very first objects formed after the Big Bang from the primordial hydrogen gas.

• The structure of the universe—The SKA will detect the ‘cosmic web’ of hydrogen and reveal the distribution of the matter in the early universe.

• Dark matter—The SKA will measure the amount of dark matter in the universe by observing the rotation of galaxies and the gravitational distortion of distant objects.

• Gravitational Waves—By timing many rapid pulsars, the SKA will be able to detect gravitational waves produced by the collisions of black holes anywhere in the universe.

• Planets around other stars—By accurate positional measurements of nearby stars, the SKA will be able to detect Jupiter-like planets and study their orbits.

Occasionally, and depending on the audience, applications beyond science were included. The mid-term review page 28 noted “The SKA also has applications in radio science communities outside astronomy, such as deep space communications and geodesy.”

5.5.11 Bologna Meeting: 2002

In January 2002 at the Bologna SKA meeting the Science Advisory committee (see Sect. 5.9.2) arranged a workshop dedicated to refining the science case for the telescope. Subgroups had been formed for the eight scientific areas identified and there were reports from the science subgroup chairs at the previous meeting in Berkeley. All groups provided reports in the SKA Memo series #5 to #13 (there is no SKA Memo #11).

• Milky Way and Local Neighbourhood Galaxies (SKA Memo #5)³⁵

• Radio Transients, Stellar End Products, and SETI (SKA Memo #6)³⁶

• Early Universe and Large-Scale Structure (SKA Memo #7)³⁷

• Galaxy Formation (SKA Memo #8)³⁸

• Active Galactic Nuclei and Supermassive Black Holes (SKA Memo #9)³⁹

• Life Cycle of Stars (SKA Memo #10)⁴⁰

• Intergalactic Medium (SKA Memo #12)⁴¹

• Spacecraft Communication (SKA Memo #13)⁴²

The Radio transient Working Group provided a detailed analysis of the forefront science to be pursued with the SKA and looked carefully at the implications for the SKA specifications. Pulsars, transients, and some SETI observations require observing modes that differ markedly from those designed for imaging modes of sources that do not vary with time. Therefore, care must be taken to incorporate these cases in the conceptual and design phases of the SKA. They also emphasised that the science predictions are based on the known populations of transient sources. The greatest return from such a survey will (should!) be the detection of currently unknown populations of sources.

The Early Universe and Large-Scale Structure Working Group focussed entirely on the newly emerging field of EoR observations (see Sects. 5.5.20 and 5.10.2). The case for observing at lower frequencies was greatly strengthened and the case for a high brightness sensitivity centrally concentrated array was promoted.

The Galaxy Formation Working Group discussed making sensitive, wide field 21 cm HI line and radio continuum surveys. They drew attention to the conflicting baseline configuration requirements between HI and continuum surveys and they also suggested, as a compromise, the centrally concentrated configurations that have now become the norm. They also remarked on the need to restructure the science case based on the important questions. An issue that is discussed further in Sect. 5.10.

The largest group included much of the traditional continuum radio astronomy community who study radio galaxies, active galactic nuclei (AGN) and the supermassive black holes in the centres of the galaxies that create the jets and extended lobes of radio continuum emission. Radio galaxies and AGN are bright and had been well studied for the previous 50 years so it was hard to identify high level science goals where the SKA would have a dramatic impact. Much of the science case involved incremental improvements. The large number of elements in any of the “large N—small D” SKA designs meant excellent quality images would be possible (see Sects. 5.7.4 and 2.2.2.4) so this was as important as high sensitivity. Understanding the energetics, stability, and internal flows of radio jets and radio galaxy lobes⁴³ requires high dynamic imaging over a wide range of angular scales and frequencies so array configurations with the widest possible range of baseline lengths were desired. One SKA specific requirement identified was the ability to distinguish thermal from nonthermal compact radio sources at high redshift which requires both high angular resolution and sensitivity.

5.5.12 Lorentz Centre Meeting: Leiden 2003

A new book on the science case was commissioned by the Science Advisory Committee at the Bologna meeting in 2002 and Schilizzi followed up by organising a meeting of all the International Science Advisory Committee Working Groups at the Lorentz Centre in Leiden from 10 to 14 November 2003. The Working Group chairs are listed in Table 5.2 and 45 astronomers from all around the world participated in this meeting. The goals were two-fold: to select key science projects, and to write the science case chapters for the new book. This was an extremely productive meeting⁴⁴ which significantly advanced the science case for the SKA as discussed in the next section.

Table 5.2 The ISAC working groups^a

5.5.13 SKA Science Book (Eds Carilli and Rawlings): 2003

Carilli and Rawlings⁴⁵ (Figs. 5.8 and 5.9) edited the publication of a complete revision of the SKA science case based on the Lorentz Centre meeting discussed in the previous section . Science with the SKA was published by Carilli & Rawlings in November 2004 and incorporated the latest results in astronomy, with emphasis on the most important outstanding problems.

Carilli and Rawlings emphasised the big picture by separating out the 5 key science projects (KSPs) that had been identified by a sub-committee chaired by Bryan Gaensler⁴⁶ (Harvard, USA): The Cradle of Life, Strong field tests of gravity, Cosmic magnetism, Galaxy evolution and cosmology and Probing the dark ages (EoR). The book then continues with another 43 chapters on different science projects with 106 contributing authors from 10 different countries.

Fig. 5.8

Chris Carilli, SWG chair 2002–2004 Credit: C. Carilli and NRAO

Fig. 5.9

Steve Rawlings looking over the GMRT, credit Katherine Blundell

It was this SKA science book, together with the 1999 Amsterdam meeting (Sect. 5.5.8), that brought the SKA to the attention of the entire astronomical community, extending it outside the community of radio astronomers.

These reports were authored by experts in a wide range of fields, some not traditional fields for radio astronomy. The incentive to contribute was driven by the thought that if an SKA was built, they wanted to ensure that their science area was covered by the design specifications. Outside the established radio astronomy community, there are contributions from the high energy particle physics community (Falcke et al., 2004) for the radio detection of high energy cosmic rays and neutrinos hitting the moon. There were also contributions on spacecraft tracking (Jones, 2004), precision astrometry (Fomalont & Reid, 2004) and planetary science (Butler et al., 2004).

Particularly noteworthy was the emergence of a range of transient science opportunities that had significant instrumental implications (see Chap. 6). The dynamic radio sky was summarised by Cordes et al. (2004) in a prescient paper which noted that extragalactic transients, which may be detectable with the high sensitivity of the SKA, are necessarily compact and would have measurable scattering and dispersion, all factors that would offer unique opportunities to probe properties of the intervening medium. The serendipitous discovery a few years later of Fast Radio Bursts (Lorimer et al., 2007), using the multibeam receiver on the Parkes radio telescope, made this prediction a reality before the SKA was built.

In their Introduction Carilli and Rawlings note “The time since the publication of the Taylor–Braun document has seen a revolution in our knowledge of the local and distant Universe. We have entered an era of ‘precision cosmology’, where the fundamental parameters (H₀, Ω_M, etc.) describing the emerging ‘standard model’ in cosmology are known to ~ ±10%. This standard model includes ‘dark energy’ and ‘dark matter’ as the two dominant energy densities in the present-day Universe. We have probed into the time of the first light in the universe, the ‘epoch of reionization’, when the UV emission from the first stars and (accreting) supermassive black holes reionizes the neutral intergalactic medium. γ-ray bursts have been shown to be the largest explosions in the universe, tracing the death of very massive stars to the earliest epochs. Supermassive black holes have gone from being a hypothetical by-product of general relativity (GR), to being a fundamental aspect of all spheroidal galaxies and how these objects formed.”⁴⁷

5.5.13.1 Connecting Quarks with the Cosmos

In 2003 there was a major review by the US National Academy of Science “Connecting Quarks with the Cosmos: 11 Science Questions for the New Century”⁴⁸. SKA had picked up four of these big questions in its science case so the SKA KSPs had meshed very well with these big questions in physics. Dark energy and dark matter had now made their appearance with increasing emphasis in the SKA science case. One example of what SKA could do well was the detection of baryon acoustic oscillations (BAO) , which are remnants of early density fluctuations in the Universe and serve as a tracer of early Universe expansion. For this, the large area neutral hydrogen surveys were now crucial and survey speed, achieved by high sensitivity and a large FoV, became an important driver of the telescope design. A sample selected by HI removes many of the biases introduced by selecting galaxies based on the integrated light from the stars formed in the galaxies; biases which are poorly understood because of the complexity of the star formation process. SKA will assemble a sufficiently large sample of galaxies to measure the BAO signal as a function of redshift and this can be used to determine the rate of evolution of the equation of state of dark energy.

5.5.13.2 Using Pulsars for Tests of General Relativity

Fig. 5.10

Mass-mass plot for the Double Pulsar system. Constraints on the system from precision timing of the two pulsars are shown. Credit: Kramer, M., et al. (2006). “Tests of General Relativity from Timing the Double Pulsar”. Science, 314(5796), 97–102

Pulsars have always been a significant component of the SKA science case because they provide unique opportunities to study neutron stars and detect gravitational waves. They also provide exquisite tests for gravitational theories.

The sheer number of pulsars that could be discovered by the SKA, in combination with the exceptional timing precision possible with SKA sensitivity, would be able to revolutionise the field of pulsar astrophysics. In 2004, just as the SKA Science Book was being written, this opportunity was greatly enhanced with the discovery of the double pulsar system, Lyne et al. (2004) and Burgay et al. (2005), a special binary neutron star system in which both neutron stars are pulsars. It was discovered using the Parkes radio telescope and its multibeam receiver. This amazing and so-far unique system gave astronomers a new probe of relativistic gravity theory. It was voted one of the top ten discoveries of 2004 by Science magazine. The system has one rapidly spinning pulsar (pulse period 22 ms) in an extremely tight binary orbit with a second slower 2.8 s pulsar. The orbital period of the binary system is just 2.4 h.

The most important application of the double pulsar system is the test of gravitational theory made possible by the detection of relativistic perturbations to the pulse arrival times. The dependence of all these effects on the masses of the stars can be predicted by Einstein’s general theory of relativity. All the curves are consistent with the predictions to better than 0.05% and this became the most precise strong-field test of general relativity at that time (Fig. 5.10) With the SKA sensitivity, large numbers of the millisecond period pulsars would be discovered: including more binaries, and possibly even including a rare pulsar orbiting a black hole which would greatly extend the strong field tests of gravity.

Carilli revisited this 2004 science case in 2019⁴⁹ and discusses how well the KSPs have stood the test of time in the light of our current view of the most exciting science.

5.5.14 OECD Global Science Forum Astronomy Workshops (2003–2004)

As described in Sect. 4.3.1, two astronomy workshops convened by the OECD Global Science Forum (GSF, the successor to the Mega-Science Forum, (see Sect. 3.2.5.2) were held, in Munich in December 2003 and Washington in April 2004.⁵⁰ The intention was to review scientific priorities and challenges in astronomy and astrophysics. The US funding agencies argued that their decadal survey process was far more effective, and that the USA had little to gain from any proposed global coordination activities. However this activity did raise the profile of the SKA as an exciting new large international project in the minds of other physicists and astronomers around the world.

5.5.15 The SKA Newsletters 2004–2012

This period is well documented in the series of SKA Newsletters which include summaries of the activities of the International Science Advisory Committee (ISAC) and Science Working Group (SWG) committees which had been set up to manage the science case (see Sects. 5.9.1 and 3.3.1.2). The activities in the period from 2004 to 2006 were dominated by the production of the Carilli and Rawlings SKA science book and by the more active promotion of the science case. Considerable effort went into giving the SKA science case more visibility in the broader astronomy and physics community. At this time the SKA was transitioning from the development of a new telescope by a group of radio astronomers to a major global scientific endeavour. This was also enhancing the visibility of all radio astronomy and was the beginning of a path to other developments, such as the SKA precursors discussed in Chaps. 4 and 6.

The other major development in this period was the finalisation of the key science goals and a key science project list (see Sect. 5.9.4). and the descriptions of the science behind these key science projects, which is summarised in the following section.

5.5.16 A Summary of the Key Science Projects (KSPs): 2006

Probing the Dark Ages

At the end of June 2005, over 120 people attended the meeting “Reionising the Universe” in Groningen, at which theorists and observers came together to discuss the many exciting developments now taking place in this new field. A subset of the contributions to the Groningen meeting have been compiled by Bryan Gaensler, who was SKA project scientist at that time.⁵¹

Strong Field Tests of Gravity

About 20 pulsar experts attended a workshop in Sydney in August 2005, including many young scientists who were becoming involved in the SKA project for the first time. In another meeting, “Gravitational Waves, Radio Pulsars and Astrometry” held in Birmingham on 30–31 March 2006, a new level of interaction developed between the radio and the gravitational wave community. At that time, both the Laser Interferometer Gravitational Wave Observatory (LIGO) and the Laser Interferometer Space Antenna (LISA) were being designed to detect gravitational waves directly while the SKA had the goal of probing gravity using techniques based on pulsar timing. Pulsar timing can probe both near field gravitational effects as well as the far field gravitational wave effects. Near field effects can be tested by observing a pulsar in a close orbit around another star. The precise evolution of the orbit can be used to place strong limits on the validity of Einstein’s general theory of relativity. The far field gravitational wave effects can be probed using observations of many pulsars seen in different directions. Correlation between pulsar timing residuals in different directions can be used to infer the presence of a gravitational wave background at the location of the earth.

Cosmic Magnetism

The magnetism team organised a conference in Bologna in August/September 2005.⁵² A wide variety of topics were discussed, covering magnetic fields from the inflation era of the Universe through to magnetic fields in nearby galaxies. Many unanswered questions in cosmology and in fundamental astrophysics are closely tied to the questions of the origin and evolution of magnetic fields. The SKA is critical to making further progress in this area.

The Cradle of Life

A special session on the importance of radio astronomy for astrobiology was included in the Astrobiology Science Conference 2006, held in Washington, DC. This was a large interdisciplinary event and attracted roughly 1000 participants. The motivation behind the radio astronomy symposium was to increase the exposure of the SKA and inform the astrobiology community about the relevant questions that can be pursued with the existing and planned tools of radio astronomy (particularly the SKA) in studying planet formation, organic biomolecules, pristine relics of our own solar system and techno-signatures (SETI) .

Galaxy Evolution, Cosmology and Dark Energy

A large meeting was organised (April 2006) in Oxford, covering cosmology, galaxy evolution and astroparticle physics. There were sessions on science with the SKA, as well as with the 1% pathfinders and the 10% SKA (SKA Phase 1).

Figure 5.11 illustrates these five KSPs (re-ordered) at the time of the February 2009 Cape Town meeting. The exploration of the unknown was added; although not formally listed as a specific KSP it was recognised as part of the design philosophy.

In 2007 during a US review by a national “Dark Energy Task Force”⁵³ it was concluded that the SKA, as well as the Large-Sized Telescope (LST) and the Joint Dark Energy Mission (JDEM) are the future major projects needed to advance our knowledge of dark energy. A corresponding UK review, carried out by the Particle Physics and Astronomy Research Council’s (PPARC) Science Committee, emphasised that the two techniques the SKA will use to study dark energy, namely weak lensing and acoustic oscillations, together hold the most promise for future studies of dark energy.

Fig. 5.11

The five KSPs from various historical presentations in 2009. The insert is a truncated version of the cover of the Elsevier book version of Carilli and Rawlings (2004). Background is an N-body simulation of the ΛCDM universe constrained to reproduce the observed large scale structure of the local universe at z = 0.84. Credit for background: Springel, White, Lemson, Kauffman, Dekel and the GIF Consortium

5.5.17 SKA Key Science Requirements Matrix 2006: Prime Science Drivers

The initial compilations of science cases made no prioritisation of the different goals and made no analysis of how they were driving the specifications. The concept of a matrix emerged in 2002^54,55 with science goals on one axis and how these goals were met by different designs on the other axis. In 2004 the specification document was updated⁵⁶ and basic design specifications were linked back to the science goals. The matrix was successful in generating a dialogue between scientists and engineers but problems with the lack of uniformity across the complex space spanned by the matrix were emerging.⁵⁷ This was exacerbated by the ongoing debate between advocates of the different SKA design concepts (see Chap. 6). However, the key science requirements matrix did provide a consolidated, although complex, description of the SKA key science goals and the SKA requirements. It was updated and significantly expanded by the SWG between 2005 and 2006, as a result of discussions at Key Science workshops and at science meetings in Pune and Paris.⁵⁸

5.5.18 SKA Reference Science Mission: 2009

By January 2009 the key science projects had evolved into the following (alphabetical) list. The structure of the science case has been changed with more emphasis on instrumental requirements rather than the broad science goals.

• Astrobiology: Search for organic molecules in molecular clouds and link them to proto-planetary discs; likely to require higher frequencies.

• Cosmic Magnetism: Understand the origin and evolution of cosmic magnetism; likely to require high polarisation purity

• Deep Continuum Field: Probe the first galaxies and protoclusters; likely to require high sensitivity, high imaging dynamic range, and long baselines.

• Deep H I Field: Track the evolution of galaxies over a significant cosmic epoch; likely to require high sensitivity and high spectral dynamic range.

• Galactic Centre survey: Probe the spacetime environment around Sgr A*, the closest super-massive black hole in the centre of the Milky Way Galaxy; likely to require high time resolution and higher frequencies to avoid scattering. This also requires a Southern Hemisphere location (see Sect. 5.8.1).

• Galactic Plane survey: Use neutron stars as probes of gravitational and nuclear physics. Test theories of gravity using ultra-relativistic binaries in the Milky Way Galaxy’s spiral disc; likely to require high time resolution.

• HI Absorption survey: Track the evolution of gas in galaxies to the earliest epochs; likely to require lower frequencies.

• Wide Area survey (a.k.a. “all-sky survey”): Various tests of theories of gravity, including studying gravitational waves using an array of millisecond pulsars and using baryon acoustic oscillations (BAO) in the galaxy distribution as a means of exploring dark energy; likely to require high survey speed and high time resolution.

Note the specification creep that is now becoming more significant with additional requirements including higher and lower frequencies, high time resolution, high survey speed, high polarisation purity and long baselines. The increase in cost, or reduction in sensitivity for a given cost, (see Sect. 4.4.3.3.1) is partly a consequence of the acceptance of these changing requirements. The scientific trade-off between the desire for increased capability even with a decrease in sensitivity may well have been justified but these very significant implications were not discussed by the science working group (see Sect. 5.9.2).

Fig. 5.12

Joseph (“Joe”) Lazio, Project Scientist (2008–2011), Director of Science [acting] (2012). Credit: J. Lazio

With Joe Lazio (Fig. 5.12) as SWG chair we see an increased emphasis on the “Exploration of the Unknown.” Lazio noted that for all components of the Reference Science Mission, it was anticipated that the observations would also be exploring the largely unknown dynamic radio sky, consistent with the SKA design philosophy which included “Exploration of the Unknown”.

5.5.19 The Science Case: 2010

Figure 5.13 still shows much the same information as in the 2006 KSPs but the presentation was now more refined and restructured with the evolution of structure in the universe as a unifying theme. The strong field tests of general relativity were further emphasised. The presentation styles and restructuring of the science management formulation had changed more dramatically than the actual science case. Visually impressive models showing how the universe evolved from its original big bang were now central to the presentation of the science case.

Fig. 5.13

The Science Case 2010. Credit: CSIRO Radio Astronomy Image Archive CRAIA-SKA010

Joe Lazio wrote a feature article for SKA Newsletter #18 in July 2010 “Science with the SKA” highlighting SKA development and future science.⁵⁹ Lazio provided an excellent review of the original motivation for the SKA and the key areas of physics and astronomy that have been chosen as priorities for the SKA. His review includes

• Probing the Dark Ages,

• Galaxy evolution, cosmology and dark energy,

• The origin and evolution of cosmic magnetism,

• Strong field tests of gravity, using pulsars and black holes

• Cradle of life,

• Exploration of the unknown

5.5.19.1 Identify Unique Radio Niches

Fig. 5.14

Bryan Gaensler. Project Scientist and Chair of the SWG 2006–2008. Credit: B. Gaensler

By 2010 it had been recognised that to compete with the big telescope projects at other wavelengths, it was important to identify niche areas in astronomy which could only be tackled by radio astronomy.⁶⁰ One such area was the study of cosmic magnetism. Cosmic magnetism had always been included as part of the science case and was vigorously promoted by Bryan Gaensler, then at the University of Sydney (Fig. 5.14). When Gaensler became chair of the Science Working Group (2006 to 2008) he led a move to elevate the study of cosmic magnetism as a special unique component of the science case. Radio astronomy has a colossal advantage over other wavebands when it comes to studying astrophysical magnetism. Radio astronomers could measure the Zeeman effect, they could measure the magnetic field directions through the polarised synchrotron emission (Fig. 5.15), and they could study the Faraday rotation of the linear polarisation as radio waves propagated through the intervening medium. Instead of being a part of other science, cosmic magnetism now became a main science driver and was indeed an excellent example of a unique niche for radio astronomy.

Fig. 5.15

Magnetic field in the grand design spiral galaxy M51. Credit: Gaensler, B., Beck, R., & Feretti, L. (2004) “The origin and evolution of cosmic magnetism” in C. Carilli, & S. Rawlings (Eds.), Science with the Square Kilometre Array, 48, 1003. Elsevier

5.5.19.2 More Focussed Science Case

The external system engineering concept design review in 2010 (see Sect. 5.9.10) included a recommendation for greater focus which resulted in an extreme reduction in the number of science goals. The factors which led up to this review have been discussed in Chap. 4, Sect. 4.5.2. This recommendation meant that many in the astronomy community felt disenfranchised. Radio continuum imaging and VLBI were no longer listed as they covered such a large range of astronomical objectives that they appear to lack focus. Although this change in focus had a positive effect on the funding agencies it resulted in an increasing disparity between the well-focussed goals needed by funding agencies and design engineers and the need for a flexible facility that could support a wide range of different observing programs and adapt to new discoveries.

5.5.20 The Epoch of Reionisation (EoR) Science Case

The style of the SKA Newsletters from # 20 (2011) changed to provide reviews of single topics by outside experts. The January 2011 Newsletter included a review of experiments exploring the Epoch of Reionization and the Dark Ages by C. Carilli, L. Greenhill, and L. Koopmans.⁶¹ This topic, now referred to as the Universe’s Dark Ages, became one of the key scientific frontiers for the SKA. It is the time soon after the Big Bang before there were any stars. The Universe was in a hot and dense state, so hot that it was completely ionised. As the Universe expanded, it also cooled, until about 400,000 years after the Big Bang when it was sufficiently cold that neutral hydrogen atoms could form. This neutral hydrogen gas filling the Universe would be the raw material from which the first stars could form. The 21 cm neutral hydrogen signal from this very distant gas is redshifted to lower frequencies. Simulations of the spatial structure were computed to illustrate how the reionisation proceeds.

The EoR is certainly a unique niche for the SKA. The Hubble Space Telescope (HST), the Atacama Large Millimeter Array (ALMA) and the James Webb Space Telescope (JWST) can study the first galaxies (assemblies of stars) but the only means by which to quantify the rapidly evolving physical conditions during the first billion years of the Universe’s history is to study the neutral hydrogen from which the first stars and galaxies formed.

5.6 International SKA Forums 2007–2012

The Funding Agencies Working Group (FAWG) established an International SKA Forum to facilitate engagement between scientists and representatives from government departments and funding agencies (see Sect. 4.2.1). The first International SKA Forum meeting took place in Manchester in October 2007. These meetings, which were held annually and rotated around the member countries, included science and engineering presentations in association with the national status reports, presentations on specific scientific topics by invited speakers, and meetings of the funding agencies.

5.6.1 The Science Case in 2011 and the Data Processing Requirements

The final International SKA Forum meeting was held in Banff, Canada, in 2011, with the case summarised neatly by Lazio:

20th Century: We discovered our place in the Universe.

21st Century: We understand the Universe we inhabit.

Box 5.5 based on Lazio’s presentation⁶² summarises how a radio wavelength observatory, the SKA, will contribute to this goal. The basic science cases for cosmology, fundamental physics and galaxy evolution remain similar, but reorganised. Topics like the Epoch of Reionisation are reworded as “how did the Universe emerge from the Dark Ages”. The life cycle of the interstellar medium and stars, the evolution of planetary systems and evidence for life on exoplanets have been added. Detection of ultra-high energy cosmic rays gets a mention as does a wider range of transient phenomena. In this presentation Lazio added details on the scientific requirements and translated these into technical requirements. The difficulty of measuring the epoch of reionisation was recognised. Lazio included a summary of the imaging requirements and the massive data processing implications for an array of this size. See Cornwell’s memo on the enormous software challenges facing the SKA.⁶³ From this time it was understood that the SKA science would be computationally limited. This was even predicted to become an issue for the SKA pathfinders and the scalability of processing solutions was a serious concern.

Box 5.5 Science Case 2011

• Cosmology

  • Era of precision cosmology, dark matter and dark energy

  • Large scale surveys in continuum and line, 1 billion galaxies

  • Detection of weak lensing

• Gravity

  • Can strong gravity be observed in action?

  • What is dark matter and dark energy? (dark energy and BAOs with H I galaxies)

• Magnetism

• Strong force

  • Nuclear equation of state

• Evolution of Galaxies and the Universe

  • How did the Universe emerge from its Dark Ages? [Epoch of reionisation renamed]

  • How did the structure of the cosmic web evolve?

  • Where are most of the metals throughout cosmic time?

  • How were galaxies assembled?

• Stars, Planets, and Life

  • How do planetary systems form and evolve?

  • What is the life-cycle of the interstellar medium and stars? (biomolecules)

  • Is there evidence for life on exoplanets? (SETI)

5.6.2 Long Baseline Science with the SKA (VLBI)

At the Banff meeting in 2011 Lisa Harvey-Smith (CSIRO, Australia) summarised the many areas of SKA research that required long baselines (>1000 km).⁶⁴ These touched on almost all aspects of the science case, including strong-field tests of gravity, gravitational-wave experiments, evolution of galaxies, galactic magnetic fields, protoplanetary discs, first-generation active galactic nuclei (AGN) and radio transients. Examples explored included high precision astrometry to measure pulsar parallax and hence distances and proper motions to constrain tests of general relativity.

5.6.3 HI Stacking

Andrew Hopkins⁶⁵ (University of Sydney) outlined the current developments in the stacking of HI emission from galaxies. If you have a sample of galaxies with known redshift the radio spectra can be aligned at the anticipated HI line frequency and averaged (stacked) to improve the sensitivity for sample average. The value of this technique is dependent on the future optical redshift surveys and Hopkin’s paper discussed opportunities in the 0.5 < z < 2 range. Demonstrations of HI stacking feasibility in the 0.2 < z < 0.7 range will be possible with the SKA pathfinders.

Introductions to the process and existing results can be found in papers by Philip Lah and summarised in his ANU PhD thesis (Lah, 2009). Note that any stacking type analysis provides essentially the same cosmological information as the cross-power spectrum.⁶⁶

5.6.4 SKA Surveys and Cosmology

SKA Newsletter #23 (November 2011) includes a review: “Paths to cosmology with the SKA” by David Bacon and Chris Blake.⁶⁷ This newsletter articles provides a good summary of the impact of deep continuum and HI surveys including subtle effects due to clustering, CMB distortion and lensing. It also discusses the Baryon Acoustic Oscillations (BAO) method which provides a cosmic yardstick and concludes that the SKA is well placed to make a major contribution.

5.7 Impact of Science Requirements on Design

Throughout the life of the project, it was well understood that the technical consequences of the developing science case had to be considered. Early in the project, this was a natural consequence of having scientists with broad knowledge of both the astronomy and the engineering designing the telescope. As time went on and both the scientists and engineers involved in the SKA were attracted from larger but more specialised communities, the integration of the astronomy and the engineering had to be more actively managed. The practical constraints required prioritisation and complex trade-offs between different options. Separate, more specialised, advisory committees were being appointed. Managing the interactions between the astronomy and the engineering groups was a complex process and a matrix of science requirements versus technical options (the compliance matrix), had been developed between 2002 and 2006 (Sect. 5.5.17). A large number of science drivers had emerged with a mixture of different specifications on the telescope design, and all these were all being considered to generate an integrated SKA specification. The matrix included the required values for different design parameters, including frequency range, sensitivity, spatial resolution, surface brightness sensitivity, field of view, multi-beaming, dynamic range, number of spectral line channels, frequency agility, total power, polarisation, and time resolution. The full details of the matrix are included in SKA Memo 83.⁶⁸ Here we summarise some of the science requirements that had a major impact on the design in ways that are discussed in Chap. 6. Comments on these design implications had already been summarised in separate memos from the Engineering Management Team (EMT)⁶⁹ and the Science Advisory Committee (ISAC).⁷⁰

5.7.1 Sensitivity

Sensitivity a factor of 10–100 times greater than any existing telescopes was always the goal and this was driven by essentially all the science cases being discussed. The issue here was the most cost-effective way to realise the required sensitivity, and this is discussed in Chap. 6.

5.7.2 SKA Survey Speed

A large field-of-view (FoV) which is achievable at these long radio wavelengths, had already been recognised as a very desirable feature. The development of the science case for cosmology and dark energy experiments (Sects. 5.5.11 and 5.6.4) made surveys and hence survey speed a critical design criterion for the SKA, with Bunton (CSIRO, Australia) and Cordes (Cornell University, USA) independently developing a figure of merit to explore trade-offs between sensitivity (Area/Tsys) and FoV.^71,72 An instrument with half the sensitivity will need four times as much integration time per field to achieve the same sensitivity, reducing its survey speed by a factor of four. Survey speed also depends directly on the imaging field of view (FoV) and on the correlator bandwidth (BW). Thus a simple measure of survey speed is the product of the FoV and the BW, weighted by the sensitivity squared.

$$ \mathrm{Survey}\ \mathrm{speed}=\mathrm{FoV}\times \mathrm{BW}\times {\left(\mathrm{Area}/\mathrm{Tsys}\right)}^2 $$

The implications of this requirement are discussed in Chap. 6. However, Cordes added an important caveat noting that this treatment does not apply to intermittent transient sources for which FoV becomes more strongly weighted.⁷³ But in 2007 the Fast Radio Burst population had not yet been discovered so the implications of the need to modify the survey speed metric did not influence design choices and the final design was not optimised for this class of object.

5.7.3 Frequency Range

The required frequency range was a critical SKA design issue, and the impact is discussed in Sects. 6.2.2.8 (exploration of the unknown) and 6.5.1 (low frequency range). The initial emphasis on high sensitivity observations of the 21 cm neutral hydrogen line, both in the local universe and at high redshift required frequencies from a few hundred MHz to 1.4 GHz and this was a wavelength range where it was thought that a large collecting area would be possible within a modest cost envelope. The other early science cases including pulsars, radio galaxies, non-thermal stellar radio emission and SETI, as well as propagation effects such as Faraday rotation, could be included with a modest extension of the upper frequency range to a few GHz.

Two factors were to dramatically change this view and have a huge impact on the development of the SKA. By the late-1990s the recognition that radio astronomy could explore the dark ages and the epoch of reionisation pushed the lower frequency down from a few hundred MHz to 30–50 MHz and this change required a split between SKA-low and SKA-mid with different antenna designs. The second factor resulted from the interest in this next-generation radio telescope in the broader astronomy community. This broadened the science case, and the initial upper frequency of a few GHz (λ = 10 cm) was soon extended to 10 GHz (λ = 3 cm). But a cut-off at 10 GHz still precludes most thermal science such as mm molecular line emission redshifted down to cm wavelengths and terrestrial planet formation so the upper frequency was further extended to 20 GHz (λ = 1.5 cm).⁷⁴

5.7.4 Image Properties

The quality of the images that the SKA will produce, and ultimately its scientific performance, will be determined in no small part by the distribution of antennas that comprise it, also known as the “array configuration‟. The array configuration requirements affect both hardware design (Chap. 6) and geographical constraints (Sect. 5.8.5).

In 2003 Andrei Lobanov⁷⁵ argued that, in addition to a significant improvement in sensitivity, high angular resolution images with high spatial dynamic range would be needed to take full advantage of the image quality being achieved in observations made at other wavelengths. Lobanov made an analysis of the Spatial Dynamic Range (SDR) for the different designs being considered at that time. As expected, the strongest requirement would be the filling factor achieved from the distribution of antennas across the array. This requirement strongly favours the large N—small D design approach (see Sect. 5.3.6) and a central concentration of antennas with optimised antenna locations. This analysis was ongoing, and a further report including simulations was provided by Lal, Lobanov and Jimenez-Monferrer in December 2008.⁷⁶

5.7.5 Simulations

A team at the University of Oxford (Wilman et al., 2008), developed a semi-empirical simulation of the extragalactic radio continuum sky to aid the design of the SKA. Their emphasis was on modelling the large-scale cosmological distribution of radio sources rather than the internal structure of individual sources. These simulations were developed under the European Comission funded SKA Design Study (SKADS) project and referred to collectively as SKADS Simulated Skies (S³). Access to these simulations was made available to the community through a set of Python-based software routines and interfaces.

The primary telescope specifications such as sensitivity, angular resolution and FoV could be specified, and the simulations could be used to quantify the achievability of various scientific goals. The simulations also provided datasets that could be used to develop data processing systems. There were limitations, such as the need to extrapolate properties of known populations, and of course any new classes of radio source could not be included. It was too difficult to realistically include the more complex instrumental behaviour so the simulations did not have impact on any design details.

5.8 Impact of Science Requirements on Siting

5.8.1 Access to the Southern Sky

The Southern Sky has special significance for astronomers because the centre of our galaxy passes overhead and while it is still visible from sites in the northern hemisphere it would be near the horizon so image quality would be degraded. The two nearest galaxies, the Large and Small Magellanic Clouds are only visible from the southern hemisphere and this provides an exceptional opportunity for a more sensitive radio telescope to detect objects in other galaxies that were previously only visible in our galaxy. At long radio wavelengths there is an additional instrumental problem in the northern hemisphere caused by the two strongest sources in the sky, Cygnus A and Cassiopeia A which are both so strong that they will be seen through the far sidelobes of the telescope degrading the dynamic range. The largest optical telescopes which are often needed to complement SKA observations are also located in the southern hemisphere. For these reasons the strong advantage of a site in southern hemisphere has always been accepted.

5.8.2 Radio Frequency Interference (RFI)

The need to avoid RFI⁷⁷ may be the highest impact scientific requirement for the SKA site selection. Beyond our solar system the universe can only be explored by analysing the radiation we receive from distant stars and galaxies. These signals which may have travelled across the universe are incredibly weak. For example, the signal from the brightest natural radio source in the sky is more than 1000 times fainter than the signal from a radio navigation satellite and the faintest signals detectable with the SKA can be a hundred million times fainter than signals from typical Low Earth Orbit (LEO) satellites! The expansion of the universe causes a Doppler shift of the frequency of signals such as the 21 cm hydrogen line at great distances. This requires changes in observing frequency and means that the narrow bands allocated by the International Telecommunications Union (ITU) for radio astronomy are completely inadequate.

As outlined by the spectrum management task force⁷⁸ the SKA can deal with RFI by the combination of:

• 1. seeking a remote location with low population density;

• 2. establishing protection and coordination of radio quiet zones around the SKA (RQZ) and,

• 3. building RFI mitigation technology into the SKA system.

The need to build the observatory in remote low-RFI locations has had a huge impact on site selection⁷⁹ and on the additional operating costs incurred at such remote locations. This will be discussed in detail in Chaps. 7 and 8. The need to establish a Radio Quiet Zone (RQZ) at the selected sites is discussed in Sect. 6.2.2.11.

A further option is to develop RFI mitigation technology and for this there is great potential for advances including adaptive nulling techniques.⁸⁰ This will also be discussed in more detail in Sect. 6.2.2.11.

The problem of RFI in radio astronomy was considered of such global importance that in January 1997 the Organization for Economic Co-operation and Development (OECD) Mega-Science forum established a working group on radio astronomy, to report on the impact of radio frequency interference on radio astronomy. As already discussed in Chap. 3, Sect. 3.2.5.2 the OECD working group’s report was prepared specifically for use by science policy makers.⁸¹ The report noted that: “The Universe beyond our Solar System can only be studied using “remote sensing” techniques, whereby electromagnetic signals emitted by distant objects, such as stars and galaxies, are captured by telescopes and subsequently analysed. Astronomical signals from even the closest stars and galaxies are very faint by the time they reach the Earth, overwhelmingly so in comparison with any man-made signal. Since the signals received at their telescopes are the only source of information for astronomers, terrestrial contamination of these signals is a very serious concern.” The OECD working group made several endorsements and recommendations to minimise the impact of RFI on future radio astronomy facilities.

5.8.3 Ionospheric Conditions

The Ionosphere causes frequency-dependent phase distortions of the incoming cosmic radiation and this can be a large effect at the low frequencies envisaged for the Epoch of Reionisation observations. If these phase gradients are linear across the array, they can be easily corrected using calibrators within the field of view, but at higher resolution (large baseline lengths) the region of the ionosphere seen by the array will vary across the field of view and the required corrections become position dependant and very much more difficult.⁸² In even more extreme situations when the phase changes are sufficiently large that the waves interfere before reaching the telescope the resulting amplitude modulation cannot be corrected. This is known as ionospheric scintillation and is strongest near the geomagnetic equator of the Earth. One region known as the “southern equatorial anomaly” is known to cause particularly severe ionospheric scintillation. Avoiding the worst ionospheric effects has serious consequences for site selection as is discussed in Sect. 7.3.8.1.

5.8.4 Tropospheric Requirements

It is well-known that ground-based astronomical observations are affected by the wavefront distortion caused by a turbulent troposphere.⁸³ Such distortion translates into a deformation of the observed source structure and higher sidelobe imperfections in the image. In most cases the effects of the troposphere can be calibrated using antenna-based self calibration algorithms,⁸⁴ which are relatively simple and computationally inexpensive at radio wavelengths. An ad hoc troposphere advisory group was established and their report⁸⁵ concluded that while tropospheric phase stability was an issue, especially at the higher frequencies; this will be mitigated by the power of the SKA to self-calibrate and the sensitivity of the telescope, and its large number of elements will make self-calibration effective even at 22 GHz. Furthermore, a spot measurement, or even an average over a year, is not a reliable long-term predictor of total water vapour content so detailed local studies will have limited value for site selection.

5.8.5 Geographic Requirements for the Antenna Configuration

The different science cases make very different demands on the array configuration which involve a trade-off between brightness sensitivity (HI and EoR detections, and transient surveys) and angular resolution (most continuum observations). Centrally condensed configurations optimise brightness sensitivity for HI and EoR surveys and minimise the number of coherent beams needed for transient searches. However, high angular resolution observations which provide finer detail in images of high brightness sources and avoid source confusion require a more uniform distribution of the array spacings out to the maximum baseline. This has resulted in a compromise with a significant fraction of the array centrally concentrated. To satisfy these requirements a central relatively flat area several kilometres in size is required as well as suitable locations for elements which provide a range of baselines up to many thousands of kilometres from the core.

Leith Godfrey, Hayley Bignall and Steven Tingay (Curtin University, Australia) made a detailed analysis of the high angular resolution science case discussed in Sect. 5.6.2⁸⁶ and noted that “The science goals and corresponding technical requirements for the high angular resolution component of the SKA are significantly different to those of the SKA core. Consequently, the requirements for remote stations must be considered separately.” They went on to define the technical requirements for the remote stations. In order to maintain an angular resolution of 0.1” at all SKA wavelengths, baselines of at least 1000 km are needed. However as discussed in Sect. 5.10.11 the long SKA baselines were not included after descoping in 2014.

5.9 Managing the Science Case

5.9.1 Science Working Group (SWG), 1994

As discussed in Chap. 3, Sect. 3.3.1.2, three advisory committees reporting to the ISSC were established by the ISSC at its meeing in Manchester in August 2000. These included a Science Working Group (SWG) , to coordinate the development of the SKA science case. The SWG initially chaired by Russ Taylor (University of Calgary, Canada), held its first meeting in Berkeley, California in July 2001. An initial membership of 24 with representation from all signatories to the MoU was established (see hba.skao.int/SKASUP4-5) with the task of defining the main SKA science drivers and developing a revised set of science requirements.

5.9.2 Science Advisory Committee, 2002

In 2002, the SWG was renamed the Science Advisory Committee (SAC) and a tentative set of sub-groups was formed ahead of a Science workshop in Bologna. At the workshop, the subgroups were charged with identifying Level I (high priority, unique to SKA) and Level II (high priority, complementary to other instruments) science requirements within their area and to determine the technical requirements such science would place on SKA, compared to the Strawman Technical Specifications⁸⁷ (see discussion in Sect. 6.2.1.3). In the two subsequent meetings in Groningen, August 2002, and in Leiden, November 2003, the SAC carried out a complete revision of the science case and this formed the input for the (Carilli & Rawlings, 2004) book “Science with the SKA” (see Sect. 5.5.13).

The autumn of 2004 saw the formation of a new SKA Science Working Group (SWG) reverting to its original name but now reporting to the newly appointed International SKA Project (ISPO) Director (see Sect. 3.3.1.3). The SWG was now part of the new management structure which also included the Engineering Working Group (EWG) and the Site Evaluation Working Group (SEWG). The remit of the SWG was wide ranging with activities planned for 2005 including the further development of the science case as discussed in the previous section, organising worldwide advocacy for the SKA, and providing information on the trade-offs between the science achievable with the SKA and site and design choice. Following publication of the SKA science book in 2004, the SWG started getting down to the detailed work on the trade-offs. The SWG started adding details and continued the development of the key science requirements ‘matrix’ introduced in Sect. 5.5.17. This matrix continued to play a key role identifying the trade-offs needed between the scientific desires of the KSPs and the harsh realities of real SKA designs.⁸⁸

Year	Science Advisory Committee/Science Working Group chairs
2000–2002	Russ Taylor (University of Calgary) served as the first chair of the newly created SWG
2002–2004	Chris Carilli (NRAO) appointed as SWG (renamed ISAC for this period) chair
2004–2006	Steve Rawlings (Oxford University) appointed SWG chair
2006–2008	Bryan Gaensler (University of Sydney) appointed chair with Joseph Lazio as vice-chair
2008–2011	Joseph Lazio (NASA Jet Propulsion Laboratory) appointed SKA Project Scientist and SWG Chair

5.9.3 Managing the Compliance Matrix

Carilli⁸⁹ chairing the International Science Advisory Committee (ISAC) ⁹⁰ was aware that the scientific working groups were in the process of refining the compliance Matrix, and he reconnected the engineering and astronomy communities by inviting the proposers of the different SKA telescope designs (see Chap. 6) to participate in the process. The ISAC identified four issues that appeared paramount at that time: high and low frequency limits, multi-beaming, response times, configuration, and field of view. Many of these Matrix entries were vigorously debated in the SKA community due to a lack of consistency in grading of the scientific requirements and, in some respects, a lack of understanding of the technology. The Matrix was revisited in the Geraldton SKA meeting in Aug 2003 to address these issues. The changing science priorities were also compromising the Matrix. For example, the elevation of the EoR case and the large volume redshift surveys to high priority made significant changes. A new “Final version of the compliance Matrix”⁹¹ was discussed extensively at the Transformational Science meetings in Pune, India in October 2005. The Matrix was used to evaluate the different design concepts and eventually the down select process in late 2005 leading to the Reference Design (see Sect. 3.4.1 and Chap. 6).

As already noted in Chap. 1, Rawlings⁹² commented on the importance of the interaction between science and engineering which was required in the development of the Matrix. These interactions between scientists and engineers were a critical element for the development of the SKA. This had always been part of the radio astronomy culture which was born from the engineering innovations of the early pioneers, however, as the SKA project grew maintaining this interaction was increasingly difficult.

5.9.4 Key Science Projects (KSPs)

In May 2003 the ISAC formed a subcommittee to identify a handful of “level 0” science goals, which could be used to attract funding and publicity, focus efforts to ensure that SKA can provides complimentary research to that proposed for other telescopes, and which can be used to optimise the SKA design.⁹³ Carilli, Chair ISAC, summarised the process of developing the KSPs as follows.⁹⁴ “A subcommittee chaired by Bryan Gaensler determined the highest priority science for the SKA. After an extensive review process by the subcommittee and the full ISAC, a final list of five topics was selected. Establishing the key science goals was a difficult process, with significant design (and possibly political) ramifications, and the Gaensler sub-committee carried through the process thoroughly, and most importantly, in a transparent and clearly unbiased manner.” The final report⁹⁵ includes a flow-down of the telescope requirements set by the key science goals. The key projects chosen were:

• Strong field tests of gravity using pulsars

• Probing the dark ages (cosmic reionisation and the first luminous objects)

• The origin and evolution of cosmic magnetism

• The cradle of life (terrestrial planet formation and astrobiology)

• The evolution of galaxies and large-scale structure

A Position Paper⁹⁶ was prepared as input to the June 2005 Heathrow meeting of the funding agencies (Sect. 3.4.1) in which the concept of Key Science Projects was further emphasised, This paper set out a small subset of prioritised science requirements that could be used for engineering design and to provide the funding agencies and the wider community with a brief statement of key science goals. As discussed in Chap. 4, Sect. 4.3.2, one of the primary roles for the funding agencies was to monitor progress in the international SKA project and balance that against their own national efforts and funding opportunities. To do this the funding agencies needed a simple focused set of science goals. This discussion was strongly influenced by the US decadal review (McKee & Taylor, 2001) and the US National Research Council’s big questions in physics (US National Research Council, 2003). For the first time these discussions now included concepts like: ensure that level-0 projects align with the broad themes and priority areas laid out by various funding agencies.

5.9.5 The Science Case for a 10% SKA: 2006

The need to develop the cases for a 10% SKA was recognised in late 2005 see Chap. 4, Sect. 4.4.1. It would need to be compelling in its own right but must also act as a strong argument for building the full SKA. In 2006 the SWG developed a science case for a SKA Phase I, with 10% of the full SKA collecting area and maximum baselines of approximately 50 km, but keeping the other specifications as laid out in the Reference Design. The SWG converged on three key topics for the SKA Phase I:

• First Light: The Epoch of Reionisation

• Building Galaxies: Hydrogen and Magnetism`

• Pulsars and the Transient Sky

These represented a sampling of the five KSPs for the full SKA, but also include some discovery science for which the SKA Phase I could carve out its own exciting niche. The SWG also identified other secondary experiments for the SKA Phase I, including weak lensing, the inter-cluster medium, spacecraft tracking, X-ray binaries, the cosmic web and interstellar scintillation.

SKA Phase 1 was to demonstrate many of the scientific and technical underpinnings for the mid- and low frequency SKA that would enable “killer science” with the full SKA. Several aspects of the SKA Phase 1 science case that had previously been developed by the SWG (circa March 2007) had been overtaken by both scientific and technical developments. Headline science themes for SKA Phase 1 were defined as neutral hydrogen from the epoch of reionisation to the present, testing general relativity theory and detecting gravitational waves using pulsar timing, and discovering transients and other new phenomena.

5.9.6 The Magnificent Memo Series

One of the challenges for SKA design and planning was to make the appropriate trade-offs between scientific requirements and engineering reality. The ISSC posed eight questions related to the impact of these trade-offs on the five KSPs and presented these questions at the Pune Transformational Science meeting in October 2005. The answers to these questions were the “Magnificent Memos Series” which were incorporated into SKA Memo 82 submitted by Bryan Gaensler, SKA Project Scientist, and Joseph Lazio, Deputy Project Scientist, on behalf of the SWG.⁹⁷ The questions addressed were:

• What key science can be delivered by SKA Phase 1 (10% collecting area)?

• What is the science case for multiple independently steerable fields of view?

• What is the impact of limiting the field of view at high angular resolution?

• What is the case for high angular resolution below a few GHz?

• What is the case for frequencies between 200 and 500 MHz?

• What is the case for high filling factor at high frequency?

• What are the options for transient detection?

• What is the impact on key science of having just a high- or just a low-frequency array?

The first and last question were deferred for separate discussions (see Sects. 5.9.5 and 5.9.9). The Magnificent Memos identified the specific science that would be affected in each case and identified many caveats, some of which were followed up in subsequent memos. This document certainly clarified the relationship between the science and the telescope design and was the basis for more realistic discussions, but it did not attempt to make any specific design recommendation, an issue that was taken up by the “Tiger Team” discussed in the next section.

5.9.7 Preliminary Specifications for the Square Kilometre Array

A Tiger Team was established by the ISSC to revise the SKA specifications and propose a realisable baseline implementation.⁹⁸ This included the identification of cost-driving specifications and the use of cost-performance estimation tools^99,100 to guide a detailed trade-off analysis as discussed in detail in Sects. 6.2.2.7 and 6.4.6.

5.9.8 The SKA Design Reference Mission (DRM): 2009

In January 2008 the SKA Specifications Review Committee¹⁰¹ (see Sect. 6.2.1.4) recommended that the Project Scientist and the SWG develop a Reference Science Mission. The term “Reference Science Mission” was chosen to align the SKA with other major projects (PanSTARRs, LSST,¹⁰² James Webb Space Telescope, etc.), which typically have “Design Reference Missions. ” The SKA Reference Science Mission was intended to lay out clearly the observational science requirements needed to achieve the SKA Science Case. The components of the science case included in the Reference Science Mission have been discussed in Sect. 5.5.18. In July 2009 the Reference Science Mission was renamed the Design Reference Mission (DRM). In 2010 the DRM underwent a technical review by the SPDO domain specialists. It was now expected to become a “living document”, responding both to scientific and technical developments. It became one of the key documents provided during the System Concept Design Review (CoDR) and together the SKA Science Case and DRM provided the overarching set of scientific requirements that the telescope was required to meet.

5.9.9 The Science Implications of an SKA “Win-Win” Siting Scenario: 2009

In October 2009 the SWG was asked by the SSEC to consider the science implications of a so-called “win-win” scenario in which SKA infrastructure is located on both candidate sites (Australia and South Africa). The SWG found that there would be little scientific advantage to a “win-win” scenario.¹⁰³ They considered three specific cases:

• A frequency split with the full SKA-mid on one site and the full SKA-lo on the other site.¹⁰⁴ The most relevant factor would be the radio frequency interference (RFI) situation and there would be little other scientific advantage to a frequency split. Some disadvantages that were noted related to the reduced capability for calibration of the ionosphere above SKA-lo and the need for transient observations at multiple frequencies at or near the same time.

• A separate large remote SKA-mid station constructed of dishes. The separation between the two candidate sites produced a significant gap in spatial frequencies (u-v plane). If the array was divided equally between the two sites, the u-v plane coverage would be unacceptable, so the SWG only considered a smaller remote station with 10% of the total collecting area. This would still be of limited utility, except for astrometric programs.

• A separate large remote SKA-mid station constructed of dense aperture arrays. Many of the same considerations applied as for a large remote SKA-mid station constructed of dishes.

5.9.10 SSEC Forms an Internal SKA Phase 1 Definition Sub-Committee: 2010

The report from an external panel carrying out the System CoDR in February 2010 (see Chap. 4, Sect. 4.4.2) included criticism that the science case was too broad. Following this report the SSEC formed an internal subcommittee with a mandate to produce an SKA Phase 1 (SKA1) Concept Design with high-level targeted science goals. The subcommittee was chaired by Mike Garrett¹⁰⁵ (ASTRON, Netherlands) and quickly came up with the following short list of key scientific drivers:

• History of neutral hydrogen: Epoch of Re-ionisation (EoR) to the current epoch;

• Pulsars for Gravity (General Relativity and the detection of gravitational waves), and

• Transient Universe (new phenomena).

5.9.11 Transition to the Pre-Construction Phase: 2011

When the Founding Board was established in April 2011 (see Sect. 4.6) it moved to complete the Joint Implementation Plan which included the plan for the pre-construction phase of the SKA. The scope of the associated Business Plan, discussed in Chap. 4, Sect. 4.7.3, included building relationships with relevant national and international astronomy organisations to leverage skills and ensure SKA Phase 1 science and opportunities were fully integrated into a global astronomy perspective. From this time, the SKA management team had a relatively light involvement in the Science case as the SKA moved into the pre-construction phase and then finally the construction phase. The SAC and its sub-committees continue to exist and to monitor changes in science opportunities and keep the astronomy community informed of progress. However as with any large-scale project it becomes increasingly difficult to respond to any changes in science requirements during the final planning and nearly impossible once the construction phase has begun.

5.10 Evolution of the Science Case and Comparison with the Current Vision

We can compare the evolving science case with the science case at the start of construction in June 2021. This is based on an old SKA web page from 2021 see (Box 5.6).

Box 5.6 Science Case 2021 (Taken from a 2021 SKA www Page)

• Galaxy evolution, cosmology, and dark energy

  • How do galaxies evolve

  • What is dark energy

• Challenging Einstein

  • Strong-field tests of gravity using pulsars and black holes

  • Gravitational wave detection by pulsar timing

• Understanding cosmic magnetism

  • Polarised synchrotron emission

  • Rotation measure

• Probing the Cosmic Dawn

  • Epoch of reionisation (EoR)

  • quasars

• The cradle of life—searching for life and planets

  • Technosignatures

  • Amino acids

  • Thermal emission from dust

• Continuum surveys

• Radio transients

  • Gamma Ray Bursts

  • Supernovae

• Solar & Heliospheric Physics

  • Solar flares

  • Coronal mass ejections (space weather)

The list has also now been expanded due to the transition from a project seeking funding, which requires a focussed science case, to a construction project at which point all possible science options may need to be considered. Many of these science drivers have been present for three decades as illustrated in Fig. 5.4 and although the science case has involved similar topics over the last 30 years, there have been big changes in detail. Detecting hydrogen at high redshift has not changed from the beginning of the SKA concept, but the way the hydrogen line observations are used has changed a lot. The earlier ideas of simply looking for clouds of collapsing gas was replaced in 2003 by a new concept to look for the change in the state of the HI (EoR) as the first stars form and this has all been named the more generic “Cosmic Dawn”. The use of the HI to trace large scale structure (e.g. BAO) emerged in 2004 and strengthens the case for survey science but since this was also part of the science case for other proposed new instruments, it is no longer emphasised as unique SKA science. Some areas that were given lower priority in the funding stage now re-emerge. These include the HI and continuum surveys and the solar observations.

As new discoveries are made at radio or other wavelengths new sciences opportunities have opened up and the emphasis has changed. Examples are: gravitational wave detection, exoplanets, and new classes of transient radio sources. The SKA precursors and SKA pathfinders are also influencing the science case as they make new discoveries (see Sect. 5.11.1).

Following the Bologna meeting of the SAC in January 2002 (Sect. 5.5.11), Chris Carilli recognised the need for changing emphasis and added the following general note¹⁰⁶ “The current document (2002) presents the SKA capabilities, and then shows all the gee-whiz stuff that can be done. In order to better impress the general astronomical community, we should start by considering the important questions facing astrophysics today, and then show how the SKA will help to solve these questions.” This change in the perceived requirements for the science case had a profound impact on the development of the science case from this time onwards, with much greater emphasis on what were considered to be the most important questions. We pick up on this issue again in Chap. 11.

We now depart from our chronological discussion and look at the way the different science cases listed in Fig. 5.4 have changed over this thirty year period of time.

5.10.1 HI at High Redshift: Evolution of Structure, Cosmology, and Dark Energy

The sensitivity needed to detect HI at high redshift was the initial driver for the SKA concept. It was also emphasised that while telescopes working at other wavelengths can detect the galaxies made of stars at high redshift, only HI observations can detect the gas from which they form. Existing all-sky neutral hydrogen line surveys only have sufficient sensitivity to reach a redshift of about z = 0.04, so estimates of the distribution of gas at higher redshift are entirely dependent on models. If the SKA could detect a large number of HI galaxies out to a redshift of z = 1.5, these models of the evolution of the universe could be tested and various cosmological tests were considered. Measurement of the HI mass function at high redshift could be used to determine the evolution of dark matter in the Universe. Such a survey could also measure the “wiggles in the power spectrum” now known as the “Baryon Acoustic Oscillations” (BAO) and this could constrain the properties of dark energy. These exciting possibilities, which would be realised through Rawlings’ dream of a “billion galaxy survey”, elevated the high redshift HI survey science case over time so cosmology and dark energy become a key theme in the SKA science case. But it should be remembered that these are very demanding requirements needing the survey speed and sensitivity of the full SKA.

Another approach is to search for HI in absorption against bright background radio sources. For such an absorption survey the sensitivity needed is independent of distance and there are no reasons why there should not be radio galaxies out to redshifts z > 7. With a lowest frequency of 130 MHz it would be possible to trace HI absorption up to redshift z = 10.¹⁰⁷ This is a less demanding experiment and does not require a centrally concentrated array.

5.10.2 Epoch of Reionisation (EoR): Probing the Cosmic Dawn

The initial thoughts on the study of the high redshift universe in 1980s and 1990s were focussed on providing enough sensitivity to detect HI emission from galaxies at high redshift. The idea was that before stars formed it might be possible to detect the collapsing gas clouds of neutral hydrogen, the Zel’dovich pancakes, but these would have very low contrast and be difficult to observe. An early VLA detection was never confirmed (see Sect. 5.5.1). Scott and Rees (1990) took a different approach by looking at the evolution of the spin temperature of the neutral hydrogen as the gas collapsed and even conjectured that the proposed GMRT in India might be able to observe hydrogen gas in this phase. A more detailed analysis of the state of the gas as galaxies formed was published by Gnedin and Ostriker (1997). Piero Madau was probably the first to introduce these ideas to the SKA community at the Oort Workshop (1997) and specifically included the possibility of using the spectral signature of the 21 cm line to probe the epoch of reionisation and heating in the early universe as discussed in Sect. 5.5.5. Theoretical studies have continued, e.g. see Pritchard and Loeb (2012) for a detailed theoretical review, and detection of the EoR has become the most important science case for SKA-low.

Very soon after the realisation by the radio astronomer, Peter Shaver (ESO), that the global redshifted spectral signal would be detectable as a sharp step in the radio spectrum at the epoch of reionisation (Shaver et al., 1999), radio astronomers started paying attention to this possibility. However, detecting the relatively strong global EoR signal has remained elusive due to contamination by foreground emission and the extreme spectral line dynamic range requirements, e.g. (Singh et al., 2022). Attention has turned back to the large imaging radio telescopes such as the SKA and its precursors and pathfinders, the Low Frequency Array (LOFAR) and the Murchison Wide-field Array (MWA). The spectral line baseline requirements become easier but only the full SKA (SKA Phase 2) will have enough sensitivity to search for the much weaker spatial structures in the EoR signal using direct imaging. However, a statistical detection of the angular and frequency power spectrum is possible with the lower sensitivity precursors. The requirements on the spectral dynamic range now sets the most stringent specifications on the SKA-low spectral bandpass and chasing the changing frequency requirements¹⁰⁸ has added design complexity (see Sect. 6.5.3). While the need for extremely accurate spectral baselines was noted, there was no corresponding instrumental analysis on how to achieve this with an array.

The SKA precursor and pathfinder telescopes around the world are making rapid progress in laying the scientific and technical foundations for EoR observations with the SKA. Ongoing efforts include the Low Frequency Array (LOFAR) in the Netherlands, the Murchison Wide-field Array (MWA) in Western Australia, and the Precision Array to Probe the Epoch of Reionisation (PAPER) which has now become HERA, the Hydrogen Epoch of Reionisation Array now being built on the SKA site in the Karoo in South Africa. The technical difficulties encountered when making these observations had been hugely underestimated and obtaining a better understanding of how to build a telescope which can achieve the required dynamic range will be essential. We will follow up on the important role being played by the SKA precursors and pathfinders in Chap. 11.

5.10.3 Gravity: Challenging Einstein

Aspects of this key theme have been consistently included in the science case from the beginning. Radio observations of pulsars provide unique opportunities to study the properties of neutron stars, they can also provide exquisite tests for gravitational theories and pulsar timing networks are being used to detect gravitational radiation (see Sect. 5.5.13.2).

The tests of Einstein’s General Theory of Relativity can be extended to the very extreme environments where gravity is exceptionally strong, such as around supermassive black holes, and these rare systems are expected to be found with the increased sensitivity and survey speed of the SKA.

In the physics community the development of instruments to make a direct detection of gravitational waves had been ongoing since the 1960s and, in 1982, the Hulse and Taylor Nobel prize winning discovery of a pulsar in a binary system indirectly confirmed the reality of the gravitational wave predictions. This reinforced the link between the precision pulsar timing community and gravitational wave detection community. In 2015, the LIGO consortium made a direct detection of gravitational waves produced by in-spiralling black-holes (Abbott et al., 2016).¹⁰⁹ This was the beginning of a new era of observational gravitational wave astronomy.

Indirect detection of gravitational waves is also possible using a network of pulsars to detect changes in the arrival time on earth of the signals emitted by millisecond pulsars. These times are modified by the effect of gravitational waves as they pass the earth and the pulsars. These pulsar timing observations are sensitive to much longer wavelength gravitational waves than LIGO and would be sensitive to a stochastic gravitational wave background that could be produced during the brief inflationary period following the Big Bang. Current pulsar observations are tantalisingly close to detecting such a background so the advances which will be provided by the SKA are keenly anticipated.

The pulsar timing requirements on the SKA are twofold. They include first finding suitable pulsars with the very stable periods needed for precision timing observations and this requires a pulsar search capability over a large field of view. Then the precision timing observations require high sensitivity when pointing at the known pulsar. These two steps require quite different specifications¹¹⁰ and these more complex and demanding requirements have been continually refined over time.

5.10.4 Cosmic Magnetism

Since the mid-1990s, cosmic magnetism has been included as one aspect of many different science cases and, from 2000, the study of magnetism was recognised as a unique radio astronomy niche for the SKA. The general topic of magnetism was identified as a KSP in 2006 and it was noted that many unanswered questions in cosmology and in fundamental astrophysics are closely tied to the questions of the origin and evolution of magnetic fields. This move was enthusiastically promoted by Bryan Gaensler who was chair of the SWG from 2006 to 2008. It is certainly the case that the role of magnetism in the evolution of the universe had been largely ignored by astronomers in the past. This is partly because it is quite hard to measure magnetic field properties and partly because when magnetic fields are included, the physics becomes much more complicated and fewer astrophysicists have the expertise needed to include magnetic fields in their models.

5.10.5 Stars

Observing radio emission from stars has been raised throughout the life of the project but has not been a consistent part of the SKA science case and has never been raised to the level of a key science project. It has always been clear that the SKA would enable new advances in the sensitivity-limited field of stellar radio astronomy as discussed in Taylor and Braun (1999). The LOFAR detections of a new population of stars with non-thermal emission (Callingham et al., 2021) has now strengthened this case. It is also possible that exoplanets may trigger radio emission processes in the parent star, perhaps heralding the dawn of a new era in exoplanet research.

5.10.6 Transient Universe: The Bursting Sky

Pulsars are a great example of rapid variability of radio sources beyond the solar system, and observing pulsars has always been an important part of the SKA science case. It is an excellent example of a niche area for radio astronomy. The sheer number of pulsars that could be discovered, in combination with the exceptional timing precision made possible with the SKA sensitivity would clearly have a revolutionary impact as discussed in Sect. 5.5.13.2.

The idea of a broader search for transient events started in 2003 with a chapter by Cordes et al. (2004) considering the science case for observations of the dynamic radio sky. They made a prescient suggestion that SKA might find a new class of extragalactic transients which could be used to probe the intergalactic medium, as had been suggested by Ginzburg (1973). Cordes et al. (2004) also correctly recognised the need for different survey speed metrics for different classes of transients and drew attention to the difference between the survey speed metric for transients and for persistent sources. This research area jumped into prominence following the discovery of a new class of radio-transient, the Fast Radio Bursts(FRB), by Lorimer et al. (2007) and finally confirmed by Thornton et al. (2013).¹¹¹ Much of the current research on FRBs is being done by the SKA precursors: ASKAP, and MeerKAT, and also the Chinese Five-hundred metre Aperture Spherical radio Telescope FAST. The field is now dominated by the results from the Canadian Hydrogen Intensity Mapping Experiment (CHIME) which was built using a cylindrical reflector proposal developed as one of the SKA design options (see Chap. 6). In future the Deep Synoptic Array (DSA-110), and eventually DSA-2000 being developed at Caltech¹¹² will be optimum transient search and localisation instruments based on the large N—small D concept promoted throughout the SKA project development. By focussing on a specialised continuum transient survey mode in the 0.7–2.0 GHz frequency range, these instruments will be relatively inexpensive compared to the far more flexible SKA observatory.

5.10.7 Solar and Heliospheric Physics

Solar radio astronomy is a broad field of research all around the world. Radio observations of solar activity are being monitored with relatively modest scale solar radio observatories so the case for using the SKA is restricted to specialised areas. Imaging the complex and variable spatial and frequency structure of solar radio flares can take advantage of the large number of elements in the SKA which provide excellent instantaneous imaging capability over a wide range of frequencies. The practical value of observations of coronal mass ejections (space weather) is increasingly acknowledged due to their impacts on earth and on orbiting spacecraft. Observations of Interplanetary Scintillation can take advantage of the wide field of view at the lower frequencies to map the structure and motion of the coronal mass ejections against a dense background of scintillating radio sources (Morgan et al., 2003).

5.10.8 Cradle of Life: Technosignatures, and SETI

One of the first concepts for a cm wavelength radio telescope with more than 1 km² collecting area was the 1971 project Cyclops “A Design Study of a System for Detecting Extraterrestrial Intelligent Life” (Oliver & Billingham, 1972) and see Chap. 2. In September 1994 Bobbie Vaile (University of Western Sydney) presented the idea of the SKA aperture tile array to Barney Oliver and Frank Drake at the SETI institute in California. This was probably the trigger for Ekers to be invited to give the Pesek Lecture in October 1995 (see Sect. 5.5.4) on the potential of a “one square kilometre array” for SETI. The cradle of life has always been included in some form in the SKA science case, but sometimes with more emphasis on detecting planetary systems, rather than technosignatures, which requires the evolution of intelligence and technology as well as having a habitable environment. Jill Tarter¹¹³ (SETI Institute) proposed that the search for extra-terrestrial intelligence (SETI) be renamed “the search for technosignatures” to broaden the approach and to remove the “not a science” stigma sometimes associated with SETI.

5.10.9 Exploration of the Unknown

Back in 1961 when Jan Oort (Leiden Observatory, Netherlands) was making a presentation to the OECD about a proposal for a future large radio telescope (the Benelux Cross which later became the Westerbork Synthesis Radio Telescope) he made the following remark which remains just as applicable today¹¹⁴: “It is an unrewarding task to outline programmes for an instrument that does not yet exist, especially if the exact design and wavelength have not been definitely fixed. It is unrewarding in several respects. In the first place, those who will work with the instrument should themselves think out their programmes, at least to a considerable extent. In the second place, as has been so regularly the case in research with new types of instruments and new methods, it may well be that the instrument will lead into new, at present unpredictable, types of research; and these might become the most important. But, in order to discuss and fix the requirements for so expensive an instrument as we are about to construct, some consideration of astronomical aims is unavoidable. As an introduction to the discussions of the instrumental design I shall therefore briefly consider some of the major programmes that would be envisaged.”

In March 2002 the International Science Advisory Committee (ISAC) radio transients working group¹¹⁵ made explicit comments on the need to explore the unknown transient population. They emphasised that science predictions are based on the known populations of transient sources. The greatest return from such a survey will (should!) be the detection of currently unknown populations of sources. As discussed in Sect. 5.10.6 one such new and unexpected population, the Fast Radio Bursts, has now been found.

In their book on discoveries in radio astronomy (Kellermann & Bouton, 2023) emphasise that astronomy is an observational science. Astronomers cannot do experiments; they can only observe. Kellermann & Bouton explore the circumstances leading to the plethora of previously unknown phenomena discovered since the beginning of radio astronomy in 1933. One extraordinary consequence, strongly emphasised throughout their book, is that the scientific discoveries for which facilities become famous are rarely those they were built for. Given the nature of many of the discoveries in radio astronomy, this outcome is not unexpected. But what is surprising is that this obvious fact has had so little influence on the discussions of future facilities and concepts like “exploring the unknown” get little emphasis, and in some cases have been actively discouraged (e.g. the US Astronomy and Astrophysics Decadal Survey—ASTRO2010, see Sect. 4.5.3.4). In Fig. 5.4 it can be seen that the science case for the exploration of the unknown has been present over the history of the SKA, but emphasis has been sporadic.

5.10.10 Impact of New Discoveries on the Science Case

Looking back over the detailed science case summaries, the influence, sometimes fleeting, of the more recent discoveries is evident. Examples already discussed include:

The optical discovery of exoplanets orbiting normal stars (Mayor & Queloz, 1995) which gave credibility to the pulsar timing searches for exoplanets orbiting neutron stars discussed in Sect. 5.5.3. The indirect detection of gravitational waves based on the timing of a binary pulsar triggered the development of pulsar timing searches for primordial gravitational waves. When Fast Radio Bursts (FRBs) were discovered in 2007 by Lorimer et al. (2007), this greatly increased the importance of the transient discovery space and since these sources were extragalactic, they added cosmological significance to transient radio astronomy science.

The summary of all the major discoveries in radio astronomy (Kellermann & Bouton, 2023) not only expands on the more recent discoveries which have affected the SKA science case, but it also provides a perspective on how the discoveries have been made.

5.10.11 Science Cases Which Have Disappeared or are no Longer Emphasised

The VLBI science enabled by long SKA baselines, as summarised in Sect. 5.6.2 was no longer feasible with the relatively short baselines in the 2014 descoped SKA Phase 1 (SKA1), although some less sensitive VLBI capability with limited baseline coverage remains possible by using the SKA core in combination with existing radio telescopes.

The emphasis on stars and planetary science slowly declined, possibly because of changing scientific interests of the astronomers writing the science case.

In the past there were many discussions about using the SKA as a receiver for planetary radar and for spacecraft communication, and these options were seriously considered by NASA. With radar, the return signals decrease with the fourth power of the distance, so even an SKA cannot make a big impact on the maximum distance that can be probed. For deep space communications a detailed analysis by Fridman, Gurvits & Pogrebenko¹¹⁶ indicated a significant potential of the SKA as a “Direct to Earth” deep space communication facility.

However, NASA planned to go to higher frequencies moving outside the range considered for SKA-low and SKA-mid¹¹⁷ and was more interested in pursuing the future of optical communications. The NASA funding structure also changed so when the cost of the deep space communications was charged to the missions, the scope for expensive deep space communications developments was limited. These projects are no longer included in the SKA science case.

Some science topics are still included in the SKA design but never get promoted, for example radio emission from Ultra High Energy (UHE) cosmic ray showers, perhaps because the community of astronomers affected is relatively small and because the advocates for this research are mostly outside the astronomy community.

Over the 30-year lifetime of the SKA project, other telescopes have been built so some of the original SKA science projects had already been done.

5.11 Impact of the SKA Project on the Science

It is interesting to realise that while the science case was developed to motivate and develop the design specifications for an SKA telescope, the SKA project has itself had significant impact on the science case. In this section we discuss some of the ways in which this has happened.

5.11.1 Role of the Precursors and Pathfinders

As discussed in Chap. 4, Sect. 4.3.3, by 2006 the SKA pathfinders had made their way onto national roadmaps (LOFAR in the Netherlands, and MeerKAT in South Africa) or were funded to do so (ASKAP in Australia). Originally, these three instruments were designated “pathfinders”, but after a surge of interest in 2008 in the designation from many existing telescopes, the small number located on the two candidate SKA sites were called “precursors” to distinguish them from the others.¹¹⁸ These facilities were initially conceived as technology demonstrators, but there were national drivers to make these significant observatories that could do useful astronomy using state-of-art technology. In 2011 John Womersley (STFC, UK) noted¹¹⁹ “The phased nature of the project also needs to be emphasised—science from pathfinders/precursors leads to science from phase I which leads to the full array in due course.”

In retrospect these SKA precursors and pathfinders greatly invigorated the radio astronomy research environment worldwide and they have made important new observations and a number of significant discoveries, some of which are illustrated in the following section. Here are some examples of the discovery of previously unknown phenomena made with the SKA precursors and the pathfinders, which already demonstrate the importance of the exploration of the unknown:

• FRBs were discovered with the existing Parkes radio telescope, but the SKA precursors (ASKAP in particular) played a major role in follow-up observations to localise these events. The most prolific FRB radio telescope now is CHIME, a facility based on the parabolic cylinder technology that was evaluated as part of the SKA project (see Sect. 6.4).

• Ionospheric ducts were discovered using the MWA (see Fig. 5.17).

• Our knowledge of the amazing population of Galactic Centre filaments was greatly enhanced following some of the first observations with the MeerKAT SKA precursor (Fig. 5.18).

• LOFAR detections of stars (Callingham et al., 2021)

• Long period radio transients found by MWA and MeerKAT (Hurley-Walker et al., 2023)

• Odd Radio Circles—ORC discovered by ASKAP and MeerKAT. Norris et al. (2021, 2022), see Fig. 5.21.

5.11.2 How the SKA Predicted the Discovery of the Fast Radio Bursts (FRBs)

This discovery of the first FRB in 2007 and the use of FRBs to make a census of the baryon content of the universe by Macquart et al. (2020) is an example of how a successful future prediction was made by the International Science Advisory Committee (ISAC) in March 2002. The ISAC report from the radio transients working group¹²⁰ (March 2002) noted: “Based on the known populations of radio transient sources, an unbiased survey of the variable radio sky could reveal populations of radio pulsars in nearby galaxies (via the emission of giant pulses like those of the Crab pulsars), possibly as distant as the Virgo Cluster. A by-product of the detection of such pulsars would be direct detection of the ionized local intergalactic medium. In turn, this would allow study of the bulk of the baryons in the local Universe.”

5.11.3 Impact of the Science Case on the SKA Project

As discussed throughout this chapter there is an underlying tension between the emphasis on a range of key science drivers, and the technology generated opportunities which have historically triggered most of the discoveries in radio astronomy; this is why (Sullivan, 2009, p. 449) classified radio astronomy as a “technoscience”. The potential science is what drives the building of the telescope and is used to justify a particular funding investment. The science case is also used to guide the design of the telescope and for the SKA a complex process was set up to explore the design impact of a multitude of science cases. This had the crucial benefit of engaging with a much broader scientific support base. So, while the continuing effort to develop the science cases did not change the basic telescope design in any fundamental way, it was necessary to involve the broader community. This resulted in expanded specifications and made the SKA a general-purpose international observatory rather than just a big new telescope, as we discuss further in Chap. 11.

Different strategies are needed to convince different target communities. The most important role of the science case was to engage with the broader science community and identify niche opportunities. Some aspects of the science case drove specifications which were essential for the engineering design and an exciting science case was motivational for both the public and all those involved in the project. Funding agencies want to see a clear process to evaluate scientific excellence and value for money. They also expect appropriate management structures and risk management. Governments consider other benefits for large scale research infrastructures projects which go beyond the science, such as discussed at the strategic workshop “Benefits of Research Infrastructures beyond Science: the example of the Square Kilometre Array (SKA)”, organised by the European Cooperation in Science & Technology (COST) and held in Rome in March 2010.¹²¹

The COST report summary included: “When decisions on large scale research infrastructures are being made, aspects beyond the respective excellent scientific cases need to be considered. These aspects should include topics like the use of sustainable energy sources, the development and building of human capacities, new communication strategies and technologies and, finally, that the project would generate incentives to enhance global and transcultural collaboration in communicating the advancement of knowledge for the benefit of mankind.”

5.12 Current Science

The authors have selected a small sample of impressive results already achieved by the SKA precursors and pathfinders that were developed as part of the SKA project. These are leading the way and provide a glimpse of the science still to come with the SKA Observatory.

5.12.1 The LOFAR Pathfinder and the Radio Galaxy Cygnus A

Cygnus A observations with LOFAR (McKean et al., 2016) show extended lobe and counter-lobe emission, consistent with previous observations. But LOFAR provides the first direct evidence for a turnover in the spectra of both ‘primary’ hot spots (see Fig. 5.16). The very rapid turnover in the hotspot spectra cannot be explained by a low-energy cut-off in the electron energy distribution, as has been previously suggested. Thermal (free–free) absorption or synchrotron self-absorption models are able to describe the low-frequency spectral shape of the hotspots; however, the implied model parameters are unlikely, and interpreting the spectra of the hotspots remains enigmatic.

Fig. 5.16

LOFAR Observation of Cygnus a at 138 MHz. Credit: J. McKean and ASTRON

5.12.2 The MWA Precursor and the Discovery of Ionospheric Ducts

While analysing the ionospheric refraction effects in MWA data, Shyeh Tjing (Cleo) Loi and Tara Murphy (University of Sydney) discovered a very regular pattern of source position offsets which were aligned with the earth’s magnetic field lines, see Fig. 5.17. This was the discovery of ionospheric ducts, and the image was used on the February 2016 cover of the Journal of Geophysical Research (Loi et al., 2015).

Fig. 5.17

MWA discovery of ionospheric ducts, February 2016 cover, Journal of Geophysical Research, Space Physics. Credit: Shyeh Tjing Loi and Tara Murphy. Front cover reproduced with permission of John Wiley & Sons

5.12.3 MeerKAT, the South African Precursor Observes the Galactic Centre

The centre of our galaxy was one of the first images obtained with MeerKAT. The 64 antenna elements in the MeerKAT array provide a level of detail never seen before. This version of the galactic centre image (Fig. 5.18) is generated from MeerKAT observations described in Heywood et al. (2022). The image uses pseudo-colour to indicate the spectral slope of the radio emission. This image reveals nearly a thousand mysterious filaments as well as circular supernova remnants and regions of on-going active star-formation.

Fig. 5.18

MeerKAT observation of the Galactic Centre at 1.28 GHz using pseudo-colours to show the large range in radio spectral index. Credit: J.C. Muñoz-Mateos, I. Heywood and SARAO. Copyright: CC-BY-NC-4.0

5.12.4 ASKAP, the Australian Precursor Finds a Rare Polar Ring Galaxy

HI surveys taking advantage of the wide-field-of-view obtained with the ASKAP focal plane array can find rare objects such as the polar ring galaxy NGC4632 shown in Fig. 5.19 (Deg et al., 2023).

Fig. 5.19

Image of the polar ring galaxy NGC4632 showing the anomalous HI component observed in the ASKAP WALLABY HI survey (diffuse purple structure). The anomalous HI is superposed on an optical image from Subaru Hyper Suprime-Cam. The disc HI has been removed. Science credit: N. Deg et al., WALLABY Survey, CSIRO S&A/ASKAP, NOAJ/Subaru. Image credit: Jayanne English (U. Manitoba)

5.12.5 FAST and Its Pulsar Surveys

The largest radio telescope in the world is now the Five-hundred-metre Aperture Spherical Telescope (FAST), located in China. This was built as a possible SKA design to demonstrate one element of a small array of very large diameter collectors (large D—small N array technology). In the first year of routine operation with its 19-beam focal plane receiver, it has a pulsar discovery rate which exceeds all previous radio telescopes (see Fig. 5.20).

Fig. 5.20

Pulsar surveys of time using different radio telescopes based on the ATNF Pulsar Catalogue (V1.70). The Chinese FAST SKA pathfinder with its 19 beam focal plane array is now dominating the pulsar discovery rate. Credit Dick Manchester (CSIRO) and Di Li (NAO, China) for updates

5.12.6 Discovering the Unknown

As we have already discussed in Sect. 5.11.1 the SKA precursors and pathfinders have already made significant new discoveries. One example is a new class of radio source never seen before. The wide-field-of-view of the ASKAP precursor included some unusual sources called Odd Radio Circles (ORC). This discovery was confirmed with the higher resolution and higher sensitivity MeerKAT data and is thought to be the result of a violent explosion one million years ago in the central faint and distant galaxy (Norris et al., 2022) (see Fig. 5.21).

Fig. 5.21

ORC (Odd Radio Circle) composite image observed with ASKAP and MeerKAT superposed on a Dark Energy Survey (DES) optical image. Science Credit: R. P. Norris (Western Sydney U.), ASKAP-EMU/CSIRO, MeerKAT, DES-NSF/AURA. Image Credit: Jayanne English (U. Manitoba)