ATNF Emerging Science Program Science and technology trends Science trends

							                         ATNF Emerging Science Program 2005/06

1. Science and technology trends:
1.1 Science trends

At the start of the third millennium, astronomy has come of age as a physical science. Fuelled
by new opportunities in technology, observational experiments such as those proposed with
the next generation of radio telescopes can now aspire to produce data sets of the size and
quality required to test the most fundamental physical theories for our understanding of the
origin of the Universe and its constituents.

Astronomers now know that many key scientific questions regarding the structure and
evolution of the Universe and its contents are intrinsically complex, with many diverse
astrophysical processes contributing to the observed trends. Consequently, many of the Big
Questions must be approached statistically, with sample of objects large enough (>106), and
covering a such broad range in frequency from radio to X- and gamma-rays, to either
disentangle the individual physical processes responsible for the trends seen in the data or to
uncover the weak signals in the data that will provide fundamental information on the nature
of our Universe.

As an example, one of the most important astrophysical problems of the current age, namely
the nature of “Dark Energy” (see Connecting Quarks with the Cosmos: Eleven Science
Questions for the New Century 2003, The National Academies Press) can be directly
addressed using the next generation of large datasets.

Recent analyses of the distances to distant supernovae, anisotropies in the Cosmic Microwave
Background, and large-scale structure in the distribution of galaxies lead to the inevitable
conclusion that the Universe not only contains a pressureless dark matter component which
dominates over the mass-energy density of ordinary (baryonic) matter, but that it also contains
an even more dominant negative pressure component, popularly known as "Dark Energy".

This dark energy may or may not be related to Einstein's well-known "cosmological
constant". Observations of the pressure-density relation (the equation of state) of dark energy
over a range of redshifts are required to understand its nature and its possible relation with
high energy physics.

Future radio telescopes (SKA and its pathfinder facilities) are powerful tools for studying
dark energy. By studying the power spectrum of the distribution of galaxies at redshifts near
unity, they can compare the geometry of the Universe with that at higher and lower redshifts,
therefore constraining its pressure-density evolution, and probing the equation of state of dark
energy.

However, such surveys will require the detection and parameterization of over a million
galaxies over a three-dimensional volume of size 2000 Gpc3. The data volume will be over
100 terapixels. The sheer data volume and the difficulty in finding and parameterizing
galaxies calls for new algorithms which will efficiently recognise the 3D signature of a galaxy
in an environment dominated by thermal noise and systematic artefacts arising from radio
frequency interference and incomplete removal of sychrotron-emitting radio continuum
sources.



                                             1/
1.2 Technology Trends

1.2.1 Aperture and Focal Plane Arrays

Aperture and Focal Plane Arrays represent a large-step push in the field of radio astronomy.
Their successful implementation will have a major impact in the field.

Aperture arrays have the potential to be the first astronomical instruments, in any wavelength
range that provide a large field of view covering a significant fraction of the sky, with
multiple simultaneous full-sensitivity beams within this field of view. The wide fields-of-
view are a key technology step in the delivering the radio surveys capable of addressing the
Dark Energy question. The SKA, particularly at lower frequencies, will almost certainly be an
aperture array.

Focal plane arrays (FPAs) have the potential to increase the efficiency, in terms of observing
time, of traditional concentrator-type radio telescopes such as parabolic dishes, by factors of
up to of order 100, while also significantly increasing their performance. An indication of the
potential of FPAs can be gained from the Parkes 13-beam 21cm multi-beam system, which is
a pre-cursor of a true FPA, but whose outstanding scientific success has placed the Parkes
telescope at the forefront of single dish radio observatories, worldwide. FPAs form an
important component of the extended New Technology Demonstrator (xNTD) design.

The successful development of aperture and focal plane array systems will depend on
breakthroughs in a number of technology areas. Antenna elements must be developed which
are capable of close packing into array systems, with suitable feeding structures, in some
cases capable of being cooled to cryogenic temperatures, and interfaced to highly integrated
receivers containing specially developed integrated circuit components. The means of
combining the elements into multiple simultaneous beams will only eventuate through the
development of innovative digital signal processing solutions using state-of-the-art processes
and techniques.

1.2.2 Virtual Observatory

A major (science & technology) trend in Astronomy is the emergence of the Virtual
Observatory (VO). The VO seeks to provide astronomers with seamless access to- and
analysis of heterogenous data sets covering the full range of the electromagnetic spectrum by
establishing standards and building layered IT tools in a homogenous interface. This provides
an enormously powerful environment for scientific discovery and underpins the future
scientific exploitation of pan-wavelength information to reach a much deeper understanding
of the physical processes responsible for many of the phenomena seen in the Universe than
has hitherto been possible. It also aims to provide the tools for data mining of large petabit
data sets.




                                            2/
2. Divisional Response:
2.1 Science and Technology Trends

ATNF is investing significantly in developing the science drivers for the next generation of
radio telescopes. This includes research in the areas of high redshift star-forming galaxies,
dark energy and pulsar physics (including the establishment of a pulsar timing array for
detection of gravity waves). This research is being led by the ATNF’s two Federation
Fellows; Ron Ekers and Dick Manchester.

2.1.2 Aperture and Focal Plane Arrays

ATNF plans to invest significant resources in the enabling technologies for the development
of multi-beaming array systems, with the aim of staying at the frontier of the field, where
systems such as the Parkes 21cm multi-beam have placed us. The enabling technologies can
be divided into three broad areas: antennas, receivers and digital signal processing.
In the antenna area, ATNF and the ICT Centre are working on the MNRF-supported New
Technology Demonstrator (NTD) program with the developing antennas/FPAs solutions for
the SKA.

ATNF will continue its work on the development of state-of-the-art monolithic microwave
integrated circuits (MMICs). MMICs will be essential building blocks in the highly
integrated receivers and processors required for array systems. Through its special
arrangement with U.S. semiconductor fabricator Velocium Products, ATNF has already
demonstrated its ability to create high performance specialised MMICs using frontline
semiconductor technologies. A project aimed at the design of a fully integrated receiver-on-a-
chip for the frequency range 500 to 1700MHz range has made good progress. This circuit is
directly applicable to systems such xNTD and could provide the backend for higher frequency
cryogenically cooled FPA receivers for the Parkes telescope.

ATNF is applying significant resources to research and development of digital signal
processing (DSP) for next generation radio telescopes, including array systems and the SKA.
Innovative techniques will be required to contain the extreme processing power needed for
broadband multi-beam array systems in realisable and practical systems.


2.2 Priorities for Investment

As a key SKA technology, the ATNF wishes to continue to its development of cooled FPA
receivers for the Parkes radiotelescope. This is outside the scope of both the NTD and xNTD
program, but cooled FPA systems will be needed on Parkes to provide the sensitivity required
to fully exploit the power of the FPA. It also wishes to continued its development of the
software required to process and search the wide-field data-cubes produced by FPA systems
as part of its contribution to the International Virtual Observatory effort and future science
exploitation of the xNTD.




                                             3/
3. Use of additional funds for emerging science in 2004-05
3.1 Aperture and Focal Plane Arrays
i) Antennas:
What the additional funds have been used for:
Resources for the line feed and FPA study.
What progress has been made:
Dr John Bunton and Prof. Sergei Vinogradov have completed a thorough electromagnetic
analysis and assessment of the practical feasibility of line feeds for cylindrical antennas. This
work was reported at the 2004 SKA meeting in Penticton, Canada, and in the publications
listed in Appendix 2. From subsequent SKA meetings, both internationally and within
Australia, it became evident that the use of parabolic reflector antennas rather than cylindrical
reflectors would find greater acceptance to the international SKA community. The effort
being put in by the ATNF towards likely SKA technology is now being focussed on focal
plane arrays with small parabolic dish antennas, and so subsequent feed array research has
shifted from 1-dimensional line feeds to 2-dimensional Focal Plane Arrays. The initial line
feed research has been of benefit to the SKAMP (SKA Molonglo Prototype) and has also had
some application to the 2-dimensional case that is being applied to the design of the major
Australian SKA prototype New Technology Demonstrator (NTD) project. This work has been
reported in references (1) and (2).

ii) Receivers:
A conceptual design for a 12mm focal plane array for the Parkes telescope.
What the additional funds have been used for:
Resources for an initial theoretical study and a conceptual design.

What progress has been made:
Two approaches to the problem have been pursued: an electromagnetic analysis of FPAs in
general, and a conceptual design study aimed specifically at the Parkes instrument.
A. Focal Plane Arrays are complex because of the relatively large coupling between both the
antenna and the receiver elements. Integrated analyses of the whole system (feeds, receivers,
primary reflector) are necessary to understand the issues involved in designing specific
instruments. Progress has been made towards this analysis. A report “An FPA measurement
model” is in preparation, and work has begun on an antenna-to-software system simulation.
This research is being conducted in conjunction with the MNRF2001-funded NTD project
that will construct a real system to allow validation of the simulations.
B. The conceptual design study has come to the following conclusions:
1. The recently extended solid surface of the Parkes dish, out to a diameter of 55M, has
   provided ATNF with a unique opportunity to implement a low-noise FPA system at high
   frequencies. It means that the illumination can be adjusted to provide a good compromise
   between an optimal F/D ratio for an efficient FPA and a guard ring which shields the FPA
   elements from ground radiation.
2. FPA elements made from aperture coupled, stacked patch antennas show promise as a
   solution for a cryogenically cooled array at 22GHz with good bandwidth and polarisation
   performance. We will build on the considerable work that has been done in the satellite
   and military fields in this area.



                                             4/
3. A 5x5 array is the smallest size that would provide an acceptable performance in terms of
   an efficient astronomical instrument. The physical size of such an array at 22GHz should
   present no problems in terms of cooling. We probably should aim for a somewhat larger
   array, e.g. 7x7.
4. Proven performance of ATNF designed MMIC LNAs at 22 GHz, and experience gained
   in packaging these devices means we are well placed to tackle the key problem of
   connecting the feed elements to the LNAs, ideally in a fully integrated structure operating
   at cryogenic temperatures.
5. Down conversion of the multiple RF signals can be achieved using proven technologies
   and commercial devices. Distributed amplification is a candidate for Local Oscillator
   routing. Both planar and hybrid integrated structures are viable.
6. A signal conditioning and digitising system mounted in the focus cabin, and using
   integrated receiver MMICs being developed in the MNRF2001 SKA program, would
   provide a fibre optic data interface to a fully digital beam former backend. Strict attention
   will need to be given to avoiding interference generation.
Overall, the concept for a 12mm FPA on the Parkes telescope appears viable and, if
implemented, would place Parkes, once again, at the forefront of single dish observatories,
worldwide.

3.2 Virtual Observatory
i) Service Grids
What the additional funds have been used for:
The emerging science funds were used to employ Dr Murphy.
What progress has been made:
Dr Tara Murphy was employed to develop the automated pipeline for the generation of radio
images from AT Compact Array data. Normally image generation is an interactive process
conducted by experts in radio aperture synthesis who provide parameters needed for the data
analysis. An important component of this work, conducted in conjunction with the ICT
Centre, provided a means of inferring those parameters from information present in the
original records. The pipeline itself has been completed for an important subset of ATCA
data types. The pipeline is connected to the newly established 1.5terabyte archive of all
ATCA data. It allows users to form images of any region of the sky observed by the ATCA
since it began operating in 1990. The pipeline’s public release on the web is currently being
planned. Dr Murphy reported her work (3) at the IVOA Small Projects Meeting, Pune, India.

ii) Data Mining
What the additional funds have been used for:
Project planning, the employment of Dr Matthew Whiting.
What progress has been made:
A position has been created and filled (Dr Matthew Whiting) to develop software that can
automatically, efficiently and robustly search multi-dimensional images, detecting,
characterizing and cataloguing the objects found. A thorough survey of source finding
techniques has been completed and an initial study of the mathematical theory has started. Dr
Whiting has attended the “Summer School in Statistics for Astronomers & Physicists” at Penn
State University to establish links with other researchers in this field. Progress has been
reported in an internal memo (4).




                                             5/
4. Proposed use of additional funds for 2005-06

4.1 Focal Plane Arrays

4.1.1 Objectives

The overall objective for 2005-6 is to continue research and development by testing FPA
models against actual measurements of a prototype focal plane array being constructed for the
NTD project, and by the detailed design of a 12mm FPA for the Parkes telescope.

i) FPA analysis
The combination of close FPA element spacing and wide frequency range along with ever
increasing needs for high dynamic range between strongest emitters to weakest emitters will
force new methods of processing the vast flow of data from telescopes using Focal Plane
Array (FPA) technology.

The close element spacing and frequency range dictate basic responses with considerable
variability and coupling between elements which will need to be measured and corrected.
The terabit/sec data flow will necessitate a rethink of processing probably more in the
direction of data flow methodologies. This combination will provide a very fertile ground for
innovation and potential leadership by CSIRO in the future SKA technologies.

A Post-doc or graduate suitably experienced in computational analysis using Matlab and other
software would be required.

Resources required:      1.0 FTE

ii) FPA for Parkes

A detailed design for a 12mm focal plane array for the Parkes telescope.

This funding would allow a continuation of the work commenced in 2004-05 and provide a
detailed design for a 12mm FPA system for the Parkes telescope. The aim is to keep Parkes
at the forefront of the field of multi-beaming on large parabolic concentrators, by being the
first to take the next big step forward.
Resources required:   0.7 FTE

4.2 Advanced radio imaging techniques for Focal Plane Arrays

4.2.1 Objectives

The objective for 2005-2006 is to initiate research and development of techniques for
astronomical imaging using Focal Plane Arrays. The performance of aperture arrays (AA) in
radio astronomy has surpassed early expectations, partly through the development of
advanced imaging algorithms such as self-calibration, mosaicing, and faceted wide-field
imaging. The fully sampled nature of modern FPAs allows adaptation of the aperture array
techniques to FPAs. Dr. Tim Cornwell will lead this effort. Dr. Cornwell has been a leader in
the development of AA techniques over the last twenty five years, and wrote an early paper


                                            6/
demonstrating the extension of the AA self-calibration algorithm to FPA systems. The goal of
our effort will be to place ATNF at the forefront of the scientific use of processing algorithms
for FPAs. Doing so will complement the objective of putting a 12mm FPA system on the
Parkes telescope, and development of NTD and SKA.

Our work will have two targets: first, the extension of current radio data reduction software to
support data from FPAs and FPA synthesis arrays (like the xNTD), and second, the
development of new algorithms for calibration, data editing, and imaging using such data. For
this we require two positions: a software developer, and a scientist conversant in both radio
astronomy and advanced data processing techniques.

Resources required:   2 FTE




                                             7/
Appendix 1: The Divisions research capabilities

ATNF aspirations

The ATNF aspires to enhance its status as a world-class radio astronomy observatory. In so
doing, its primary aims are to provide world-class facilities for radio astronomy and to
support and foster a vigorous astronomical community within Australia.
Until 2009 the Australia Telescope Compact Array will be the world’s largest radio telescope
with access to Southern skies that is sensitive to radio signals at millimetre wavelengths. In
addition, by 2006/07 the ATNF will have improved the current performance of the Compact
Array for wide-band observations by a factor of 10; further enhancing the scientific output of
the Array and its use for internationally-competitive research programs. Users of the Compact
Array will also have important scientific leverage in accessing the international Atacama
Large Millimeter Array (ALMA), when ALMA is commissioned at the end of the decade.
The ATNF also aspires to be one of the leaders in the international effort to design, build and
operate the next generation of radio-astronomy facilities through its involvement with the
Square Kilometre Array (SKA). The SKA is a billion-dollar project that will provide a radio
telescope with a collecting area of about one square kilometre, making it a hundred times
more sensitive than any existing radio telescope. The ATNF will seek to facilitate, wherever
possible, scientific, technical and political engagement in the SKA by all other stakeholders
(CSIRO, Universities, Industry and Government) within Australia.
By the early years of next decade the ATNF, together with Australian and international
partners, aims to have built and be operating a major “pathfinder” instrument for the SKA
(ISKAP). As part of the roadmap towards ISKAP, ATNF plans to work with other CSIRO
divisions and Australian Universities to develop infrastructure (the extended New Technology
Demonstrator or xNTD) at the proposed Australian site for the SKA. This will enable
Australia to consolidate its position in the SKA by establishing local capabilities and technical
leadership influencing key decisions of site choice (2006) and technology choice (2009) over
the coming years. Once completed in 2008/9 the xNTD will be operated as part of ATNF and
deliver key science outcomes to the Australian community in survey astronomy.
During the next decade the ATNF will continue to play a key role in frontier astrophysical
research with ISKAP that will further influence the design of SKA, in particular relating to
wide-field applications such as the nature of dark energy, the evolution of galaxies in the early
universe, and the use of pulsar observations as probes to test general relativity.
The ATNF also aspires to consolidate its position as one of the world’s major suppliers of
radioastronomy services; including instrumentation, astronomical data products and
spacecraft tracking. Through its ongoing technology program, the ATNF will grow and
develop its existing capabilities in strategic areas of antenna design, receiver technology and
signal processing and will invest in emerging areas such as the Virtual Observatory. It will
continue to build strategic links with industry partners to maximise the returns from its
technology development to Australian industries. It will work with international agencies,
including NASA and ESA, to play a key role in the development of the next generation Deep
Space Network. The ATNF will also explore a strategic partnership with other Australian
astronomy institutions (including the Anglo-Australian Observatory, the ANU Research
School for Astronomy and Astrophysics and the CSIRO ICT Centre) to build critical mass in
the area of astronomical instrumentation.


                                             8/
Divisional Capabilities by Theme
                     Capability                                                          Total

                     Telescope    Receiver          Signal       cm-wave     mm-wave
Theme
                     Operations   Technology        Processing   Astronomy   Astronomy

Approx No of EFTs*   61           29                19           14          4           127

% of total           48%          23%               15%          11%         3%          100%

Theme 1

Theme 2

Theme 3

Other 1


Key: Proportion of Capability allocated to each Theme.
          ‘High’                  ‘Medium’                       ‘Low’




Appendix 2: Emerging science outputs in 2004-05
(1) D.B Hayman, T.S. Bird, K.P. Esselle, P.J. Hall, "Encircled Power Study of Focal Plane
Field for Estimating Focal Plane Array Size," to be presented at the 2005 IEEE AP-S
International Symposium on Antennas and Propagation and USNC/URSI National Radio
Science Meeting to be held in Washington, DC, USA on July 3-8, 2004

(2) J.S. Kot, J.D. Bunton, D.B. Hayman, S. Jackson, “Large Array with Focal Plane Array
Feeds for the SKA”, to be presented at IEEE MTT Symposium on “Very Large Microwave
Arrays for Radio Astronomy and Space Communications” at IEEE IMS2005 at Long Beach
CA, June 2005.

(3) T. Murphy, “Progress in the Australian Virtual Observatory” presented at the IVOA Small
Projects Meeting, Pune, India, October 2004.

(4) M. Whiting, “What we know about source detection and its theory”, Internal ATNF
report, June 2005.




                                               9/
Appendix 3: Advancing Frontiers of Science Role

Theme                 Research area                Base Expenditure Base + Supplement

Astrophysics          1 Science
                      1.1 High redshift            $250k             $250k
                      Universe/dark energy
                      1.2 Pulsars                  $250k             $250k
Technologies for      2. Focal Plane Arrays
radioastronomy
                      2.1 Analysis and modelling   $55k              $136k
                      2.2 Receiver design          $500k             $587k
                      2.3 Advanced imaging         $250k             $470k
                      2.4 Signal Processing        $557k             $557k
                      3 Virtual Observatory
                      3.1 Data Mining              $110k             $110k
                      Total                        $1972k            $2360k
                                                                     $388k (supplement
                                                                     only)

Base expenditure in 1 largely funded from Federation Fellowship and CSIRO matching funds
Base expenditure in 2 and 3 largely funded from external sources e.g. MNRF-2001.

Funding received: $350k




                                         10/