MWA: The Mileura Widefield Array

Link to MWA Web site

I. General project/facility description

1. Overview of the facility/project
   The MWA/LFD is a university based, innovative instrument development program that may serve as a pathfinder for the Square Kilometer Array (SKA; discussed separately in this report). Although a low-cost demonstrator instrument, the MWA/LFD will provide world-class science capability for studies of the Epoch of Reionization (EOR), transient radio emission and high precision remote sensing of the heliosphere. It will have an effective collective area of ~8000 square meters at 200 MHz, and will have capability in the range from 80-300 MHz. Operations are planned to start in 2007.
   Mileura is a remote livestock ranch in outback Western Australia, remarkably free of man-made RFI because of its location far from population centers in an area being formally designated as a radio quiet zone. It is the centimeter/meter wave equivalent of a high, dry and dark site.

2. Managing institution and organization
   The MWA/LFD is led by MIT, with involvement both of the Center for Space Research (CSR) on campus and the Haystack Observatory. The project leader/manager is located at Haystack and workpackages are shared between the US and Australia. The EOR scientific collaboration, focused on the primary astronomical scientific goal, is led by CSR. The LFD science program is based on active involvment of all academic partners, currently (Jan 2005) including participants from MIT, CfA, Caltech, Cornell, NRAO, NRL, UCSD, several institutions in Australia as well as Japan and Germany.
   The detailed management plan involves the MWA joint development with the ATNF-led NTD project allowing for efficient sharing of resources, work packages and developments. Extensive communication and coordination between the separate LFD and NTD project management teams is built into the overall structure. The MWA Executive Group includes the Directors of the Haystack Observatory and the ATNF. Numerous other senior experienced personnel oversee specific tasks.

3. Funding source(s)
   The LFD is a collaboration between several partner institutions and organizations, all of whom plan to provide significant funding toward the total expected cost (LFD only) of $10M. The largest single component of funding is $4.25M requested from the NSF in Sept 2004. Most of the remaining funds from partners are already secure, pending approval of the NSF grant. Due to the significant heliospheric and space weather component of the LWA science goals, part of the NSF funding has been requested from ATM and part from AST. ATM has already funded a study at Haystack and CSR on space weather applications of the MWA. A $2.0M grant from the NSF ITR program in 2001-4 supported design work on the LOFAR project, focused on frequencies above 80 MHz, and on low angular resolution applications for the core region of the array. This work was transferred to the LFD, and both the technical and scientific design of the LFD lie on the original development path. In addition to the NSF funds, MIT has invested $0.8M in institutional funds, so that $2.8M has been spent on the LFD to date.
4. Construction history and cost
   Future facility; NYA
   Construction is not expected to start before the end of 2005. The projected construction cost is ~$10M, including development, hardware, software, infrastructure and 1 year of operations and science analysis. These costs are based on coordination with the approved Australian NTD project.

5. Operational history and cost
   Future facility; NYA
   The US share of operating costs for the full MWA (not the demonstrator) in the next decade is estimated to be ~$2M/year, assuming substantial US investment in a low frequency capability.

II. Technical details

1. Specifics of telescope/instrument
   The LFD will contain 8000 dual-polarization dipole antennas optimized for the 80-300 MHz frequency range, arranged as 500 "tiles", each a 4 X 4 array of dipoles. The array will have no moving parts, and all telescope functions including pointing will be performed by electronic manipulation of dipole signals, each of which contains information from ~4 steradians of sky centered on the zenith. Each tile will perform an analog beamforming operation, narrowing the field of view to a fully steerable ~25 degrees at 150 MHz. The 500 tiles will be scattered across a roughly 1.5 km region, forming an array with very high imaging quality, and a field of view of several hundred square degrees at a resolution of several arcminutes. FPGA-based massively parallel digital hardware will select and condition a 32 MHz instantaneous bandwidth, and perform cross-correlation and digital array beamforming. Software for array calibration, as well as to support specific science goals will be developed.

2. New capabilities anticipated/planned in next 5-10 years
   The LFD array design is a major departure from conventional approaches, with simple brute-force computing favored over fine-tuning and precision in the antennas and receivers. As such, the LFD plays a unique and vital technology development and demonstration role for the future of radio astronomy.
   The successful completion and operation of the LFD would, by mid-2008, act as a springboard to implementation of the full MWA as an international user facility in the next decade. The ATNF goal is to build a sensitive high-redshift HI survey instrument for pursuing Dark Energy research, operating down to around 500 MHz. Assuming this comes to pass, the LFD side of the MWA project may lead to several science driven technology developments, which in order of perceived likelihood are:
   • Greatly increased collecting area and sensitivity (factor > 10)
   • Inclusion of long baselines to expand the range of science targets
   • Expansion of digital capabilities (bandwidth, number of beams, etc)
   • Expansion of frequency coverage up to ~500 MHz
   • Expansion of frequency coverage below 80 MHz
   The total cost of a low-frequency component of the full MWA implementing most of the above features, and if built as a standalone facility, has been estimated at $60-70M. A major US involvement, including cost sharing from existing and/or new partners, might require NSF investment in the $20M range in the next 5-10 years. This low cost of a major new research facility is a consequence of the inexpensive nature of low-frequency analog hardware, and the design emphasis on digital systems subject to Moore's Law. The most costly single item will be software related to calibration and widefield imaging, an area where synergy with Australian widefield efforts will be maximized.
   This work constitutes important technology development for future large radio telescope arrays of which the SKA (discussed separately) is the pre-eminent example. A single design approach cannot efficiently cover the full SKA frequency range, which spans a factor of several hundred. The US SKA technology development program (SKA-TDP) is focused on solutions based on small paraboloids, best suited to higher frequencies. The LFD focus instead explores the important 100-600 MHz range, where phased array technology offers key advantages over steerable paraboloids. Radio array design is a rapidly moving field driven by digital advances, and the LFD demonstrator approach maximizes options for efficient and cost-effective future growth.

III. User profile

1. % of "open skies" time
   The LFD will become operational in late 2007, and is projected to run in low-cost campaign mode for one year. During that time, experiments will be scheduled, executed and results analyzed, by science collaborations made up of researchers from LFD partner institutions as well as outside researchers whose collaboration has been prearranged. As the project moves forward, user-oriented software will be developed as funds permit, building upon experience from operating the demonstrator. We anticipate that open skies time will be introduced in 2008-9, and will steadily be increased toward 100% as the full MWA user facility is implemented.

2. Institutional affiliations of users
   At present, there are three key science areas and associated science collaborations, dealing with EOR, transients and heliospheric research. Formal collaborations are still developing, but currently (Dec 2004), 7 different US institutions, 5 Australian institutions, one European and one Japanese institution are represented. Informal expressions of interest span a much wider range of institutions. Project policy is to actively broaden such participation in LFD science based on contribution of resources as well as interest.

3. Student access, involvement, usage
   The LFD will be a training ground for US students with both technical and scientific interests. Students will have access to data through their affiliation with supervisors who are established members of the LFD collaborations. One MIT student is currently involved in early deployment of prototype systems and will spend time in Australia. Another student is starting work on cosmological simulations at the CfA.
   In general, the LFD represents a good opportunity for multiple student thesis projects, with such a large volume and variety of data products to be generated. Particularly, LFD project policy is to strongly encourage graduate students in instrument development and subsequent research, emphasizing the revitalization of technological expertise.

IV. Science Overview

1. Current forefront scientific programs
   Future facility; NA.
   Although the instrument is not yet built, active research associated with the project is being conducted, primarily on the Epoch of Reionization and on heliospheric science. Current work focuses on theoretical expectations, and understanding how to design the LFD for maximum scientific impact.

2. Major discoveries (through 1999)
   Future facility; NA

3. Science highlights of last 5 years
   Future facility; NA

4. Main future science questions to be addressed
   
   • Detecting and characterizing the redshifted 21 cm HI signal from the cosmological epochs of reheating and reionization (EOR), between redshifts of 6.5 and 16. By virtue of its wide field of view, the LFD will have sufficient sensitivity to measure the predicted power spectrum of fluctuations due to structure in the neutral IGM, to the presence of bubbles caused by the first ionizing sources, and to spin temperature gradients generated by the reheating process. It will also be able to image "holes" in the neutral IGM caused by quasar-generated Strömgren spheres. For EOR studies, control of systematics and removal of foregrounds are the primary technical challenges, and are central to the LFD design

   • Characterizing the dynamic radio sky, with a sensitivity 6 orders of magnitude better than previous searches in this frequency band. The LFD design includes the All Sky Monitor, with continuous monitoring for transient sources on all timescales longer than 0.5 seconds, yielding excellent discovery potential. As for EOR and as for previous instruments in this band, systematic errors, not collecting area will be the limiting factor, and the LFD design is focused on minimizing them.

   • Remotely sensing the solar wind plasma and disturbances in it, and placing constraints on the magnetic fields, via scintillation and Faraday rotation observations. The volume and nature of data to be generated by the LFD can lead to significantly improved space weather prediction. The capability of constraining magnetic fields is unique and essential for such prediction.

   • Several secondary science goals, including studies of pulsars, the ISM, the ionosphere, recombination lines, and solar burst imaging.

5. Synergies with other major forefront facilities
   
   • VLA: provides useful information on the 327 MHz and 74 MHz sky, as well as ionospheric behavior. VLA data are already in use for algorithm testing.
   • ATA: Scientifically, LFD and ATA cover complementary frequencies ranges, beneficial to transient source studies. Some of the technical challenges have similar solutions and discussions regarding sharing of IP for digital filterbank implementations in FPGA devices are ongoing, for example.
   • SWIFT, HETE-2, GLAST, LSST, LIGO: These facilities will act as triggers for MWA observing. The electronic agility of the LFD will allow efficient followup.

6. Unique contributions
   The key projects are all unique to the LFD and cannot be replicated by other existing or planned instruments on this timescale and with these performance levels. The combination of unique attibutes such as relatively unexplored frequency range, a continuously and fully accessible field of view of several hundred square degrees, high surface brightness sensitivity, high fractional processed bandwidth, fully electronic operation with multibeaming, electronic pointing and tuning, and superb instantanous point spread function plus tight control of systematics, all in one instrument, delivers unique research capability.

V. Education/Outreach activities

1. Visitor facility
   
2. Student programs
   
3. Other (as apply)
   

VI. Documentation/website URLs

1. URL of facility website
   http://web.haystack.mit.edu/MWA
   Note that this website is under development.
2. URL of EPO website
   
   • Haystack Observatory EPO website, covering a wide range of activities and initiatives into which LFD-related materials are fully integrated.
   • MIT Center for Space Research EPO website, which will also include LFD future activities.
3. URL(s) of any brief overviews of project/facility
   LFD Overview
4. URL(s) of miscellaneous documentation
   See above.