An Australian-developed algorithm has allowed researchers at the CSIRO to map, for the first time, Centaurus A, an enormous galaxy which emits a radio glow covering an area 200 times bigger than the full Moon.
Developed by the Australian Square Kilometre Array Pathfinder (ASKAP) telescope team, the algorithm allows researchers to resolve imaging issues caused by a massive differential between the “brightness” of the radio waves emitted at the core and edge of the Centaurus A galaxy, based some 14 million light-years away.
“In the centre of the galaxy is a supermassive black hole, 50 million times the mass of our sun, which is incredibly bright,” explains Ilana Feain, of CSIRO’s Australia Telescope National Facility (ATNF).
“Because the centre is so bright we have a dynamic range issue, which means that when using conventional image processing tools the image is saturated in the bright regions and we are unable to detect the faint emission regions around these very bright areas. To resolve this, we needed very sophisticated software processing algorithms and these are something that have only recently been developed.”
Compiling 406 images taken over 1200 hours by the Compact Array radio-telescope based in Narrabrai, NSW, the research team was able to “mosaic” the images together over the course of 10,000 hours of processing time on an Opteron-based Sun cluster.
With the completed image, supplemented by data from the 64-metre radio telescope dish located in Parkes, NSW, the research team can now begin to understand how the most massive galaxies have formed and evolved with time, Feain says.
“The energy released from supermassive black holes at the centre of massive galaxies like Centaurus A is intimately related to the formation of galaxies,” she says.
“But the actual physics of how the energy from a black hole is coupled to the forming galaxy is not well understood. On one hand we see evidence of energy from supermassive black holes being capable of heating and blowing out the gas and dust from a forming galaxy, effectively halting it from forming more stars. On the other hand, as is the case in Centaurus A, we can see that the energy from the black hole has caused gas and dust clouds in the galaxy to heat and collapse and form more stars. We don’t know which scenario is more important on a global scale of massive galaxy formation.”
According to Tim Cornwell, project lead and computing engineer for ASKAP, cracking the imaging challenges of Centaurus A has proved to be a useful test case for future challenges the CSIRO expects to face with the ASKAP telescope.
Due for completion in 2012, ASKAP will act as a path-finding instrument and showcase for Australia’s claim to host the Square Kilometre Array — a new generation radio telescope with a discovery potential 10,000 times greater than the best present-day instruments.
ASKAP, the CSIRO says, will give astronomers remarkable insights into the formation of the early universe and to test theories of cosmic magnetism and predictions from Einstein’s theory of relativity.
ASKAP will comprise an array of 36 antennas each 12-metres in diameter, capable of high dynamic range imaging and using wide-field-of-view phased array feeds, currently under development by the CSIRO.
“The problem with current radio telescopes is that they can only see a narrow field of view, so if you want to image the entire sky you have to make about 100,000 ‘pointings’ and that’s just too slow,” Cornwell says.
Using ASKAP, an image of the entire sky in could be produced in just three or four hours, allowing researchers to produce a new map every night showing changes day by day.
“You could also survey all the neutral hydrogen in the nearby galaxies over the course of a year to give you a census of all the galaxies in our neighbourhood,” Cornwell says. “That’s important for understanding the local structure of the universe.”
According to the CSIRO, in one week ASKAP will generate more information than is currently contained on the whole World Wide Web; in one month it will generate more information than is contained in the world’s academic libraries.
To handle the phenomenal amount of data the Pathfinder will generate — expected to be about 20 petabytes per year — the CSIRO will look to deploy a 100 teraflop cluster based in either Perth or Geraldton, Cornwell says.
“Much of the processing we do is embarrassingly parallel, so for this application we don’t need a lot of interconnect,” he says. “With a dedicated supercomputer you are largely paying for high speed interconnect. We just want the overall processing capability.”
If Australia is successful in its bid for the Square Kilometre Array, then it will look to deploy an exaflop worth of processing power to handle the additional load, Cornwell says.
Much of the software being used to manage the collection and processing of the raw data from the Pathfinder and ASKAP is being written in-house with C++ and MPI.
However, for the telescope monitoring software, the CSIRO is using EPICS — a set of open source software tools and applications which provide a software infrastructure for use in building distributed control systems to operate devices such as particle accelerators and major telescopes
For connectivity, a dedicated fibre link, being built by the CSIRO and managed by AARNet, will connect the telescope site at Boolardy, WA to the CSIRO at its Geraldton facility. The National Broadband Network will then link Geraldton to Perth and wider Australian and international networks.
As reported by CIO in June, Australia is also leading astronomic research with the SkyMapper observatory, tasked with scanning the night skies to create the Southern Sky Survey.
A deep digital map of the southern sky, the Southern Sky Survey — with a little help from the National Computational Infrastructure (NCI) National Facility — will allow astronomers to study interstellar objects ranging from nearby asteroids to super-distant objects like quasars.