ICRAR Projects
The all-sky signal from the Epoch of Reionisation.
Dr Randall Wayth
Detecting the faint signal from the Epoch of Reionisation (EoR) in the early universe is a major goal of modern cosmology. This project will use data from the ICRAR Epoch of Reionisation global signal project to search for the faint radio signal from the EoR. This signal is present over the entire sky and can be detected, in principle, by a single precision antenna and receiver system. Such a system is under construction at ICRAR.
The project will require the student to develop a software pipeline to process large volumes of radio astronomy data. The pipeline will include stages to excise radio-frequency interference (RFI), calibrate the antenna and sky model and (hopefully!) extract the EoR signal and measurement errors on the signal.
View the project proposal (pdf).
A Census of Nearby Galaxies and their properties
Prof Simon Driver
In this project you will use data from the Galaxy And Mass Assembly (GAMA) survey to extract a nearby sample of several hundred galaxies spanning a wide mass range and explore their basic properties. A key focus is to determine a new classification scheme for dwarf galaxies. At the moment we have over 10 types of classifications of dwarf systems but do we need this man. By assembling a small catalogue of nearby galaxies with size, shape, mass and chemical abundance information you will explore the multi-dimensional parameter space and look for natural sub-groupings to help define an automated classification system.
The project requires "big picture" thinking, use of standard astronomy software packages and the development of code in either Fortran or the R programming language.
View the project proposal (pdf).
Connection between radio and gamma radiation in active galaxies
Dr Nadia Kudryavtseva with Dr Jean-Pierre Macquart
Active galactic nuclei are extremely bright sources at all wavelengths from radio to gamma radiation, which are variable on timescales from minutes to tens of years. It is believed that the enormous brightness of these sources is due to the accretion of matter on to a super-massive black hole in the center of a galaxy. Active galaxies eject matter in the form of jets, which moving outward with speeds close to the speed of light. The jets emit high energy gamma radiation, which was recently detected for numerous active galaxies by the space Fermi/Lat gamma-ray telescope.
Various physical mechanisms are proposed for the generation of the gamma emission, but we still do not know what causes it and where exactly the gamma-rays are produced. This project will focus on the connection between the radio jet power, magnetic field and the gamma-ray luminosity.
You will produce maps for the whole 2cm MOJAVE sample (1 map per galaxy) with semi-automatic mapping software. You will estimate the flux-density and polarised flux in fixed bins in the jet (1 pc, 5 pc, 10 pc bins), and then search for the correlation between the total radiated power and the polarised emission in different regions in the jet. Comparing this with the Fermi gamma-ray light curves and jet flux-density variability will determine the long-sought location of the gamma-ray emission and will help to us understand what generates it.
The successful completion of the project will lead to a refereed publication in an astronomical scientific journal.
View the project proposal (pdf).
Development of Radio Astronomy Receiver Instrumentation
Dr. Budi Juswardy with Prof Peter Hall
Developing cost-effective and high-performance receiver systems is a particular challenge in radio astronomy, where highly-sensitive, wideband, low-noise receiving chains, with low-power consumption are required. ICRAR is currently developing and evaluating suitable components to realise low-frequency aperture array radio receivers working at 70MHz - 450MHz. Various aspects of the receiver front-end sub-system design, such as antenna, low noise amplifier (LNA), driver circuitry, fibre optic links, protection devices (ESD diodes), and off-grid power generation (based on photovoltaic panel) will be examined. The project will give an insight on the design of radio instrumentation for the $2 billion square kilometre array (SKA) project. The project would suit an engineering/applied physics student interested in participating in the development of radio astronomy instrumentation.
View the project proposal (pdf).
Dust extinction and emission in HIPASS galaxies
Prof Gerhardt Meurer and Dr Ed Elson
The student will correlate our list of HI selected galaxies observed in the ultraviolet and H-alpha with the source catalogue from the Japanese Akari satellite which detects thermal emission from warm dust. While ultraviolet and H-alpha emission are commonly used to estimate the star formation rate in galaxies, they can be severely affected by dust absorption. The far infrared emission that Akari measures is sensitive to the dust grains that absorb this radiation thus in principle this will allow us to recover the flux lost to dust absorption. After correlating the catalogues at different wavelengths, we will use the catalogue to constrain a model of the dust absorption properties, which will allow better dust corrections of the multi-wavelength data, and more accurate estimates of the star formation rate of galaxies.
View the project proposal (pdf).
Imaging of Ultra-wide field of view
Dr John Morgan and Dr Leith Godfrey
In radio interferometry, the resolution of an array at a given frequency is proportional to the baseline length. Very Long Baseline Interferometry (VLBI) uses radio telescopes with intercontinental baselines to produce the highest resolution images in radio astronomy.
For the parabolic dishes typically used for interferometry at centimetre wavelengths, the ratio of the field of view to the resolution (the number of “pixels” across the field of view) is the same as the ratio of the longest baseline to the width of the largest dish.
For smaller radio interferometers (such as the Australia Telescope Compact Array) which have dishes a few 10s of metres in diameter and maximum baseline lengths of a few kilometres, it is common to image the entire field of view of the instrument, resulting in an image around a megapixel in size. However with the huge resolution of VLBI, imaging the same patch of sky will produce a far larger high-resolution image, around 1 to 100 gigapixels!
Techniques have been developed to repeatedly average the original visibility dataset (up to a Terabyte in size) and allow it to be processed in smaller chunks. There are many good reasons for doing this and one bad one: the data reduction software that we use is old and not designed for working on such large datasets. In this project we will explore the use of high-performance computing to do data analysis on the large (unaveraged) dataset directly. This is an open-ended project which can be adapted a great deal to suit the interests of an individual student. A project plan will be decided at the end of week 2.
View the project proposal (pdf).
Multi-wavelength properties of star forming galaxies
Prof Gerhardt Meurer with Dr Ed Elson
As a follow-up to the HI selected all sky survey HIPASS, we have started the surveys SINGG and SUNGG to observe these radio selected galaxies in the light of star formation tracers H-alpha line emission and ultraviolet continuum emission. I have developed a pipeline to improve the processing of these star formation images. The student will run the pipeline and write new software to extend its capability to include linking results obtained at other wavelengths into a larger database and to write web pages to present the data in a uniform manner.
View the project proposal (pdf).
MWA images of big, bright radio sources
Dr Randall Wayth with Dr Stephen Ord and Dr Nadia Kudryavtseva
The Murchison Widefield Array (MWA) radio telescope has a huge field of view compared to conventional radio telescopes – around 30 degrees across – which makes it possible for the telescope to make images of very large structures on the sky in a single pointing.
The goal of this project is to use recent data from the MWA to form images of large radio sources. As well as making “pretty pictures”, the images will be scientifically valuable. The images can be compared to data from other telescopes/frequencies to measure essential properties of the objects such as spectral index and polarisation, which are related to the physics of the emission mechanism in the object.
This project will be an excellent introduction to synthesis imaging with radio telescope arrays. The project can optionally include investigation of non-conventional imaging and deconvolution techniques depending on student interest.
View the project proposal (pdf).
The nature of relativistic jets in X-ray binaries
Dr James Miller-Jones with Dr Roberto Soria
X-ray binaries are double star systems comprising a compact stellar remnant (black hole or neutron star), in a close orbit with a less-evolved star. Matter is transferred from the donor star into the strong gravitational field of the compact stellar remnant. To conserve angular momentum, this matter forms an accretion disc, whose inner regions heat up to the point at which they emit X-rays. A significant fraction of the gravitational energy liberated by the infalling matter can be channelled into outflowing jets that originate close to the black hole or neutron star and are accelerated to speeds close to that of light. The relativistic electrons in these jets emit synchrotron radiation, which we detect with radio telescopes. It is still unclear what events in the inner regions of the accretion flow lead to the formation of these jets. The student will work with Dr. Miller-Jones and Dr. Soria to explore the properties of the radio jets from an X-ray binary, observing them evolve in time, and relating variations in the radio emission from the jets to changes in the X-ray emission from the accretion flow, to improve our understanding of how the jets form and evolve. The project will primarily involve processing and analysing data from one of the world's premier radio telescopes (the ATCA or LBA in Australia, or the EVLA or VLBA in the USA). The exact data set to be analysed will be chosen in consultation with the student.
View the project proposal (pdf).
What makes quasars twinkle?
Dr Hayley Bignall with Dr Cormac Reynolds and Dr Jean-Pierre Macquart
In the same way that stars “twinkle” due to turbulence in the Earth's atmosphere, whereas planets do not due to their larger angular sizes, compact radio sources “twinkle” at centimetre wavelengths due to turbulence in the interstellar medium – our Galaxy's “atmosphere”.
This “twinkling”, or interstellar scintillation, depends on both the line of sight through the Galaxy as well as the angular size of the radio source. In order to scintillate, a source must contain a component of angular size no larger than several tens of microarcseconds – seeing with this resolution is equivalent to being able to spot a coin on the moon! Interestingly, very distant (high redshift) quasars show reduced levels of scintillation, indicating that they have larger apparent angular sizes, compared with their lower redshift counterparts.
Very Long Baseline Interferometry (VLBI) provides a means of directly imaging milliarcsecond-scale structure. Earlier comparisons showed that scintillating sources tend to be more compact on milliarcsecond scales than sources which don't scintillate. This project will compare VLBI properties, from the literature and recent archival data, with published scintillation-induced variability of a large sample of quasars, in order to investigate whether the high redshift quasars have intrinsically bigger jets than their low redshift counterparts. This project will involve collating results from published data and developing scripts to read in, plot, compare and statistically analyse the data. Optionally, the project can involve additional analysis of recent, unpublished VLBI data, or comparison with multiwavelength source properties.
View the project proposal (pdf).
Search for diffused cluster emission in TGSS data
Dr Sabyasachi Pal
Diffused radio emission is observed in some galaxy clusters. This phenomenon is more evident in the low radio frequency regime (100-1000 MHz) of the radio spectra. TGSS is a major survey to map the whole northern hemisphere (and most parts of the southern hemisphere) by the Giant Meterwave Radio Telescope (GMRT) at 150 MHz. We will use TGSS images to search for diffused radio emission near the centre of the galaxy clusters.
View the project proposal (pdf).
VLBI observations of Venus Express
Dr John Morgan
In December 2008 a lunar occultation of Venus occurred. Three Italian radio telescopes simultaneously observed a radio signal from Venus Express over a time range that covered both immersion and emersion of the signal for all radio telescopes. Careful correlation and comparison of the arrival time of the wave at the three telescopes could allow the detection of the Lunar atmosphere.
The data are available and their analysis would suit a technically-inclined student with an interest in high-performance computing and radio interferometry. The extremely bright and narrow-band signal allows extremely precise measurements of small frequency shifts to be made. Previous students have made changes to the DiFX software correlator to optimise it for data reduction of this type. Further exploration of these techniques is warranted and may have interesting implications for a range of observations including SETI.
Several different avenues can be explored depending on the interest of the student, emphasising the high-performance computing, algorithm development or observational aspects. A final project plan will be decided at the start of week 3.
View the project proposal (pdf).
ICRAR/iVEC Joint Projects
Compressive sampling for multi-resolution images (MWA/ASKAP/SKA).
A/Prof Slava Kitaeff
Compressive sampling is new and rapidly developing field in signal and image processing. It’s applicability to some radio astronomy techniques has been discussed. This project is to investigate a suitability of compressive sampling for providing multi-resolution images from the data cubes. The project will require studying the literature, developing a program code, and benchmarking.
View the project proposal (pdf).
A database driven I/O server for serving data slices to distributed applications
W/Prof. Andreas Wicenec with Derek Gerstmann, A/Prof. Chen Wu and A/Prof. Kevin Vinsen
The latest radio astronomical and simulated astrophysical multi-dimensional datasets can easily exceed many terabytes (TB) in size. Handling such large amounts of data efficiently poses many interesting problems for I/O systems, for both the hardware and software layers involved. The goal of this project is to implement a basic server infrastructure which supports a parallel and/or distributed application to request slices of multi-TB data sets stored in a distributed hierarchical storage system. The information about the actual location(s) of a given slice will managed via an in-memory database, and the proposed server system will be used to select the most appropriate copy of the requested information residing on the fastest layer of the storage hierarchy, based on locality and coherency. The implementation can be done in either of the three languages mentioned below. The available test data sets will be in HDF5 format, the server will be implemented as a RESTfull HTTP server. We are working in an agile test-driven environment and this project should be implemented using these paradigms as well.
View the project proposal (pdf).
GPU Porting Software Engineering Process
A/Prof Kevin Vinsen
Massively parallel software is the cornerstone of modern simulation techniques. The associated software engineering processes required to support this have been left behind. As most ports are done with a very small budget, high ceremony process is not practical or affordable. This project will look at what light weight processes are required to port code from a traditional HPC environment to a GPU cluster.
View the project proposal (pdf).
An HDF5 to FITS round-trip converter
W/Prof. Andreas Wicenec with A/Prof. Chen Wu
Since many years astronomical data sets are stored in a format called FITS (Flexible Image Transport System). This format, as the name implies, has initially been developed to allow transferring big multi-dimensional images from the observatories to the astronomers. Being very flexible it is nowadays used to transport almost any kind of gridded and array oriented data. HDF5 (Hierarchical Data Format) is a much more recent development and allows the storage and self-consistent description of complex data sets. Every subset can be a complex data set by itself, or it can also be array or gridded data. HDF5 is widely used in the computational simulation community. The goal of this project is to develop a converter between HDF5 and FITS, which should allow maintaining as much of the meta-data as possible. In the ideal case a conversion from HDF5 to FITS and back should result in the same file, only subject to a different order of the meta data keywords.
View the project proposal (pdf).
PCA Tomography for Radio astronomical data cubes
W/Prof. Andreas Wicenec with Dr Christopher Harris
Recently (Ricci 2011, http://arxiv.org/abs/1104.5511) Principle Component Analysis (PCA) techniques have been employed to analyse three dimensional data cubes from Integral Field Unit (IFU) instruments. The results are very interesting and encouraging and it would be very interesting to see the application of this technique on spectral cubes from radio observations and investigate the potential for scientific exploitation of these data cubes. In the course of the project proposed here the student should first get a good overview of the OpenCL language, MPI and highly parallel computing. Then the algorithms behind PCA tomography and the structure and content of radio astronomical data cubes should be analysed. Based on this PCA tomography should be implemented in OpenCL and MPI with the goal to run it on a GPU cluster. In the course of this project the code will only be tested on smaller data sets and on single nodes.
View the project proposal (pdf).
Footnote links and information
This page:
- Last updated
- Wednesday, 22 June, 2011 1:30 PM
- Website feedback
- info@icrar.org
- Page ID
- 1668087