Ocean modelling

Associate Professor Jochen Kaempf, Lecturer in Oceanography, Flinders University

In South Australia no issue is more important than water. To ensure drinking water is available in times of severe drought the State Government has invested $228 million in building a seawater desalination plant at Port Stanvac, Adelaide.

Associate Professor Jochen Kaempf, Lecturer in Oceanography, and his research group at Flinders University have been using eResearch SA’s supercomputing facilities for hydrodynamic modelling purposes in a rich variety of applications.

For example, Jochen’s model simulations have enhanced our understanding of the risks that seawater desalination poses to the marine environment.

Using eResearch SA’s high performance computing facilities, Jochen has been able to model the mixing of desalination brine with seawater to measure the impact on the dilution of salinity in sea water, and monitor the risk of marine pollution.

His model’s predictions currently play an important role in assessing the viability of BHP Billiton’s proposal to build a seawater desalination plant in Upper Spencer Gulf as part of the Olympic Dam expansion.

Jochen’s current focus is to develop new methods to quantify marine connectivity in the ocean, which is fundamental to understanding the ways marine systems operate including their response to environmental change. Overall, his research contributes to the implementation of sound, sustainable management of marine resources in Australia.

More information contact:
Assoc. Prof. Jochen Kaempf (email: jochen.kaempf@flinders.edu.au)

Mixed-metal and nano-sized catalytic clusters

Researchers from the School of Chemistry and Physics at the University of Adelaide have been using eResearch SA’s supercomputers to investigate the properties of catalysts that could be used to improve some of the world’s most significant chemical processes.

The development of more efficient catalysts holds huge potential gains for both industry and the environment. Catalysts speed up chemical reactions, significantly reducting capital costs for businesses, improving chemical and energy efficiency, and reducing the impact on the environment.

Around 80% of all current industrial chemical processes use catalysts. The world’s largest and arguably most important catalysed industrial process is the Haber-Bosch process for producing ammonia from N₂ and H₂. It alone consumes a huge 1% of the world's energy supply, meaning the potential gains from the discovery of more efficient catalysts are great.

Recent research effort has focussed on improving catalytic activity through the use of nano-sized metal particles. Nano-sized metal clusters containing 3–40 atoms have been shown to induce catalysed activity at significantly lower temperatures than bigger metallic surfaces. By combining the advantages of mixed-metal catalysts and nano-sized metal clusters, researchers think it may be possible to achieve unprecedented control over the activity, efficiency and selectivity of metallic catalysts.

To produce these improvements, researchers are working to understand the microscopic details of the catalytic process and, in particular, how molecular interactions with mixed-metal clusters change as a function of nanoparticle size and composition. Research currently being carried out using eResearch SA’s facilities addresses this; the goal is to undertake systematic experimental and computational investigations into the chemical and physical properties of mixed-metal clusters and their interactions with several important molecules such as N₂, CO, and CO₂.

One day this might even help researchers solve the holy grail of catalytic processes—how to convert atmospheric CO₂ gas simply and easily into another chemical compound, thus offering the world a clear climate change solution.

Fluid mechanics and turbulence modelling

Researchers from the School of Mathematical Sciences at the University of Adelaide are using eResearch SA's facilities to explore fluid mechanics and undertake sophisticated turbulence modelling.

Turbulence is characterised by unsteady three-dimensional fluid motion over a wide range of spatial and temporal scales. It is a fundamental problem in many fields, including astrophysics, oceanography, meteorology, combustion, aeronautics and engineering.

The equations governing turbulent flow have been known for over a century, however there is no known general solution. Instead, computers can be used to obtain approximate solutions. The problem is that in most cases of practical interest, the range of scales is so enormous that the computational problem cannot be solved on even the largest supercomputers. Therefore, simplified models are needed.

The goal of the researchers is to produce tractable models capable of reliably predicting turbulent flows. This involves the use of both state-of-the-art turbulence models and large-scale computational resources.

eResearch SA's supercomputer Corvus is being used to run code for this research project—the scale of the calculations is such that they could not be run on ordinary machines.

The image to the right is a visualisation of a turbulent mixing layer—the flow formed between two streams of different velocity. The image shows the mixed fluid only, coloured according to composition.

Optimising optical fibre design

The Institute for Photonics & Advanced Sensing (IPAS) brings together physicists, chemists and biologists to pursue a new transdisciplinary approach to science.

The Institute is developing novel photonic, sensing and measurement technologies that will change the way science is done within traditional discipline areas, stimulating the creation of new industries, and inspiring a new generation of scientists to be engaged in solving real-world problems.

At the heart of IPAS is photonics, the science of light. IPAS uses novel fibre optics and laser systems to explore the world.

Working with discipline specialists from chemistry, biology and all other sciences linked by a common purpose of evolving new research methodologies, IPAS is developing practical solutions which solve society’s problems.

IPAS research targets applications in five key market areas:

• Defence and national security
• Environmental monitoring
• Mining
• Preventative health
• Food and wine.

IPAS major research strengths span:

• the production of novel optical materials
• new developments in surface science for sensing
• optical fibre sensors
• lasers and optical systems
• laser radar (LIDAR) and environmental and atmospheric monitoring
• theory and computational modelling
• fibre laser development
• nonlinear fibres and devices
• luminescence for radiation monitoring and archaeological dating

The research IPAS carries out on eResearch SA’s high-performance computing facilities is finite element based modelling (FEM) of our microstructured optical fibres (MOF’s).

Having access to high-performance computing facilities enables IPAS to accurately and rapidly simulate wave propagation in glass with complex structures. This allows IPAS to model important fibre properties such as dispersion, nonlinearity and confinement loss, which allows optimisation of fibre design parameters before fibres are produced in the IPAS fibre production facility.

The ability to rapidly model fibres gives IPAS an understanding of the physical parameter space they need to be in to achieve the Institute’s research goals.

The access to eResearch SA’s high-performance computing resources are used not only in the design of new fibres but also with the analysis of existing fibres, optimising the design and fabrication process.

Understanding the universe

Tera-electron volt (TeV) gamma-rays are extremely high energy photons that are generated by astrophysical phenomena such as supernova explosions, black hole environments, and pulsars. The fact that we detect gamma-rays from such objects tells us there are very energetic physical processes occurring in these objects.

TeV telescopes allow us to detect the gamma-ray photons indirectly as they enter and interact with Earth’s atmosphere. Data from the telescopes yield information such as the direction of the source and the energy of the photons. These observations are important because they reveal information about the universe that cannot be measured with other types of telescope.

Dr Gavin Rowell, of the High Energy Astrophysics Group in the University of Adelaide’s School of Chemistry and Physics, has been studying this field of astronomy for 20 years and has worked in Japan, Germany and Australia.

He has been working with the High Energy Astrophysics Group on the design of an array of up to 50 TeV gamma-ray telescopes, each approximately 6 meters in diameter. This array is designed to open up a new window in TeV gamma-ray astronomy. Gavin’s plan to build a telescope array in outback Australia marks significant progress by Australian researchers in this area of astronomy.

Gavin and his group use Hydra, one of eResearch SA’s high-performance computers, to simulate the interaction that occurs when TeV gamma-rays and ‘cosmic-rays’ interact with our atmosphere, and to model the optimal design of the array of these telescopes so that they can discern gamma-rays from cosmic-rays.

Cosmic-rays cause complications in the analysis of data from a gamma-ray telescope because they produce signals which are similar to those of gamma-rays.

Using Hydra to model how the telescope and data analysis respond to both gamma-ray and cosmic-ray events means that the design of the telescope array is as efficient as possible, giving Gavin’s team new information about some of the universe’s most remarkable mysteries.