Spatio-temporal geodata mining


Observations suggest that fundamental evolutionary cycles on Earth, including major sea-level fluctuations and pulses of deep Earth resource formation are linked to tectonic and deep mantle processes controlling supercontinent formation, dispersion and amalgamation. However, most current geodynamic models cover merely a tiny 3% portion of Earth's history. To understand the Earth as a dynamic and complex system on which we are dependent, we need to investigate the "deep" geological time, explore at least two cycles of supercontinent formation and dispersion, and provide significant advancec in answering fundamental questions regarding: The relative role of top-down (driven by tectonic plates) and bottom-up (driven by deep mantle convection) processes in causing major supercontinent/superocean cycles on Earth; the effect of different plate configurations before and after the Pangean supercontinent on large-scale mantle convection flow and surface topography; and the relation between the time-dependence of subduction geodynamics and ore deposit formation along convergent plate margins. To achieve these advances, we need to reconstruct the Earth over a time period of more than 1 billion years before the present. A predictive framework integrating paleogeographic information with tectonic and geodynamic simulation tools will be able to reveal long-term, non-linear feedbacks between processes in Earth's interior, the crust, and the surface.

The aims of this project are:

1. The identification of niche spatio-temporal geological environments for the formation of mineral and energy systems. Existing plate models form the basis for reconstructing age-coded geological data back in time, factoring in time-dependence affecting both their characteristics and location and considering (a) Varying degrees of uncertainty and a fragmented geological record, (b) Heterogeneous data types, (c) Spatio-temporal variability, and (d) Large high-resolution datasets.

2. The application of data-mining tools to accumulate diverse sources of geological data for tectonic plate modelling into deep-Earth time. Accurate plate models only exist for a fraction the Earth's history due to a preservation bias in the geological record. Pursuing plate modeling further back in time relies on the accumulation of fragments of geological data from different sources. We will use a novel data mining methodology to bring together the wide variety of fragmented evidences from different sources into a single "coherence optimization framework". As such optimal use of existing data can be made when comparing alternative plate configuration hypothesis.

3. Multivariate exploitation of small-scale geophysical data for mineral deposit prospectivity. Data mining methodologies are becoming increasingly popular as part of exploration workflows. State of the art approaches tackle the combination of multi-modal datasets by using assumptions such as statistical independence of sources and use simple fusion rules and multivariate pattern recognition to handle data that is scaled differently, and collaborative filtering to unify collections of data associated with a targeted object.


Professor Dietmar Muller, Dr Rohitash Chandra

Research Location

School of Geosciences

Program Type



This project is a collaboration between the Centre for Translational Data Science and the EarthByte Group at the School of Geosciences, designed to develop and apply Bayesian machine learning tools to the analysis big and complex geodata and to the rigorous testing of tectonic and dynamic Earth models against observations. The project is designed to lead to an improved understanding of how the plate-mantle system has evolved over the last billion years, providing an improved framework for deep Earth resource formation, particularly ore deposits.

Additional Information

There is the potential for research travel money through the Edgeworth David Travelling Scholarship, which is offered to several students per year within the School of Geosciences, and through a funded SIEF project in this research area. 
In addition to the academic requirements set out in the Science Postgraduate Handbook, you may be required to satisfy a number of inherent requirements to complete this degree. Example of inherent requirement may include:

  • Confidential disclosure and registration of a disability that may hinder your performance in your degree;
  • Confidential disclosure of a pre-existing or current medical condition that may hinder your performance in your degree (e.g. heart disease, pace-maker, significant immune suppression, diabetes, vertigo, etc.);
  • Ability to perform independently and/or with minimal supervision;
  • Ability to undertake certain physical tasks (e.g. heavy lifting);
  • Ability to undertake observatory, sensory and communication tasks;
  • Ability to spend time at remote sites (e.g. One Tree Island, Narrabri and Camden);
  • Ability to work in confined spaces or at heights; Ability to operate heavy machinery (e.g. farming equipment);
  • Hold or acquire an Australian driver’s licence;
  • Hold a current scuba diving license;
  • Hold a current Working with Children Check;
  • Meet initial and ongoing immunisation requirements (e.g. Q-Fever, Vaccinia virus, Hepatitis, etc.)
You must consult with your nominated supervisor regarding any identified inherent requirements before completing your application.

Want to find out more?

Contact us to find out what’s involved in applying for a PhD. Domestic students and International students

Contact Research Expert to find out more about participating in this opportunity.

Browse for other opportunities within the School of Geosciences .


spatio, temporal, geodata, mining, geology, geosciences, dietmar muller, Machine learning, Bayesian optimisation, high-performance computing

Opportunity ID

The opportunity ID for this research opportunity is: 1829

Other opportunities with Professor Dietmar Muller

Other opportunities with Dr Rohitash Chandra