A three-dimensional (3D) map of the Cooper Basin region has been produced from 3D inversions of Bouguer gravity data using geological data to constrain the inversions. The 3D map delineates regions of low density within the basement of the Cooper/Eromanga Basins that are inferred to be granitic bodies. This 3D data release constitutes the second version of the 3D map of the Cooper Basin region. It builds on Version 1 of the Cooper Basin Region Geological map, released in 2009.

The Cooper Basin region is coincident with a prominent geothermal anomaly and forms part of a broad area of anomalously high heat flow. High-heat-producing granites, including granodiorite of the Big Lake Suite (BLS) at the base of the Cooper and Eromanga Basins sequences combined with thick Cooper/Eromanga sedimentary sequences that provide a thermal blanketing effect, result in temperatures as high as 270° C at depths <5 km. The location and characteristics of other granitic bodies are poorly understood and accurately identifying them is an important first step towards future geothermal exploration in this region.

3D Bouguer gravity field inversion modelling was carried out using the UBC inversion software. An initial gravity inversion was performed using seismic horizons to constrain the 3D distribution of the Cooper/Eromanga Basin sediments. Densities, derived from seismic velocities from a refraction seismic survey in the region, were assigned to the Cooper/Eromanga sediments in order to constrain their gravity contribution. A series of Iso-surfaces were generated, enclosing low density lobes within the basement of the initial sediment-constrained inversion model. Gravity 'worms' were used to pick the iso-surfaces that approximate the lateral sub-sediment extent of potential granites within the basement. A series of subsequent granite-constrained inversions were generated by assigning different maximum cut-off depths to the lobes. The inversion model that produced the most 'neutral' result had a maximum cut-off depth of 10 km.

The 3D map was then used to predict temperatures throughout the volume of the map. Thermal properties were sourced from the literature and from direct measurements. Forward predictions of temperatures were carried out using the Simulator for HEat and MAss Transport (SHEMAT) software package. Thermal properties were iteratively updated until a satisfactory match was achieved between the model and temperature measurements. The resulting temperature distribution gives strongly elevated temperatures over the BLS, as well as broader regions of elevated temperature in the northwest of the study area toward Mt Isa, under the Adavale Basin in the north-east of the study area, and south-east of the BLS.

Uncertainty was analysed using a stochastic modelling technique. A sensitivity analysis was first performed to select the parameters which, when varied, had the greatest effect on the predicted temperatures. These parameters are: thermal conductivity of the basin sediments, heat production of the basement and granite units, and basal heat flux. Stochastic models were then run, giving the standard deviation of the temperature at each point in the model. The resulting standard deviation distribution shows that areas of highest predicted temperature are also areas of highest error. However, when the standard deviation values are converted to percentage error, a different pattern emerges: Highest error values are observed where the Cooper Basin sediments are thickest. Lower error values are observed over the BLS and in the southeast of the model area.