Anatomy of a heat flow anomaly: a South Australian perspective

Mike Sandiford, Jeremy Meyer, Narelle Neumann, Sandra McLaren
Department of Geology, University of Adelaide.

Some 23 existing heat flow measurements from South Australia show a significant heat flow anomaly located on, but significantly broader than, the Flinders Ranges. The average measured surface heat flow for this region is ~90 mWm-2 and is almost twice the measured heat flow in the central and western parts of the Gawler craton (Cull, 1982). An obvious pertinent question concerning this dataset relates to its accuracy. We have used studies of the coherence between gravity and topography to provide an independent test on the thermal state of the lithosphere, since the effective elastic thickness of the lithosphere is generally regarded as a measure of the depth to some critical isotherm (~500°C). The coherence study shows an increase in the effective elastic thickness from 36+2-6 km in the southern Flinders Ranges to 92+50-16 km in the Gawler Craton, implying crustal thermal gradients vary by a factor of at least 2 across this area. Estimates for the northern Flinders Ranges are poorly constrained at 9660 km, probably reflecting the incorporation of domains with very different effective thicknesses in the analysis.

Accepting that the heat flow data contain useful information, what then can we infer about the thermal state of the South Australian lithosphere? The standard procedure for relating observed variations in surface heat flow to deeper thermal regimes is to regress surface heat flow-heat production data to obtain estimates of the both reduced or mantle heat flow (qr) and the characteristic length scale (hr) for the distribution of heat production in the crust. For the South Australian dataset this approach yields estimates for qr of ~30 mW m-2 (e.g., Sass & Lachenbruch, 1989), implying upper mantle thermal gradients of ~10°C/km and lithospheric thickness no greater than about130 km. This interpretation is inconsistent with the SKIPPY evidence for lithospheric thicknesses in excess of 200 km throughout this region (e.g., Zielhuis & van der Hilst, 1996). One important point to emerge from the coherence study is that, for the measured surface heat flows, the South Australian lithosphere appears abnormally strong, and certainly much stronger than the African lithosphere at any given surface heat flow (Hartley et al., 1975). Indeed, the estimates for elastic thickness imply that the near surface thermal gradients associated with the surface heat flow measurements cannot propagate to depths greater than about 5-10 km in the crust, and that mantle or reduced component of the heat flow is very low (of the order of 10-15 mWm-2). We show that the conventional regression technique greatly overestimates qr when there is substantial lateral variations in crustal heat sources, as suggested by detailed surface heat flow measurements in the Roxby Downs region (Houseman et al., 1989) which show variations in surface heat flow between 70-120 mWm-2 at the 10-50 km scale.

The implied low values of qr require that the remainder (~75-80 mWm-2) of the surface heat flow in this province is contributed by crustal sources, with the total crustal budget being some 2-3 times normal Proterozoic crust. Not surprisingly the heat flow anomaly coincides with a U-rich province hosting, amongst others, the massive Cu-U-Au-REE deposit at Roxby Downs, and the processes by which such crustal-scale enrichment's are achieved is, in itself, an outstanding geochemical problem with profound metallogenic implications. Broadly the heat flow anomaly corresponds to a region in which the main crust forming events occurred in the late Palaeoproterozoic through to Mesoproterozoic (~1850-1590 Ma), and can be distinguished from the central and western Gawler Craton in which relicts of an older Archaean-earliest Proterozoic(~2500-2400 Ma) crust forming event are widespread. The Mesoproterozoic granitic gneisses in this region often show extraordinary enrichment’s in heat producing elements (for example the Yerila Granite near Mount Painter has heat production of ~40 mWm-3), and provide the obvious candidate for carriers of the anomalous crustal heat sources. Indeed, the observed surface heat production levels in the Mesoproterozoic sequences at Mount Painter are so high that they easily account for all excess heat flow providing the surface exposures are representative of the upper 5 km of the crust.

Our emerging picture for this anomalous heat flow province is that it reflects exceptionally high, but laterally heterogeneous, concentration s of heat producing elements at shallow crustal levels. This view has a number of important geophysical implications for our understanding of the South Australian lithosphere including the framework for tectonic reactivation which has repeatedly affected the Flinders Ranges (and continues today in the currently enhanced seismic activity in this region). Simple thermo-mechanical models of the lithosphere show that the upper mantle thermal structure is extremely sensitive to the depth of burial of such an enriched radioactive layer, with Moho temperatures changing by up to 40°C per km of burial of such a layer. This analysis may provide an insight into the reason why the Flinders Ranges is a region of active deformation of the Australian continent. Not only is it an unduly hot region, but it is also a region where the anomalous heat producing rocks have been buried to ~5 km beneath the sedimentary successions of the Adelaide Geosyncline. We would predict that the mantle temperatures beneath the Flinders Ranges would be ~100°C hotter than adjacent areas bounding the Flinders Ranges with equivalent heat flow (and more than 300°C hotter than the central and western Gawler Craton); a prediction that may well be testable by seismic experiments designed to test velocity structure of the South Australian upper mantle at horizontal scales of less than 100 km.

References cited

Cull, J.P., 1982, An appraisal of Australian heat flow data, BMR Journal of Australian geology & geophysics, 7, 11-21.

Hartley, R., Watts, A.B., & Farihead, J.D., 1996, Isostasy in Africa, Earth and Planetary Science Letters, 137, 1-18.

Houseman, G.A., Cull, J.P, Muir, P.M., Paterson, H.L., 1989, Geothermal signatures of uranium ore deposits on the Stuart Shelf of South Australia. Geophysics, 54, 158-170.

Sass J.H. and Lachenbruch, A.H. 1975, Thermal regime of the Australian Continental Crust, In: ed. McElhinny, M.W., The Earth: its origin, evolution and structure, Academic Press, pp 301.

Zielhuis, A. and van der Hilst, R.D., 1996, Uppper-mantle shear velocity beneath eastern Australia from inversion of waveforms from Skippy portable arrays. Geophysical Journal International, 127, 1-16.