Heat flow in South Africa

Abstract It has been known for many years that the increase of temperature with depth on the Witwatersrand is exceptionally slow. The normal gradient in Europe and America is about 33° C/km., whilst that in the gold mines near Johannesburg is only about 10° C/km. It has been generally supposed that the conductivities did not differ by any large factor,* and that the low-temperature gradient indicated a heat flow much less than that in other parts of the world. As there is reason to suppose (Jeffreys 1929) that most of the heat is generated by the radioactivity of a layer of granite underlying the continents, and since the principle of isostasy requires this layer to be thicker under the African plateau than elsewhere, a greater heat flow would be expected in Africa than in Europe. The conductivity data are meagre and of doubtful reliability, and it seemed desirable to make a systematic study of the question. The temperatures and conductivities in a number of English bores have been studied by Benfield in this laboratory, and detailed and accurate temperature measurements have been made by Dr Krige (1939) and Mr Weiss (1938) in deep boreholes in South Africa. When the author was invited to visit Johannesburg as a guest of the Bernard Price Institute of Geophysical Research he took the opportunity to measure the conductivities of specimens of rock from some of the boreholes in which temperature measurements had been made. It is the purpose of this paper to describe these measurements and to discuss the results. As the conditions were exceptionally favourable, both for the temperature and for the conductivity measurements, the problem has been investigated in considerable detail. It was thought that, in addition to the direct interest of the results, they may be useful in indicating the disturbances to be feared in cases where no such detailed study is possible.