Halokinesis and thermal convection

Abstract

THEORETICAL, numerical and experimental studies^1–3 have successfully modelled realistic analogues of natural salt domes which have risen from deep source layers through formerly superincumbent layers of rock, by a process known as halokinesis. In such studies the deformation has been approximated to the gravitational stabilisation of Rayleigh-Taylor instabilities in which dense viscous cover layers originally overlay a less dense effectively viscous salt layer. Various examples of salt deformation have been reported in the past decade which seem to have involved salt too shallow for the overlying sediments to have either compacted to densities greater than that of crystalline salt⁴ or to load the (actually plastic) salt above its assumed yield point⁵. Thus in part of the North Sea the Zechstein salt seems to have started doming in Muschelkalk times when buried by only 610–760 m of (possibly soft) argillaceous sediments⁶. Similarly, in part of the Mississippi basin, salt seems to have moved beneath an overburden thickness of 300–600 m (ref. 4). Furthermore, small salt domes seem to be actively developing beneath only 150–300 m of valley-fill alluvium in the 2,000–2,600 m thick tabular Luke salt body in Arizona^7,8. Such cases of halokinesis, being premature in terms of the likelihood of unstable compositional density gradients, suggest that one or more factors previously neglected may be involved in at least some salt deformations⁴. Here the degree of mechanical instability due to geothermal gradients within thick layers of salt and in rock sequences with salt at the bottom is examined. I find that heat may have a more significant effect on halokinesis than that of merely reducing the strength of salt⁵, and that cyclic thermal convection may occur within some salt bodies and that some salt domes may develop as thermal plumes.