BEYOND KTB - ELECTRICAL-CONDUCTIVITY OF THE DEEP CONTINENTAL-CRUST

Great strides have been made in understanding the upper part of the cr ust by in-situ logging in, and laboratory experiments on core recovere d from super-deep bore-holes such as the KTB. These boreholes do not e xtend into the lower crust, and can contribute little to the elucidati on of mechanisms that produce the high electrical conductivities that are commonly observed therein by magneto-telluric (MT) methods. Labora tory studies at simulated lower crustal conditions of temperature, pre ssure and saturation, on electrolyte saturated rocks thought to have b een derived from the lower crust, have not been possible up until now due to their experimental difficulty. It is necessary to subject elect rolyte-saturated rock samples to independently controlled confining an d pore-fluid pressure, which implies that the rock be sleeved in some impermeable but deformable material, that can withstand the very high temperatures required. Metals are the only materials capable of being used, but these cause great difficulties for cell sealing and conducti vity measurement. In this paper we describe recent breakthroughs in ex perimental work, specifically the development of two new types of soph isticated metal/ceramic seal, and a conductivity measurement technique that enables the measurement of saturated rock conductivity in the pr esence of a highly conducting metallic sleeve. The advances in experim ental technique have enabled us to obtain data on the electrical condu ctivity of brine saturated basic, acidic and graphite-bearing rocks at lower crustal temperatures and raised pressures. These data have faci litated the comparison of MT derived crustal electrical conductivity p rofiles with profiles obtained from laboratory experiments for the fir st time. Initial modelling shows a good agreement between laboratory d erived and MT derived profiles only if the mid-crust is composed of am phibolite pervaded by aqueous fluids, and the lower crust is composed of granulite that is saturated with aqueous fluids and/or contains int erconnected grain surface films of graphite. The experimental data are consistent with a three layer crust consisting of an aqueous fluid sa turated acidic uppermost layer, above an aqueous fluid saturated amphi bolite mid-crust, and a granulite lowermost crust, which may or may no t be saturated with aqueous fluids, but if not, requires the presence of an additional conduction mechanism such as conduction through thin graphite films.