Electrical and seismic anisotropy of the lithosphere with the focus on central southern Africa

Abstract

The aim of this study was to gain a better geological understanding of the southern African region through the use of magnetotelluric (MT) and seismic techniques. Specifically, the newly collected southern African magnetotelluric experiment (SAMTEX) data are analysed for directionality using a tensor decomposition technique. Instead of conducting the analysis for given periods, as is commonly done, the data are analysed for approximate depths due to the variable electromagnetic penetration across the region. I also re-analyse previously collected southern African seismic experiment (SASE) data for shear wave splitting of teleseismic events using standard processing techniques. These analyses provide information on the electrical and seismic anisotropy properties of the region, which may then be related to tectonics and geological structure. It is found that MT conductive direction results, for both crustal and lithospheric mantle depths, are significantly more complex than has previously been observed in other regions. The complexity is attributed to be due to strong effects of large-scale conductivity heterogeneities on the conductive directions measured. The reanalysis of some of the SASE stations for shear wave splitting has produced near-identical results to those previously measured, and I was not able to conclusively demonstrate the presence or absence of 2-layer anisotropy. A previously unnoticed relationship is observed between thick lithosphere, and regions of well correlated seismic fast axis directions and plate motion directions. Combined with the observations of vertical variations in conductive directions of the MT results, this has led to a new model being proposed to explain the anisotropy results observed in the region. The model suggests both lithospheric and asthenospheric contributions to seismic anisotropy, with a significantly stronger anisotropic layer below thicker lithosphere, which is proposed to be due to stronger lattice preferred orientation (LPO) of olivine as a result of increased flow velocity below the thicker lithospheric keel. This model is supported by other geoscientific observations, including the results from the lithospheric and asthenospheric MT analysis. No strong correlation between the measured MT and seismic anisotropy parameters is observed, likely because the electrical anisotropy is strongly effected by structure, and the seismic anisotropy is predominantly a result of LPO of olivine (in places, quite strongly vertically varying).