Seismological Investigations of the Continental Upper Mantle

A key element in our understanding of continental deformation and dynamics is our knowledge of the detailed structure and properties of the continental upper mantle. Very little of the Earth's mantle can be directly sampled, and geophysical techniques provide the most comprehensive approach to understanding the structure of the lithosphere across the continents. Beneath the continents the large differences in seismic wavespeed observed between old cratonic regions and younger terranes can be predominately explained by variations in temperature. However, sharp lateral variations in seismic velocity, as seen in some regional and local experiments, cannot be explained by differences in thermal structure. Recent results from surface wave tomography suggest that as well as the rapid lateral changes in velocity, the vertical variations in shear wavespeed are not compatible with simple compositional models and cratonic geotherms. Below a number of old continental regions seismic velocities beneath the Moho are slower than expected and there appears to be a positive velocity gradient in the uppermost mantle. In order to investigate this problem it is necessary to develop methods to make a combined interpretation using different techniques. The anomalous structure is highlighted in surface wave tomography, however these studies have a limited vertical resolution insufficient to image any very rapid transition in physical structure. The limitation can be addressed by combining surface wave tomography with receiver functions. The receiver function makes use of seismic waves which are converted at velocity contrasts and will therefore highlight discontinuities in physical properties. In regions with closely spaced seismic stations it should be possible to investigate not only the vertical variations, but also any lateral discontinuities in physical properties. The proposal will investigate the structure and properties of the continental upper mantle through a joint interpretation of surface wave dispersion and receiver functions beneath permanent seismic stations in Africa and Australia. Comparison of velocity models beneath different tectonic settings is important to enable a better understanding of the physical evolution of the continental lithosphere. Are the slow velocity and positive gradient observed in all settings, or a particular feature of a tectonic process? In southern Africa the methods can also be tested on data from a temporary seismic array. Array data should yield more detailed information on continental structure, the better resolution allowing the characterisation of lateral changes in structure. However, it is also important to assess whether a 1-2 year temporary deployment of seismometers can provide as reliable a model as those obtained on many years of permanent station data. Within southern Africa an additional geophysical technique will be incorporated into the interpretation. In contrast to seismological techniques, methods using magnetotelluric data are predominately a function of the connectivity of minor phase properties in a rock (e.g., low order partial melt, fluids, carbon in graphite form) and thus give complementary information on the structure of the lithosphere. The results from magnetotelluric studies can therefore be combined with the seismological models and further enhance our knowledge of the continental lithosphere. The aim of this proposal is to develop methods that will allow a quantitative interpretation of seismological models in terms of the physical properties of the uppermost mantle. Is it possible to accurately observe the depth range of anomalous shear wavespeeds, and can they be associated with a particular mineralogy? The overall results of the study will greatly improve our knowledge of the present state of the continental lithosphere.