Upper mantle structure of shear-waves velocities and stratification of anisotropy in the Afar Hotspot region

Abstract

The Afar area is one of the biggest continental hotspots active since about 30 Ma. It may be the surface expression of a mantle “plume” related to the African Superswell. Central Africa is also characterized by extensive intraplate volcanism. Around the same time (30 Ma), volcanic activity re-started in several regions of the African plate and hotspots such as Darfur, Tibesti, Hoggar and Mount Cameroon, characterized by a significant though modest volcanic production. The interactions of mantle upwelling with asthenosphere, lithosphere and crust remain unclear and seismic anisotropy might help in investigating these complex interactions. We used data from the global seismological permanent FDSN networks (GEOSCOPE, IRIS, MedNet, GEO- FON, etc.), from the temporary PASSCAL experiments in Tanzania and Saudi Arabia and a French deployment of 5 portable broadband stations surrounding the Afar Hotspot. A classical two-step tomographic inversion from surface waves performed in the Horn of Africa with selected Rayleigh wave and Love wave seismograms leads to a 3D-model of both S V velocities and azimuthal anisotropy, as well as radial S H / S V anisotropy, with a lateral resolution of 500 km. The region is characterized by low shear-wave velocities beneath the Afar Hotspot, the Red Sea, the Gulf of Aden and East of the Tanzania Craton to 400 km depth. High velocities are present in the Eastern Arabia and the Tanzania Craton. The results of this study enable us to rule out a possible feeding of the Central Africa hotspots from the “Afar plume” above 150–200 km. The azimuthal anisotropy displays a complex pattern near the Afar Hotspot. Radial anisotropy, although poorly resolved laterally, exhibits S H slower than S V waves down to about 150 km depth, and a reverse pattern below. Both azimuthal and radial anisotropies show a stratification of anisotropy at depth, corresponding to different physical processes. These results suggest that the Afar hotspot has a different and deeper origin than the other African hotspots (Darfur, Tibesti, Hoggar). These latter hotspots can be traced down to 200 km from S-wave velocity but have no visible effect on radial and azimuthal anisotropy.