S-wave receiver function studies in African sedimentary basins

Date
2017
Authors
Inguane, Helio Filemone
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Sedimentary basins are the result of prolonged subsidence of the Earth’s surface. They occupy 45% of the African surface. Knowledge of their area and depth is important because they often contain mineral, energy and groundwater resources. The transition between crust and mantle (the Moho) is believed to hold important clues to the Earth’s evolution and has been the subject of many studies, including P-wave Receiver Function (PRF) studies, to determine the structure and composition of the crust and uppermost mantle in Southern and Eastern Africa. The PRF method relies on the partial conversion of P-waves produced by teleseismic earthquakes to S-wave waves at the Moho. The travel-time delay between the direct P-wave and the Ps phase is used to deduce the thickness and average velocity of the crust. However, the PRF technique fails in regions where there is strong intracrustal layering (such as sedimentary basins), because the reverberations produced by the layers arrive at the seismometer simultaneously with the Ps phase. Here the S-wave Receiver Function (SRF) method works better, as the Sp phase arrives before any reverberations produced by intracrustal layering. In this study I have used the SRF method to investigate crustal structure beneath sedimentary basins in Southern and Eastern Africa. The aim of this research was to constrain the crustal thickness and shear wave velocity in seven sedimentary basins in Eastern and Southern Africa using S-wave Receiver Functions (SRFs). Teleseismic earthquakes with magnitude ≥5.5 and 60 to 82 degree epicentral distance were used to generate the SRFs using data acquired by seismic stations that were deployed between 2007 and 2013 in three rift basins (Lake Albert, Lake Edward and Rukwa) and four pull–apart basins (Mandawa, Mozambique, Rovuma and Ruvu). A moveout correction was made to align the SRFs obtained from different earthquakes, enabling them to be stacked to reduce random noise and enhance the signal-to-noise ratio of the SP phase and the accuracy of the pick of the SP arrival time. The SP arrival time uncertainties, typically 0.05 s of time error, were estimated for each station using the bootstrapping method. The surface wave group velocity models for each station (at 10, 15, 20, 25 and 30s periods) were used to constrain the depth–velocity models. The grid search modeling was performed using the DISPER80 package. The following crustal thicknesses (H) and average crustal shear velocities (Vs) were obtained:  Lake Albert and Lake Edward rift basins situated within the Mesoproterozoic Ruwenzori orogenic belt: H of 38.8 ± 2.4 km and 33.83 ± 0.9 km, respectively; Vs of 3.72 km/s and 3.73 km/s, respectively;  Rovuma, Mandawa and Ruvu pull–apart basins within the Neoproterozoic Mozambique mobile belt: H of 32.73 ± 1.8 km, 37.79 ± 2.2 km and 39.63 ± 2.2 km, respectively; Vs of 3.68 km/s, 3.76 km/s and 3.79 km/s, respectively; and  Phanerozoic Mozambique pull-apart basin: H of 36.9 ± 2.1 km and Vs of 3.7 km/s. These results were compared with previous studies. The crustal thickness reported in a global review of Proterozoic terrains (Durrheim and Mooney, 1994) ranged between 40-55 km, while Rudnick and Fountain (1995) reported an average thickness of 43 km. For stations located in basins in the Mesoproterozoic Ruwenzori orogenic belt, this study produced estimates of H and Vs of 36.3 ± 2.4 km and 3.7 km/s, respectively. Vs is similar to estimates by Julià et al. (2005) and Tugume et al. (2013) for stations in the same region, while H is a few kilometers thinner. This study obtained H of 32-40 km beneath Neoproterozoic pull-apart Tanzanian coastal basins located within the Mozambique mobile belt (Chatellier and Slevin, 1988), while Tugume et al. (2013) estimated the crust of the adjacent Tanzanian craton to be 39 km thick. In the Phanerozoic Mozambique basin, this study found H and Vs of 36.9 ± 2.1 km and 3.7 km/s, respectively. Kgaswane et al. (2009), using joint inversion of receiver functions and Rayleigh wave dispersion, estimated the H and Vs for the northeast Limpopo belt (west of the Mozambique basin) to be 40 ± 3 km and 3.7 km/s, respectively. In general, this study found thinner crust and slower Vs than previous studies. However, it is important to note that the station locations were different. Previous studies analysed P-wave Receiver Functions (PRFs) recorded by stations located in the interior of the continent and near to the Tanzania craton, while this study analysed S-wave Receiver Functions (SRFs) recorded by stations located in rift and coastal sedimentary basins. It is likely that the crust thinned during extension and continental break-up.
Description
A dissertation submitted in fulfilment of requirements for the degree of Master of Science to the Faculty of Science, University of the Witwatersrand, Johannesburg, March 2017.
Keywords
Citation
Collections