Processing and interpretation of the Bushveld seismic survey P1-86

Abstract

An investigation into the processing and interpretation of the P1-86 reflection seismic data, acquired by the Geological Survey of South Africa in 1986, is shown in this thesis. The seismic line is geographically located west of Pretoria, from Sandpits to Swartdamstat. A non-migrated stacked section was presented in Du Plessis and Levitt (1987). The presence of structural complexities and high amplitude diffractions/out-of-plane reflections on the original seismic section prompted a decision to process the line using the methodology of interpretive data processing (Hinds et al., 1989). It was postulated that interpretive processing and migration of the data may improve the imaging of some of the structural complexities, noted in the original investigation by Du Plessis and Levitt (1987), and enhance seismic events in regions of low signal-to-noise ratio. At each step in the processing sequence, the effects of the processing on the data was evaluated. The reprocessed brute stack section revealed additional structural detail. Phase-shift migration of the stacked section restored complex reflection events to true spatial positions. Non-collapsed diffraction energy is attributed to out-of-the-plane reflections and noise. It is suggested that these diffractions may result from complex 3-D geological structure. A geological model was postulated using the interpretation derived from the final migrated and non-migrated stacked time sections. It is proposed that the Transvaal Sequence and the upper zone of the Rustenburg Layered Suite are regionally continuous across the survey. The RLS and Transvaal Sequence are identified as northerly dipping events which appear to undergo structural updoming towards the northern part of the survey. The marginal norites are proposed to pinch out, and the main zone is proposed to thin, over the anticlinal structure. The postulated updoming is possibly related to similar structures in the Brits area (Roberts, 1981), a few kilometers to the west of the survey. The anticlinal structure resulting from the postulated updoming may have caused a tensional regime in the rocks of the Transvaal Sequence and RLS. This may have resulted in complex faulting. Signal interference in this region of faulting, caused by conflicting reflectors (Ohlovich, 1964), is proposed to be one of the causes of the low signal-to-noise ratio in the northern region of the survey. A model of the subsurface imaged by the seismic survey was fitted to the regional gravity data using two-dimensional forward modeling (Talwani et al., 1959). The good fit of the calculated gravity anomaly values to the observed gravity anomaly values supports the feasibility of the seismically-derived geological model.