Seismic Monitoring is a technique that uses vibration monitors in the detection and analysis of dynamic stress waves, generated from failure of rock mass. This technique was applied at the Selibe-Phikwe Cu-Ni mine (BCL) and surrounding areas to study the seismic activity. Shaking of the ground destabilizes both manmade and natural structures like, buildings, pillars and faults, through weakening and causing cracks on buildings, crushing pillars and awakening dormant faults. The main aim of the research project was to provide information about the seismic activity taking place within and around the Selibe-Phikwe Cu-Ni mining area with specific objectives of determining the locations of the seismic events, determining the seismic source mechanisms and other source parameters and providing a seismic hazard assessment. Data for this project was acquired from the Botswana Geological Institute (BGI). BGI had set up a temporary seismic monitoring system using seismic equipment from the Institute of Mine Seismology (IMS). These included; surface mount sensors that record vibrations, an enclosure with seismic monitoring equipment that digitizes and processes data as well as a Wi-Fi radio for communication to the server. Data processing was accomplished using IMS TraceTM and IMS VantageTM software. IMS TraceTM was used for calculating source parameters, picking of phase arrivals and calculation of the location of the source energy (earthquake) while IMS VantageTM is an analysis and visualization toolkit that was used to understand and explore seismicity both temporally and spatially within the mine. The data analyzed was for a period of 33 months (from 27th June 2019 to 11th March 2022). A total of 266,931 events were processed; these were vii located in the south-eastern extension of the mine with majority of the events having occurred in 2019. Local and moment magnitude of all the events ranges from -2.7 to 3.6. Most large events occur below the 980m level. There are more small events (<0) than large events (>0). A correlation between the number of events per month and pumping activities was established. Events of magnitude >1.5 were used for moment tensor decomposition. The moment tensor decomposition of the amplitude and polarities together with the full waveform for the same event yielded a more diviatoric component with a much higher double couple in both cases. The source type plots showed a neither implosive nor explosive component and axes orientations of the two, were not very well constrained. An instability analysis was carried out that showed an increase in Cumulative Apparent Volume which indicates an increase in fracturing of parts of the rock mass within the south-eastern part of the mine, and as they become unstable and there might occur large scale rock mass deformation. Log Energy Index was greater than zero which indicates accumulation of stress, this can be regarded as a strain softening stage and it should be a warning indicator for potential damage. The frequency magnitude analysis showed a much gentler cumulation which indicates a higher proportion of large events in the study area; this therefore indicates a significantly noticeable degree of seismic hazard.

Thesis (MSc in Geological Sciences)---Botswana International University of Science and Technology, 2024