Interpretation of rock mass yield using apparent stress of microseismic events – examples from Glencore’s Nickel Rim South Mine, Sudbury, Ontario

Seismic events are manifestations of sudden, inelastic deformations in a rock mass. The characteristics of these deformations are reflected in quantitative descriptors of seismic events, known as seismic source parameters. The seismic source parameter apparent stress (σa) is a relative estimate of co-seismic stress change. This parameter has been used by other authors to infer local stress conditions in a rock mass and how they vary in space and time. This thesis investigates short term space-time variations in apparent stress in spatially-isolated volumes of a rock mass following open stope blasts. The results are used to infer how the new excavation affects the nearby rock mass. These excavations are expected to cause local stress increases as stress redistributes around the void, and stress decreases as rock mass near the excavation yields and sheds stress. Data for this research is taken from Glencore’s Nickel Rim South Mine (NRSM) in Sudbury, Ontario. This modern operation has a state-of-the-art microseismic monitoring array that consistently locates events with less than 5 m of residual error and provides accurate source parameter estimates for large and small events. NRSM uses a bulk open stoping mining method, which creates large excavations on the order of several thousand tonnes in a matter of a few seconds. The large stress changes and loss of confinement produced by these blasts can generate large seismic responses which provide an abundance of information about rock mass behaviour. Analysis of apparent stress over time is performed spatially isolating a group of events and tracking apparent stress variation in a moving window. This analysis shows that apparent stress: • Rapidly increases for a matter of minutes immediately following the blast. • Gradually decreases for a matter of hours to a lower level than before the blast. • Slowly increases over a matter of days to weeks. Based on the mechanics of rock masses near excavations and use of apparent stress to infer local stress conditions, these variations are interpreted to correspond with: • Mining-induced stress increasing with the creation of a new excavation. • Stress decreasing as the rock mass near the stope yields and/or enters its post-peak strength. • Stress increasing farther from the stope as a result of the more damaged rock mass closer to the stope shedding stress onto less damaged regions. Variations in event-stope distances are also analysed to re-enforce the inferences made based on the variations in apparent stress over time. The analysis of event-stope distances shows that events: • With high apparent stress tend to locate farther from the stopes than events with low apparent stress. • Initially move away from the stopes for a matter of hours after the blast. • Gradually move back towards the stopes for a matter of hours to days. • Gradually move away from the stopes again for a matter of days to weeks. The inferred rock mass behaviour based on these variations are: • Higher stress conditions in more confined and less damaged regions of the rock mass. • Stress redistributing around the new excavation and creating a new zone of active rock mass yield. • Yield occurring in a more consistent region of the rock mass as redistribution of stress outside the more damaged regions slows. • Growth of damaged zones as rock mass yield progresses over time These observed variations and inferred behaviour are used to propose a conceptual model for rock mass yield which describes the formation of an excavation damage zone following a blast. Further study based on this research may use more detailed aspects of the variations to calibrate constitutive behaviour of rock masses (making it a useful tool for numerical model calibration), or relate variations in apparent stress to the occurrence of large events (making it a useful tool for seismic hazard forecasting).

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.