Distribution Characteristics of Mining-Induced Seismicity Revealed by 3-D Ray-Tracing Relocation and the FCM Clustering Method

Abstract

Induced seismicity in underground mining zones reflects inner changes of the rock structure caused by human activities, thus providing potential information for hazard assessment. Precise hypocenter parameters and magnitudes of the induced micro-earthquakes as well as their spatial distributions are essential for understanding the seismicity. We first accurately relocate 834 micro-earthquakes recorded by a 28-sensor micro-seismic monitoring system from an underground phosphate mine, using an iterative pseudo-bending ray-tracing method and a 3-D velocity structure obtained by seismic tomography. We then determine the local magnitudes of the events by a well-recalibrated formula, and finally analyze the seismicity for different parts of the mine divided by the fuzzy c-means clustering method according to the spatial distribution of the micro-earthquakes. The bimodal distribution of the recalibrated local magnitudes indicates two fundamentally different processes of rock failure may exist in the mine. They are possibly fracture-dominated rupture and friction-dominated slip, i.e., tensile failure and shear failure of the surrounding rock. The clustering result shows that mining practices cause very different seismic features in different parts of the mine. Furthermore, we find that most of the large events are located in low-V zones, and that events occurred in the lower velocity zones tend to have a correspondingly larger magnitude, as revealed by the seismicity analysis and seismic tomography.