Paper detail

On comparison of simulated and observed seismicity

Numerical simulation of seismicity has been successfully developed and used for the two last decades. Presently, the general theory of modeling and the progress in computational techniques provide wide options for simulation of seismic and aseismic events with various source mechanisms accounting for blocky structure of rock mass, inclusions, faults, cracks, complicated contact conditions and various mechanical properties of rock. Meanwhile, in practical applications, the input data are limited and uncertain. The data on observed seismicity are also often limited with a few parameters, like coordinates and time. The paper aims to agree the input and output data, used in and provided by numerical simulations, with uncertain and limited data of direct observations. For the input parameters, we suggest their minimal set, which complies with commonly available data. For output seismic parameters, we distinguish three major groups, which are provided by field observations. The first group includes the common (minimal) data on distributions of the event location. These distributions are of special value for improving the input data on geometrical features of a problem. The second group employs the data (commonly available, as well) on the event magnitude. These distributions are of exceptional need for evaluating the risk of strong events. The third group employs data on the event source mechanism. It is based on the tensor of seismic moment/potency, provided by advanced mining seismic systems. This group includes distributions of the geometrical parameters of the event source (orientation of nodal planes, B, P and T directions). It is especially important when establishing and using the connection between stresses and seismicity. The exposition is illustrated by considering an example of long-wall mining in a coal seam.