Construction of a 3D velocity model for microseismic event location on a monitored rock slope

Summary We are planning a tomographic survey to characterize an unstable rock mass that has been monitored with a microseismic sensor network since 2013. An automatic algorithm is used to classify the events filtering noise and interferences. To interpret the events for monitoring the evolution of the fracture system within the rock mass, we are now developing a hypocenter location procedure. The algorithm needs a 3D velocity model. In this work, we discuss the preliminary source tests performed on site to select a portable source suitable for the harsh environment and the size of the monitored area. The small database generated with these tests has been also used to perform a location exercise using a constant velocity. The results are encouraging but it is clear that a 3D velocity model is required to improve the accuracy. Finally, the database has been used to perform a tomographic exercise to extract a preliminary 3D velocity model. The result seems meaningful and shows that despite the difficulties in deploying and moving sources and sensors (a 24-geophone spread will be combined with the microseismic network to increase the ray coverage) in the harsh environment of the rock cliff, the tomographic survey is promising.