Integration of Lidar-Based Structural Input and Discrete Fracture Network Generation for Underground Applications

In this study the authors present an approach of establishing and validating discrete fracture networks (DFNs) for underground projects using LiDAR (Light Detection and Ranging) as the source data. With the use of LiDAR in geotechnical and geological engineering becoming increasingly popular, it is necessary to establish the interactive application of this technology with other tools. Such a tool is the generation of DFNs and their integration into the geomechanical design, with a specific focus on underground projects such as tunnels, caverns, repositories etc. This paper attempts to show an approach in which LiDAR data from the Brockville Tunnel, located in Ontario, Canada, is used as the source for the determination of input parameters of DFN modelling based on manual and automatic mapping techniques. Having determined a representative set of input parameters, a deterministic DFN model is created in order to calibrate other modelling parameters associated with the generation process, leading to the creation of multiple DFN models. By employing the representative elementary volume (REV) concept, these models are used in order to examine the effect of the different joint sets on the estimated REV, and to introduce an approach of determining the required number of DFN realizations and the size of the DFN models.