Three-dimensional statistical modelling of a fractured rock mass—an example from the Fanay-Augères mine

摘要: Abstract This paper describes a new technique for stochastic modelling of 3-D fracture networks. We use geostatistical simulation methods to reproduce features the spatial structure rock, such as variation density and orientation in space. For an example method, we mapped data from Fanay-Augeres mine, Limousin, France. Two different sections drift wall, S1 S2, were mapped. The section is wet, S2 dry. each case, fractures are divided into five sets set modelled separately. represented discs placed randomly diameter disc chosen independently fixed probability distribution determined trace length distribution. A point process, called parent-daughter used locating discs. process gives clumping or swarming not found usual Poisson model. characterized fluctuation about mean set. has that simulated with geostatistics. Geostatistical simulations two systems made 68 m cube which contains over 5 million fractures. connectivities assessed see if there any correlation hydrologic observations. connectivity appears be larger than S2. smaller overall but big These result more consistent field observation wet reason this difference apparently related presence fault rock surrounding S1, thus indicating need understand role faults zones characterizing hydrology fractured rock.