The control of stress history and flaw distribution on the evolution of polygonal fracture networks

Abstract

We use boundary element models of fracture propagation and linkage to investigate the factors controlling the development of two-dimensional, multi-directional–polygonal fracture networks, characterised by a large number of abutting intersections between fractures. The position and orientation of a number of fracture seeds are prescribed in the model, which propagate when the applied stress reaches a critical value, according to linear elastic fracture mechanics theory. The applied boundary condition is a remote, isotropic, horizontal tension, where the stress is increased at a steady rate throughout each model to simulate continued fracture growth. Realistic polygonal systems are developed with the boundary element model simulations, which are comparable with those observed in natural systems (such as those found within Eocene and Oligocene mudrocks in the North Sea and on the surface of Mars). If conditions exist where a small number of fracture seeds propagate and develop significant structures before others, then these will dominate the resulting fracture network geometry. Not only do such early structures represent the largest fractures in the system, they also significantly modify the stress field around them preventing some other seeds from developing, and influencing the propagation paths of nearby fractures. Fracture seeding distribution and the rate at which the stress is increased are found to be the most significant parameters affecting the development of fracture network geometry. These results suggest that the geometry of evolving fracture networks should be considered not only in terms of the mechanical properties of the deforming material, but also in terms of the stress rate driving deformation.