Localization of volcanic activity: 2. Effects of pre-existing structure

The capture of vertically rising dikes by faults is examined with a combination of an analytic solution, numerical simulations and field observations. One of the numerical solutions and the field observations also provide insight into the intrusion of sills into the hanging wall of faults that are reactivated when lubricated by magma. The approximate analytic solution indicates that dike capture by a fault will be limited primarily to steeply-dipping faults. Capture will become easier, meaning that it will be possible for faults with shallower dips, as the intersection of the dike with the fault occurs at smaller depths. Capture will also be easier as it is harder for crack to grow in the hanging wall; this may be interpreted in terms of a higher tensile strength or in terms of shorter pre-existing cracks or higher fracture toughness. Numerical analysis using a discrete element code confirms this conclusion but also shows that the analytic solutions overestimate the difficulty of intruding a pre-existing fault, the maximum deviation occurring for faults dipping at 45°. This finding that dike capture by faults should be restricted to high-angle faults and shallow depths is supported by field observations of exhumed Miocene basaltic volcanic terrain in central Nevada, USA. Differences in style seen in other regions are attributed to both the rock properties and the regional stress field. The numerical solutions also indicate that synintrusive fault slip is likely to lead to formation of sills in the hanging wall at depths of a few hundred meters or less: a finding that is also consistent with the field observations in Nevada.