The development of fault network geometry

Day, Gavin Kenneth (1998) The development of fault network geometry. University of Southampton, Doctoral Thesis.

Abstract

A fault network is a pattern of interacting or linked faults that accommodate triaxial or biaxial strain. The stages leading to the formation of a fault network were investigated from field examples that show displacements that can be accurately determined. The field examples come from localities at Ross Point, Gipsy Point in S. Scotland, and Nash Point and Southerndown, S. Wales. Three hand samples were also analysed from Raeberry, S. Scotland. The fault networks contain both isolated faults, which have their nearest neighbour at some distance from them, and linked faults. Fault networks become connected when they contain throughgoing faults. These were studied using displacement-distance (d-x) plots, cumulative displacement-distance, statistical models, box-counting, first order information dimension and correlation dimension methods.

Co-linear faults showed tip-to-tip fault linkage. Isolated faults followed the relationship of displacement (d_max ) and length (l), where d_max/l^E with E = 1. Maps of two normal fault networks at Ross Point (where faults are only locally connected) and Nash Point (where most faults form a connected fault network) show different behaviour. A log-normal frequency distribution model describes the maximum displacement distribution of the faults in the unconnected fault network at Ross Point and a normal frequency distribution for the spacing distribution. The cumulative displacement-distance (Σd-x) plots describe a homogeneous strain distribution. The faults in the connected fault network at Nash Point show a power-law frequency distribution model for the maximum displacement distribution and a log-normal frequency distribution for the spacing distribution. The Σd-x plots described an inhomogeneous strain distribution. The geometry of these fault networks is interpreted to have caused these differences in behaviour. Isolated strike-slip faults at Southerndown and Gipsy Point had similar d-x profiles to the normal faults.

Fractal analysis of these and other natural fracture patterns showed the cross-over length (r_c) had a relationship with minimum resolved displacement (r_min) where r_c ∝ (r_min)ⁿ_. Fracture density (s) had a relationship with minimum resolved displacement s ∝ (r_min)ⁿ was inferred to affect the fracture density which in turn affects the cross-over length.

This record has no associated files available for download.

Contributors

Download statistics