Influence of Pre-existing Weakness on Normal Fault Growth: Implications From Discrete Element Modeling

Abstract

Fault evolution in multiphase rift basin is complicated, and not well understood because the first-phase structures are normally deeply buried. The effect of pre-existing weak structures, such as faults, on normal fault growth is investigated by discrete element modeling, focusing on fault geometry, fault interactions and fault network evolution in three dimensions. We use a three-layer discrete element model, comprising a lower ductile layer and two brittle layers, with a weak planar pre-existing structure at 60° to extension direction in the lower brittle layer. The evolution of normal fault network is divided into three stages: (i) reactivation of pre-existing structure and nucleation of new faults; (ii) propagation and interaction; and (iii) linkage of pre-existing structure and new faults. During the first stage (0-10% extension), most (80% area and 100% length) of the pre-existing structure is reactivated. New faults also nucleate at various locations in the upper brittle layer, forming extension-perpendicular conjugate fault sets. During the second stage (10%-20% extension), a few ‘teeth-like’ extensions or fringes grow upward from the upper tip of the reactivated structure, into the previously un-deformed upper brittle layer. In addition, the reactivated structure starts to become the dominant fault, influencing the strike of new faults nucleating near it. During the third stage (20%-25% extension), the ‘teeth-like’ extensions or fringes hard-link with extension-perpendicular faults in the upper brittle layer, creating a complex, twisted fault geometry. Following linkage, the displacement maximum on the twisted fault migrates from the extension-perpendicular segments to the branch line. Four styles of fault interaction (plus no interaction) eventually develop between the reactivated structure and surrounding new faults: i) isolated, ii) abutting, iii) twisting, and iv) conjugate. Our study demonstrates that a reactivated structure can propagate into the cover in an irregular (fringe-like) pattern, and influences the orientation and geometry of new faults in its proximity. In a rift setting, this would imply a greater variation in fault density, orientation and geometry compared to normal fault growth without pre-existing structural weaknesses. The study has implications for fault growth models, and for the geometry and compartmentalization of tilted fault block traps in rift basins.

AAPG Datapages/Search and Discovery Article #90291 ©2017 AAPG Annual Convention and Exhibition, Houston, Texas, April 2-5, 2017