2003 Seattle Annual Meeting (November 2–5, 2003)

Paper No. 3
Presentation Time: 1:30 PM


AYDIN, A., DAVATZES, N. and FLODIN, E., Geological & Environmental Sci, Stanford Univ, Stanford, CA 94305, aydin@stanford.edu

The concept of conjugate faults composed of two intersecting sets with opposite shear sense is central to Mohr/Coulomb-based Anderson models of faults (extant models hereafter). Both geometric and kinematic attributes of faults, together with the notion that the greatest compression bisects the dihedral intersection angle, have been exploited for a wide range of purposes including identification and prediction of fault patterns and inversion of fault data for paleostress. Our new field data and corollary conceptual model for fault evolution in the brittle realm challenge the foundation of traditional concept of conjugate faults. We decipher conjugate fault development by successive shearing of joints and splay joints in the Mesozoic Navajo/Aztec sandstones in two settings: Normal faults in Waterpocket monocline, Utah, and predominantly strike-slip faults in Valley of Fire, Nevada. Faults formed by this process are distinguished from the predictions of the extant models by: [1] The type and mode of component structures (initial mode-I failure then shearing versus shear failure in extant models); [2] the temporal and spatial relationships among component structures (splays and their subsequent shearing derived by stress concentration and stress/material rotation versus faults forming instantaneously in two orientations at arbitrary locations); and [3] continuous interaction and interplay among component structures at various scale (interactive development of hierarchical faults controlling the physical and stochastic attributes of the system such as spacing and length distribution versus faults as passive features which have no consequences for the final attributes of the system).

All of these differences reflect the fact that faults in the physical world have tip lines with associated stress concentration in contrast to faults without stress perturbation in a homogeneous stress field in the extant models. We conclude that interpreting faults and predicting their properties without a physical understanding of their evolution is a risky endeavor and should be avoided.