Paper No. 13
Presentation Time: 4:45 PM
VISCOUS LAYER MODEL FOR A FAULT-PROPAGATION FOLD
Fault propagation is explicitly added to the viscous layer model for deformation above a fault cutting rigid blocks [Patton & Fletcher, 1998; Johnson and Johnson, 2001]. As in the trishear model, discrete domains in which the velocity field is prescribed or determined move through the rock volume as the fault tip propagates. A horizontal plane pinned to the fault tip divides the volume in two such domains. In the lower domain, the velocity field is a prescribed continuous field corresponding to block motion accommodated by shear on a narrow shear zone whose width and dip are specified. The velocity field in the upper domain is that determined in a viscous fluid layer for the prescribed basal velocity and conditions of zero traction at its upper surface. Material rising through the horizontal domain boundary becomes part of the viscous layer; material falling below this surface participates in the prescribed rigid block/shear zone velocity field. Deformation and structural form is obtained by integration over a prescribed program of propagation to slip ratio (P/S). The fault-tip stress distribution may be explicitly evaluated in the viscous layer, and used in an empirical model to interpret P/S in natural or model structures. This hybrid kinematic-mechanical model has the unsatisfactory elements of kinematic and some mechanical models of so-called fault-propagation folds: (i) an ad hoc definition of rigid and deformable regions; (ii) lack of attention to the large-scale deformation which determines fault slip and propagation, of which the region of the fold and fault tip is a small part; and (iii) an assumption of constant rheological behavior, despite large inhomogeneous deformation. With these caveats, the model structures provide useful insight into effects of fault zone dip, P/S, and anisotropy on structural form, the role of the free surface on deformation as the fault propagates, and the interpretation of small-scale brittle features that can be associated with the stress distribution. RCF was supported by NSF OPP-9815160.