We construct a series of distinct-element models that consist of bonded assemblies of elastic particles that simulate the brittle deformation associated with large-scale, fault-related folding over a rigid footwall. The initial rock mass is simulated by a series of discrete, circular, elastic, frictional particles, bonded in shear and tension and capable of progressive fracture during loading. This produces discontinuities within the simulated rock mass: Regions of intact material are separated by discrete faults or fault zones. The simulations are fully dynamic. Material properties are assigned as microproperties of and between particles; elastic stiffness, friction, and bond strength (shear and tensional) describe particle interactions. Because the macroscopic elastic, failure, and flow properties are not directly determined by the microproperties, uniaxial and biaxial compression tests must be conducted on a representative specimen to determine macroscopic parameters, such as unconfined compressive strength, Young's modulus, friction angle, and cohesion.

To determine the effect of growth strata on the evolution of the underlying fold, growth strata with thicknesses equivalent to 0%, 25%, 50%, 75%, and 100% of the elevation of the associated anticline above the upper footwall flat are added to the upper surface of the models. The presence or absence of growth strata has a significant effect on the final geometry of the associated fault-related fold. The growth strata provide resistance to forelandward translation, causing folds to tighten from open, gentle ramp anticlines with no or thin growth strata to strongly overturned fault-propagation-type folds with thick growth sequences. The nature and intensity of fracturing is strongly affected by growth-strata thickness. All models have effectively identical early stages, as a result of thin or no growth, and initial deformation at the thrust ramp is always by back thrusting. This changes with increased burial. (1) With no growth strata, back thrusting continues, although fault dips become shallower with time. (2) With intermediate thicknesses of growth strata, back thrusting gives way to, or alternates with, bedding-parallel shear. (3) With thick growth strata and thus high confining pressure, brittle deformation is inhibited and replaced by elastic deformation of the particles. These models produce realistic-looking packages of growth strata. Well-developed growth triangles are common in backlimbs, and are cut by the faults that originate in the pregrowth strata. The growth strata commonly show angular unconformities and stratigraphic pinch-outs. Overturned forelimb growth strata are significantly deformed, recording brittle deformation associated with progressive growth of the underlying fold.