Experimental Results

The experimental models exhibited a wide spectrum of ramp/ flat thrust accommodation styles mainly controlled by scale-effects and the rheological stratification, as discussed below.

The scale effect of ramp-flat accommodation
In a first model ( Fig.4a) a single plastilina block with induced ramp was detached from a rigid base to simulate the scale effect and rheological control of ramp-flat thrust accommodation. This was done simply by end-loading models with similar initial set-up but under different g-values i.e. under normal gravity and alternatively in a centrifuge (cf. Figs.4 b & c). In the model end-loaded under normal gravity, because strength was not properly scaled for gravity, the hangingwall block translated forward without bending and unbending to the form of the footwall. As a result void spaces developed during translation above fault bends.

By comparison in the centrifuged model (Fig.4c), the special geometric features of ductile fault bend folding developed from the necessity of making geometric adjustments in the hanging wall to conform to the shape of the underlying flat-ramp-flat footwall. This required a yield stress of the hanging wall blocks in near balance with the gravity stress. In this particular model, because the footwall was deformable a footwall synform developed in response to the load imposed by the advancing hanging wall.

Figure 4. a) Initial set-up of a single plastilina layer detached above a rigid base. b) and c) represent deformed models under normal gravity and in a centrifuge, respectively. (Select image for enlargement)

Ramp-flat accommodation in a competent layer beneath ductile strata
Figures 5 and 6 show various 2-layer models which show how the initial ramps changed shape while slip accumulated along the flat and ramp sector, mainly controlled by the rheological stratification. In a first arrangement (Fig. 5a) the plastilina material, overlain by a soft (Fig. 5b) or stiff (Fig.5c) non-Newtonian viscous material was detached above a rigid base. Slip along the rigid base and along the ramp caused the competent plastilina hangingwall to be driven into the overlying, less competent layer. The soft overburden material thinned above the ramp anticline and thickened in the area surrounding the anticline (Fig. 5b). In this model, because the base was rigid there was a physical continuity between the ramp and the flat gliding horizon, movement was accommdated in the manner of fault bend folding. Moreover, the load imposed by the advancing hangingwall created a footwall syncline.

By comparison, when the competent plastina layer was overlain by stiff ductile strata (Fig. 5c), the overthrust was small compared to the total shortening of the stratigraphic package. The stiff overburden material acted as a more or less rigid lid hindering forwards displacement along the ramp surface. In consequence the initial ramp became modified to a listric shape with deformation. In these two-layer models isostatic adjustment was not aided by a density instability because the overlying materials were less dense than the hangingwall/footwall plastilina blocks.

Figure 5. a) Initial set-up of two-layer models detached above a rigid base, while overlain by soft, respectively stiff ductile strata. b) and c) represent the deformed models. (Select image for enlargement)

Ramp-flat accommodation above ductile substrates
In other rheologically stratified two-layer models (Fig.6a), the same materials as the previous tests (Fig .5) were used, except that the hangingwall/footwall plastilina blocks rested on ductile layers of various competence. In these models, as compared to the previous ones (Fig. 5 b and c) there was no direct physical continuity between the ramp surface and the flat basal decollement. When the hangingwall/footwall plastilina blocks overthrusted a weak viscous substratum (Fig.6b), the soft material was mobile enough to get injected into the ramp during forwards transport of the hangingwall. In addition, forward transport of the hangingwall downflexed the footwall which created a shallower ramp along which movement of the hangingwall could easily be accommodated. In these models adjustment of the plastilina hangingwall/footwall blocks took place in the presence of buoyancy forces. The original 30§ ramp became rotated to near horizontal position during translation of the hangingwall above the footwall. Such a two-layer model where a soft substratum layer is transported from synclines to anticlinal core as the upper competent layer ramps up is supported by geological data. For example, salt and evaporite-cored anticlines beneath developing ramps are well known in the Appalachian plateau (e.g. Wiltschko and Chapple, 1977) and in the Jura (e.g. Suter, 1981; Jordan & Noack, 1992).

By comparison when the competent material was underlain by a stiff ductile layer (Fig 6c), the initial straight ramp became wedge-shaped with deformation. In addition, the overthrusting hangingwall accummulated higer layer parallel shortening strain as compared to the previous model, with no or little penetration of the ductile substratum beneath the ramp. The wedge fault geometry seems to require low stiffness contrast between the competent ramping layer and the surroundings.

Figure 6. as in 5 except that the deformed models are underlain by ductile strata. (Select image for enlargement)

Ramp-flat accommodation in embedded ductile strata
Figures 7a-c illustrate a three-layer stratigraphic arrangement where the plastilia single layer is sandwiched between ductile units of various competence. The greater the contrast in stiffness, the higher the tendency for the sandwiched competent member to develop buckle folds rather than migrate forward in ramp-flat thrust style (Fig. 7b). In other words, the induced ramp had little or no effect in guiding the subsequent development of the overall structure. Rather. the competent layer buckled and the adjacent soft matrix merely responded to the deflecting stiff layer by offering an overall resistance to its deflection (Fig.7 b). Wavelength selection of the stiff member was largely a function of the thickness ratio and viscosity contrast between the stiff member and the softer matrix. The soft matrix thickened beneath growing antiforms and thinned beneath synforms, controlled by the folding instability of the competent plastilina member.

In contrast, the matrix materials was stiff, possessing yield strength. This provided reduced stiffness contrast with the competent embedded member, the initial straight ramp changed shape to a wedge fault, where a footwall synform mirrored the hangingwall antiform (Fig. 7c), in the manner suggested by Ramsay (1992) for the Kimmeridge model. The slip-dependent stretching along the fault surface most likely determined the overall geometry of the wedge fault (Fig.7c).

Figure 7. a) Initial set-up of three-layer models where the competent single layer is sandwiched between soft (b) and stiff (c) ductile strata. (Select image for enlargement)