Experiments

We show the results of two experiments with identical boundary conditions, except for the shape of the basal layer of viscous PDMS (Fig. 1). In experiment 73 the model had two lateral PDMS-sand boundaries parallel to, and one frontal boundary perpendicular to, the shortening direction (Fig. 1a). In experiment 75, one side of the PDMS layer was oblique to the shortening direction (at an angle of 35°), while the other side was parallel to it (Fig. 1b). Purely for descriptive purposes we refer to the sand-corundum-sand layers as the brittle domain, and to the sand-corundum-sand layers with a basal viscous PDMS layer as the brittle-viscous domain. In both experiments the structural evolution in the brittle domain differed markedly from the one in the brittle-viscous domain. X-ray CT volumetric data were acquired from the area indicated in Fig. 1.

Figure 2. Movie of cross-sectional evolution of experiment 73 illustrating the difference in structural style between brittle and brittle-viscous domain. Each section represents a 2 mm thick X-ray computerized tomography (CT) slice that was acquired parallel to the shortening direction. The upper two sections in each frame cross the brittle domain, the lower two sections the brittle-viscous domain. Initial width of the model was 27 cm; initial height was 3 cm. Shortening increment between frames is 1 cm. PDMS = dark brown, quartz sand = light brown, corundum powder = yellow.(Select image to view animation)

Experiment 73 – Evolution of structures in cross-sections through the brittle and brittle-viscous domain
Progressive shortening was initially accommodated by thrust faults with opposite vergence (Fig. 2). The downward converging thrust faults defined a pop-up structure that rooted near the base of the lowermost sand layer in the brittle domain and at the top of the viscous layer in the brittle-viscous domain. The backthrusts dip steeper than the forward thrusts. As deformation increased, forward thrusts were predominantly active. Backthrusts showed little movement and were passively transported over the forward thrust. Fault-bend folds formed as the thrust sheet moved over the ramp of the forward thrust. In the brittle-viscous domain, viscous material moved upward along the ramp.

In the brittle domain, progressive deformation caused the basal detachment to be activated in front of the pop-up structure and a new thrust imbricate formed. Activity along the older forward thrust ceased and fault movement began to occur along the newly developed forward thrust in its footwall. With increasing deformation, another in-sequence imbricate thrust formed.

In the brittle-viscous domain, however, progressive deformation resulted in the development of a new frontal thrust far away from the mobile wall at the forward boundary between basal PDMS and sand. Displacement along the forward thrust of the pop-up structure near the mobile wall ceased. Because of continuing movement of the mobile wall, the forward thrust of the pop-up became progressively steeper and the triangular block bounded by forward thrust, backthrust and mobile wall underwent a rotation about a horizontal axis. This boundary effect caused bulging and extension in the upper part of the fault-bend fold and small normal faults formed in its outer arc. As progressive shortening increased, backthrusts formed at the lower bend of the active frontal ramp in the brittle-viscous domain. In this domain, an out-of-sequence thrust appeared in the region between the existing forward thrusts. Movement along the out-of-sequence thrust took place at the same time as displacement along the foremost forward thrust. Displacement along this out-of-sequence thrust was also coeval with movement along backthrusts of the frontal pop-up structure and resulted in a complex interference pattern.

Figure 3. Line drawings after plan view photographs of experiment 73 illustrating progressive development of surface structures at 2, 4, 6 and 8 cm of shortening. Transfer zones connecting frontal ramps develop parallel to the shortening direction. (Click for Enlargement)

Experiment 73 in surface view and the development of a transfer zone
In surface view, thrust faults developed first in the brittle domain and propagated along strike into the brittle-viscous domain. Downward-converging thrust faults bound a pop-up structure that strikes parallel to the mobile wall (Fig. 3). Since forward thrusts in the brittle-viscous domain rooted at the top of the viscous layer and dipped slightly steeper than in the brittle domain, the thrust front surfaced slightly closer to the mobile wall in the brittle-viscous domain (Fig. 3a). With progressive shortening, a forward thrust originated at the frontal boundary between brittle and viscous material (Fig. 3b). Deformation then took place in two completely different parts of the model: close to the mobile wall in the brittle domain, and near the frontal termination of the basal PDMS in the brittle-viscous domain. (Fig. 3c,d).

Figure 4. Movie of 3D evolution with time of experiment 73 illustrating the formation of a transfer zone. Each frame shows a perspective three-dimensional view that consists of 88 serial cross-sections each representing a 2 mm thick CT slice. Brittle domain closest to viewer. Initial width and height of the model was 27 cm and 3 cm, respectively. Shortening increment between frames is 1 cm. (Select image to view animation)

Figure 5. Movie of 3D evolution with time of experiment 73 illustrating the formation of a transfer zone. Each frame shows a perspective three-dimensional view that consists of 88 serial cross-sections each representing a 2 mm thick CT slice. Brittle-viscous domain closest to viewer. Shortening increment between frames is 1 cm. (Select image to view animation)