Schreurs, G., Hänni, R. and Vock, P.  2002. The influence of brittle-viscous multilayers on faulting during rifting: an analogue modelling approach. Schellart, W. P. and Passchier, C. 2002. Analogue modelling of large-scale tectonic processes. Journal of the Virtual Explorer, 7, 87-94.
The influence of brittle-viscous multilayers on faulting during rifting:
an analogue modelling approach

Experimental results

Single stage rifting - Experiment 77

The evolution of the fault pattern in the single stage rifting experiment as seen in surface view (Fig. 3 & 4) is closely related to the distribution of brittle and viscous layers at depth. In domain with a basal viscous layer overlain by brittle layers, the spacing between conjugate normal faults was larger than in the domain that had a second, upper viscous layer embedded in brittle layers (see experimental set-up; Fig. 1). As normal faults propagated along strike they either linked or overlapped with other sub-parallel oriented faults. In some cases lateral propagation of different normal fault segments induced a change in fault orientation in the area where they met. With progressive extension, new faults appeared within previously formed grabens. Extensional transfer zones mark the transition between widely spaced conjugate faults and narrowly spaced conjugate faults and closely mimic the position of the underlying interbedded weak layer. Where this boundary is initially parallel to the extensional direction, the alignment of transfer zones is also more or less parallel to it; where this border is initially at an angle of 30° to the extension direction, the transfer zones also forms at an angle of 30°. Note that normal faults in the domain above the interbedded weak layer did not form instantaneously across the entire long dimension of this domain. As a result some transfer zones also formed in this domain.

Figure 3. Structures in experiment 077 after 9% extension as seen in surface view, vertical sections and in cut-out view. Black dashed lines in surface view indicate major transfer zones located above the interbedded viscous layer. Dashed red rectangles indicates areas analysed by X-ray tomography.

Figure 4. Movie showing surface evolution of experiment 077. Note the difference in graben width between domain with a second, interbedded viscous layer (see Fig. 1 for its location) and adjacent domains with only a basal viscous layer. Initial grid spacing on surface was 4 cm. First frame is after 5 mm of extension. Subsequent frames are at 5 mm extension increments.(Select image to view movie)

Fig. 5 shows the evolution of structures in four cross-sections through part of the model. The fault evolution was very different in domains with and without an interbedded viscous layer. In those parts of the model with an interbedded weak layer progressive extension resulted in decoupled conjugate normal fault systems in upper and lower brittle strata. The horizontal distance between graben-bounding normal faults was small and the offset along individual faults was limited. In contrast, in the domain without a second interbedded viscous layer, the grabens were fewer, deeper and wider, and faults extended all the way down to the top of the basal viscous layer. Initial graben width reflects the depth to the viscous layer. A larger vertical distance to the viscous layer results in a wider graben. Fault offset along existing faults increased with progressive extension. In large grabens new steep normal faults formed that are antithetic to the bounding faults and merged with them at depth. Although normal faults at early stages of extension were straight in cross-section, they became listric with progressive deformation and fault-bounded blocks rotated about a horizontal axis. In the domain with an interbedded viscous layer, conjugate normal fault systems in the lower brittle compartment formed at different locations than in the upper brittle compartment. The formation of conjugate normal faults in the lower compartment caused a downbending of the overlying brittle compartment and lateral flow of the interbedded viscous layer. At the final stages of extension, parts of the upper brittle compartment were in contact with the lower brittle compartment.

Figure 5. Successive vertical CT images through the brittle-viscous model (domain 077a; see Fig. 3): two sections cross the domain with two viscous layers (upper two images), and two the domain with only a basal viscous layer overlain by brittle layers (lower two images). Corundum is dark grey; sand is medium-grey, and viscous PDMS is light-grey. Initial width of model was 22.5 cm and initial height 3.5 cm. Extension increment between frames is 5 mm. (Select image to view movie)

Fig. 6 shows the 3D evolution of part of the model with time. Progressive extension manifested itself differently in domains with and without an interbedded viscous layer. In the latter domain grabens are fewer, wider and deeper. Transfer zones form in the transition zone between the two domains. Fig. 7 illustrates the drastic lateral changes in the structural style across the transition zone by 80 serial vertical cross-sections at 9% extension. Oblique transfer zones that strike parallel to the boundary of the interbedded viscous layer (Fig. 8) can be followed at depth in Figs. 8 and Fig. 9. The latter figure consists of a series of horizontal sections at 9% extension. Fig. 10 shows the structures in experiment 077 at 18% extension, while Fig. 11 and 12 show serial horizontal sections through transfer zones at the same stage of deformation. Note how some of the faults in the lower brittle compartment – below the interbedded viscous layer - die out laterally towards the domain with only one basal viscous layer.

Figure 6. Movie of 3D evolution with time of experiment 077 (domain 077a; see Fig. 3) showing the development of a transfer zone parallel to the extension direction. Each frame shows a cut-out 3D view that consists of 80 serial cross sections each representing a 2 mm thick CT slice. Extension increment between frames is 1 cm. Viscous silicone (PDMS) is red, quartz sand is orange and corundum powder is a dark yellow to orange. (Select image to view movie)

Figure 7. Movie showing the geometry of part of experiment 077 by 80 serial vertical sections after 9% extension. Note the drastic change in structural style as soon as the interbedded viscous layer appears.(Select image to view movie)

Figure 8. Horizontal section through 3D cut-out perspective view showing oblique transfer zone after 9% extension (domain 077a; cf. Fig. 3).

Figure 9. Successive horizontal sections through 3-D perspective view after 9% extension (domain 077b). The location of the oblique extensional transfer zone is linked to the oblique boundary of the interbedded viscous layer. (Select image to view movie)

Figure 10. Structures in experiment 77 after 18% extension illustrated by surface photograph, vertical sections and two cut-out 3-D perspective views. Transfer zones are related to the shape and position of the interbedded viscous layer.

Figure 11. Successive horizontal sections through 3-D perspective view after 18% extension (domain 077b; see Fig. 10). (Select image to view movie)

Figure 12. Successive horizontal sections through 3-D perspective view after 18% extension (domain 077a; see Fig. 10). (Select image to view movie)

 

previous | index | next