Strainsoftening feedbacks

Once strain starts to localize (Section 8.6.2), several mechanisms may enhance crustal weakening and reduce the amount of work required to continue the deformation. Two of the most influential of these strain-softening mechanisms involve increased pore fluid pressure, which results from crustal thickening, and the vertical advection of heat, which results from concentrated surface erosion and the exhumation of deep crustal rocks. These processes may cause strain to continue to localize as deformation progresses, resulting in a positive feedback. The transpressional plate boundary on the South Island of New Zealand illustrates how these strain-softening feedbacks allow a dipping fault plane to accommodate large amounts of strain.

One of the principal results of the SIGHT program (Section 8.3.3) is an image of a low velocity zone below the surface trace of the Alpine Fault (Fig. 8.2b). In addition to low seismic wave speeds, this zone includes an elongate region of very low (40 ohm-m) resistivity in the middle to lower crust that generally parallels the dip of the Alpine Fault (Fig. 8.22). Magnetotelluric soundings show that the region forms part of a U-shaped pattern of elevated conductivity that rises northwestward toward the trace of the Alpine Fault, attains a near-vertical orientation at ~10 km depth, and approaches the surface about 510 km southeast of the fault trace (Wannamaker et al., 2002). Stern et al. (2001) concluded that the low velocities and resistivities result from the release of fluids during deformation and prograde metamor-phism in thickening continental crust (Koons et al., 1998). In support of this interpretation, areas of hydrothermal veining and gold mineralization of deep crustal origin coincide with the shallow continuation of the conductive zone (Wannamaker et al., 2002). Similar steeply dipping conductive features coincide with active strike-slip faults in other settings, including the San Andreas Fault (Unsworth & Bedro-sian, 2004), the Eastern California Shear Zone, and the southern Walker Lane (Park & Wernicke, 2003). These observations suggest that elevated pore fluid o West Coast §2 km-i m 0 ■

shot 1

East Coast

Ci CD

East Coast shot 1

0 20 40 60 80 100 120 140 160 km = Contours of resistivity

Teleseismic waves from western Pacific

Figure 8.22 Crustal structure below the Alpine Fault (AF) showing region of low P-wave velocities and low resistivity that satisfies wide-angle reflections and teleseismic delays (image provided by T. Stern and modified from Stern et al., 2002). Contours of wave speed shown by solid and dashed lines (km sShading is resistivity ranging from 40 ohm-m for darkest zone to 600 ohm-m for lightest. Zones of strong crustal reflectivity (A, B, C) are from Stern et al. (2001). Dashed lines represent ray path for wide-angle reflections and P-wave delays.

pressures characterize the Alpine Fault and other major strike-slip fault zones.

Laboratory experiments on the mechanics of faulting show that high fluid pressures in the crust result in a reduction in the magnitude of differential stress required to slip on a fault (Section 2.10.2). In New Zealand, this reduction in the crustal strength is implied by an unusually thin (8 km) seismogenic layer, which coincides with the top of the low velocity zone beneath the Alpine Fault (Leitner et al., 2001; Stern et al., 2001). These relationships suggest that high fluid pressures have reduced the amount of work required for deformation on the Alpine Fault, allowing large magnitudes of slip. As the convergent component of deformation on the South Island increased during the late Cenozoic, the magnitude of crustal thickening and fluid release also increased, resulting in a positive feedback that led to a further focusing of strain in the fault zone.

In addition to high fluid pressures, surface uplift and enhanced erosional activity may result in a strain-softening feedback. The removal of surface material due to high rates of erosion unloads the lithosphere and causes the upward advection of heat as deep crustal rocks are exhumed (Koons, 1987; Batt & Braun, 1999; Willett, 1999). If the exhumation is faster than the rate at which the advected heat diffuses into the surrounding region, then the temperature of the shallow crust rises (Beaumont et al., 1996). This thermal disturbance weakens the lithosphere because of the high sensitivity of rock strength to temperature (Section 2.10).

In the case of the Southern Alps, moisture-laden winds coming from the west have concentrated erosion on the western side of the mountains, resulting in rapid (5-10 mm a-1) surface uplift, an asymmetric topographic profile, and the exhumation of deep crustal rocks on the southeast side of the mountain range (Fig. 8.23a). Thermochronologic data and exposures of metamor-phic rock show an increase in the depth of exhumation toward the Alpine Fault from the southeast (Kamp et al., 1992; Tippet & Kamp, 1993). Surface uplift and exhumation have progressively localized near the Alpine Fault since the Early Miocene, resulting in the exposure

2000

Precipitation

' i !\ Exhumation -15 j j \ ^

1 1 y. Elevation

5

is

7000 6000 5000 4000 3000 2000 1000 0

0 1 50 100 150

Alpine fault Distance from coast (km)

7000 6000 5000 4000 3000 2000 1000 0

v ..Surface exposure

0 1 50 100 150

Alpine fault Distance from coast (km)

v ..Surface exposure

r

25 km

Step-up shear

Retro-shear zone

Initiation of uplift Pro-shear zone

30 km

Step-up shear

Wind zones

30 km

Low strength décollement

Sub-continental lithospheric mantle

Continental crust

Low strength décollement

Retro-shear zone

Initiation of uplift Pro-shear zone

30 km

Continental crust

Figure 8.23 (a) Average elevation, precipitation and exhumation along a transect of the central Southern Alps (modified from Willett, 1999, by permission of the American Geophysical Union. Copyright © 1999 American Geophysical Union). Exhumation is from Tippett & Kamp (1993). Numerical model setup (b) and results (c) illustrating the thermal evolution of a 30-km-thick upper crust and the exhumation history (white arrows) of a particle passing through a convergent orogenic system modeled on the Southern Alps (modified from Batt & Braun, 1999, and Batt et al., 2004, by permission of Blackwell Publishing and the American Geophysical Union, respectively. Copyright © 2004 American Geophysical Union). Convergence involved a rate of 10mm a-1 over 10 Ma. Horizontal and vertical scales are equal. Black region marks the peak strain rates and is interpreted to equate with the Alpine Fault for the Southern Alps. Dashed envelope above the model represents the approximate volume of eroded material lost from the system. (d) Strain rates after Batt & Braun (1999) showing pro- and retro-shear zones.

Sub-continental lithospheric mantle

Figure 8.23 (a) Average elevation, precipitation and exhumation along a transect of the central Southern Alps (modified from Willett, 1999, by permission of the American Geophysical Union. Copyright © 1999 American Geophysical Union). Exhumation is from Tippett & Kamp (1993). Numerical model setup (b) and results (c) illustrating the thermal evolution of a 30-km-thick upper crust and the exhumation history (white arrows) of a particle passing through a convergent orogenic system modeled on the Southern Alps (modified from Batt & Braun, 1999, and Batt et al., 2004, by permission of Blackwell Publishing and the American Geophysical Union, respectively. Copyright © 2004 American Geophysical Union). Convergence involved a rate of 10mm a-1 over 10 Ma. Horizontal and vertical scales are equal. Black region marks the peak strain rates and is interpreted to equate with the Alpine Fault for the Southern Alps. Dashed envelope above the model represents the approximate volume of eroded material lost from the system. (d) Strain rates after Batt & Braun (1999) showing pro- and retro-shear zones.

2000

30 km

Wind zones of rocks that once resided at mid-crustal depths (Batt et al., 2004).

To investigate how erosion, exhumation, and heat advection cause these asymmetries and result in the localization of strain on a dipping fault plane, researchers have developed numerical experiments of plate convergence and transpression (Koons, 1987; Beaumont et al., 1996; Batt & Braun, 1999; Willet, 1999). In most of these experiments, crustal deformation is driven by underthrusting the mantle lithosphere of one plate beneath an adjacent, stationary plate (Fig. 8.23b). As mantle lithosphere subducts, the crust accommodates the convergence by deforming. A doubly vergent accre-tionary wedge develops, whose geometry is determined by the internal strength of the crust and mantle, the coefficient of friction on the basal detachment (Dahlen & Barr, 1989), and patterns of erosion at the surface (Willett, 1992; Naylor et al., 2005).

Figure 8.23c,d show the results of an experiment applied to the Southern Alps. In this case, the moving and stationary blocks represent the Pacific and Australian plates, respectively. Initial conditions include a 30-km-thick crust with a feldspar-dominated rheology and a fixed temperature of 500°C at its base (Batt & Braun, 1999; Batt et al., 2004). Over a period of 10 Ma, two ductile shear zones form and define a doubly vergent wedge that becomes progressively more asymmetric through time (Fig. 8.23c,d). A retro-shear zone develops into a major, crustal-scale thrust. A pro-shear zone also forms but does not accumulate significant strain. Surface erosion and crustal exhumation are concentrated between the two shear zones, reaching maxima at the retro-shear zone. The effects of these processes are illustrated in Fig. 8.23c by the white arrows, which show the exhumation trajectory of a selected particle. The dashed envelope above the model represents the approximate volume of eroded material. As heat is advected upwards in response to the exhumation the mechanical behavior of the deforming region changes. The heat decreases the strength of the retro-shear zone, which brings hot material from the base of the crust to the surface, and weakens the fault. This preferential weakening of the retro-shear zone relative to the pro-shear zone increases the localization of strain on the former and enhances the asymmetry of the model (Fig. 8.23d).

The results of this experiment explain how erosion, exhumation, and thermal weakening result in a concentration of strain along a dipping thrust surface in the upper crust during continental collision. The model predictions match many of the patterns observed in the

Southern Alps. Nevertheless, discrepancies also exist. For example, despite the thermal weakening and strain localization caused by exhumation and thermal advec-tion, the retro-shear zone in Fig. 8.23b remains several kilometers thick and does not narrow toward the surface. Batt & Braun (1999) speculated that this lack of fit between the model and observations in New Zealand reflects the absence of strain-induced weakening, high fluid pressures, and other processes that affect strain localization (e.g. Section 7.6.1). Nevertheless, the model explains the prominence of the Alpine Fault as a discrete, dipping surface that accommodates large amounts of slip in the Southern Alps.

To determine whether positive strain-softening feedbacks allow the Alpine Fault to accommodate obliqueslip along a single dipping fault, Koons et al. (2003) developed a three-dimensional numerical description of transpression for two end-member cases. In both cases, a three-layered Pacific plate is dragged along its base toward an elastic block located on the left side of the model (Fig. 8.24a). The elastic block simulates the behavior of the strong, relatively rigid Australian plate; the crustal layers of the Pacific plate accommodate the majority of the strain. A pressure-dependent MohrCoulomb rheology simulates brittle behavior in a strong upper crust. Ductile deformation in a weak lower crust is described using a thermally activated plastic rheology. As in most other models of this type, a zone of basal shear separates the lower crust from Pacific mantle lithosphere. Oblique plate convergence results in velocities of 40 mm a-1 parallel to and 10 mm a-1 normal to a vertical plate boundary. Maintaining the western slope at a constant elevation simulates asymmetric erosion at the surface.

In the first experiment (Figs. 8.24a-f), the Pacific plate exhibits a horizontally layered crust. As deformation proceeds, two well-defined fault zones extend down from the plate boundary through the upper crust, forming a doubly vergent wedge. This wedge includes a vertical fault that accommodates lateral (strike-slip) movement and an east-dipping convergent (thrust) fault along which deep crustal rocks are exhumed (Fig. 8.24f). In the second experiment (Figs. 8.24g-l), the Pacific plate exhibits a thermally perturbed crust in which advection of hot rock has weakened the upper crust and elevated the 350°C isotherm to within the upper 10 km of the crust. In this model, strain is concentrated within the thermally perturbed region. Through the upper crust, the lateral and convergent components of strain occur along the same

T Pacific

0 40 80 120 160 200

160 200

400 km

(i) Lateral strain rates

E it 10

0 40 80 120 160 200

(b) (c) Lateral strain rates

(b) (c) Lateral strain rates

E it 10

160 200

400 km

I 10

ffl 50 Q

I 10

ffl 50 Q

(i) Lateral strain rates y

Distance (km) Distance (km)

(d) (e) Convergent strain rates (j) (k) Convergent strain rates

J 10

Q 50

J 10

Q 50

0 40 80 120 160 200 0 40 Distance (km)

0 40 80 120 160 200 0 40 Distance (km)

0 40 80 120 160 200 0 40 Distance (km)

0 40 80 120 160 200 0 40 Distance (km)

Initial stage: no thermal anomaly

Evolved stage:

Initial stage: no thermal anomaly

Evolved stage:

Figure 8.24 Mechanical model of oblique compression between two plates involving a two-layered crust above mantle lithosphere (images provided by P. Koons and modified from Koons et al., 2003, with permission from the Geological Society of America). (a) Initial model where crust is dragged along its base at velocities of 40 mm a~' parallel to the y-axis (Vy) and 10 mm a~' parallel to the x-axis (Vx). Crustal rheology is horizontal and not yet perturbed by advection. (b) Vertical profile constructed parallel to x-axis showing the component of motion parallel to the y-axis (velocities in mm a~') and (c) plot of lateral strain rates. (d) Vertical profile parallel to the xz plane showing distribution of vertical motion and (e) plot of convergent strain rates. (f) In this model lateral and convergent components are accommodated on two separate structures. (g-k) Results with velocity profiles and strain rate analysis for a model that includes thermal weakening associated with exhumation and concentrated erosion. At this stage both lateral and convergent components of motion are accommodated along a single dipping structure in the upper crust (l) and may separate in the lower crust.

Ul eastward-dipping fault surface. In the lower crust, the two components separate, producing two zones of deformation (Fig. 8.24l). These results illustrate how an evolving thermal structure resulting from asymmetric erosion and exhumation stabilizes the lateral and convergent components of oblique collision along a single dipping fault. They also suggest that a partitioning of deformation onto separate strike-slip and dip-slip faults is favored where thermal weakening is absent.

Boating Secrets Uncovered

Boating Secrets Uncovered

If you're wanting to learn about boating. Then this may be the most important letter you'll ever read! You Are Going To Get An In-Depth Look At One Of The Most Remarkable Boating Guides There Is Available On The Market Today. It doesn't matter if you are just for the first time looking into going boating, this boating guide will get you on the right track to a fun filled experience.

Get My Free Ebook


Post a comment