Every year, a few earthquakes startle the residents of California. Some tremors do little more than rattle a few dishes, but occasionally the jolting of a great earthquake tumbles buildings, ruptures roads, and triggers landslides. Most earthquakes in California signify that a sudden increment of slip has occurred somewhere along the San Andreas Fault zone. This strike-slip (or strike-slip) fault zone, which contains countless individual faults, accommodates northward motion of the Pacific Plate relative to the North American Plate (Figure 14.1). Most North American residents have heard of the San Andreas Fault—but it’s not the only major strike-slip fault zone on this planet! Strike-slip faults cut both continental and oceanic crust in many places. Examples in continental crust include the Queen Charlotte Fault of western Canada, the Alpine Fault of New Zealand, the faults bordering the Dead Sea, the Anatolian Faults of northern Turkey, the Chaman Fault in Pakistan, and the Red River and Altyn Tach Faults of China. Strike-slip faults in oceanic crust most commonly occur along mid-ocean ridges, where they trend perpendicular to the ridge axis and link segments of the ridge. But there are important examples that link nonaligned segments of trenches as well.

FIGURE 14.1. (a) Regional map of the plate boundary between the North American and Pacific Plates. The San Andreas Fault is the strike slip fault zone that defines this boundary in California. Inset showing the major strike slip faults. J & F = Juan de Fuca Plate. Preserving many key characteristics, we refer to features of the San Andreas Fault throughout this chapter. (b) Terrain image of the San Andreas Fault. [Google Earth] [13.1]

As we noted earlier, a strike-slip fault, in the strict sense, is a fault on which all displacement occurs in a direction parallel to the Earth’s surface; that is, slip lineations on a strike-slip fault are horizontal. Given this constraint, slip parallels the strike of the fault surface, so these faults are commonly called strike-slip faults, and sometimes lateral slip faults. Strictly speaking, strike-slip displacement does not produce uplift or subsidence. In the real world, strike-slip movement is commonly accompanied by a component of shortening or extension. Specifically, transpression occurs where there is a combination of strike-slip movement and shortening, and can produce uplift along the fault. Transtension occurs where there is a combination of strike-slip movement and extension, and can produce subsidence along the fault.

In this chapter, we discuss the nature of deformation within strike-slip fault zones (both oceanic and continental), and we review the tectonic settings in which they develop. You will find that a wide variety of complex subsidiary structures develop in strike-slip fault zones that we capture by the term wrench tectonics. We begin the chapter by explaining the kinematic distinction between transfer faults and transcurrent faults, the two basic classes of strike-slip faults. Full terminology of strike-slip faults is summarized in Supporting Material.

CLASSES OF STRIKE-SLIP FAULTS

When examining the role that strike-slip faults play in crustal deformation, we find it convenient to distinguish between two kinematic classes of faults: transfer faults and transcurrent faults. A fault in one class differs from a fault in the other class in terms of the geometry of its endpoints (the locations along strike where the fault terminates), the way that slip magnitude varies along the fault’s length, and the way that the fault’s geometry evolves through time. The key distinguishing elements are:

Strike-slip fault - A fault on which displacement is (mostly) parallel to fault strike, in present-day surface coordinates. The term is purely geometric, and has no genetic, tectonic, or size connotation. We distinguish two kinematic classes:

Transcurrent fault characteristics are:

it dies out along its length;
the displacement across is less than length of the fault;
the length of the fault increases with time and continued movement;
displacement is greatest at center of fault trace and decreases toward its ends.

Transfer fault characteristics are:

once formed, displacement across can be constant along length of the fault;
displacement across can be much greater than length of the active fault;
its length can be constant, increase, or decrease with time.
it starts/terminates at another fault (e.g., plate boundary)

To make the distinctions more concrete, we’ll look at the specific characteristics in more detail below.

Transfer Faults

Transfer faults became prominent in the literature as part of the plate tectonics revolution when Tuzo Wilson coined the term transform faults in the early 1960s, in reference to the third category of plate boundary (distinct from convergent and divergent plate boundaries). The kinematic properties of transform faults are not restricted to plate boundary settings, however, so we use the more general term transfer faults for this kinematic class, and use the term transform fault for those associated with plate boundaries.

FIGURE 14.2. Plate boundary transfer faults, or transform faults. (a) Map sketch of a transfer fault linking two convergent plate boundaries. The fault trace occurs only between X and Y. (b) Map sketch showing a transfer fault linking two ridge segments. The length of a transform fault can change with time. (c) Transform length stays constant if spreading rates on ridge segments at both endpoints are the same. (d) Transform length decreases if the spreading rate at one endpoint is less than the subduction rate at the other. (e) The amount of displacement remains the same along the length of a transform if the length of the transform stays constant or decreases. Displacement at X is the same as the displacement at Y. (f) If the transform length changes with time, then the amount of slip varies along the length. At time 1, the fault is fairly short. At time 2, the length of the fault is longer. Displacement at X, in the middle of the fault, is greater than displacement at point Y, near an endpoint. [19.3, 19.4, 19.5]

While some strike-slip faults are, indeed, plate boundaries, the term transfer fault is used to describe any strike-slip fault that has the following characteristics:

• The active portion of a transfer fault terminates at discrete endpoints. At the endpoints, the transfer intersects other structures (Figure 14.2). For example, the transfer can terminate at a shortening structure (e.g., convergent boundary, thrust fault, or stylolite) or at an extensional structure (e.g., divergent boundary, normal fault, or vein).

• The length of a transfer fault can be constant, or it can increase or decrease over time. For example, in Figure 14.2c, the spreading rate on ridge segment A is the same as the spreading rate on ridge segment B. As a consequence, the length of the transform fault connecting these ridge segments remains constant over time. In contrast, Figure 14.2d shows a transfer fault where one end terminates at a ridge segment, but the other end terminates at a trench. If the rate of subduction at the trench exceeds the rate of spreading on the ridge, then the length of the transfer fault connecting them decreases over time. In Figure 14.2f, the length of the transform fault between two triple junctions labeled (T1 and T2) increases with time if the triple junctions move apart.

• In cases where the fault length is constant or decreasing, the amount of displacement along the length of a transfer fault is constant. For example, the displacement at point X on the fault in Figure 14.2e is the same as the displacement at point Y. If the length of a transfer fault increases over time, however, the amounts of displacement on the younger portions of the fault are less than the amounts on the older portions (Figure 14.2f).

• Displacement across a transfer fault can be greater than the length of the fault itself. For example, consider a 10-km long transform fault that links two ridge segments. If more than 10 km of spreading occurs on the ridge segments, then there will be more than 10 km of slip on the transfer fault.

FIGURE 14.3. (a) Sketch map showing a transfer fault linking two rift segments. (b) Later, the rift segments have evolved into new mid-ocean ridges, and the transfer fault has evolved into an oceanic transform fault. Note that this fault still has the same length as when it originated. [19.7]

Transfer faults occur in a variety of settings and at a variety of scales (from mesoscopic to regional). At a mesoscopic scale, transfer faults can link non-coplanar stylolites or non-coplanar veins. In Chapter X we saw how transfer faults link segments of rifts, oriented at a high angle to the rift segments and terminating at normal faults. Where rifting is successful, so that a new plate boundary forms, transfer faults in the rift link ocean spreading segments of a divergent plate boundary (Figure 14.3). In this setting we use the name transform fault, as originally proposed by J. Tuzo Wilson.[1] Transforms can occur in oceanic and continental crust, so we typically add this modifier. For example, the San Andreas Fault (California), the North Anatolian Fault (Turkey) and the Alpine Fault (New Zealand) are examples of major continental transforms.

Let’s see how transform faults display the kinematic characteristics we described above by looking at an example, a 10-km long transform fault linking two segments of the Mid-Atlantic Ridge (Figure 14.4). Each end of the transform fault terminates abruptly at a ridge segment (i.e., at a zone of extension) trending at a high angle to the transform. If the spreading rate on each ridge segment is the same, then the length of the transform stays constant over time, regardless of the amount of slip that has occurred along it. Further, the same amount of slip occurs everywhere along the length of the fault. (This makes sense if you remember that the transform fault initiated at the same time as the ridge segments—it was “born” with the length that it now has. Note also that the sense of slip on the faults must be compatible with the spreading directions on the ridge segments.) The amount of slip along an oceanic transform fault can be much greater than the length of the fault. For example, if there has been 1000 km of spreading on the ridge segments, there must be 1000 km of displacement on the 10-km long transform.

FIGURE 14.4. Evolution of an oceanic transfer (or transform) fault (TF; red). The amount of slip on a transform can exceed the length of the transform. (a) At time 1, the transform is 10 km long. Its length does not change over time, even though spreading at the ridges occurs continuously. (b) At time 2, the amount of displacement on the fault is already 1.5 × the length of the fault. (c) At time 3, the displacement on the fault is 3 × the length of the fault. The continuation of the transform fault is, by definition, inactive, but marked by a topographic lineament on the ocean floor, called the fracture zone (FZ; blue). [19.6]

Transcurrent Faults

The second kinematic class of strike-slip fault, transcurrent faults differ from transfer faults in a number of ways:

• Transcurrent faults die out along their length. This means that at the endpoint of a transcurrent fault, the fault does not terminate abruptly at another fault, but either splays into an array of smaller faults (sometimes called a horsetail), or simply disappears into a zone of plastic strain. Typically, fault splays comprising a horsetail are curved. Depending on the direction of curvature with respect to the sense of displacement, thrust or normal components of displacement occur in faults of the horsetail, and these movements will be accompanied either by folding and uplift where there is a thrust component, or by tilting and subsidence where there is a normal component. (Figure 14.5).

• Transcurrent faults initiate at a point and grow along their length as displacement increases (Figure 14.6a). As a consequence, short faults have a small amount of displacement, while long faults have a large amount of displacement. Thus, fault displacement (as measured in map view) is proportional to fault length.

FIGURE 14.5. Terminations along transcurrent faults. (a) Here, the fault terminates in a horsetail composed of a fan of normal faults. (b) Here, the fault terminates in a horsetail composed of an imbricate fan of thrust faults. [19.8]

• Displacement across a transcurrent fault is greatest near the center of its trace and decreases to zero at the endpoints of the fault (Figure 14.6b).

• The displacement on a transcurrent fault is always less than the length of the fault (Figure 14.6b).

FIGURE 14.6. (a) Growth of a transcurrent fault. As time passes, the fault lengthens, and the displacement on the fault increases. In the process, some horsetail splays are abandoned. (b) Map sketch illustrating how the displacement varies along a transcurrent fault (horsetails are not shown). The heavier line is the fault, and the thin lines are marker lines. At time 1, the fault is short and has offset only marker 4. At time 2, the fault has grown, and now offsets markers 3, 4, and 5. Note that the markers just beyond the tips of the fault are starting to bend, prior to rupturing. Note also that the displacement on the fault is less than the length of the fault. [19.9]

Transcurrent faults typically develop in continental crust as a means of accommodating development of regional strain. For example, slip on a conjugate system of mesoscopic transcurrent faults results in shortening of a block of crust in the direction parallel to the line bisecting the acute angle between the faults. As is the case with transform faults, transcurrent faults can form at any scale, from mesoscopic to regional.

STRIKE-SLIP SYSTEMS

Major strike-slip faults, meaning ones that have trace lengths ranging from tens to thousands of kilometers, are not simple planar surfaces. Typically, they have locally curved traces, divide into anastomosing (braided) splays, include several parallel branches, and/or occur in association with subsidiary faults and folds. In this section, we focus on the structural complexities of large continental strike-slip faults and explore why these complexities occur.

Distributed Deformation in Strike-Slip Zones

Regional-scale strike-slip deformation in continental crust does not produce a single, simple fault plane. Rather, such shear produces a broad zone containing numerous individual strike-slip faults of varying lengths, as well as other structures such as normal and reverse faults, and folds. Let’s first look at the array of individual strike-slip faults that occurs in a large continental strike-slip fault zone. As an example, consider the San Andreas Fault zone of California. The San Andreas Fault “proper” is one of about 10 major strike-slip faults, and literally thousands of minor strike-slip faults, that slice up western California. In some localities, faults have sinuous traces, so neighboring faults merge and bifurcate to define, overall, an anastomosing array. Elsewhere, the zone includes several subparallel faults. Locally, one fault dies out where another, parallel, but non-coplanar one initiates. The region between the endpoints of two parallel but non-coplanar faults is called a stepover (Figure 14.7). Localized faulting occurs in the stepover region to accommodate the transfer of slip.

FIGURE 14.7. A stepover along a strike-slip fault. (a) Sketch map showing how regional dextral shear can be distributed along fault segments that are not coplanar. Slip is relayed from one segment to another at a stepover. (b) At a restraining stepover, compression and thrusting occur. (c) At a releasing stepover, extension and subsidence occur. [19.11]

Continental strike-slip fault zones also may contain en echelon arrays of thrust faults, folds, and normal faults, as well as subsidiary strike-slip faults (Figure 14.8). Typically, the thrust faults and folds trend at an angle of 45° or less to the main fault, and the acute angle between subsidiary thrust faults and the main fault (or between the fold hinges and the main fault) opens in the direction of shear. The normal faults also trend at an angle of 45° or more with respect to the main fault, but the acute angle between subsidiary normal faults and the main fault opens opposite to the direction of shear. Most subsidiary strike-slip faults trend at a shallow angle to the main fault.

FIGURE 14.8. Arrays of subsidiary structures associated with right-strike (dextral) shear. (a) Subsidiary strike-slip faults (Riedel shears: R and R′); (b) en echelon folds, and en echelon thrusts; (c) en echelon folds which formed and then were later offset by shear on a strike-slip fault; (d) en echelon normal faults and veins. [19.12]

Transpression and Transtension

The trace of the San Andreas Fault in southern California is not just a featureless line on the surface of the Earth. At some localities, the trace lies within a train of marshy depressions (called sag ponds); elsewhere, the trace is marked by 50-m high ridges (called pressure ridges) that contain tight folds (Figure 14.9).

$C:\Users\vdpluijm\AppData\Local\Microsoft\Windows\INetCache\Content.Word\19_17c.jpg$

(c)

FIGURE 14.9. (a) Air photo showing the trace of the San Andreas Fault, north of San Francisco (Tomales Bay). Note that the faulting has locally caused a water-filled depression to form. (b) Photograph of pressure ridges along the San Andreas Fault, San Luis Obispo County, California. (c) A cross section of a pressure ridge in a road cut across the San Andreas Fault near Palmdale, California. [GE: W 34°33'40.77" N 118°07'57.63" W] [19.17]

The presence of such topographic and structural features tells us that motion along the fault is not perfectly strike-slip. Ridges form in response to transpression, a combination of strike-slip displacement and compression that yields a component of shortening across the fault. This shortening causes thrusting and uplift within or adjacent to the fault zone. Notably, where the fault zone contains a broad band of weak fault rocks (gouge and breccia), transpression squeezes the fault rocks up into a fault-parallel ridge, much like a layer of sand squeezes up between two wood blocks that are pushed together (Figure 14.10a).

FIGURE 14.10. Wood-and-sand model illustrating the concept of transpression and transtension. (a) When wood blocks shear and squeeze together (transpression), intervening sand is pushed up; (b) When blocks shear and pull apart (transtension), sand sags. [19.18]

Topographic depressions reflect transtension, a combination of strike slip and extension. The extensional component causes normal faulting and subsidence (Figure 14.10b). Transpression or transtension can occur along the entire length of a fault zone, if the zone trends oblique to the vectors describing the relative movement of blocks juxtaposed by the fault. Such a situation develops where global patterns of plate motion change subsequent to the formation of the fault, for a fault is a material plane in the Earth and cannot change attitude relative to adjacent rock once it has formed.

The dimensions of transpressive or transtensile structures forming along a strike-slip fault depend on the amount of cross-fault shortening or extension, respectively. Where relatively little transpressive or transtensile deformation has occurred, cross-fault displacement results in relatively small pressure ridges or sags, with relief that is less than a couple of hundred meters. If, however, transpression or transtension has taken place over millions of years, significant mountain ranges develop adjacent to the fault. For example, transpression along the Alpine Fault of New Zealand has resulted in uplift of the Southern Alps, a range of mountains that reaches an elevation of > 3.5 km above sea level Figure 14.11.

FIGURE 14.11. The Alpine Fault in New

Zealand is a major strike-slip fault with significant topography. (a) This right-strike continental transform links the Macquarie Trench (M) with the Tonga-Kermadec Trench (TK). (b) Corresponding satellite view of New Zealand’s South island. [Google] [19.2]

Alpine Fault landsat

Seismic-reflection studies of large continental strike-slip faults indicate that subsidiary faults in transpressional or transtensional zones within strike-slip systems are concave downwards, and merge at depth into the main vertical fault plane (Figure 14.12). Thus, in cross section, large continental strike-slip faults resemble flowers in profile, with the petals splaying outwards from the top of a stalk. This configuration of faults is, therefore, referred to as a flower structure. In transpressive zones, a positive flower structure develops, in which the slip on subsidiary faults has a thrust-sense component (Figure 14.12a); whereas in transtensile zones, a negative flower structure develops, in which the slip on subsidiary faults has a normal-sense component (Figure 14.12b).

FIGURE 14.12. (a) Schematic block diagram of a positive flower structure. (b) Block diagram of a negative flower structure. [19.19]

Restraining and Releasing Bends

In many locations, transpression and transtension take place at distinct bends in the trace of a strike-slip fault. A fault bend, in the context of strike-slip faults, is a portion of a fault where the strike of the fault changes. Elsewhere, we also used fault bend when discussing dip-slip faults, which is a location where a fault’s dip changes. To see what happens, picture a fault that strikes east-west, parallel to the trend of the regional displacement (Figure 14.13a and b).


	FIGURE 14.13. Map-view models of fault bends along strike-slip faults. The “edges” of the crustal blocks are provided for reference. (a) Releasing bend at which normal faults and a pull-apart basin have formed. (b) Restraining bend at which thrust faults have formed. (c) Application of this model to the San Andreas Fault north of Los Angeles (LA). The dashed lines outline imaginary reference blocks. The San Andreas Fault bends along the margin of the Mojave Desert. [19.20]

Along segments of the fault that strike exactly east-west, motion on the fault can be accommodated by strike-slip motion alone, for the fault plane parallels the regional displacement vectors. But at fault bends, where the strike of the fault deviates from east-west, the fault plane does not parallel regional displacement vectors, and there must be either transpression or transtension, depending on the orientation of the bend. Let’s look at the geometry of these bends in more detail.

A bend at which transpression takes place is called a restraining bend (Figure 14.13b), because the fault segment in the bend inhibits motion. In other words, at restraining bends, pieces of crust on opposite side of the fault push together, causing crustal shortening. Along regional scale restraining bends, a fold-thrust belt can form, causing the uplift of a transverse mountain range (i.e., a mountain range trending at an angle to the regional trace of the fault). For example, the Transverse Ranges just north of Los Angeles in southern California developed because of shortening across a large restraining bend along the San Andreas Fault (Figure 14.13c).

A bend at which transtension occurs is called a releasing bend because, at such bends, opposing walls of the fault pull away from each other (Figure 14.13a). As a consequence of this motion, normal faults develop, and the block of crust adjacent to the bend subsides. Displacement at a releasing bend yields a negative flower structure or, in cases where the bend is large, and large amounts of extension have taken place, a pull-apart basin. A pull-apart basin is a rhomboid-shaped depression, formed along a releasing bend and filled with sediment eroded from its margin. The dimension and the amount of subsidence in a pull-apart basin depends on the size of the bend and on the amount of extension. Notably, formation of small pull-apart basins involves brittle faulting only in the upper crust, but formation of large pull-apart basins involves thinning of the lithospheric mantle, so that after extension ceases, the floor of the basin thermally subsides. Examples of present-day pull-apart basins include the Dead Sea at the border between Israel and Jordan, and Death Valley in eastern California. In both of these basins, the land surface lies below sea-level, creating an environment in which summer temperatures become deadly hot.

Both restraining bends and releasing bends can exist simultaneously at different locations along the same fault. As a consequence, a region of crust moving along the fault may at one time be subjected to transtension and then, at a later time, be subjected to transpression. When this happens, a negative flower structure, or pull-apart basin formed at a releasing bend, becomes inverted and changes into a positive flower structure, and normal faults bordering a pull-apart basin become reverse faults. Inversion causes sediment that had been deposited in negative flower structures or pull-apart basins to thrust up and over the margins of a strike-slip fault zone.

Strike-Slip Duplexes

A strike-slip duplex consists of an array of several faults that parallel a bend in a strike-slip fault (Figure 14.14). The map-view geometry of a strike-slip duplex resembles the cross-sectional geometries of a thrust-fault duplex or a normal-fault duplex that we explore in other chapters. Strike-slip duplexes that formed at restraining bends are also called transpressive duplexes, whereas those formed at releasing bends are called transtensile duplexes.

FIGURE 14.14. Map-view sketch of strike-slip duplexes formed along a dextral strike-slip fault. [19.21]

Understanding the Complexity of Strike-Slip Zones

Why do continental strike-slip fault zones contain so many subsidiary faults? To understand this complexity, we’ll review an experiment that we introduced earlier. Take a homogeneous clay slab and place it over two adjacent wooden blocks (Figure 14.15). The clay cake represents the weaker uppermost crust, and the wooden blocks represent stronger crust at depth. Now, push one of the blocks horizontally so that it shears past its neighbor. As the blocks move relative to one another, the clay cake begins to deform, partly by plastic mechanisms and partly by brittle failure. Initially, this brittle deformation yields arrays of small strike-slip faults called Riedel shears. Two sets of Riedel shears, labeled R and R', develop. We can picture these shears as forming a conjugate system relative to the far-field maximum compressive stress driving development of the overall fault zone. Eventually, P shears develop, which finally link with Riedel shears to form a throughgoing strike-slip fault. With this model, we can speculate that some of the subsidiary faults in a strike-slip zone initiated as Riedel shears or as P shears.

FIGURE 14.15. Laboratory model of strike-slip fault development. (a) Before deformation, a clay cake rests on two wooden blocks that are pressed together. The clay represents the brittle uppermost crust. The vertical boundary between the two blocks represents a dextral strike-slip fault. (b) As deformation begins, Riedel shears develop in the clay cake. (c) A map view of the top surface also shows a later stage of deformation, in which Riedel shears have been linked by P fractures, and a throughgoing fault has developed. [19.14]

The scenario above suggests that, even if the upper crust were homogeneous, one might expect strike-slip zones to contain subsidiary strike-slip faults. In reality, the upper crust is heterogeneous. Crust may contain a variety of different rock types with different strengths, and the contacts between rock units may occur in a variety of orientations. In addition, the crust contains preexisting planar weaknesses such as joints, old faults, and foliations. All these heterogeneities cause stress concentrations and local changes in stress trajectories. As a result, faults may locally bend, and they may split to form two strands on either side of a stronger block. In sum, the process by which the fault is formed, as well as the crust’s heterogeneity, ensures that strike-slip fault zones include a variety of subsidiary fault splays.

FIGURE 14.16. Subsidiary structures along a strike-slip fault. (a) A map view of dextral simple shear. A square becomes a parallelogram, and a circle in the square becomes an ellipse. (b) R and R′ are conjugate shear fractures formed in rock cylinder subjected to an axial stress, at predicted angles. (c) The strain ellipse showing that folds and thrusts form perpendicular to the shortening direction, while normal faults and veins form perpendicular to the extension direction. R and R′ shears form at an acute angle to the shortening direction. (d) You can simulate wrench kinematics and the structures of strike-slip deformation with a sheet of paper. [19.15]

To picture why en echelon arrays of thrusts, folds, and normal faults have the orientations that they do, picture a block of crust that is undergoing simple shear in map view. An imaginary square superimposed on the zone transforms into a rhomb, and an imaginary circle transforms into a strain ellipse (Figure 14.16a). In the direction parallel to the short axis of the ellipse, the crust shortens, so thrusts and folds develop perpendicular to this axis. In the direction parallel to the long axis of the ellipse, the crust stretches, so normal faults and veins develop perpendicular to this axis. As deformation progresses, the main strike-slip fault eventually slices the block in two, and the two halves of the subsidiary faults and folds are displaced with respect to one another. R and R′ Riedel shears can be understood as conjugate shear fractures whose acute bisector is parallel to σ1, the regional maximum principle stress (Figure 14.16b and c), with angle of ~30o as predicted by Coulomb failure theory. You can also simulate deformation in a strike-slip fault zone by shearing a piece of paper between your hands (Figure 14.16d), with the ridges that rise in the center of the paper representing folds, before the paper tears.

SETTINGS OF CONTINENTAL STRIKE-SLIP ZONES

Until now, we have focused our discussion on characteristic structural features that occur in strike-slip fault zones. Now, we broaden our perspective and turn our attention to the tectonics of strike-slip faulting. Specifically, we describe the various plate settings at which strike-slip fault zones develop.

Continental Transfer Faults

The traces of reverse faults trending nearly perpendicular to the regional transport direction dominate the map pattern of fold-thrust belts. Locally, however, these belts also contain strike-slip faults whose traces trend nearly parallel to the regional transport direction. Some of these transfer faults, known as lateral ramps, cut a thrust sheet into pieces that move relative to one another. In places where lateral ramps have a near-vertical dip, they are also called tear faults. Lateral ramps or tear faults can accommodate along-strike changes in the position of a frontal ramp with respect to the foreland. Examples of lateral ramps bound the two ends of the Pine Mountain thrust sheet in the southern Appalachians (Figure 14.17).

(a)

(b)

FIGURE 14.17. Transfer faults in thrust systems. (a) Map of the Pine Mountain thrust system in the southern

Appalachians (eastern USA), showing lateral ramps (Jacksboro Fault and Russell

Fork Fault), the term more commonly used for transfer faults in this setting. (b) Block diagram indicating the concept of a lateral ramp or tear fault (box). [19.26]

Analogous to contractional settings in fold-thrust belts, extensional systems include transfer faults. Rifts typically consist of a chain of distinct segments, with adjacent segments differing in amount of extension. Also, the axis of one segment may not align with the axis of the adjacent segment in the chain. Each segment in a rift is linked to its neighbor by an accommodation zone that contains strike-slip faults. Where transtension or transpression occurs, flower structures may develop along transfer faults.

FIGURE 14.18. Block model of the Garlock Fault, southern California (USA), accommodating the offset from the extended Basin and Range Province to the north and the Mojave Desert to the south. [19.28]

The Garlock Fault in southern California is one of the largest examples of a strike-slip within a rift environment (Figure 14.18). This fault, a ∼250-km long left-strike slip fault, forms the northern border of the Mojave Desert and intersects the San Andreas Fault at its western end. It defines the boundary between two portions of the Basin and Range Rift: the “Central Basin and Range” lying to the north, and the Mojave block lying to the south. The Garlock Fault exists because these two portions have not developed the same amount of extensional strain. Specifically, a greater amount of extension has occurred in the Central Basin and Range than in the Mojave block. To accommodate this difference, the crust to the north of the fault has slid westward, relative to the crust to the south. Notably, this motion pushed the portion of the San Andreas Fault north of its intersection with the Garlock Fault to the west, creating a major restraining bend in the San Andreas Fault zone. The presence of this bend is responsible for development of the Transverse Ranges just north of Los Angeles. Motion between the Pacific and North American Plates across the Transverse Ranges bend in the San Andreas Fault also generates compressive stress within the Mojave block. This stress has caused some of the normal faults formed by rifting in the block in the past to be reactivated today as strike-slip faults.

Continental Transcurrent Faults

Strike-slip faults form along convergent plate boundaries where the vector describing the relative motion between the subducting and overriding plates is not perpendicular to the trend of the convergent margin (Figure 14.19a). At such oblique-convergent margins, the relative motion between the two plates can be partitioned into a component of dip-slip motion (thrusting) perpendicular to the margin, and a component of horizontal shear (strike-slip faulting) parallel to the margin. Present-day examples illustrate that the strike-slip faults of oblique-convergent plate boundaries develop in a variety of locations across the margin, including the accretionary wedge, the volcanic arc, and the back-arc region. Partitioning of relative movement into dip-slip and strike-slip components accompanies the oblique collision of two buoyant lithospheric masses (Figure 14.19b). The strike-slip component of motion displaces fragments of crust laterally along the orogen.

FIGURE 14.19. (a) Strike-slip faulting at an oblique convergent margin. Note that the faults occur at various locations across the width of the margin. The large arrow indicates the relative movement of the downgoing plate. (b) Strike-slip faulting at an oblique collisional margin. Note that in this portrayal, the faults terminate at depth at a mid-crustal detachment horizon. [19.24]

When the colliding mass is a small exotic terrane (see Chapter 14), the terrane may be sliced up by strike-slip faults after it docks (Figure 14.20). For example, Wrangelia, an exotic crustal block that was incorporated into the western margin of North America during Mesozoic oblique convergence, was sliced by strike-slip faults into fragments that were then transported along the margin. As a result, bits and pieces of Wrangelia occur in a discontinuous chain that can be traced from Idaho to Alaska (W in Figure 14.20b).

FIGURE 14.20. (a) Map-view sketches showing progressive stages during oblique docking of an exotic terrane. Note how the terrane is sliced by faults subsequent to docking, and slivers slip along the length of the orogen. (b) Map of the North American Cordillera, showing the regions that consist of accreted crust, which is subdivided into numerous exotic terranes. [19.24, 17.29]

cordilleranterranes(BP)

Strike-slip faulting also develops at collisional margins where one continent indents the other, as we’ll explore in the next chapter. The vise created when two continents converge can cause blocks of crust to be squeezed laterally out of the zone of collision, a process called lateral escape. The boundaries of these escaping blocks are strike-slip faults. For example, during the Cenozoic collision of India with Asia, India pushed northward into Asia, resulting in a fan-shaped pattern of strike-slip faults accommodating the escape of blocks of Asia eastward. A similar phenomenon is currently happening in the eastern Mediterranean, where the northward movement of the Arabian Peninsula along the Dead Sea transform has resulted in the westward escape of Turkey, squeezed like a watermelon seed between the North Anatolian Fault and the East Anatolian Fault.

Settings of Oceanic Strike-Slip Zones

All ocean ridges, plate boundaries along which seafloor spreading occurs, consist of non-coplanar segments that range in length from tens to hundreds of kilometers. Each segment is linked to its neighbor by an oceanic transfer fault, typically called oceanic transform fault in this setting, and an (inactive) extension that leaves a topographic scar on the ocean floor, called a fracture zone.

Oceanic Transform Faults

Oceanic transform faults are plate boundaries, with strike-slip deformation accommodating the motion of one oceanic plate laterally past another as seafloor spreading progresses (Figure 14.21). The length of oceanic transform faults varies from 10 to 1000 km. As measured across strike, they are relatively narrow (less than a few kilometers wide); but even considering this narrow width, they are so abundant that between 1% and 10% of the oceanic lithosphere has been affected by deformation related to strike-slip faulting. It is important to remember from our earlier discussion of kinematics that transform faults occur only between ridge segments, recorded by characteristic earthquake activity; they do not extend beyond them. There are, however, pronounced topographic lineaments, known as fracture zones, which extend beyond the tips of the active transform segments.

http://media.marine-geo.org/files/images/2007-MGDS-237.preview.jpg

FIGURE 14.21. Map of the Clipperton fracture zone (FZ) and transform zone (TZ) of the East Pacific Rise (EPR). Note intersection highs at ridge tips, and trough and ridges along the transform zone. Contours in meters below sea level. EPR = East Pacific Rise. (b) Topography of the Clipperton fracture zone and transform of the East Pacific Rise. Note intersection highs at ridge tips, and trough and ridges along the transform zone. [marine-geo.org]

GE: 10°11'43.36" N 103°48'46.01" W [19.29]

Oceanic Fracture Zones

Seismicity in the ocean floor shows that earthquakes occur along the ridge segments and along the transform faults between them, but that they are very rare along fracture zones. Thus, fracture zones are not active fault zones. Instead, they are the expression of once active transforms that juxtapose different age ocean floor. Also, when you cross a transform boundary, you pass from one plate to another, but when you cross a fracture zone, you stay on the same plate.

Satellite gravity measurements, side-scan sonar images, dredge hauls, cores, and submarine photographs lead to the conclusion that oceanic transform zones and fracture zones are not featureless lines on the surface of the sea floor. Rather, they contain escarpments, ridges, and narrow troughs (Figure 14.21). The bathymetric complexity of oceanic transform zones and fracture zones develops in response to a variety of phenomena. First, since the depth of ocean floor depends on the age of the underlying lithosphere and the ocean floor on one side of a zone is not the same age as the ocean floor on the other side (except at the point on a transform fault halfway between two ridge tips), there must be a change in ocean-floor depth across a zone. Second, in places where transforms are not precisely small circles around an Euler pole (see Chapter 11), transpression and transtension generate a flower structure; the flower structure becomes inactive once a transform zone becomes a fracture zone, but the faults comprising it do not disappear. Third, slip along a transform zone pervasively fractures the crust. Seawater circulates through the fracture network and reacts with crustal basalt, altering olivine to form a hydrated mineral called serpentine. This process increases the volume of the crust, because serpentine is less dense than olivine, and thus causes the sea floor to rise. Fourth, heat radiating from the tip of a ridge segment can cause isostatic uplift of the lithosphere beyond the tip. This uplifted lithosphere gradually drifts away from the ridge termination, as sea-floor spreading progresses, forming an elongate ridge (an intersection high) bordering the fracture zone (Figure 14.21). At a distance from the ridge, the intersection high cools and subsides.

FIGURE 14.22. Today’s plate boundary transfer faults (or transforms). Ocean ridges (double lines), trenches (lines with teeth on overriding plate) and transfer faults (single lines) mark the plate boundaries. In a few places, plate boundaries are ill-defined, as marked by dashed lines. [14.14]

Notably, the development of escarpments in oceanic transform zones and fracture zones, coupled with the weakening of rock by pervasive fracturing and serpentinization, sets the stage for submarine slope failure. The debris that tumbles down the escarpments collects in thick piles consisting of sedimentary breccia. Thus, in contrast to normal ocean crust, the oceanic crust of transform zones and fracture zones typically has a coating of sedimentary breccia. A more detailed examination of a map with global plate boundaries (Figure 14.22) reveals that not all oceanic transform faults offset ridge segments. Some are plate boundaries that link subduction zones, others link subduction zones to ridges. Examples include the north and south border of the Scotia Plate between Antarctica and South America, and the northern border of the Caribbean Plate. Another spectacular example, the Alpine Fault of New Zealand's south island, connects along strike to opposing oceanic subduction zones.

Back to Wrench Tectonics Contents

Back to PSG&T Home

[1] Tuzo Wilson proposed this new fault class in 1965.