1.3.1 Transform Faults Linking Ridge with Ridge

This type of transform (Figure 1.2a) separates blocks of oceanic lithosphere in intermediate and some fast and some slow spreading systems, where the described spreading rate categories are using criteria from Small (Reference Small, Buck, Delaney, Karson and Lagabrielle1998), Dick et al. (Reference Dick, Lin and Schouten2003), and Kriner et al. (Reference Kriner, Pockalny and Larson2006). The structural architecture in this case can be characterized as an orthogonal pattern of transforms (e.g., Freund and Merzer, Reference Freund and Merzer1976; Figures 1.2b, 3.2a, and 3.2e; S3 architecture in Figure 1.2c). This pattern evolves in the following different lithospheric settings:

1. 1) some slow spreading systems that develop a new oceanic lithosphere by magma-assisted spreading; and

2. 2) intermediate and fast systems that develop a new oceanic lithosphere by magma-assisted spreading.

Figure 1.2 (a) Basic types of transforms based on the possible configurations of neighbor plate boundaries (Wilson, Reference Wilson1965).

Reprinted by permission from Springer Nature: Nature. Wilson, J. T. A new class of faults and their bearing on continental drift. 207, 343–347. ©1965.

Figure 1.2 (b) Free-air gravity anomaly map of the southern Central Atlantic and western Equatorial Atlantic imaging a system of Mid-Atlantic Ridge segments and linking ridge transforms. Abbreviations: OFZ – oceanic fracture zone, R – spreading ridge segment, RT – ridge transform.

(modified from Nemčok et al., Reference Nemčok, Rybár, Odegard, Nemčok, Rybár, Sinha, Hermeston and Ledvényiová2015a)

Figure 1.2 (c) Dependence of structural architecture of the spreading system on the spreading rate as indicated by numerical modeling (Puthe and Gerya, Reference Puthe and Gerya2014). Architectures S1, S2, S3, and S4 represent rates of 10 mmy^–1, 20 mmy^–1, 40 mmy^–1, and 80 mmy^–1, respectively. See text for further explanation.

Reprinted from Gondwana Research, 25. Puthe, C. and Gerya, T. Dependence of mid-ocean ridge morphology on spreading rate in numerical 3-D models. 270–283. ©2014, with permission from Elsevier.

It needs to be emphasized that the structural architecture is rather different in spreading systems that are either faster or slower than the aforementioned systems (Figure 1.2c). For example, observations and modeling indicate that ultra-slow and some slow spreading systems do not develop well-pronounced transforms, but rather curved spreading centers (Shemenda and Grocholsky, Reference Shemenda and Grocholsky1994) or oblique amagmatic spreading sections (Dick et al., Reference Dick, Lin and Schouten2003; Puthe and Gerya, Reference Puthe and Gerya2014) instead of transforms (S1 architecture in Figure 1.2c). Oblique shear zones were described, for example, from some segments of the slow spreading Mid-Atlantic Ridge (Schouten and White, Reference Schouten and White1980; Sempere et al., Reference Sempere, Lin, Brown, Schouten and Purdy1993). This tectonic setting evolves in pre-existing lithosphere affected by stretching.

In the case of some modeled fast spreading systems, transforms are initially possible but do not represent a steady-state feature, because they become replaced by curved ridge sections after a few million years (Gerya, Reference Gerya, Duarte and Schellart2016; S4 architecture in Figure 1.2c). Furthermore, the models of ultra-fast systems are characterized by the absence of transforms at rates exceeding 140 mmy^–1 (Naar and Hey, Reference Naar and Hey1989; Kriner et al., Reference Kriner, Pockalny and Larson2006). Ridge offsets are accommodated by microplates, propagating rifts, and overlapping spreading centers (Kriner et al., Reference Kriner, Pockalny and Larson2006). Overlapping centers are described, for example, from some segments of the fast spreading East Pacific Rise (Macdonald and Fox, Reference Macdonald and Fox1983; Lonsdale, Reference Lonsdale1985). Both fast and ultra-fast systems evolve in magma-assisted crust and lithospheric mantle affected by dominating buoyancy forces and subordinate stretching.

The aforementioned non-transform discontinuities from both very slow and very fast spreading systems can be characterized as follows. Those not exceeding a length of a few tens of kilometers are short-lived and known to be able to migrate along the strike of the spreading system (Macdonald et al., Reference Macdonald, Scheirer and Carbotte1991). The longer ones are more stable, able to last several million years. Unlike ridge transforms, their outboard extensions do not form small circles reflecting the pole of relative motion (Detrick et al., Reference Detrick, White and Purdy1993).

A steady-state transform representing intermediate, some fast and some slow spreading systems is composed of a zone of distinct troughs and ridges, which can be more than several tens of kilometers wide (Figures 3.5c, 3.13a–c, 3.16, and 7.23). Their elevation difference from the top of the undeformed oceanic crust can reach several kilometers (Figures 3.13a, 3.13c, 7.23). Individual strike-slip faults of the transform represent a narrow fault zone along which oceanic plate segments are passing each other horizontally (Detrick et al., Reference Detrick, White and Purdy1993). It is relatively stable for millions of years. Ridge transform faults of intermediate, fast and some slow spreading systems are commonly perpendicular to the spreading ridge segments and parallel to the spreading–controlling extension (Oldenburg and Brune, Reference Oldenburg and Brune1972; Gerya, Reference Gerya, Duarte and Schellart2016; Figure 1.1) once they reach the steady-state development (Figure 1.2d). They play an important role in affecting the thermal regime, mantle flow, melting, and crystallization at mid-oceanic ridges (Fox and Gallo, Reference Fox and Gallo1984; Parmentier and Forsyth, Reference Parmentier and Forsyth1985; Phipps Morgan and Forsyth, Reference Phipps Morgan and Forsyth1988). Active, ridge-connecting transforms, which are usually tens to hundreds of kilometers long, extend further out from between the spreading ridges as inactive oceanic fracture zones of similar geometry (Detrick et al., Reference Detrick, White and Purdy1993; Gregg et al., Reference Gregg, Lin, Behn and Montesi2007; Figures 1.2b and 1.3). These extensions, represented either by zones of weakness or locked fault systems, can be occasionally reactivated in cases of spreading vector and rate changes, and main plate motion changes (Gomes et al., Reference Gomes, Gomes, Palma, Jinno, de Souza, Mohriak and Talwani2000; Robinson, Reference Robinson2011). Furthermore, because the thermal contraction continues as the oceanic lithosphere drifts away from the spreading ridge, some dip-slip displacement can be expected along oceanic fracture zones. In these cases, the older lithosphere forms the hanging wall. In most cases oceanic fracture zones are aseismic. They can be traced for thousands of kilometers (Figure 1.2b) onto the ridge flanks as small circles reflecting both the direction of plate motion and the divergence of plates on either side of the ridge, as a function of rotation around an Euler Pole (Menard and Chase, Reference Menard, Chase and Maxwell1970).

Figure 1.2 (d) Numerically simulated development of the ridge transform fault (Gerya, Reference Gerya2013a), representing (1) propagating spreading centers reaching each other, (2) overlapping centers, (3) proto-transform development stage, (4) transform nucleation stage, and (5) transform growth stage.

Republished with permission of Springer Nature from Initiation of transform faults at rifted continental margins: 3D petrological–thermomechanical modeling and comparison to the Woodlark Basin. Gerya, T., Petrology, 21(6), 550–560. ©2013. Permission conveyed through Copyright Clearance Center, Inc.

Figure 1.3 Bathymetric and residual mantle Bouguer anomaly maps of fast slipping Siqueiros and slow slipping Atlantis ridge transforms (Gregg et al., Reference Gregg, Lin, Behn and Montesi2007). (a) Bathymetric map of the Siqueiros Transform. Solid black lines – structural architecture of oceanic crust, dashed black lines – transform and oceanic fracture zone locations, circles – seamount locations, white line – plate boundary. NVZ – neo-volcanic zone, OSC – overlapping spreading center. (b) Bathymetric map of the Atlantis Transform. Black and white lines indicate transform and spreading centers. NTO – non-transform offset. (c) Residual mantle Bouguer anomaly map of the Siqueiros Transform. (d) Residual mantle Bouguer anomaly map of the Atlantis Transform.

Reprinted by permission from Springer Nature: Nature Letters. Gregg, P. M., Lin, J., Behn, M. D. and Montesi, L. G. J. Spreading rate dependence of gravity anomalies along oceanic transform faults. Nature Letters, 448, 12, 183–187. ©2007.

In contrast to continental strike-slip faults, the ridge transforms may show opposite displacement sense within a pattern characterized by the same strike. Furthermore, the displacement sense of the ridge transform can be opposite to what is apparent from the spreading ridge offset.

The thermal regime and mechanical behavior of ridge transforms are distinctively different from those of continental transforms. This is due to heat conduction and advection from the warmer, younger plate into the older, colder plate (Louden and Forsyth, Reference Louden and Forsyth1976; Roland et al., Reference Roland, Behn and Hirth2010; Kolandaivelu et al., Reference Kolandaivelu, Harris and Lowell2017) and the elastic layer development, which both cause ridge transforms to have unusual topographic and gravimetric structure (Sandwell and Schubert, Reference Sandwell and Schubert1982; Sandwell, Reference Sandwell1984; Pockalny et al., Reference Pockalny, Gente and Buck1996).

As documented by submersible studies of the Oceanographer and Vema ridge transforms (OTTER Scientific Team, 1985; Auzende et al., Reference Auzende, Bideau and Bonatti1989), the walls of both transform faults and their oceanic fracture zones represent steep outcrops of basaltic, gabbroic, and ultramafic rocks. Furthermore, the crustal architecture of ridge transform faults and their oceanic fracture zones differ from that of normal oceanic crust (Detrick et al., Reference Detrick, White and Purdy1993, and references therein). Slow slipping Atlantic Transforms and their fracture zones cut the oceanic crust, which is thinner than normal (Figure 1.4). Mantle Bouguer anomaly images of these transforms are represented by anomalies that are more positive than anomalies of neighboring spreading centers (Kuo and Forsyth, Reference Kuo and Forsyth1988; Lin et al., Reference Lin, Purdy, Schouten, Sempere and Zervas1990; Lin and Phipps Morgan, Reference Lin and Phipps Morgan1992; Detrick et al., Reference Detrick, Needham and Renard1995).

Figure 1.4 Schematic cross section across a typical North Atlantic oceanic fracture zone (Detrick et al., Reference Detrick, White and Purdy1993). A – normal oceanic crust. B – thin crust of the oceanic fracture zone, lacking the seismic layer representing the gabbro layer and characterized by unusually low compressional wave velocities. C – the thinnest crust underlying the transform valley. D – gradually thinning crust up to 20 km from the oceanic fracture zone. E – transverse ridge bordering the oceanic fracture zone, characterized by relatively normal compressional wave velocities, relatively normal crustal thickness, and Moho upwarping. F – partially serpentinized upper mantle underlying some oceanic fracture zones.

Reprinted with permission from John Wiley and Sons. Detrick, R. S., R. S. White, and G. M. Purdy (Reference Detrick, White and Purdy1993), Crustal structure of North Atlantic Fracture Zones, Rev. Geophys., 31(4), 439–458.

Mantle Bouguer anomaly maps of the slow spreading Mid-Atlantic Ridge indicate a significant crustal structure asymmetry at ridge transform intersections, characterized by thinner and thicker crust at inside and outside corners, respectively (Detrick et al., Reference Detrick, White and Purdy1993). The inside corners frequently contain oceanic core complexes exhumed in footwalls of upward convex detachment faults (e.g., Cann et al., Reference Cann, Blackman and Smith1997; Cannat et al., Reference Cannat, Lagabrielle and Bougault1997; Ranero and Reston, Reference Ranero and Reston1999; Tucholke et al., Reference Tucholke, Fujioka, Ishihara, Hirth and Kinoshita2001; MacLeod et al., Reference MacLeod, Escartín and Banerji2002; Escartín et al., Reference Escartín, Mével, MacLeod and McCaig2003). Typically, the crust of these oceanic fracture zones (Figure 1.4):

1. 1) is very thin;

2. 2) is characterized by very low compressional wave velocities;

3. 3) is characterized by rather high velocity gradients;

4. 4) lacks a gabbro layer;

5. 5) is sometimes thinnest underneath the deepest trough, sometimes underneath the trough flanks;

6. 6) thins very abruptly between the normal oceanic crust and oceanic fracture zone; and

7. 7) contains transverse ridges with relatively normal crustal thickness that represent distinct topographic highs and are underlain by upwarped Moho.

The thinner crust of transform faults and non-transform offsets at slow spreading systems (see also Kuo and Forsyth, Reference Kuo and Forsyth1988; Lin et al., Reference Lin, Purdy, Schouten, Sempere and Zervas1990; Blackman and Forsyth, Reference Blackman and Forsyth1991; Lin and Phipps Morgan, Reference Lin and Phipps Morgan1992; Tolstoy et al., Reference Tolstoy, Harding and Orcutt1993) is interpreted as the result of reduced magma supply at the ends of individual spreading segments due to mantle upwelling targeting segment centers (Fox and Gallo, Reference Fox and Gallo1984; White et al., Reference White, Detrick, Sinha and Cormier1984; Phipps Morgan and Forsyth, Reference Phipps Morgan and Forsyth1988; Lin et al., Reference Lin, Purdy, Schouten, Sempere and Zervas1990; Sparks and Parmentier, Reference Sparks and Parmentier1993; Canales et al., Reference Canales, Collins, Escartín and Detrick2000) and stretching dominating in the inside corners of their intersections (Mutter and Karson, Reference Mutter and Karson1992).

Furthermore, a transform-controlled juxtaposition of the older, colder lithosphere to a warm spreading center causes an important perturbation of the thermal regime and affects the extruded lava composition (Bender et al., Reference Bender, Hodges and Bence1978, Reference Bender, Langmuir and Hanson1984; Langmuir and Bender, Reference Langmuir and Bender1984; Reynolds et al., Reference Reynolds, Langmuir, Bender, Kastens and Ryan1992; Reynolds and Langmuir, Reference Reynolds and Langmuir1997; Gregg et al., Reference Gregg, Behn, Lin and Grove2009).

In contrast to slow slipping ridge transforms, fast slipping Pacific Transforms and their fracture zones cut the oceanic crust, which is thicker than normal, while the thickness can vary from transform segment to transform segment (Fox and Gallo, Reference Fox and Gallo1984; Macdonald et al., Reference Macdonald, Fox and Perram1988; Canales et al., Reference Canales, Detrick, Toomey and Wilcock2003). Mantle Bouguer anomaly images of these transforms are represented by anomalies that are more negative than anomalies of neighboring spreading centers (Gregg et al., Reference Gregg, Lin, Behn and Montesi2007, Reference Gregg, Behn, Lin and Grove2009). Apart from the subordinate effect of increased fracture-related porosity and mantle serpentinization on the aforementioned gravity image, local crustal thickness changes play the dominant role (Gregg et al., Reference Gregg, Lin, Behn and Montesi2007). The increased crustal thickness can be explained by a combination of (Menard and Atwater, Reference Menard and Atwater1969; Fornari et al., Reference Fornari, Gallo and Edwards1989; Perfit et al., Reference Perfit, Fornani and Ridley1996; Begnaud et al., Reference Begnaud, McCain, Barth, Orcutt and Harding1997; Karson et al., Reference Karson, Tivey and Delaney2002; Gregg et al., Reference Gregg, Behn, Lin and Grove2009):

1. 1) lateral magma transport from the spreading ridge into the transform fault zone;

2. 2) magma-rich accretion of oceanic crust at intra-transform spreading centers;

3. 3) lava pooling in transform troughs; and

4. 4) general leaky magmatic accretion along the entire transform length.

Numerical simulations by Gregg et al. (Reference Gregg, Behn, Lin and Grove2009) indicate that the key control of the crustal thickness variations associated with a specific pattern of intra-transform spreading centers and linking transform faults is the volume of mantle over which melt is pooled. The longer the intra-transform center length, the greater the extent of melting and the smaller the average melting pressure.

The Atlantis Transform, which offsets the Mid-Atlantic Ridge at about 30°N, is an example of a slow slipping transform (24 mmy^–1) (Figure 1.3). Its structural architecture is represented by a single strike-slip fault zone (Figure 1.5a). This is analogous to other slow slipping transforms, although exceptions, such as the slow slipping Romanche and Andrew Bain Transforms in the Equatorial Atlantic and Southwest Indian Ocean, respectively, are characterized by wide complex deformation zones resembling those of continental transforms (Ligi et al., Reference Ligi, Bonatti, Gasperini and Poliakov2002; Sclater et al., Reference Sclater, Grindlay, Madsen and Rommevaux-Jestin2005). The calculated depth of serpentinization at the Atlantis Transform reaches mantle levels (Gregg et al., Reference Gregg, Lin, Behn and Montesi2007).

(a) Model of the slow spreading ridge, characterized by mantle upwelling focused at spreading ridge segment centers. This results in thinner oceanic crust at segment ends and transforms.

(b) Model of the fast spreading ridge, characterized by relatively homogeneous mantle upwelling. The magma-rich character of ridge segments even at their ends is the likely cause of the overshooting ridges and dyke emplacement into the neighboring portion of the transform. Note crustal thickness variations along the transform caused by the crustal accretion at all individual intra-transform ridges.

Reprinted by permission from Springer Nature: Nature Letters. Gregg, P. M., Lin, J., Behn, M. D. and Montesi, L. G. J. Spreading rate dependence of gravity anomalies along oceanic transform faults. Nature Letters, 448, 12, 183–187. ©2007.

Figure 1.5 Models of oceanic crust accretion, fracturation, and serpentinization at ridge transform intersections together with structural architecture of ridge transform faults (Gregg et al., Reference Gregg, Lin, Behn and Montesi2007). White arrow – spreading vector, black arrow – mantle flow trajectory, gray – lithosphere defined by the basal temperature of 700°C, dotted pattern – crust, wiggly black-white lines – zone of fractured lithosphere, light grey polygons – rock volumes affected by serpentinization.

An example of a transform affected by a hotspot is seen in the north of Iceland. Here the extensional transform zone links the sub-aerially exposed neo-volcanic zone of northeast Iceland with the Kolbeinsey Ridge, located to the north in the Norwegian–Greenland Sea, in a left-lateral step (Figure 8.2a). The Tjörnes Fracture Zone is a 75 km wide zone of extensional grabens and strike-slip faults (Hannington et al., Reference Hannington, Herzig and Stoffers2001). Its structural architecture is complex, resembling that of fast slipping transforms, yet its strike is at a low angle to the σ₃ direction. The reason for its anomalous architecture is that it links an ultra-slow but hotspot-affected Kolbeinsey Ridge, with a spreading rate of 15.8 mmy^–1 (Pálmason, Reference Pálmason1973; DeMets et al., Reference De Mets, Gordon, Argus and Stein1994), with the strongly hotspot-affected North Iceland rift zone, rifting at an average rate of about 20 mmy^–1, where a single fissure swarm, such as the Krafla swarm, can episodically reach an extension rate of 60 mmy^–1 (Árnadóttir et al., Reference Árnadóttir, Geirsson and Jiang2008). Due to the hotspot influence, the linked spreading centers are magma-rich, which results in a potential for overshooting ridges. One example is the Manareyjar Ridge (Figure 8.4), which occurs inside the Tjörnes Transform zone, between its two major internal strike-slip faults. Being connected to the Theistareykir fissure swarm of the Northern Iceland rift zone, it does not follow the strike of the transform zone but is located at a high angle to the boundary strike-slip faults. Its location is most likely affected by the normal faults and sinistral strike-slip faults bounding the crustal blocks moving by a bookshelf mechanism, being bounded by three major dextral strike-slip faults dividing the Tjörnes ridge transform into three parallel internal zones (e.g., Rongvaldsson et al., Reference Rongvaldsson, Gudmundsson and Slunga1998; Riedel et al., Reference Riedel, Schmidt, Botz and Theilen2001; Stefansson et al., Reference Stefansson, Gudmundsson and Halldorsson2008). Volcanic activity along the entire transform zone is indicated by volcanoes located inside pull-apart basins.

The Siqueiros Transform, which offsets the East Pacific Rise, is an example of a fast slipping transform (118 mmy^–1) (Figure 1.3). Its structural architecture is markedly different from that characterizing slow slipping transforms (Figure 1.5b). It contains five individual faults separating four intra-transform spreading centers (Fornari et al., Reference Fornari, Gallo and Edwards1989), analogous to other fast transforms (Menard, Reference Menard1967; Menard and Atwater, Reference Menard and Atwater1969; Searle, Reference Searle1983; Fox and Gallo, Reference Fox and Gallo1984; Gregg et al., Reference Gregg, Lin and Smith2006). The intersections of its bounding faults with neighboring spreading ridge segments contain overshooting ridge tips typical for magma-rich fast spreading systems (Fornari et al., Reference Fornari, Gallo and Edwards1989; Figure 1.5b). The curved geometry of ridges can be interpreted as a result of either local perturbation of the stress field at ridge transform intersections or preferential dyke propagation along the pre-existing transform-related fracture systems (Gregg et al., Reference Gregg, Lin, Behn and Montesi2007). Fresh basalt samples and spreading-related magnetic anomalies at this transform and several others in the East Pacific region, such as the Quebrada, Discovery, Gofar, and Garrett Transforms, indicate active accretion taking place at intra-transform spreading centers (Fornari et al., Reference Fornari, Gallo and Edwards1989; Carbotte and Macdonald, Reference Carbotte and Macdonald1990; Hekinian et al., Reference Hekinian, Bideau, Cannat, Francheteau and Hebert1992, Reference Hekinian, Bideau, Hebert and Niu1995; Perfit et al., Reference Perfit, Fornani and Ridley1996; Nagle et al., Reference Nagle, Pickle, Saal, Hauri and Forsyth2007). With regard to estimated depth extent of the serpentinization along fast slipping transforms, the calculated 500°C isotherm, which represents the upper limit of serpentine stability (Ulmer and Trommsdorff, Reference Ulmer and Trommsdorff1995), does not reach down to mantle depths and remains within the crust (Gregg et al., Reference Gregg, Lin, Behn and Montesi2007).

There has been much debate about the influence of pre-breakup continental fault zones on post-breakup ridge transform faults. Some studies interpret the ridge transform faults as being predisposed by the locations of the pre-existing continental zones of weakness (e.g., Wilson, Reference Wilson1965; Mascle, Reference Mascle1976; Wright, Reference Wright1976; Sibuet and Mascle, Reference Sibuet and Mascle1978; Rosendahl, Reference Rosendahl1987; Bellahsen et al., Reference Bellahsen, Leroy and Autin2013; Gibson et al., Reference Gibson, Totterdell and White2013). Our experience with long transforms that underwent an oceanic–continental stage of development, such as the Zenith–Wallaby–Perth (Figures 3.2d–e), Romanche (Figures 2.6a–b, 2.10b, 2.17, 3.5c, and 4.9), St Paul (Figures 2.6a–b, 2.10b, 3.5c, and 4.9), and Coromandal (Figures 2.12a, 6.9a–b, and 6.15) transforms, indicates that they went through the interaction of Andersonian (see Anderson, Reference Anderson1951) or modified Andersonian faulting on their continental side with non-Andersonian faulting on their oceanic side. This interaction must, to some extent, have controlled the propagation trend of the ridge transform segment located immediately outside of the most external continental–oceanic segment.

Other studies interpret ridge transform fault propagation as unaffected by pre-existing anisotropy (e.g., Taylor et al., Reference Taylor, Goodlife and Martinez2009; Basile, Reference Basile2015). They document that the initial spreading offsets are often of a non-transform type, and that transform faults develop either after or during spreading nucleation. Natural examples of this stage are observed in the Woodlark Basin (Taylor et al., Reference Taylor, Goodlife and Martinez2009). It seems that such a propagation mode can characterize the portion of the post-breakup system, which either contains relatively short oceanic–continental transforms or none at all.

Additional support for transform fault propagation being unaffected by pre-existing anisotropy comes from three-dimensional numerical modeling simulating advanced slow spreading and initial intermediately fast spreading systems (Gerya, Reference Gerya2013a), which suggests that (Figure 1.2d):

1. 1) spreading segments nucleate in an en echelon pattern in overlapping rift units at the start of the initial disorganized spreading stage;

2. 2) subsequently, still during the disorganized spreading stage, proto-transforms sometimes nucleate as oblique to a spreading vector rather than parallel to a spreading vector;

3. 3) subsequently, still during the disorganized spreading stage, the initially oblique proto-transforms rotate into ridge-perpendicular position; and

4. 4) subsequently, during the organized spreading stage, the spreading vector-parallel transform direction becomes a rule governed by space accommodation during the mature ridge transform spreading that is reached after about the first five million years of spreading.

Further modeling support for spontaneous oceanic transform development comes from Gerya (Reference Gerya, Duarte and Schellart2016).

The proto-transform rotation is controlled by the long-lasting weakness of proto-transforms (Gerya, Reference Gerya, Duarte and Schellart2016). It is the main reason for the continuously maintained kinematic linkage between the two sea-floor spreading ridges. This rotation can be described by the function (Gerya, Reference Gerya, Duarte and Schellart2016):

$tan\alpha =tan{\alpha }_{\mathrm{o}}\left[1-v_{\mathrm{s}}\left(t-t_{\mathrm{o}}\right)/L\right],$(1-1)

where α is the deviation angle of the proto-transform from the spreading direction, α_o is the initial deviation angle of the proto-transform from the spreading direction, v_s is the spreading rate, t is the time at which the deviation angle is calculated, t_o is the time of the proto-transform nucleation, and L is the spreading center offset. Function 1-1 indicates that the spreading–parallel proto-transform orientation is reached at $t=t_{\mathrm{o}}+L/v_{\mathrm{s}}.$

The aforementioned model indicates that the orthogonal transform spreading center pattern is controlled by the 180° rotational symmetry for open-space occupation for coevally acting accretion and sliding of new plate material (Gerya, Reference Gerya, Duarte and Schellart2016). This pattern represents the only steady-state pattern that is thermo-mechanically consistent.

As already indicated, transforms that connect ridges typically develop only in advanced slow and initial intermediately fast spreading systems. Using the same modeling approach as used by Gerya (Reference Gerya2010, Reference Gerya2013a), Puthe and Gerya (Reference Puthe and Gerya2014) published modeling results that indicate that spreading center patterns depend on the spreading rate. Based on modeling results, they can be divided into four scenarios (Figure 1.2c):

1. 1) a stable system developing alternating new magmatic and amagmatic plate zones (S1 in Figure 1.2c) in the case of ultra-slow spreading;

2. 2) an asymmetrically spreading system developing transient features – that is, proto-transforms (S2 in Figure 1.2c) – in some cases of slow spreading;

3. 3) a stable orthogonal spreading system (S3 in Figure 1.2c) in the case of some slow and some intermediate spreading; and

4. 4) a stable system developing curved ridges (S4 in Figure 1.2c) in the case of fast and some intermediate spreading.

Oceanic fracture zones in the fully oceanic domain – that is, where they traverse only oceanic crust far away from continents – are essentially unprospective for hydrocarbons. However, where these fractures approach continental margins at acute angles, they can have significant relevance for petroleum. A prime example occurs in offshore Guyana, where the recent spate of oil discoveries in Cretaceous turbidites, including Liza and related fields such as Yellowtail and Payara, appear to have a close relationship with fracture zones. The fields occur where the thick, primarily Cretaceous sedimentary wedge oversteps the Guyana continental margin and drapes several ridges of the oceanic fracture zone. The fields lie where thick continental crust is juxtaposed against normal oceanic crust. Farther outboard, normal oceanic crust is juxtaposed with the zone of prominent ridges and troughs representing the oceanic fracture zone. The fields also seem to coincide with areas where the transforms are “leaky” and associated with volcanism. The association between fracture zones, volcanism, and stacked hydrocarbon plays is interesting and merits further research. It is plausible that, in this area, turbidite sands were deposited by the small changes in gradient created by the fracture zones. Compaction over buried ridges associated with fracture zones also nucleated later normal faulting in the sediments, which may have aided vertical migration. Further discussion of this field complex may be found in Chapter 15.

1.3.2 Transforms Evolving from the Continental Transform Stage, Through the Oceanic–Continental Transform Stage, into the Transform Margin Stage

This type of transform represents a fault zone along which different plates, microplates – or at least portions of plates with different accretion age – are passing each other horizontally (Figure 1.6a). Some of them can be exceptionally long and active for a distinctly long time period. Their interesting feature is that they evolve from continental transforms into transforms separating different types of lithosphere.

Figure 1.6 (a) Sketch of the strike-slip fault zone with pull-apart terrains evolving into the ocean. A checkpoint is indicated by a small circle. Stages: I, rifting of continental lithosphere and the checkpoint situated at the dextral transform fault. II, the checkpoint still adjacent to continental lithosphere but a landward portion of the future transform margin adjacent progressively to normal continental crust, then thinned continental crust and then oceanic crust. III, the checkpoint adjacent to the spreading ridge. IV, the checkpoint adjacent to oceanic crust and an inactive transform fault. d is the distance traveled by the spreading center and l is the total length of the transform. d = μ/2 + t, where μ is the half spreading rate and t is the time length of ridge traveling.

(Nemčok et al., Reference Nemčok, Rybár, Sinha, Hermeston, Ledvényiová, Nemčok, Rybár, Sinha, Hermeston and Ledvényiová2016a)

The kinematic development of such a transform, which helps to break up the continent and subsequently accommodates the lateral clearance of the drifting-away daughter continents, may consist of three stages (e.g. Mascle and Blarez, Reference Mascle and Blarez1987; Figure 1.6a):

1. 1) active continental stage;

2. 2) active continental–oceanic stage; and

3. 3) passive margin stage.

The time spans and the termination of individual stages vary along the strike of the transform margin. The termination of the continental stage in a specific area of the future transform margin depends on the time required for the adjacent continent to clear this area and for the juxtaposition of this area to laterally clearing oceanic crust (Figure 1.6a). The passive margin stage begins in specific area once the spreading center clears this area.

Examples of transforms that evolved from the continental stage to passive margin stage include those of the African Equatorial Atlantic (Figure 1.6b) and West Australian margins. The largest transforms of these two example margins, and most other large transform margins worldwide, provide evidence that continental transforms that evolved into continent–ocean transform faults were affected by pre-existing continental zones of weakness (Nemčok et al., Reference Nemčok, Ledvényiová and Henk2021a). Such evidence comes from either transforms dividing segments of the ocean with different accretion time spans, or transforms dividing different plates and microplates. Some of these bound large marginal plateaus with breakup ages different from those of neighboring margin segments (e.g., Demerara, Guinea, and Exmouth plateaus; Marinho et al., Reference Marinho, Mascle and Wannesson1988; Stagg et al., Reference Stagg, Alcock and Bernadel2004; Nemčok et al., Reference Nemčok, Rybár, Odegard, Nemčok, Rybár, Sinha, Hermeston and Ledvényiová2015a). Some of them link orthogonal margins with different ages of breakup (e.g., Davie, Romanche, Coromandal, Gettysburg–Tarfaya and Cobequid–Chedabucto–Gibraltar Transforms; Nemčok et al., Reference Nemčok, Stuart, Segall, Post, Rosen and Olson2005a, Reference Nemčok, Henk, Allen, Sikora, Stuart, Mohriak, Danforth and Post2012a; Schettino and Turco, Reference Schettino and Turco2009; Sinha et al., Reference Sinha, Nemčok, Choudhuri, Sinha, Pundarika Rao, Nemčok, Rybár, Sinha, Hermeston and Ledvényiová2016).

Figure 1.6 (b) Structural architecture of the Romanche Transform margin of the Ivory Coast, Ghana, and Togo, which evolved from the continental transform stage and the subsequent ocean–continent stage into the transform margin development stage.

(modified from CGMW, 1990; Antobreh et al., Reference Antobreh, Faleide, Tsikalas and Planke2009; Nemčok et al., Reference Nemčok, Henk, Allen, Sikora, Stuart, Mohriak, Danforth and Post2012a, Reference Nemčok, Sinha, Doré, Nemčok, Rybár, Sinha, Hermeston and Ledvényiová2016b). Republished with permission of Elsevier from Rift-shear architecture and tectonic development of the Ghana margin deduced from multichannel seismic reflection and potential field data. Antobreh, A. A., Faleide, J. I., Tsikalas, F. and Planke, S. Marine and Petroleum Geology, 26, 345–368. ©2009. Permission conveyed through Copyright Clearance Center, Inc. International Geological Map of Africa at scale 1:5 M, 3rd edition (1985–1990). Coordination: G. Choubert and A. Faure-Muret, cartographic synthesis: P. Chanteux, copyright: CGMW-UNESCO, Paris, 1990

Evidence on the post-breakup kinematic adjustments of large continental–oceanic transforms (Nemčok et al., Reference Nemčok, Ledvényiová and Henk2021a) indicates that these faults can undergo different kinematic regimes through time.

Hydrocarbon discoveries in this setting are found, for example, on the African side of the Equatorial Atlantic, although not all of them are commercial fields (Figure 1.6c). Some fields occur in pull-apart margin segments located between two transform margin segments. Typical examples are the Akasa, Banda, Belier, Eland, Enyenra, Gazelle, Ibex, Jubilee, Kudu, Makore, Ntomme, Odum, Okure, Orca, Sankofa, Seme, Teak, Tweneboa, and Wava fields in offshore Benin, Ghana, and Ivory Coast. Other fields come from the transform margin segments, such as the Acajou, Baobab, Espoir, Foxtrot, Lion, Lome, Panthere, Paon, Saltpond fields in offshore Ghana, Ivory Coast, and Togo.

Figure 1.6 (c) Fault map from Figure 1.6(b) together with hydrocarbon field locations. Hydrocarbon discoveries: A – Almond, Ac – Acajou, B – Beech, Ba – Baobab, Be – Belier, El – Eland, Es – Espoir, Ey – Enyenra, F – Foxtrot, G – Gazelle, J – Jubilee, I – Ibex, K – Kudu, L – Lion, Lo – Lome, M – Mahogany, N – Ntome, O – Odum, Ok – Okure, Or – Orca, P – Panthere, Pa – Paon, S – Sankofa, Se – Seme, Sp – Saltpond, T – Tano, Tw – Tweneboa, W – Wawa.

Background topography and bathymetry maps are from GEBCO (2019). Republished with permission of Elsevier from Rift-shear architecture and tectonic development of the Ghana margin deduced from multichannel seismic reflection and potential field data. Antobreh, A. A., Faleide, J. I., Tsikalas, F. and Planke, S. Marine and Petroleum Geology, 26, 345–368. ©2009. Permission conveyed through Copyright Clearance Center, Inc. International Geological Map of Africa at scale 1:5 M, 3rd edition (1985–1990). Coordination: G. Choubert and A. Faure-Muret, cartographic synthesis: P. Chanteux, copyright: CGMW-UNESCO, Paris, 1990.

1.3.3 Transforms Linking Ridge with Convergent Plate Boundary

The best-known example of a transform that links a ridge with a convergent plate boundary and deforms mostly continental lithosphere is the San Andreas Fault (Figures 1.7a–b). The fault is located between the Mendocino triple junction and the East Pacific Rise spreading system of the Gulf of California (Elders, Reference Elders and Elders1979; Elders et al., Reference Elders, Rex, Meidav, Robinson and Biehler1972; Elders and Sass, Reference Elders and Sass1988; Figure 1.7a). Its structural architecture is complex. It contains segments with different deformation regimes, including locked segments, aseismically creeping segments, and transitional segments between the aforementioned two end members.

Figure 1.7 (a) Simplified map of the San Andreas Transform linking the system of spreading centers in the Gulf of California with the subduction zone to the north of the Mendocino triple junction (Elders et al., Reference Elders, Rex, Meidav, Robinson and Biehler1972; Elders, Reference Elders and Elders1979; Elders and Sass, Reference Elders and Sass1988).

Reprinted with permission from John Wiley and Sons. Elders, W. A. and Sass, J. H. Reference Elders and Sass1988, The Salton Sea Scientific Drilling Project: Journal of Geophysical Research, B, Solid Earth and Planets, v. 93, no. 11, p. 12953–12968.

Figure 1.7 (b) Present-day characteristics of the San Andreas Transform (Gerya, Reference Gerya, Duarte and Schellart2016). Abbreviations: ECSZ – Eastern California Shear Zone, GF – Garlock Fault, SAF – San Andreas Fault, SJF – San Jacinto Fault. (a) Predicted slip rates. (b) Predicted maximum shear stress. (c) Predicted plastic strain rate outside the GF, SAF, and SJF. (d) Predicted uplift rate. Circles document seismicity.

Reprinted with permission from John Wiley and Sons. Gerya, T. Reference Gerya, Duarte and Schellart2016. Origin, evolution, seismicity, and models of oceanic and continental transform boundaries. In: Duarte, J. C. and Schellart, W. (Eds), Plate Boundaries and natural hazards. Geophysical Monograph, 219, pp. 39–76.

The Carrizo Plain in central California is an example of a locked segment. It experiences a transpressive regime and, therefore, has some associated thrust faults. The last rupture of this section was the M7.9, 1857 Fort Tejon earthquake. Limited micro-seismicity has been recorded since the advent of modern instrumentation (Hill et al., Reference Hill, Eaton and Jones1990). Although it does not represent the total depth extent of brittle fracturing along this fault segment, magnetotelluric imaging allows one to interpret a 4–5 km depth extent of the increased fracture porosity zone (Unsworth et al., Reference Unsworth, Egbert and Booker1999; Figure 8.7).

A creeping segment example is seen in the area near Hollister, California, where the fault creeps aseismically at a rate of about 10–15 mmy^–1 (Burford and Harsh, Reference Burford and Harsh1980; Evans et al., Reference Evans, Burford and King1981). A combination of seismic velocity distribution and magnetotelluric imaging allows one to interpret a 10 km depth of the increased fracture porosity zone (Thurber et al., Reference Thurber, Roecker and Ellworth1997; Bedrosian et al., Reference Bedrosian, Unsworth, Egbert and Thurber2004; Figures 8.8 and 8.9). Synthetic fault zone trapped waves allow estimation of the width of the shallow portion of this San Andreas segment to be about 120 m (Li et al., Reference Li, Ellsworth, Thurber, Malin and Aki1997). However, the entire zone of increased fracture porosity indicated by magnetotelluric data is about 7 km wide. It occurs between the San Andreas and Calaveras Faults (Bedrosian et al., Reference Bedrosian, Unsworth and Egbert2002, Reference Bedrosian, Unsworth, Egbert and Thurber2004; Figure 8.8).

A transitional segment is observed in the Parkfield area in California, located between the aforementioned locked and creeping segments. This section of the San Andreas Fault experiences moderate earthquakes of about M6 about every 22 years, and extensive micro-earthquake activity at about 3 km depth (Bakun and Lindh, Reference Bakun and Lindh1985). Magnetotelluric imaging combined with seismic velocity distribution allows one to interpret a 4–5 km depth extent of the increased fracture porosity zone (Unsworth and Bedrosian, Reference Unsworth and Bedrosian2004; Figure 8.10). This zone is about 600 m wide on the surface.

Continental transforms can be associated with important hydrocarbon reserves (Harding, Reference Harding1974, Reference Harding1990; Christie-Blick and Biddle, Reference Christie-Blick, Biddle, Biddle and Christie-Blick1985; Harding et al., Reference Harding, Vierbuchen, Christie-Blick, Biddle and Christie-Blick1985; Lowell, Reference Lowell1985). The San Andreas Transform provides a series of examples. For example, a system of anticlines in its southern San Joaquin Basin hosts a number of fields (Wilcox et al., Reference Wilcox, Harding and Seely1973; Harding, Reference Harding1976; Namson and Davis, Reference Namson and Davis1988; Medwedeff, Reference Medwedeff1989, Reference Medwedeff, Mitra and Fisher1992; Hardy et al., Reference Hardy, Poblet, McClay, Waltham, Buchanan and Nieuwland1996; Mueller and Suppe, Reference Mueller and Suppe1997; Figure 1.7c). Another prolific region, the Los Angeles Basin, is located further south. It can be roughly described as an irregular Late Miocene–Early Pleistocene pull-apart basin that underwent an important compressional event, responsible for hydrocarbon trapping in anticlines and faulted anticlines (Yerkes et al., Reference Yerkes, McCulloch, Schoellhamer and Vedder1965; Biddle, Reference Biddle and Biddle1991).

Figure 1.7 (c) San Andreas Fault together with hydrocarbon fields hosted by its anticlines in the southern San Joaquin Basin (modified from Namson and Davis, Reference Namson and Davis1988).

Modified from Namson, J. S. and Davis, T. L. Reference Namson and Davis1988. Seismically active fold and thrust belt in the San Joaquin Valley, central California. Geological Society of America Bulletin, 100, 257–273.

1.3.4 Transforms Linking Convergent Plate Boundaries

An example illustrating this transform type is the dextral transform fault zone forming the southern boundary of the Caribbean Plate (Figure 1.8a). In fact, the characteristic portion of this zone is only its extent to the east of Puerto Cabelo in Venezuela – that is, the portion that contains the San Sebastian Fault, El Pilar Fault, and other strike-slip faults further east toward the Barbados accretionary prism, all with more-or-less E–W strikes. This is because the western, North Andean portion of this transform fault zone bends around the westward-moving South American Plate, and, as a result, gained transpressional character.

Figure 1.8 (a) Seismotectonic map of the Caribbean Plate.

(Calais, Reference Calais2018)

The North Andean dextral transpression is a result of the interaction of five different plates (Cortés and Angelier, Reference Cortés and Angelier2005; Figure 1.8a). The Cocos Plate moves northeastward at a rate of 71 mmy^–1. The Nazca Plate near Colombia, central Peru, and northern Chile moves eastward, east–northeastward, and east–northeastward at rates of 54, 22, and 32 mmy^–1, respectively. Bayona (2011, pers. com.) also characterizes the Colombian segment of the Nazca Plate by the east–northeastward movement. On the contrary, the North American Plate moves to the west–southwest. The South American Plate moves to the west and represents an important resistance against the eastward advance of the North Andean retro-wedge. Its movement rates are 47 and 48 mmy^–1 in central Peru and northern Chile, respectively (Moretti, 2011, pers. com.). This westward movement is associated with an intra-Middle Miocene increase in Mid-Atlantic Ridge sea-floor spreading rate (Cobbold, 2011, pers. com.). The increased rate controls a large-scale buckling recognized in platform areas of the South American Plate, which is under horizontal loading (Moretti, 2011, pers. com.).

The aforementioned eastern portion of the transform zone is rather wide and composed of numerous dextral strike-slip faults and their associated fault families (e.g., Audemard et al., Reference Audemard, Romero, Rendon and Cano2005; Figure 1.8b).

Figure 1.8 (b) Map of the Quaternary faults of the Venezuelan portion of the deformation zone representing the boundary between the Caribbean and South American Plates (Audemard et al., Reference Audemard, Romero, Rendon and Cano2005). Colombian faults: 2 – Santa Marta-Bucaramanga, 16 – Guaicaramo Thrust, 17 – Cusiana-Yopal thrust system, 36 – Soata, 37 – Cobugon. Venezuelan faults: 1 – Oca-Ancon, 2 – Lagarto, 3 – Urumaco, 4 – Rio Seco, 5 – Matapalo, 6 – Western Paraguana, 7 – Cumaraguas, 8 – Cabo San Rona-Puerto Escondido, 9 – Adicora, 11 – Taima-Taima/Chuchure, 12 – Carrizal, 13 – El Hatilo, 14 – La Soledad, 16 – Araurima.

Reprinted from Earth-Science Reviews, 69, Audemard, F. A., Romero, G., Rendon, H. and Cano, V. Ramprasad. Quaternary fault kinematics and stress tensors along the southern Caribbean from fault-slip data and focal mechanism solutions, 181–233. ©2005, with permission from Elsevier.

1.3.5 Trench-Linked Strike-Slip Faults

Trench-linked strike-slip faults form yet another group of strike-slip faults that cut through the entire lithosphere (Figure 1.1), and result from oblique subduction or accommodate changes in subduction direction. They represent fault zones parallel to the trench (Sylvester, Reference Sylvester1988) and located:

1. 1) inside the magmatic arc, such as the Sumatran Fault (McCaffrey et al., Reference McCaffrey, Zwick and Bock2000; Figure 1.9) and the Philippine Fault (Barrier et al., Reference Barrier, Huchon and Aurelio1991); or

2. 2) inside the fore-arc, such as the Hawkey Ridge Fault in the Aleutians (Scholl, Reference Scholl1999), faults of the southern Ryukyu arc (Lallemand and Chemenda, Reference Lallemand and Chemenda1999), and the Atacama Fault in Chile (McCaffrey, Reference McCaffrey, Hall and Blundell1996); or

3. 3) along the boundary between the fore-arc and hinterland plates, such as the South Japan Sea Fault Zone (Itoh et al., Reference Itoh, Tsutsumi, Yamamoto and Arato2002).

Figure 1.9 Tectonic map of the Andaman Sea region showing the Sagaing Transform linking the Eastern Himalayan syntaxis of the India–Eurasia collisional zone with spreading centers of the Central Region of the Andaman Sea.

(Morley, Reference Morley, Nemčok, Rybár, Sinha, Hermeston and Ledvényiová2015)

Trench-linked strike-slip faults can accumulate displacements of several hundred kilometers, accommodating portions of the deformation of the overriding plate behind the trench (Woodcock, Reference Woodcock1986). These faults are commonly short-lived features. Similarly to the previous transform category, they frequently reactivate pre-existing faults and become subsequently reactivated by other tectonic events.

For example, the recently active dextral South Japan Sea Fault Zone is associated with oblique Quaternary subduction of the Philippine Sea Plate under the Eurasian Plate (Itoh et al., Reference Itoh, Tsutsumi, Yamamoto and Arato2002). It reactivated a lithospheric weak zone, the development history of which started with Early Miocene back-arc rifting at the southern margin of the Japan Sea (Otofuji et al., Reference Otofuji, Hayashida, Torii, Nasu, Uyeda, Kushiro, Kobayashi and Kagami1985; Tamaki et al., Reference Tamaki, Suyehiro, Allan, Ingle and Pisciotto1992) and continued with Late Miocene–Early Pliocene inversion (Itoh and Nagasaki, Reference Itoh and Nagasaki1996).

In the case of the Atacama Fault, the earliest, Jurassic–Early Cretaceous, deformation was a system of both dip and sinistral strike-slip ductile shear zones in the magmatic arc, developing under amphibolite metamorphic conditions (Mpodozis and Ramos, Reference Mpodozis, Ramos, Ericksen, Cañas-Pinochet and Reinemund1990; Scheuber and Andriessen, Reference Scheuber and Andriessen1990; Brown et al., Reference Brown, Diaz and Grocott1993). These conditions developed in a transtensional trench-linked fault zone (Reutter et al., Reference Reutter, Heinsohn, Scheuber and Wigger1991).

This deformation was overprinted by brittle sinistral strike-slip faulting of the trench-linked strike-slip fault zone, which was located oceanward of the magmatic arc because this arc has migrated eastward since the Jurassic–Early Cretaceous (Arabasz, Reference Arabasz1971; Brown et al., Reference Brown, Diaz and Grocott1993). The fault zone was subsequently reactivated by late Oligocene–Early Miocene contraction (Olson, Reference Olson1989) and the Quaternary extension (Arabasz, Reference Arabasz1971; Naranjo, Reference Naranjo1987; Gonzáles et al., Reference Gonzáles, Dunai, Carrizo and Allmendinger2006).

The development of trench-linked strike-slip faults is related to strain partitioning from obliquely converging plates. Their kinematics are determined from systematic studies of earthquake slip vectors (Fitch, Reference Fitch1972; Jarrad, Reference Jarrad1986), GPS-derived displacement vectors (Bock et al., Reference Bock, McCaffrey, Rais and Murata1990; McCaffrey et al., Reference McCaffrey, Bock and Rais1990), plate convergence vectors estimated from poles of rotation, paleostresses calculated from outcrop fault-striae data (Nemčok, Reference Nemčok1993; Gayer et al., Reference Gayer, Hathaway, Nemčok, Holdsworth, Strachan and Dewey1998), and finite-element modeling (McCaffrey et al., Reference McCaffrey, Zwick and Bock2000). They are primarily controlled by the stress distribution along the interface between upper and lower plates (McCaffrey et al., Reference McCaffrey, Zwick and Bock2000). Considering the upper plate as being sufficiently heterogeneous to approach a material continuum, the most typical location for trench-linked strike-slip faults should be the upper plate above the down-dip boundary of the largest interplate stress. When such a system evolves, it should develop multiple strike-slip faults further oceanward (McCaffrey et al., Reference McCaffrey, Zwick and Bock2000). Secondary localization controls of trench-linked strike-slip faults include weak crustal and mantle zones related to the magmatic arc, and suitably oriented pre-existing weak deformation zones (McCaffrey et al., Reference McCaffrey, Zwick and Bock2000; Itoh et al., Reference Itoh, Tsutsumi, Yamamoto and Arato2002).

In the case of strongly arcuate convergent systems, such as the Carpathians advancing onto the European Platform, in which the subduction vector changed from SW-, to WSW-, to W-directed (e.g., Nemčok et al., Reference Nemčok, Pospíšil, Lexa and Donelick1998a), trench-linked strike-slip faults change their location inside the overall complex fault pattern in relatively short time intervals (Figure 1.10).

Reprinted by permission of the AAPG whose permission is required for further use.

Figure 1.10 Fault kinematics and timing in the ALCAPA microplate of the Carpathians for (a) Ottnangian, (b) Karpatian, (c) middle Badenian, and (d) Sarmatian (Nemčok et al., Reference Nemčok, Pogácsás, Pospíšil, Pícha and Golonka2006a). Thick black solid lines indicate main strike-slip faults, and thick black dashed lines indicate main normal faults active during their respective period. Ba – Balaton Fault Zone, Bu – Buda Fault Zone, Ce – Čertovica Fault Zone, CH – Central Hungarian Fault Zone, CJ – Cerna–Jiu Fault, CS – Central Slovakian Fault System, Da – Darnó Fault Zone, Dr – Drava Fault Zone, DV – Dragos–Voda Faults, EC – fault contact of External and Internal Flysch in the Eastern Carpathians, Ga – Gailtal Fault, HM – fault zones of the Central Hungarian Mountains, Hu – Hurbanovo Fault, La – Lavanttal Fault Zone, Ma – Malé Karpaty marginal faults, MM – Mur–Murz Fault Zone, Mu – Muráň Fault, OL – Ondava–Laborec Fault System, PB – faults of both Matra Mountains and north of the Pannonian Basin, PJ – Pottendorf–Jablonica Fault Zone, Po – Pohorelá Fault, Ra – Rába Fault Zone, Sa – Sava Fault Zone, SB – Schrattenberg–Bulhary Fault Zone, SC – Southern Carpathian Fault, SE – Salzachtal–Ennstal–Mariazell–Puchberg Fault System, Sl – Slánske Vrchy Fault System, So – Somes Fault, St – Steinberg Fault, Fa – Farské Fault Zone, ST – South Transylvanian Fault, Ta – Tatry Fault, Tr – Trebišov Fault System, Vi – Vihorlat Fault Zone. AAPG © 2006. Nemčok, M., Pogácsás, G., and Pospíšil, L. Activity timing of the main tectonic systems in the Carpathian–Pannonian region in relation to the roll-back destruction of the lithosphere. In: Pícha, F. and Golonka, J. (Eds.), The Carpathians and their foreland: geology and hydrocarbon resources, AAPG Memoir, 84, pp. 743–766.

Examples of hydrocarbon fields in this transform setting are found in the Central Sumatran Basin (Williams and Eubank, Reference Williams, Eubank and Lambiase1995), including the Balam, Beruk, Damar, Duri, Hitam, Kotabatak, Libo, Lingai, Minas, Pematang, Petapahan, Pungut, Suram, Tandun, Topaz, and Waduk oil and gas fields. They are usually associated with basin-dividing structural highs, and anticlines along their flanks, charged from lacustrine source rocks in surrounding structural lows.

1.4.1 Indent- and Lateral Extrusion-Linked Strike-Slip Faults

Indent- and lateral extrusion-linked strike-slip faults (Figure 1.1) are represented by strike-slip faults separating crustal blocks escaping from high to low potential energy conditions. They usually form complex fault arrays. They undergo geometric changes as the escaping block evolves.

Lateral extrusion-related strike-slip fault examples come from the Carpathian–Pannonian region. They were coeval with late Oligocene–Early Miocene large-scale lateral, west-to-east-directed material movement from the Eastern Alps into the remnant Carpathian Flysch Basin (Ratschbacher et al., Reference Ratschbacher, Frisch and Linzer1991a, Reference Ratschbacher, Merle, Davy and Cobboldb), which was affected by both extrusion from the west and subduction rollback from the northeast and east (Nemčok et al., Reference Nemčok, Pospíšil, Lexa and Donelick1998a; Sperner et al., Reference Sperner, Ratschbacher and Nemčok2002).

The Carpathian orogen developed by northeastward and eastward migration of an accretionary wedge in front of the ALCAPA and Tisza–Dacia microplates (Csontos et al., Reference Csontos, Nagymarosy, Horváth and Kováč1992). Extrusion-related faults accommodated separate movements of both microplates and their various internal crustal blocks (Figure 1.10). Their characteristic feature is their constant rearrangements into different regional patterns accompanied by changes of kinematics during the late Oligocene–Early Miocene time interval (Figure 1.10a), and their subsequent rearrangements and reactivation when Carpathian development was progressively dominated by subduction rollback during the Karpatian–Sarmatian time period (Nemčok et al., Reference Nemčok, Pogácsás, Pospíšil, Pícha and Golonka2006a; Figure 1.10b). Some of them were active during the Karpatian–Sarmatian time period as trench-parallel strike-slip faults. As the subduction kept changing into collision from west to east along the arc, and accretionary wedge large-scale advance kept ceasing from west to east, earlier strike-slip faults were undergoing reactivation as normal faults (Figures 1.10c–d).

Indented by the bend between the Eastern and Southern Carpathians, the Carpathian foreland underwent late Sarmatian–Meotian sinistral strike-slip (Figure 1.11), accompanied by subordinate thrusting (Matenco et al., Reference Matenco, Bertotti, Cloetingh and Dinu2003). Its development was relatively short-lived because the subduction–collision transition there was accompanied by subducting oceanic slab breakoff (e.g., Nemčok et al., Reference Nemčok, Pospíšil, Lexa and Donelick1998a).

Figure 1.11 Late Miocene faults of the Carpathian orogen and its indented foreland (modified from Matenco et al., Reference Matenco, Bertotti, Cloetingh and Dinu2003).

Reprinted from Sedimentary Geology, 156, Matenco, L., Bertotti, G., Cloetingh S., and Dinu, C. Subsidence analysis and tectonic evolution of the external Carpathian–Moesian Platform region during Neogene times 71–94. ©2003, with permission from Elsevier.

Former pull-apart basins located on both sides of the ALCAPA microplate, the Vienna and Danube Basins along its northwestern side, and Drava and Sava Basins along its southeastern side (e.g., Nemčok et al., Reference Nemčok, Pospíšil, Lexa and Donelick1998a), serve as examples of hydrocarbon-hosting basins in this type of strike-slip setting (e.g., Szalay and Koncz, Reference Szalay, Koncz and Spencer1993; Pogácsás et al., Reference Pogacsás, Mattick, Tari, Varnai, Teleki, Mattick and Kokay1994; INA, 2000a, b; Tari and Horváth, Reference Tari, Horváth, Golonka and Pícha2006). Because the pull-apart basins were short-lived features, subsequently disrupted by normal faulting (Nemčok et al., Reference Nemčok, Pogácsás, Pospíšil, Pícha and Golonka2006a), their source rocks reside in both syn- and post-pull-apart strata, and both are known to charge syn- and post-pull-apart reservoir rocks, as documented in the Danube Basin. Most typical fields, such as Beničanci, Mosti, and Šandrovac fields in the Drava Basin, and Dugo Selo, Jamarica, Mramor Brdo, Stružec, Žutica fields of the Sava Basin are hosted by faulted blocks, faulted anticlines, and anticlines.

Further examples of lateral extrusion-related faults must have been initial stages of the development of the Northern and Eastern Anatolian faults that were needed for the separation of the westward escaping Anatolian wedge from the rest of the collisional zone developing between Eurasia and the Arabian indenter. Here, one can see the interplay of both tectonic escape of crustal blocks from the collisional site to its lateral margins (e.g., Molnar and Tapponier, Reference Molnar and Tapponier1975) and extensional collapse, which causes thickened, topographically elevated, and, thus, gravitationally unstable continental crust to flow toward lower levels (e.g., England and Houseman, Reference England and Houseman1988).

The North and East Anatolian Faults do not qualify as intraplate faults in their present-day stage. Despite the gravity collapse of the overthickened collisional zone between Eurasia and the Arabian indenter representing a significant component of their controlling dynamics, they are both interpreted to be transforms. While the North Anatolian Fault represents a northern boundary of the escaping Anatolian Block, separating it from Eurasia, the East Anatolian Fault represents a segment of the boundary between Eurasia and Arabia (Figure 1.12).

Figure 1.12 The Chaman Transform fault linking the overridden plate of the collision zone with the overriding plate of the subduction zone, North Anatolian Fault representing a boundary between Anatolia and Eurasia, and East Anatolian Fault representing the boundary between Anatolia and Arabia.

(modified from Khan et al., Reference Khan, Bhatti, Baitu and Sarwar2009)

The Northern and Eastern Anatolian Faults accommodated the switch from the north-northwestward thrusting to the westward lateral escape of the Anatolian wedge (McKenzie, Reference McKenzie1972; Şengör and Yılmaz, Reference Şengör and Yılmaz1981; Dewey et al., Reference Dewey, Hempton, Kidd, Şaroğlu, Şengör, Coward and Ries1986). Timings for the escape quoted in the literature include Late Miocene (i.e. Sarmatian, 12–13 Ma) (Dewey and Şengör, Reference Dewey and Şengör1979; Şengör et al., Reference Şengör, Görür, Şaroğlu, Biddle and Christie-Blick1985; Hempton, Reference Hempton1987; Le Pichon et al., Reference Le Pichon, Chamot-Rooke, Lallemant, Noomen and Veis1995), Early Pliocene (5 Ma) (Tatar, Reference Tatar1975; Barka and Hancock, Reference Barka, Hancock, Dixon and Robertson1984; Barka and Kadinsky-Cade, Reference Barka and Kadinsky-Cade1988; Koçyiğit, Reference Koçyiğit1989; Barka, Reference Barka1992; Westaway, Reference Westaway1994) and Late Pliocene (2.5 Ma) (Şaroğlu, Reference Şaroğlu1988). Our interpretation of the last thrusting at the Pontides front in the Black Sea (see also evidence for Late Miocene sediments sealing the frontal thrusts in Kaymakci et al. (Reference Kaymakci, Inceöz, Ertepinar, Koç, Sosson, Kaymakci, Stephenson, Bergerat and Starostenko2010)), and analysis of timing constraints from the literature, points toward escape in the Sarmatian.

In the middle Neogene, portions of the northern and southern orogenic systems located to the east of the Black Sea were approaching each other, until their fronts met in Georgia and the collisional zone between Eurasia and Arabia progressively developed a considerably thickened continental crust. The Middle and Late Miocene fault activity maps (Nemčok et al., Reference Nemčok, Stuart and Odegard2009) indicate that the crust in the future Anatolian region must have been overthickened because it started to undergo extension. The extension first appeared in the easternmost portion of the future Anatolian wedge during the Middle Miocene (Figure 1.13a). It was accommodated by the NNE–SSW striking system of normal faults. The nucleation and development of these faults progressively extended toward the west until they deformed the whole present-day Anatolian region, their strikes varying aerially between NE–SW, NNE–SSW, and NW–SE (Figure 1.13b). During the Late Miocene, the normal faults formed a regional pattern located among the progressively developing strike-slip fault segments of the future Northern and Eastern Anatolian strike-slip zones separating the Anatolian wedge from its surroundings. This “preparation” of the overthickened crust for the lateral escape ended between the Sarmatian and Pliocene.

Figure 1.13 Middle Miocene (a) and Late Miocene (b) faults of the future Anatolian wedge region.

(Nemčok et al., Reference Nemčok, Stuart and Odegard2009)

Strike-slip faults bounding the laterally extruding lithospheric or crustal blocks are typically characterized by a displacement increase from the indentation point to the free interface. The Anatolian wedge serves as a good example. While the GPS-derived horizontal movement components in its eastern portion next to the Arabian indenter are about 18 mmy^–1, they are about 40 mmy^–1 in its western portion moving into the free interface (see Yilmaz, Reference Yilmaz2011).

The example of the Korudag Anticlinorium in this setting documents how this petroleum-bearing anticlinorium was shaped by multiple tectonic events. Some of them were related to the lateral extrusion-linked fault propagation. The subduction- and collision-related Early Miocene anticlinorium was further modified by the Middle Miocene propagation of the oldest splay of the North Anatolian Fault and Quaternary deformation along the North Anatolian Fault (Sen and Yillar, Reference Sen and Yillar2008). Both the anticlinorium and parallel anticlines around it host the Cayirdere, Degirmenkoy, Karacali, North Marmara, Sevindik, Seymen, Terkirdag, Vakiflar, and Yulafli fields.

1.4.2 Transfer Faults

Transfer faults offset extensional, contractional, or strike-slip structures. They have larger extent than tear faults, accommodating the transfer of slip between two neighboring similar structures. Transfer faults are the hard-linked version of more diffuse transfer structures, which include features such as relay ramps, linking transfer faults, en echelon step faulting, tilted fault blocks, and zigzag fault arrays in extensional terrains (Moustafa, Reference Moustafa2002).

When residing in an extensional terrain, strike-slip transfer faults provide a hard linkage of rift units of the same or opposite polarity and entire rift zones (Bosworth, Reference Bosworth1985). Their role is to accommodate differences in structural styles and strain amounts along the strike of the extensional system or the direction of extension (e.g., Gibbs, Reference Gibbs1984; Lister et al., Reference Lister, Etheridge and Symonds1986; Bosworth, Reference Bosworth and Lambiase1995; McClay and Khalil, Reference McClay and Khalil1998). Depending on their geometry with respect to the controlling stress regime, they can be divided into pure strike-slip, transtensional, and transpressional (Gibbs, Reference Gibbs1984; Lister et al., Reference Lister, Etheridge and Symonds1986).

The reason these faults have such variable kinematic regimes is that their propagation is frequently affected by pre-existing zones of weakness (Versfelt and Rosendahl, Reference Versfelt and Rosendahl1989). Examples of such propagation come from the Gulf of Suez–Red Sea rift system (Younes and McClay, Reference Younes and McClay2002; Figure 1.14a). Detailed interpretation of these faults has been performed on outcrops of pre-rift granites and foliated metamorphic rocks around the Darag, October, Zeit, and Duwi half-graben basins.

Figure 1.14 (a) Location map of the Darag, October, Zeit, and Duwi half-graben basins of the Gulf of Suez–Red Sea rift system. Black bold arrows indicate dip direction of half-grabens. Large white arrows indicate the direction of late Oligocene–Early Miocene extension. Solid lines are normal faults. Major Precambrian shear zones are shown in stippled lines. AZA – Zaafarana accommodation zone, MAZ – Morgan accommodation zone, DAZ – Duwi accommodation zone.

(Younes and McClay, Reference Younes and McClay2002)

As rifting in this area evolved, fault patterns propagated from pre-rift rocks to syn-rift sediments. North–south to northwest–southeast striking faults, which were perpendicular to the regional extension, had normal fault displacements (Angelier, Reference Angelier1985; Steckler et al., Reference Steckler, Berthelot, Lybéris and Le Pichon1988; Khalil, Reference Khalil1998). Most of them were not affected by the pre-existing anisotropy. Northeast–southwest and WNW–ESE striking faults were characterized by sinistral and dextral oblique-slip displacements (Prat et al., Reference Prat, Montenat, Ott d’Estevou and Bolze1986; Chorowicz, Reference Chorowicz1989; Abu El Karamat and Fouda, Reference Abu El Karamat and Fouda1990). Their development started with reactivation of the weak zones in the basement (Lybéris, Reference Lybéris1988; Jarriage et al., Reference Jarriage, Ott dÉstevou and Burollet1990; Younes, Reference Younes1996) prior to or coeval with propagation of the rift-bounding faults, and finished with arrest of the border faults and transfer of displacement (Younes and McClay, Reference Younes and McClay2002). The arrest of the propagating rift-bounding faults is documented by a diagram showing their decreasing throws as they approach the Hamrawin transfer fault from both sides (Younes and McClay, Reference Younes and McClay2002; Figure 1.14b). Both normal faults eventually die out at this fault.

Figure 1.14 (b) Fault displacement profile along the main fault system of the Duwi accommodation zone. The shaded area indicates an overlap area between the Duwi transfer zone and the northernmost tip of the Hamrawin shear zone. Displacement is measured from cross sections, taking into account the thickness variation of pre-rift strata and slope of the top of the basement. Note that these are minimum values, as in most places the pre-rift sedimentary section has been eroded from the footwall.

(Younes and McClay, Reference Younes and McClay2002)

The interaction of propagating rift-bounding faults with propagating transfer faults depends on the geometries and kinematics at their intersection (Younes and McClay, Reference Younes and McClay2002; Figure 1.14c). If their slip components were divergent, tensile stresses at the intersection become magnified, allowing the rift-bounding fault to kink or jog. The kinks have potential to evolve into transfer faults linking the main intersecting faults. If the slip components are convergent, the propagation of the rift-bounding fault either terminates abruptly at the reactivated shear zone or cuts through it.

Figure 1.14 (c) Kinematics of the reactivation of pre-existing Precambrian shear zones during late Oligocene–Early Miocene rifting in the Gulf of Suez–Red Sea region. Obtuse angles between rift-bounding faults and reactivated shear zones promote extension, acute ones contraction. See text for explanation.

(Younes and McClay, Reference Younes and McClay2002)

The Gulf of Suez–Red Sea examples show that transfer faults that reactivated pre-existing weak zones can have various kinematic characters that depend on the relationship of their controlling stress field and geometry of weak zones. However, transpressional transfer faults are relatively rare in nature.

An example of such a fault is the Jeremboabo sinistral transpressional fault in the failed Late Jurassic–Early Cretaceous Jatobá–Tucano–Recôncavo rift zone of northeast Brazil (Destro et al., Reference Destro, Szatmari, Alkmim and Magnavita2003). The fault is located on the flexural side of the rift zone, where it links smaller rift-bounding faults. Its displacement sense, determined from striated faults in outcrops of Berriasian–Valanginian syn-rift sediments, is consistent with the displacement sense of linked bounding faults on both sides. Its displacement was sinistral oblique, with a strong reverse component. The transfer character of the Jeremboabo Fault is further documented as based on its linkage with the major dextral Caritá transfer fault, located between rift units of the Jatobá–Tucano–Recôncavo rift zone that have opposite polarity. The reason for its relatively unusual transpressional character is the transfer fault reactivation of the pre-existing zone of weakness.

Transfer faults can be represented by single faults, such as those from the Gulf of Suez–Red Sea rift region (Younes and McClay, Reference Younes and McClay2002) or they can be represented by wide zones of strike-slip faults with pull-apart basins among them, as in those of the Gettysburg–Tarfaya transfer zone on US and Moroccan Central Atlantic margins (Nemčok et al., Reference Nemčok, Stuart, Segall, Post, Rosen and Olson2005a and references therein). A further well-documented case is the Garlock Fault in southern California, which functions as a transfer between the extended Basin and Range province in the north and the unextended Mojave Desert area in the south (Sylvester, Reference Sylvester1988).

When residing in contractional terrains, strike-slip transfer faults provide a hard linkage of thrust units of the same or opposite polarity, and can bound entire orogenic salients and reentrants. The Potwar Plateau in Pakistan, with its dextral Jhelum and sinistral Kala Bagh strike-slip faults, represents the first example (Figure 1.15a). The primary function of these two faults is to accommodate the south-southeastward advance of the orogenic salient, which is more than 100 km in front of the frontal thrusts of its neighbor thrust belt segments.

Figure 1.15 (a) Transfer faults bounding the Potwar Plateau, Pakistan (modified from Khan et al., Reference Khan, Ahmed, Raza and Arif1986; Quadri and Quadri, Reference Quadri and Quadri1996; Jaswal et al., Reference Jaswal, Lillie and Lawrence1997; Aamir and Siddiqui, Reference Aamir and Siddiqui2006; Moghal et al., Reference Moghal, Saqi, Hameed and Bugti2007).

Republished with permission of AAPG from Geology of Petroleum in Kohat-Potwar Depression, Pakistan. Khan, M. A., Ahmed, R., Raza, H., Arif, K. The American Association of Petroleum Geologists Bulletin, 70, 396–414. ©1986. Permission conveyed through Copyright Clearance Center, Inc. Republished with permission of AAPG from Structure and Evolution of the Northern Potwar Deformed Zone, Pakistan. Jaswal, T., Lillie, R. and Lawrence, R. The American Association of Petroleum Geologists, 70, 396–414. ©1997. Permission conveyed through Copyright Clearance Center, Inc. Republished with permission of the Society of Exploration Geophysicists from Interpretation and visualization of thrust sheets in a triangle zone in eastern Potwar, Pakistan. Aamir, M. and Siddiqui, M. M. The Leading Edge, 2206, 24–37. ©2006. Permission conveyed through Copyright Clearance Center, Inc.

The Sulaiman Range in Pakistan is a second example. The range is an orogenic salient, which required a relatively simple dextral strike-slip accommodation along its western side. However, its eastern side is much more complex and wider, represented by an entire system of sinistral strike-slip faults, oblique-slip faults, and thrusts. Various transpressional structures among these faults host a variety of condensate fields (Figure 1.15b). The largest strike-slip fault at the eastern side is the NNE–SSW striking fault that extends from the Savi Ragha field to the south of the Giandari well. Another important fault is the Kingri Fault, which has a NNW–SSE strike. The former separates the Zindapir Folded Zone – represented by gas-producing or oil-and-gas-producing anticlines such as Zindapir, Afiband, Rodho, and Dhodak – from the rest of the Sulaiman thrustbelt. Its southern compressional horse-tail structure is formed by a system of splaying-off anticlines, including the Loti and Sui gas fields and the dry Giandari anticline.

Figure 1.15 (b) Transfer faults bounding the Sulaiman Range, Pakistan.

(Nazeer et al., Reference Nazeer, Solangi and Brohi2013)

Other examples of hydrocarbon fields in this setting come from the northwest–southeast striking diffuse transfer zones that segment the Faroe–Shetland Basin. These zones include the Judd Fault and its smaller analogs. They help to set up some of the Paleocene fields. Farther north in the Barents Sea, a northwest–southeast striking transfer zone, generally considered to be a reactivated offshore extension of the Paleozoic Trollfjord–Komagelv Fault, provides a hard linkage between two different hydrocarbon provinces, the Hammerfest and Nordkapp Basins (e.g., Gabrielsen and Færseth, Reference Gabrielsen and Færseth1989).

1.4.3 Tear Faults

Tear faults represent the third group of strike-slip faults that are crustal deformational features. They represent relatively small-scale faults accommodating either inhomogeneous shortening or inhomogeneous extension along systems of primary faults and folds (Twiss and Moores, Reference Twiss and Moores2007). They are usually steep and parallel to the regional displacement vector.

Good natural examples of this type of fault come from thin-skinned thrust belts, such as the Jacksboro and Russel Fork Faults bordering the Pine Mountain thrust sheet in the Appalachians (Mitra, Reference Mitra1988a), multiple unspecified tear faults in the Jura Mountains (Laubscher, Reference Laubscher1972), multiple unspecified tear faults in the frontal thrust sheets of the southwestern British Variscan orogen in South Wales (Figure 1.16a), and multiple unspecified tear faults in the frontal portion of the Western Carpathians (Maheľ, Reference Maheľ1973).

Figure 1.16 (a) Geological map of the portion of the Variscan external margin in southwestern Britain showing a system of tear faults and Variscan tectonic transport directions at numbered outcrops. CCD – Careg Cennen Disturbance, LLD – Llanon Disturbance, ND – Neath Disturbance, SVD – Swansea Valley Disturbance, TD – Trimsaran Disturbance.

(Gayer et al., Reference Gayer, Hathaway, Nemčok, Holdsworth, Strachan and Dewey1998)

Tilting is typical of tear faults inside thrust sheets, together with tilting of the thrust sheet they segment. Examples include the sinistral strike-slip fault that segments the Forebalkan Unit of the Eastern Balkans (Figure 1.16b). A similar relationship characterizes tear faults segmenting hanging walls of listric normal faults.

Figure 1.16 (b) Sinistral tear fault (dip direction/dip-trend/plunge = 105°/90°–18°/16°) deforming the Paleocene Komarevo Formation of the Forebalkan Unit near Dobromir Village, Eastern Balkans.

As the hosting thrust sheet or hanging wall becomes inactive, tear faulting stops and these faults can passively rotate together with the sheet or hanging wall, resulting in unusual geometries, especially if the thrust sheet occupies the rear of the accretionary wedge where the thrust sheets passively rotate to vertical and even overturned positions.

Tear faults can have a significant modifying presence on petroleum systems in foreland fold-and-thrust belts. An excellent example of this role is seen in the southern sub-Andean fold-and-thrust of Bolivia (Pereira et al., Reference Pereira, Vergani, Cambon, Zamora, McClay and Ramos2018). Tear faults in this region are shown to act as both barriers and conduits for secondary migration. In some cases, they appear to control remigration of reservoired hydrocarbons from leaky thrust sheets to shallower levels in the Cenozoic succession, as seen for example in the Tajibo gas field, some 15 km west of the thrust-induced Charagua Range. The role of tear faults and other fractures that cut across the thrust belt is particularly interesting at this location, since most of the source rocks in the region reached their critical generation moment before the Andean orogeny, and remigration is, therefore, required to charge the newly developed structures.

1.4.5 Small-Scale Conjugate Strike-Slip Fault Systems in Individual Thrust Sheets

This type of strike-slip fault represents the fourth group of strike-slip faults that are crustal deformational features. Being relatively common in anticlines of thrust belts, they are Andersonian conjugate faults (Anderson, Reference Anderson1951) that accommodate coeval strike-parallel stretching and orthogonal shortening. Examples from the Sulaiman Range in Pakistan are shown in Figure 1.17.

Figure 1.17 A system of small-scale conjugate strike-slip faults in the frontal Zin-Loti anticline of the Sulaiman Range, Pakistan.

(Khan et al., Reference Khan, Bhatti, Baitu and Sarwar2009)

1.5.1 Structural Architecture of Strike-Slip Faults

Comparing results of numerous natural case studies and analog material experiments (Cloos, Reference Cloos1928; Riedel, Reference Riedel1929; Hills, Reference Hills1963; Tchalenko, Reference Tchalenko1970; Tchalenko and Ambraseys, Reference Tchalenko and Ambraseys1970; Sylvester and Smith, Reference Sylvester and Smith1976; Harding and Lowell, Reference Harding and Lowell1979; Bartlett et al., Reference Bartlett, Friedman and Logan1981; Christie-Blick and Biddle, Reference Christie-Blick, Biddle, Biddle and Christie-Blick1985; Naylor et al., Reference Naylor, Mandl and Sijpesteijn1986), a list can be made of structural elements of strike-slip fault zones (Figure 1.18). It includes the following:

1. 1) synthetic strike-slip faults – R (Riedel) shears;

2. 2) fault splays at ends of R shears;

3. 3) antithetic strike-slip faults – R′ (conjugated Riedel) shears;

4. 4) secondary synthetic strike-slip faults – P shears;

5. 5) tensile fractures – T fractures;

6. 6) normal faults;

7. 7) folds;

8. 8) antithetic strike-slip faults – X shears symmetric with R′ shears;

9. 9) shear lenses having long axes parallel to principal displacement zone; and

10. 10) strike-slip faults parallel with principal displacement zone – Y shears.

Mechanical theory for simple shear (Cloos, Reference Cloos1928; Riedel, Reference Riedel1929), using Coulomb’s criterion, allows one to calculate the exact geometry of individual shears with respect to the principal displacement zone, as controlled by the angle of internal friction of the deforming material, φ. Riedel shears are located at an angle of φ/2 from the direction of movement, oriented by the acute angle toward the movement direction. R′ shears lie at an angle of 90° – φ/2 from the movement direction. P shears are propagated at an angle of φ/2 from the direction of movement, being symmetrical to R shears.

The aforementioned analog material experiments and case studies have documented how strike-slip faults progressively develop by a sequence of deformation events, developing different architectural elements from the list above (Figure 1.18a). Each new generation of shears is better suited for the accommodation of the involved block displacements – that is, they are capable of taking over a larger proportion of the overall displacement. Progressive linkage of architectural elements eventually results in a system of isolated lenses of passive material roughly parallel to the principal displacement zone. After this stage, the strike-slip fault zone reaches the residual deformation stage, which takes place at reduced friction, and almost all its displacements are accommodated along a single principal displacement zone. For any normal stress, the shear stress, which was at its peak at the Riedel shear development stage, has decreased to a lower value sufficient for overcoming the residual strength of the fault zone.

Figure 1.18 (a) Development of pure strike-slip fault (Naylor et al., Reference Naylor, Mandl and Sijpesteijn1986).

Reprinted from the Journal of Structural Geology, 8(7), M. A Naylor, G. Mandl, C. H. K Supesteijn. Fault geometries in basement-induced wrench faulting under different initial stress states. 737–752. ©1986, with permission from Elsevier.

Folds associated with strike-slip faults are arranged in an en echelon pattern with respect to the principal displacement zone. They plunge in both directions away from the principal displacement zone, and have steeply dipping axial planes. They commonly develop in the sedimentary cover, decoupled from the basement deformed by the strike-slip fault. As the strike-slip fault propagation continues into the sedimentary cover, their two halves get cut and displaced.

A progressive addition of new generations of architectural elements in developing strike-slip fault zones is frequently affected by reactivation of pre-existing zones of weakness. Even the previously created architectural elements of the developing zone represent an anisotropy, which has a progressively growing influence on the evolution of the strike-slip fault system (Woodcock and Fischer, Reference Woodcock and Fischer1986).

Linkage of various types of shears results in upward-diverging shears linked into a relatively narrow zone toward the trace of the strike-slip fault zone at depth (Christie-Blick and Biddle, Reference Christie-Blick, Biddle, Biddle and Christie-Blick1985). Such features are best imaged in reflection seismic profiles and are known as flower structures (Gregory in Harding and Lowell, Reference Harding and Lowell1979) or palm-tree structures (Sylvester and Smith, Reference Sylvester and Smith1976; Figure 1.18b).

Figure 1.18 (b) Line drawing of the NNE–SSW striking reflection seismic profile through the Elk Hills anticline, Southern San Joaquin Basin, showing a positive flower structure.

(Nemčok and Schamel, Reference Nemčok and Schamel2000)

Because a large strike-slip fault zone grows by the nucleation and linkage of various architectural elements developing at various locations along its strike, the zone usually contains multiple strike-slip fault segments. They have different relationships to the overall displacement and controlling stresses and include (Crowell, Reference Crowell and Dickinson1974; Woodcock and Fischer, Reference Woodcock and Fischer1986; Gamond, Reference Gamond1987; Figure 1.19a):

1. 1) releasing bends;

2. 2) restraining bends;

3. 3) strike-slip duplexes;

4. 4) strike-slip fans or horse-tail structures;

5. 5) releasing offsets;

6. 6) restraining offsets;

7. 7) contractional bridges; and

8. 8) extensional bridges.

Figure 1.19 (a) Structures in the strike-slip fault (Woodcock and Fischer, Reference Woodcock1986).

Reprinted from the Journal of Structural Geology, 8, Woodcock, N. H. and Fischer, M. Strike-slip duplexes. 725–735. ©1986, with permission from Elsevier.

All are characterized by a distinct vertical displacement amplitude (Woodcock and Fischer, Reference Woodcock and Fischer1986). As documented by the Dasht-e-Bayaz Fault case study (Tchalenko and Ambraseys, Reference Tchalenko and Ambraseys1970), releasing and restraining areas are characterized by subsidence and uplift, respectively. The Transverse Ranges to the north of Los Angeles along the San Andreas Fault are an example of such a restraining bend.

Strike-slip duplexes – that is, fault-bound lenses imbricated between overlapping en echelon fault segments – do not function like duplexes in thrust belts or extensional terrains. In the case of thrust and extensional duplexes, the plane strain is maintained in the vertical plane by the deformational reactions of the surface to the thickening or thinning of the duplex. However, in the case of strike-slip faults, the analogous maintenance of plane strain in the horizontal plane is accompanied by lateral deformation of surrounding rocks. The plane strain in this case breaks down due to the vertical response of surrounding areas. Furthermore, vertically oriented strike-slip faults lack the gravity potential perpendicular to the fault plane that influences the sequence of fault development in duplexes with vertical movement.

Duplexes of systems with vertical movements usually maintain a similar character along one main fault zone, while both types of duplexes can coexist in a strike-slip fault zone. This is because duplexes of strike-slip fault zones can be created by several different kinematic processes (Woodcock and Fischer, Reference Woodcock and Fischer1986; Swanson, Reference Swanson1988, Reference Swanson1989, Reference Swanson2005; Marsh et al., Reference Marsh, Pachan and Sims1990), while duplexes of thrust terrains are usually the result of a single process – that is, progressive footwall collapse taking place in the direction of the hanging wall movement.

Because of the zero volume change, shortening of a contractional duplex must be balanced by growth of its width, and lengthening of the extensional duplex must be balanced by reduction of its width (Hempton and Neher, Reference Hempton and Neher1986; Woodcock and Fischer, Reference Woodcock and Fischer1986). These shape changes of growing duplexes require accommodation in surrounding rocks, represented by uplift and subsidence of adjacent blocks.

Duplexes can travel along the principal displacement zone. Extreme examples of such large displacement, exotic duplexes, are analogous to far-traveled blocks in thrust terrains (Woodcock and Fischer, Reference Woodcock and Fischer1986). The Ancenis Terrane in the Hercynian Belt of Armorica provides an example of such an exotic duplex that has traveled along the dextral northern boundary of Gondwana (Shelley and Bossiere, Reference Shelley and Bossiere2001).

Duplexes, depending on their character, are characterized by different flower structures in reflection seismic images. Restraining duplexes create a positive flower structure with upward convex reflectors among faults (Figure 1.18b) and releasing duplexes create a negative flower structure with upward concave reflectors among faults. An example of the former occurs along the Atacama fault system in northern Chile (Cembrano et al., Reference Cembrano, Gonzales and Arancibia2005). Examples of the latter occur in the Lambertville sill in New Jersey (Laney and Gates, Reference Laney and Gates1996) and in the oil field-hosting Quealy duplex (Stone, Reference Stone1995).

Kinematic development of a strike-slip fault zone requires the presence of displacement dissipation structures at both ends. The most typical structural patterns are horse-tail structures, known under various names (e.g., Wilcox et al., Reference Wilcox, Harding and Seely1973; Crowell, Reference Crowell and Dickinson1974; Christie-Blick and Biddle, Reference Christie-Blick, Biddle, Biddle and Christie-Blick1985; Woodcock and Fischer, Reference Woodcock and Fischer1986; Sylvester, Reference Sylvester1988; Figure 1.19a). A good example of a mesoscale horse-tail structure associated with dextral strike-slip in the Lower Jurassic section of the Bristol Channel is shown in Figure 1.19b. Large-scale examples are associated with the Lavic Lake Fault in California and several faults in offshore Guyana (Figure 1.19c).

Figure 1.19 (b) Two horse-tail structures on receding sides of two dextral strike-slip faults the Liassic section in the Bristol Channel, UK.

Figure 1.19 (c) Horse-tail structures on the retreating sides of sinistral strike-slip faults at Guyana Transform margin imaged by the 3D seismic volume. Their horse-tail structures contain numerous small pull-apart basins. Mapped horizon represents the top of the strata having the age somewhere between the upper Barremian and the boundary between the Aptian and Albian. Note that most of the strike-slip faults reached only early stages of their evolution.

(Nemčok et al., Reference Nemčok, Rybár, Odegard, Nemčok, Rybár, Sinha, Hermeston and Ledvényiová2015a)

Not all naturally occurring strike-slip fault zones have pure strike-slip displacement. Depending on the relationship of their geometry to the controlling stress field, they can be divided into (Wilcox et al., Reference Wilcox, Harding and Seely1973; Naylor et al., Reference Naylor, Mandl and Sijpesteijn1986):

1. 1) pure strike-slip faults;

2. 2) transtensional (divergent) strike-slip faults; and

3. 3) transpressional (convergent) strike-slip faults.

Analogously to the aforementioned types of strike-slip duplexes, their flower structures vary (Harding et al., Reference Harding, Vierbuchen, Christie-Blick, Biddle and Christie-Blick1985). Transtensional strike-slip fault zones are characterized by negative flower structures with dominating synforms. Transpressional ones are characterized by positive structures with prevalent antiforms.

Analog material experiments (Naylor et al., Reference Naylor, Mandl and Sijpesteijn1986) have shown that these three types of strike-slip faults grow by the nucleation and linkage of different combinations and sequences of architectural elements. This results in their different character, allowing one to recognize them in reflection seismic images.

In the case of a pure strike-slip fault, the principal stresses of the controlling stress field can be characterized as ${\mathrm{\sigma }}_{\mathrm{H}1}={\mathrm{\sigma }}_{\mathrm{H}2}<{\mathrm{\sigma }}_{\mathrm{V}},$ where σ_H1 and σ_H2 are horizontal and σ_V is vertical. The resultant architecture of this type is shown in Figure 1.18a.

In the case of a transtensional strike-slip fault, the controlling stress field can be characterized by ${\mathrm{\sigma }}_{\text{HMax}}>{\mathrm{\sigma }}_{\mathrm{V}}>{\mathrm{\sigma }}_{\text{HMin}}$. In this setting, the σ₁ stress is parallel to the principal displacement zone in the initial stage of the experiment. The finite strain in this setting is partitioned between the simple shear component representing the strike-slip faulting regime along the fault and the pure shear component representing the extension perpendicular to the fault.

The transtensional strike-slip fault is characterized by the simplest and straightest fault zone (Figure 1.20). R shears in en echelon arrangement are rare. A simple, long strike-slip fault is common. It propagates soon after a small displacement of the basal plates of the experimental apparatus. Minor branching of the fault zone can occur, associated with propagation of P shears. Shears in vertical cross section have inclinations larger than 70°. Their traces in the section are usually straight. The cross section through the zone usually contains only a single fault or a single fault can branch into two faults in close proximity to the surface.

Figure 1.20 Transtensional strike-slip fault (Naylor et al., Reference Naylor, Mandl and Sijpesteijn1986), characterized by (a) map views and (b) cross sectional views.

Reprinted from the Journal of Structural Geology, 8(7), M. A. Naylor, G. Mandl, C. H. K. Supesteijn. Fault geometries in basement-induced wrench faulting under different initial stress states. 737–752. ©1986, with permission from Elsevier.

Transtension is also known to have caused topographic depressions parallel to the fault strike. Narrow elongated basins in the Tomales Bay north of San Francisco, and in the Buckeye and Meadow Parks in southern San Francisco, are good examples.

In the case of transpressional strike-slip faults, the controlling stress field can be characterized by σ_HMax > σ_V > σ_HMin. In this setting, the σ₁ stress is perpendicular to the principal displacement zone in the early stage of the experiment. The finite strain in this setting is partitioned between the simple shear component representing the strike-slip faulting regime along the fault and the pure shear component representing the compression perpendicular to the fault (Sanderson and Marchini, Reference Sanderson and Marchini1984).

The transpressional strike-slip fault is characterized by a wide and complex fault zone (Figure 1.21). The first R shears are at angles of more than 60° to the principal displacement zone. They can rotate around thrusts, which are parallel to the principal displacement zone. Increasing displacement sequentially generates shears oriented at progressively smaller angles to the principal displacement zone. The final zone eventually also includes P shears. The faults in cross section have dips around 25° in the vicinity of the base and 40–90° on the surface. The helicoidal aspect of R shears is, thus, very enhanced.

Figure 1.21 Transpressional strike-slip fault (Naylor et al., Reference Naylor, Mandl and Sijpesteijn1986), characterized by (a) map views and (b) cross sectional views.

Reprinted from the Journal of Structural Geology, 8(7), M. A. Naylor, G. Mandl, C. H. K. Supesteijn. Fault geometries in basement-induced wrench faulting under different initial stress states. 737–752. ©1986, with permission from Elsevier.

Transpression is also known to have caused topographic highs parallel to the fault strike. Narrow elongated pressure ridges along the San Andreas Fault in San Luis Obispo County and along the Lavic Lake Fault in California are good examples.