Abstract For as long as geologists have looked at deformed rocks, they have grappled to understand the mechanical origins of deformation. Natural systems are inherently complex so that, for many, purely geometric and kinematic approaches have sufficed. However, we know that stress within the brittle upper crust controls the nucleation, growth and reactivation of faults and fractures, induces seismic activity, affects the transport of magma and modulates structural permeability, thereby influencing the redistribution of hydrothermal and hydrocarbon fluids. An endeavour of structural geology and seismotectonics is therefore to reconstruct states of stress and their evolution over geological time from observations of the final products of rock deformation. Experimentalists endeavour to recreate structures observed in nature under controlled stress conditions. Earth scientists studying earthquakes attempt to monitor or deduce stress changes in the Earth as it actively deforms. All are building upon the pioneering researches and concepts of Ernest Masson Anderson dating back to the start of the 20th century. His insights, encapsulated in a small number of research papers and in the book The Dynamics of Faulting and Dyke Formation with Applications to Britain , continue to influence investigations in structural geology, seismology, rock mechanics, processes of hydrothermal mineralization and physical volcanology. This volume celebrates this legacy.

Abstract The initial M w 7.1 Darfield earthquake sequence was centred west of Christchurch City in the South Island of New Zealand but aftershocks, including a highly destructive M w 6.3 event, eventually extended eastwards across the city to the coast. The mainshock gave rise to right-lateral strike-slip of up to 5 m along the segmented rupture trace of a subvertical fault trending 085±5° across the Canterbury Plains for c . 30 km, in agreement with teleseismic focal mechanisms. Near-field data however suggest that the mainshock was composite, initiating with reverse-slip north of the surface rupture. Stress determinations for the central South Island show maximum compressive stress σ 1 to be horizontal and oriented 115±5°. The principal dextral rupture therefore lies at c . 30° to regional σ 1 , the classic ‘Andersonian’ orientation for a low-displacement wrench fault. An aftershock lineament trending c . 145° possibly represents a conjugate left-lateral strike-slip structure. This stress field is also consistent with predominantly reverse-slip reactivation of NNE–NE faults along the Southern Alps range front. The main strike-slip fault appears to have a low cumulative displacement and may represent either a fairly newly formed fault in the regional stress field, or an existing subvertical fault that happens to be optimally oriented for frictional reactivation.

Abstract Plate boundary forces acting within the Cocos Plate that is being subducted at a rate of 8.5–9.0 cm a −1 towards N32°E below the Caribbean Plate and the Panama microplate are found responsible for contemporaneous superimposed compressive, wrench and extensive fault patterns in Central Costa Rica. The stress inversion of fault-slip planes and focal mechanisms reveals a prevailing convergence-imposed N20°–45°E almost horizontal compression. Ellipsoid R values [ R =(σ 1 –σ 2)/(σ 2 –σ 3)] in the range of 0.3–0.05 and 0.8–0.93 are responsible for the permutation of σ 2 to σ 3 and σ 2 to σ 1 , respectively, and show typical Andersonian configurations with one stress axis vertical or close to it. Coulomb failure stress (CFS) analysis reveals that up to 5 bars (0.5 MPa) of tectonic loading are being imposed on east–west thrusts and on critically oriented conjugate NW- and NE-trending strike-slip faults. Non-optimally oriented structures are potential targets for reactivation even with 2 bars (0.2 MPa) of load. Triggering and interaction with volcanic activity is highly suspected in one documented recent case. When the regional fault population was tested for its slip tendency (τ / σ n), a good correlation with CFS results was found.

Abstract A dip histogram for intracontinental M >5.5 reverse-slip ruptures reveals a trimodal distribution with a dominant Andersonian peak (fault dip, δ =30±5°) flanked by subsidiary clusters at δ =10±5° and 50±5°, and no dips greater than 60°. For a simple compressional regime (σ v = σ 3), the dominant peak is in accord with the reshear of optimally oriented faults with a friction coefficient of μ s =0.6±0.2, implying frictional lock-up at δ =60±10° consistent with the observed upper dip bound. The low-dip cluster (δ =10±5°) is dominated by thrusting in the frontal Himalaya and may incorporate staircase thrust systems in cover sequences with deflections along bedding anisotropy. The cluster of moderate-to-steep reverse fault ruptures (δ =50±5°) is likely dominated by compressional inversion of inherited normal faults. In both circumstances, however, there appears to be competition between Andersonian thrusts in various stages of development and non-optimal failure planes dipping at either high or low angles. A delicate balance between levels of differential stress and fluid-pressure determines whether or not a poorly oriented thrust or reverse fault reactivates in preference to the development of new, favourably oriented Andersonian thrusts.

Abstract Compressional tectonic inversions are classically represented in 2D brittle failure mode (BFM) plots that illustrate the change in differential stress (σ 1 − σ 3) versus the pore-fluid pressure during orogenic shortening. In these BFM plots, the tectonic switch between extension and compression occurs at a differential stress state of zero. However, mostly anisotropic conditions are present in the Earth's crust, making isotropic stress conditions highly questionable. In this study, theoretical 3D stress-state reconstructions are proposed to illustrate the complexity of triaxial stress transitions during compressional inversion of Andersonian stress regimes. These reconstructions are based on successive late burial and early tectonic quartz veins which reflect early Variscan tectonic inversion in the Rhenohercynian foreland fold-and-thrust belt (High-Ardenne Slate Belt, Belgium, Germany). This theoretical exercise predicts that, no matter the geometry of the basin or the orientation of shortening, a transitional ‘wrench’ tectonic regime should always occur between extension and compression. To date, this intermediate regime has never been observed in structures in a shortened basin affected by tectonic inversion. Our study implies that stress transitions are therefore more complex than classically represented in 2D. Ideally, a transitional ‘wrench’ regime should be implemented in BFM plots at the switch between the extensional and compressional regimes.

Abstract The Ceduna Sub-basin is located within the Bight Basin on the Australian southern margin. Recent structural analysis using newly acquired two-dimensional (2D) and three-dimensional (3D) seismic data demonstrates two Late Cretaceous delta–deepwater fold–thrust belts (DDWFTBs), which are overlain by Cenozoic sediments. The present-day normal fault stress regime identified in the Bight Basin indicates that the maximum horizontal stress (S Hmax) is margin parallel; Andersonain faulting theory therefore suggests the delta-top extensional faults are oriented favourably for reactivation. A breached hydrocarbon trap encountered in the Jerboa-1 well demonstrates this fault reactivation. Faults interpreted from 3D seismic data were modelled using the Poly3D © geomechanical code to determine the risk of reactivation. Results indicate delta-top extensional faults that dip 40–70° are at moderate–high risk of reactivation, while variations in the orientation of the fault planes results in an increased risk of reactivation. Two pulses of inversion are identified in the Ceduna Sub-basin and correlate with the onset of rifting and fault reactivation in the Santonian. We propose a ridge-push mechanism for this stress which selectively reactivates extensional faults on the delta-top, forming inversion anticlines that are prospective for hydrocarbon exploration.

Abstract Parts of the Australian continent, including the Otway Basin of the southern Australian margin, exhibit unusually high levels of neotectonic deformation for a so-called stable continental region. The onset of deformation in the Otway Basin is marked by a regional Miocene–Pliocene unconformity and inversion and exhumation of the Cretaceous–Cenozoic basin fill by up to c . 1 km. While it is generally agreed that this deformation is controlled by a mildly compressional intraplate stress field generated by the interaction of distant plate-boundary forces, it is less clear whether the present-day record of deformation manifested by seismicity is representative of the longer-term geological record of deformation. We present estimates of strain rates in the eastern Otway Basin since 10 Ma based on seismic moment release, geological observations, exhumation measurements and structural restorations. Our results demonstrate significant temporal variation in bulk crustal strain rates, from a peak of c . 2×10 −16 s −1 in the Miocene–Pliocene to c . 1.09×10 −17 s −1 at the present day, and indicate that the observed exhumation can be accounted for solely by crustal shortening. The Miocene–Pliocene peak in tectonic activity, along with the orthogonal alignment of inverted post-Miocene structures to measured and predicted maximum horizontal stress orientations, validates the notion that plate-boundary forces are capable of generating mild but appreciable deformation and uplift within continental interiors.

Abstract Overpressure within a circular magmatic chamber embedded in an elastic half space is a widely used model in volcanology. However, this overpressure is generally assumed to be bounded by the bedrock tensile strength since gravity is neglected. Critical overpressure for wall failure is thus greater. It is shown analytically and numerically that wall failure occurs in shear rather than in tension, because the Mohr–Coulomb yield stress is less than the tensile yield stress. Numerical modelling of progressively increasing overpressure shows that bedrock failure develops in three stages: (1) tensile failure at the ground surface; (2) shear failure at the chamber wall; and (3) fault connection from the chamber wall to the ground surface. Predictions of surface deformation and stress with the theory of elasticity break down at stage 3. For wall tensile failure to occur at small overpressure, a state of lithostatic pore-fluid pressure is required in the bedrock which cancels the effect of gravity. Modelled eccentric shear band geometries are consistent with theoretical solutions from engineering plasticity and compare well with shear structures bordering exhumed intrusions. This study shows that the measured ground surface deformation may be misinterpreted when neither plasticity nor pore-fluid pressure is accounted for. Supplementary material: The numerical benchmark data are available at: http://www.geolsoc.org.uk/SUP18517 .

Abstract A common feature of thrust-related anticlines developing in thrust wedges is the presence of extensional structures paralleling the fold axial trend (i.e. longitudinal structures). These form in response to hinge-perpendicular stretching and indicate that the minimum stress component orients parallel to the regional shortening direction, that is, the maximum and minimum stress components locally invert. Under the assumption that the regional stress component paralleling the shortening direction is almost constant during folding, such an inversion requires a large local stress drop. In the thrust-related Boltaña anticline, folding was both predated and accompanied by the development of longitudinal extensional faults. Meso-scale contractional structures are only rarely found. Where contractional structures are present, cross-cutting relationships indicate that these structures episodically developed within a mainly extensional framework which was established during flexural bending of the foredeep and continued during thrust-related folding. We conclude that the coexistence of extensional and compressional structures relates to a stress fluctuation, with normal faulting and discontinuous fold growth occurring during ‘extensional’ and ‘compressional’ stages, respectively.

Abstract Delta–deepwater fold–thrust belts are linked systems of extension and compression. Margin-parallel maximum horizontal stresses (extension) on the delta top are generated by gravitational collapse of accumulating sediment, and drive downdip margin-normal maximum horizontal stresses (compression) in the deepwater fold–thrust belt (or delta toe). This maximum horizontal stress rotation has been observed in a number of delta systems. Maximum horizontal stress orientations, determined from 32 petroleum wells in the Gulf of Mexico, are broadly margin-parallel on the delta top with a mean orientation of 060 and a standard deviation of 49°. However, several orientations show up to 60° deflection from the regional margin-parallel orientation. Three-dimensional (3D) seismic data from the Gulf of Mexico delta top demonstrate the presence of salt diapirs piercing the overlying deltaic sediments. These salt diapirs are adjacent to wells (within 500 m) that demonstrate deflected stress orientations. The maximum horizontal stresses are deflected to become parallel to the interface between the salt and sediment. Two cases are presented that account for the alignment of maximum horizontal stresses parallel to this interface: (1) the contrast between geomechanical properties of the deltaic sediments and adjacent salt diapirs; and (2) gravitational collapse of deltaic sediments down the flanks of salt diapirs.

Abstract This study examines present-day stress orientations from borehole breakout and drilling-induced fractures in 57 boreholes in the Nile Delta. A total of 588 breakouts and 68 drilling-induced fractures from 50 wells reveal sharply contrasting present-day maximum horizontal stress (S Hmax) orientations across the Nile Delta. A typical deltaic margin-parallel S Hmax exists in parts of the Nile Delta that are below or absent from evaporites (NNE–SSW in the west, east–west in the central Nile, ESE–WNW in the east). However, a largely margin-normal (NNE–SSW) S Hmax is observed in sequences underlain by evaporites in the eastern Nile Delta. The margin-normal supra-salt S Hmax orientations are often subperpendicular to the strike of nearby active extensional faults, rather than being parallel to the faults as predicted by Andersonian criteria. The high angle between S Hmax and strike of these extensional faults represents a new type of non-Andersonian faulting that is even less-suitably oriented for shear failure than previously described anomalous faulting such as low-angle normal faults and highly oblique strike-slip faults (e.g. San Andreas). While the mechanics of these non-Andersonian faults remains uncertain, it is suggested that the margin-normal supra-salt orientation generated by basal forces imparted upon rafted blocks sliding down seawards-dipping evaporites.

Abstract Delta–deepwater fold–thrust belts (DDWFTBs) develop over low-angle detachment faults which link extension to downslope contraction. Detachment faults have been examined in previous studies for the Amazon Fan, Niger, Nile, Angola, Baram and Bight Basin DDWFTBs. The driving mechanisms for the movement along the detachment remain uncertain, however. Previous authors have attributed the movement along detachment faults to high pore-fluid pressure, which reduces the effective normal stress acting on a fault surface thereby encouraging sliding along the fault. However, high pore-fluid pressure has not been directly confirmed in many of these faults due to a lack of well data in detachment surfaces. In this study, finite element modelling was used to test the effects of pore-fluid pressure, coefficient of friction, sediment rigidity and sediment wedge angle on sliding along the detachment. The modelling suggests that increased pore-fluid pressures and decreased coefficients of friction increase slip along a detachment. At hydrostatic pore-fluid pressures, sediment rigidity and sediment wedge angle have relatively little effect on the movement of the sediment wedge along the detachment. Modelling of these conditions using ABAQUS™ improves our understanding of the nature and mechanics of DDWFTBs and their underlying detachments.

Abstract The weakness of fault zones is generally explained by invoking an elevated fluid pressure or the presence of extremely weak minerals in a continuous fault gouge horizon. This allows for faults to slip under an unfavourable normal to shear stress ratio, in contrast to E. M. Anderson's theory of faulting. However, these mechanisms do not explain why faults should nucleate in such an orientation as to make them misoriented and non-Andersonian. Here we present a weakening mechanism, involving the mechanical anisotropy of phyllosilicate-bearing mylonite belts, which is likely to influence the nucleation of faults in addition to their subsequent activity. Considering three natural examples from the Alps (the Simplon, Brenner and Sprechenstein-Mules fault zones) and a review of laboratory tests on anisotropic rocks, we apply anisotropic slip tendency analysis and show that misoriented weak faults can nucleate along a sub-planar phyllosilicate-rich mylonitic foliation, constituting a large-scale mechanical anisotropy belt and preventing the development of Andersonian optimally oriented faults.

Abstract The Law of Effective Stress has found wide application in structural geology, rock mechanics and petroleum geology. The commonly used form of this law relies on an assumption of isotropic porosity. The porosity in and around fluid-saturated fault zones is likely to be dominated by tectonically induced cracks of various shapes and sizes. Previously published field and laboratory data show that these cracks occur in distinct patterns of preferred orientation, and that these patterns vary around the fault zone. This paper uses the more general form of the Law of Effective Stress which incorporates anisotropic poroelasticity to model the geomechanical response of fault zones surrounded by patterns of oriented cracks. Predictions of fault stability in response to fluid pressure changes are shown to depend on both the nature (or symmetry) of the crack pattern and the orientation of the crack patterns with respect to the in situ stress. More complete data on the porosity of natural fault zones will enable more accurate predictions of fault stability in the subsurface.

Abstract The dilatancy–diffusion hypothesis was one of the first attempts to predict the form of potential geophysical signals that may precede earthquakes, and hence provide a possible physical basis for earthquake prediction. The basic hypothesis has stood up well in the laboratory, where catastrophic failure of intact rocks has been observed to be associated with geophysical signals associated both with dilatancy and pore pressure changes. In contrast, the precursors invoked to determine the predicted earthquake time and event magnitude have not stood up to independent scrutiny. There are several reasons for the lack of simple scaling between the laboratory and the field scales, but key differences are those of scale in time and space and in material boundary conditions, coupled with the sheer complexity and non-linearity of the processes involved. ‘Upscaling’ is recognized as a difficult task in multi-scale complex systems generally and in oil and gas reservoir engineering specifically. It may however provide a clue as to why simple local laws for dilatancy and diffusion do not scale simply to bulk properties at a greater scale, even when the fracture system that controls the mechanical and hydraulic properties of the reservoir rock is itself scale-invariant.

Abstract IT has been known for long that faults arrange themselves naturally into different classes, which have originated under different conditions of pressure in the rock mass. The object of the present paper is to show a little more clearly the connection between any system of faults and the system of forces which gave rise to it. It can be shown mathematically that any system of forces, acting within a rock which for the time being is in equilibrium, resolves itself at any particular point into three pressures or tensions (or both combined), acting across three planes which are at right angles to one another.

Abstract Geologists have long grappled with understanding the mechanical origins of rock deformation. Stress regimes control the nucleation, growth and reactivation of faults and fractures; induce seismic activity; affect the transport of magma; and modulate structural permeability, thereby influencing the redistribution of hydrothermal and hydrocarbon fluids. Experimentalists endeavour to recreate deformation structures observed in nature under controlled stress conditions. Earth scientists studying earthquakes will attempt to monitor or deduce stress changes in the Earth as it actively deforms. All are building upon the pioneering research and concepts of Ernest Masson Anderson, dating back to the start of the twentieth century. This volume celebrates Anderson’s legacy, with 14 original research papers that examine faulting and seismic hazard; structural inheritance; the role of local and regional stress fields; low angle faults and the role of pore fluids; supplemented by reviews of Andersonian approaches and a reprint of his classic paper of 1905.

ABSTRACT Intracontinental reverse-fault ruptures (magnitude, M > 5.5) in the upper, seismogenic crust have dips, d, that lie in the range 10-60° and are broadly compatible with expectations from laboratory measurements of rock friction. Two distinct peaks occur, at δ = 30 ±5° and δ = 50 ±5°. Although the lower peak, at d δ≈ 30°, corresponds to expected optimal orientation for frictional reactivation under horizontal compression and may be attributed to ramp failure, the peak at d δ ≈ 50° is attributed to the compressional reactivation of inherited normal faults during positive inversion. A notable contrast with active subduction earthquakes and a data set from the Himalayan frontal thrust system is a general scarcity of low-angle thrusts with δ ≈10°. Frictional-mechanics analysis of reverse-fault reactivation endorses long-standing suggestions that active thrust systems are likely to be fluid overpressured. However, such analysis emphasizes that both steeper reverse faults (δ = 50 ±5°) and very low angle thrusts (d < 10°) must be overpressured locally with respect to their surroundings (with P f → σ 3 ) to become reactivated in preference to the formation or reactivation of more favorably oriented faults. Maximum sustainable overpressures are, however, limited to sublitho-static values by the presence of faults that are oriented at angles that are less than those required for frictional lockup (i.e., δ < 60°) in the prevailing stress field. Dynamic processes of fluid redistribution tied to the earthquake stress cycle include upward migration of overpressured fluids through fault-valve action, especially on steeper reverse faults (δ > 60°), and episodic to-and-fro migration of fluids, along strike, induced by mean-stress cycling coupled with strong σ 2 directional permeability. Remarkable similarities between steep reverse-slip structures hosting mesozonal gold-quartz vein systems and comparable structural assemblages developed at much higher structural levels in sedimentary basins suggest that extreme valving action (involving redistribution of large volumes of overpressured fluids) may also play a role in hydrocarbon migration, especially in regions undergoing positive tectonic inversion.