Alexei N. B. Poliakov

Modes of faulting at mid-ocean ridges

Abyssal-hill-bounding faults that pervade the oceanic crust are the most common tectonic feature on the surface of the Earth. The recognition that these faults form at plate spreading centres came with the plate tectonic revolution. Recent observations reveal a large range of fault sizes and orientations; numerical models of plate separation, dyke intrusion and faulting require at least two distinct mechanisms of fault formation at ridges to explain these observations. Plate unbending with distance from the top of an axial high reproduces the observed dip directions and offsets of faults formed at fast-spreading centres. Conversely, plate stretching, with differing amounts of constant-rate magmatic dyke intrusion, can explain the great variety of fault offset seen at slow-spreading ridges. Very-large-offset normal faults only form when about half the plate separation at a ridge is accommodated by dyke intrusion.

Self-consistent rolling-hinge model for the evolution of large-offset low-angle normal faults

The nature of the physical processes responsible for the formation of continental and oceanic metamorphic core complexes is widely debated. The controversy focuses primarily on whether the low-angle normal faults observed in these environments formed and slipped at low angles or were rotated from an original high-angle orientation after large offsets. We describe a self-consistent numerical model for the extension of a brittle layer that can spontaneously produce normal-fault structures. In our formulation, a fault or faults form because strength is locally reduced with increasing strain. If the reduction in fault strength is <∼10% of the total strength of the layer, then faults lock after an offset smaller than the layer thickness and new faults form. Larger strength reduction leads to single faults that continue to slip no matter how large the fault offset. If the strength reduction occurs by the loss of cohesion, then we see the unlimited offset faults for layers <11–22 km thick for reasonable values of cohesion. The key result of this study is that structures very similar to those observed in both oceanic and continental core complexes are produced by rotation of the inactive part of the model fault after very large offset.

Factors controlling normal fault offset in an ideal brittle layer

We study the physical processes controlling the development and evolution of normal faults by analyzing numerical experiments of extension of an ideal two-dimensional elastic-plastic (brittle) layer floating on an inviscid fluid. The yield stress of the layer is the sum of the layer cohesion and its frictional stress. Faults are initiated by a small plastic flaw in the layer. We get finite fault offset when we make fault cohesion decrease with strain. Even in this highly idealized system we vary six physical parameters: the initial cohesion of the layer, the thickness of the layer, the rate of cohesion reduction with plastic strain, the friction coefficient, the flaw size and the fault width. We obtain two main types of faulting behavior: (1) multiple major faults with small offset and (2) single major fault that can develop very large offset. We show that only two parameters control these different types of faulting patterns: (1) the brittle layer thickness for a given cohesion and (2) the rate of cohesion reduction with strain. For a large brittle layer thickness (> 22 km with 44 MPa of cohesion), extension always leads to multiple faults distributed over the width of the layer. For a smaller brittle layer thickness the fault pattern is dependent on the rate of fault weakening: a very slow rate of weakening leads to a very large offset fault and a fast rate of weakening leads to an asymmetric graben and eventually to a very large offset fault. When the offset is very large, the model produces major features of the pattern of topography and faulting seen in some metamorphic core complexes.

A simple parameterization of strain localization in the ductile regime due to grain size reduction: A case study for olivine

We propose a simple parameterization of the transition between dislocation creep and grain-size-sensitive creep under conditions characteristic of the lithospheric mantle and derived from the results of laboratory experiments on olivine-rich rocks. Through numerical modeling and linear stability analysis, we determine the conditions under which this transition takes place and potentially leads to strain localization. We pay particular attention to the effect of cooling rate and strain rate which are likely to be dominant parameters in actively deforming tectonic areas. We conclude that at constant temperature, strain localization can only take place if the rheology of the material is nonlinearly related to grain size; that strain localization is facilitated by syndeformation cooling; that there is only a narrow region in the strain rate versus cooling rate parameter space where localization is likely to take place; and that grain growth inhibits strain localization at fast cooling rates but may lead to “grain growth localization” at low cooling rates. We draw attention to the potential consequences of our analysis of strain localization for the style of plate motions at the Earths surface.

Abyssal hills formed by stretching oceanic lithosphere

Tectonic plates are formed and move apart at mid-ocean ridges. Some portion of this plate-separation process can occur by stretching of the crust, resulting in a complex pattern of extensional faults. Abyssal hills, the most ubiquitous topographic features on Earth, are thought to be a product of this faulting,. Here we report the results of a self-consistent numerical model of lithospheric formation and stretching that includes spontaneous fault creation. In this model, an axial valley develops where the fault activity is most concentrated. The ‘frozen’ fault-generated topography, rafted out of the axial valley, is visually and statistically similar to observed abyssal hills formed at many slower-spreading ridges. Faults appear to be replaced by new faults because their offset changes the local stress field. We accordingly need no temporal variation in magmatism, as required by some previous models, to control the spacing or offset of faults. Our model results suggest instead that the irregularity of abyssal hill relief may result from a self-organized critical stress state at spreading centres.

Do faults trigger folding in the lithosphere

A number of observations reveal large periodic undulations within the oceanic and continental lithospheres. The question if these observations are the result of large-scale compressive instabilities, i.e. buckling, remains open. In this study, we support the buckling hypothesis by direct numerical modeling. We compare our results with the data on three most proeminent cases of the oceanic and continental folding-like deformation (Indian Ocean, Western Gobi (Central Asia) and Central Australia). We demonstrate that under reasonable tectonic stresses, folds can develop from brittle faults cutting through the brittle parts of a lithosphere. The predicted wavelengths and finite growth rates are in agreement with observations. We also show that within a continental lithosphere with thermal age greater than 400 My, either a bi-harmonic mode (two superimposed wavelengths, crustal and mantle one) or a coupled mode (mono-layer deformation) of inelastic folding can develop, depending on the strength and thickness of the lower crust.

Prediction of faulting from the theories of elasticity and plasticity: what are the limits?

Abstract Elasticity, rigid-plasticity and elasto-plasticity are the simplest constitutive models used to describe the initiation and evolution of faulting. However, in practice, the limits of their application are not always clear. In this paper, we test the behaviour of these different models using as examples tectonic problems of indentation of a die, compression with basal shear, bending of a plate and normal faulting around a dike. By comparing the results of these tests, we formulate some guidelines that may be useful for the selection of an appropriate constitutive model of faulting. The theory of elasticity reasonably predicts the initiation of the fault pattern but gives erroneous results for large strains. The theory of rigid-plasticity is more appropriate for large deformations, where the geometry of faults can be found by the method of characteristics. This method works well for zones of failure that are not severely constrained by elastic material outside e.g. when faults are connected to the free-surface, a viscous substratum or a zone of weakness. Non-associated elasto-plasticity is the most complete theory among those considered in this paper. It describes the evolution of faults from the initiation of localized deformations to the formation of a complicated fault network.

Oceanic broad multifault transform plate boundaries

Oceanic transform plate boundaries consist of a single, narrow (a few kilometers wide) strike-slip seismic zone offsetting two mid-ocean ridge segments. However, we define here a new class of oceanic transform boundaries, with broad complex multifault zones of deformation, similar to some continental strike-slip systems. Examples are the 750-km- long, 120-km-wide Andrew Bain transform on the Southwest Indian Ridge, and the Romanche transform, where the Mid-Atlantic Ridge is offset by a lens-shaped, ∼900-km- long, ∼100-km-wide sliver of deformed lithosphere bound by two major transform valleys. One of the valleys is seismically highly active and constitutes the present-day principal transform boundary. However, strike-slip seismic events also occur in the second valley and elsewhere in the deformed zone. Some of these events may be triggered by earthquakes from the principal boundary. Numerical modeling predicts the development of wide multiple transform boundaries when the age offset is above a threshold value of ∼30 m.y., i.e., in extra-long (>500 km) slow-slip transforms. Multiple boundaries develop so that strike-slip ruptures avoid very thick and strong lithosphere.