Rethinking conditions necessary for pseudotachylyte formation: Observations from the Otago schists, South Island, New Zealand (original) (raw)

Elsevier

Tectonophysics

Abstract

Pseudotachylytes and two distinct types of cataclasite in the Otago Schist at Tucker Hill, South Island, New Zealand, provide evidence for both seismic slip and aseismic creep on a normal fault zone during regional crustal extension in late Cretaceous time. Regional geologic evidence suggests that the fault had its present low-angle dip (ca. 10°) at the time it was active. ‘Type A’ cataclasites, presumably aseismic, can be recognized by bi-fractal grain size distributions, monomict composition, angular clasts of uniform textural maturity, and a crude fabric defined by oriented grains and transgranular fractures. ‘Type B’ cataclasites, possibly cosesimic, have characteristics consistent with fluidized grain flow, including heterogeneous clast shapes and types, a bimodal grain size distribution, intrusive relationships with other rocks, and the absence of any fabric or transecting fractures. Pseudotachylyte, which occurs as fault veins, injection veins and more complex types of intrusive structures, consistently cuts across and invades Type A cataclasites but is both intrusive into and included as clasts in Type B cataclasites.

These relationships are consistent with a fault evolution model in which the development of a damage zone through aseismic cataclasis (Type A) facilitates the formation of pseudotachylyte in a subsequent seismic event by providing a permeable matrix through which fluids can drain in the early stages of slip, thereby maintaining frictional contact between rock surfaces. The formation of pseudotachylyte, in turn, may seal the fault zone and lead to thermal pressurization in a later seismic cycle, forming fluidized (Type B) cataclasites. Seismic slip on the low-angle normal fault zone at Tucker Hill may have occurred by two distinct modes of dynamic weakening — melt lubrication and thermal pressurization — in successive seismic events.

Although there is a perception among geologists that pseudotachylyte is most likely to form in intact, crystalline rocks, geophysical models of fault zones clearly demonstrate that pseudotachylyte formation is actually suppressed in low-permeability rock because any fluids present would be unable to escape the fault zone and thermal pressurization would rapidly reduce frictional resistance. The paradigmatic occurrences of pseudotachylyte in otherwise pristine crystalline rocks probably represent somewhat exceptional circumstances (single rupture events at very high effective stress in dry rock). Coseismic frictional melts may actually be more common in hydrated rocks like the schist at Tucker Hill, but harder to recognize and also vulnerable to overprinting as a fault zone matures. In such rocks, pseudotachylyte may represent an intermediate stage in the evolution of a fault zone, the period between the formation of a high-permeability damage zone and the development of a low-permeability fault core.

Introduction

Pseudotachylyte — glassy rock representing frictionally-generated melt — is considered the only unambiguous indicator of seismic slip on ancient faults (Magloughlin and Spray, 1992, Cowan, 1999), and as such it provides a rare window into processes that occur at the source of an earthquake. The radiated energy of an earthquake, recorded instrumentally, probably represents only about 10% of the total energy released in a seismic slip event (McGarr, 1999), with the remainder absorbed along the fault by fracturing and frictional dissipation. Of this absorbed energy, less than 3% is thought to be consumed by fracturing (Pittarello et al., 2008), and thus frictional heating is by far the largest component of the earthquake energy budget.

Assuming adiabatic conditions (appropriate for seismic events, for which slip duration is much shorter than the time scale for thermal diffusion), the relationship between the heat generated per unit area (Q) along a fault and the thickness (z) of any resulting frictional melt is:Q(per unit area)=τfD=zρcΔT+Δhfus1−Φwhere τ_f is shear resistance, D is coseismic displacement; ρ is rock density, c is specific heat capacity, Δ_T is the temperature rise required for melting, Δ_h_fus is the heat of fusion, and Φ is the fraction of unmelted material within the pseudotachylyte (O'Hara, 2001). For values of shear resistance (10–100

MPa) and coseismic slip (0.1–1

m) corresponding to moderate-sized (M6–7) earthquakes, Eq. (1) predicts that such events should produce enough heat to form pseudotachylyte lenses on the order of 0.1–1-cm thick (Sibson and Toy, 2006). Larger earthquakes would be even more likely to yield significant amounts of frictional melt.

Yet pseudotachylyte is comparatively rare. Although evidence of frictional melting has been found in a wide array of igneous and metamorphic rock types from a large range of inferred depths (from 2 to >

50

km) (Sibson and Toy, 2006), the occurrence of pseudotachylyte in ancient fault zones is the exception rather than the rule. This in itself is an important observation about coseismic processes, indicating that in most fault zones, one or more factors must act to suppress frictional melting. Among these is the thickness of the work zone during the slip event (Spray, 1995, Kanamori et al., 1998). The instantaneous local temperature rise along a fault zone for a slip event of duration t_e is a linear function of the shear strain rate (d_ε/d_t_), which is equivalent to the slip rate (v) divided by the thickness of the processed rock (w):ΔT=τfdε/dtte/cρ=τfte/cρvw

This means that melting is less likely in faults with wide damage zones. However, the absence of melts even along faults where slip has been highly localized indicates that there must be other variables that inhibit frictional melting.

The consensus among geologists has been that fluids in fault zones are the primary explanation for the relative rarity of pseudotachlytes. In a review of the published studies of pseudotachylyte occurrences, Sibson and Toy (2006) stated: “Thermal pressurization of fault fluids during seismic slip inhibits melt generation…Pseudotachylyte represents high-stress (τ

>

100

MPa) rupturing associated with fault initiation or reactivation in dry, intact (or metamorphically reconstituted) crystalline crust. Its scarcity is accounted for by the progressive infiltration of aqueous fluids into evolving fault zones.” Although many of the classic occurrences of pseudotachylyte fit this description (e.g., Swanson, 1992, Di Toro and Pennachioni, 2005), and their formation has been replicated in geophysical models (e.g., Fialko, 2004), pseudotachylytes have also been found in rocks that were not obviously ‘dry’ (e.g., in subduction settings [Austrheim and Anderson, 2004, Ujiie et al., 2007]) — or ‘intact’ (e.g., strongly cataclasized granite [Otsuki et al., 2003]) when the frictional melts formed. These occurrences suggest that there must be another set of conditions that favor the generation of pseudotachylytes in some fault zones.

The purpose of this paper is to describe the relationships among texturally distinct types of fault rocks, including pseudotachylyte, within a shallowly dipping normal fault zone in schists in central Otago, South Island, New Zealand and to suggest a new model for the geologic conditions under which pseudotachylyte may form. The rocks in the study area indicate that fault rock permeability and transient fluid pressure effects — with and without the formation of frictional melt — dominated the co- and inter-seismic behavior of the fault zone, and that the textures produced by these dynamic processes dictated the subsequent evolution of the fault over several earthquake cycles. These rocks may also provide insights into the puzzle of slip on low-angle normal faults, and the mechanisms of dynamic weakening along fault zones during seismic events.

Section snippets

Geologic context

Central Otago on the South Island of New Zealand is underlain by a thick sequence of lower Mesozoic greenschist-facies psammites and pelites known as the Otago Schist, part of the late Permian to Jurassic Haast Schist Group (Mortimer, 1993). Regional differences in lithology and whole rock geochemistry of the Otago schists have made it possible to delineate terrane boundaries within these strongly deformed, predominantly turbiditic rocks (Mortimer and Rosen, 1992). The present study area, at

Field description

There are at least five subhorizontal to gently (<

15°) north-dipping, pseudotachylyte-bearing fault zones exposed in an area of about 6 hectares on Tucker Hill (Fig. 2). Barker, 2003, Barker, 2005 carried out detailed petrological and geochemical analyses of samples of the pseudotachylytes and documented chilled margins, embayed lithic clasts, microlites of newly formed minerals, and probable amygdules, evidence that these are melt-generated rocks derived from the host schists.

The

Interpretation and discussion

Classic studies of pseudotachylytes in crystalline rocks, with sharply defined fault veins and relatively simple injection vein geometries (e.g., Swanson, 1992, Di Toro et al., 2005), have strongly shaped geological thinking about the conditions necessary for coseismic frictional melting. Among geologists, there is a common perception that pseudotachylytes can only form in low-porosity, intact rock (Sibson and Toy, 2006) or in fractured rock that has been healed by the growth of new minerals

Conclusions

Seismic slip on a low-angle normal fault zone in the Otago schists appears to have been accomplished by two distinct and possibly alternating mechanisms of dynamic slip weakening: melt lubrication and thermal pressurization. Formation of a permeable damage zone by cataclasis may have been a prerequisite to pseudotachylyte formation by making it possible for the fault zone to be drained of pore fluids at the onset of slip. Once the pseudotachylyte formed, it too was drawn away from the fault

Acknowledgments

This work was made possible by sabbatical support from Lawrence University and a U.S. Senior Scholar grant from Fulbright New Zealand. I thank Virginia Toy and Richard Norris (University of Otago) for introducing me to the field area and for interesting conversations about pseudotachylytes; Shaun Barker (University of British Columbia) for helpful correspondence; and Emily Thiem and Olav, Finn and Karl Bjørnerud for assistance in the field. Christie Rowe and an anonymous reviewer provided

References (66)

Pseudotachylytes in an ancient accretionary complex and implications for melt lubrication during subduction zone earthquakes

J.Struct. Geol.

(2007)

Pseudotachylyte-generating faults in central Otago, New Zealand

Tectonophysics

(2005)

J. Struct. Geol.

(2004)

Do faults preserve a record of seismic slip? A field geologist's view

J. Struct. Geol.

(1999)

Superheated friction-induced melts in zoned pseudotachylytes within the Adamello tonalites (Italian Southern Alps)

J. Struct. Geol.

(2004)

Fault plane processes and mesoscopic structure of a strong-type seismogenic fault in tonalites (Adamello batholith, Southern Alps)

Tectonophysics

(2005)

Can pseudotachylytes be used to infer earthquake source parameters? An example of limitations in the study of exhumed faults

Tectonophysics

(2005)

Grain size distributions of fault rocks: a comparison between experimentally and naturally deformed granitoids

J. Struct. Geol.

(2007)

Microstructural and chemical changes associated with cataclasis and frictional melting at shallowcrustal levels: the cataclasite–pseudotachylyte connection

Tectonophysics

(1992)

Frictional melting processes and products in geological materials: introduction and discussion

Tectonophysics

(1992)

Comminution and fluidization of granular fault materials: implications for fault slip behaviour

Tectonophysics

(2003)

Very high strains recorded in mylonites along the Alpine Fault, New Zealand: implications for the deep structure of plate boundary faults

J. Struct. Geol.

(2003)

A pseudotachylyte geothermometer

J. Struct. Geol.

(2001)

An argument against a crush origin for pseudotachylytes based on the analysis of clast-size distribution

J. Struct. Geol.

(1992)

Recognizing the seismic cycle along ancient faults: CO2-induced fluidization of breccias in the footwall of a sealing low-angle normal fault

J. Struct. Geol.

(2008)

Fault structure, wear mechanisms and rupture processes in pseudotachylyte generation

Tectonophysics

(1992)

Internal structure and permeability of major strike–slip fault zones: the Median Tectonic Line in Mie Prefecture

Southwest Japan, J. of Struct. Geol.

(2003)

Potassium–argon age studies of a low-grade, progressively metamorphosed greywacke sequence, Dansey Pass, South Island, New Zealand

J. Geol. Soc. London

(1985)

A fault constitutive relation accounting for thermal pressurization of pore fluid

J. Geophys. Res.

(2002)

Pseudotachylytes from Corsica: fossil earthquakes from a subduction complex

Terra Nova

(2004)

Experiments on a Gravity-free Dispersion of Large Solid Spheres in a Newtonian Fluid under Shear

Cretaceous age, composition and microstrucuture of pseudotachlylyte in the Otago Schist, New Zealand

NZ J. Geol. Geophys.

(2010)

Textural zones in the Haast Schist: a reappraisal

NZ J. Geol. Geophys.

(2001)

Flowage differentiation: limitation of the Bagnold effect to narrow intrusions

Contrib to Mineralog. and Petrol.

(1976)

Sheath-like fold structures in pseudotachylytes

J. Struct. Geol.

(1992)

Progressive metamorphism from prehnite–pumpellyite to greenschist facies in the Danseys Pass area, Otago, New Zealand

Bull.of the Geol. Soc. of America

(1972)

Processes leading to eclogitization (densification) of subducted and tectonically buried crust

J. Geophys. Res.

(2002)

Stokesian dynamics

Annu. Rev. Fluid Mech.

(1988)

A geological fingerprint of low-viscosity fault fluids mobilized during an earthquake

J. Geophys. Res.

(2009)

Structure and geochemistry of the Rise and Shine Shear Zone mesothermal gold system, Otago Schist, New Zealand

NZ J. Geol. Geophys.

(2006)

Tectonic significance of Cretaceous bivergent extensional shear zones in the Torlesse accretionary wedge, central Otago, New Zealand

NZ J. Geol. Geophys.

(2002)

Differential stress magnitudes during regional deformation and metamorphism: upper bounds imposed by tensile fracturing

Geology

(1983)

Cited by (19)

2016, Tectonophysics
For these reasons, we speculate that the presence of fluids prevents pseudotachylite formation in the case of the granitoid rocks of the study area providing insights into alternative processes for energy dissipation during seismic events. Particularly fluidization during embrittlement has been associated with seismicity in previous studies (Smith et al., 2008; Brodsky et al., 2009; Bjørnerud, 2010; Rowe and Griffith, 2015), where prolonged periods of slow viscous granular flow are interpreted to be interrupted by rapid seismic embrittlement. The observation of cyclic occurrences of frictional events, as shown in our study area, provides additional evidence for seismic activity (Scholz, 1998; Gratier et al., 2002; Handy and Brun, 2004; Matysiak and Trepmann, 2012; Wintsch and Yeh, 2013).

2013, Journal of Structural Geology
Examination of the microstructures within an individual clast reveal its seismic origin, but the wider zone of cataclasite records a subsequent deformation event (Rowe et al., 2011). Other examples of clasts of pseudotachylyte within gouge or cataclasites are reported from a variety of settings (e.g. Macaudière and Brown, 1982; Magloughlin and Spray, 1992; Otsuki et al., 2003; Lin et al., 2005; Kirkpatrick and Shipton, 2009; Bjørnerud, 2010; Mambane et al., 2011). These examples show clasts of pseudotachylyte occur within gouge layers alongside intact pseudotachylyte veins (e.g. Fig. 4B), and as isolated fragments in cataclasites where no continuous pseudotachylyte fault veins are preserved nearby. View all citing articles on Scopus

View full text

Copyright © 2010 Elsevier B.V. All rights reserved.