high velocity frictional properties of alpine fault rocks mechanical data microstructural analysis and implications for rupture propagation

Accepted Manuscript High-velocity frictional properties of Alpine Fault rocks: Mechanical data, microstructural analysis, and implications for rupture propagation Carolyn Boulton, Lu Yao, Daniel R Faulkner, John Townend, Virginia G Toy, Rupert Sutherland, Shengli Ma, Toshihiko Shimamoto PII: S0191-8141(17)30040-8 DOI: 10.1016/j.jsg.2017.02.003 Reference: SG 3445 To appear in: Journal of Structural Geology Received Date: August 2016 Revised Date: 30 January 2017 Accepted Date: February 2017 Please cite this article as: Boulton, C., Yao, L., Faulkner, D.R., Townend, J., Toy, V.G., Sutherland, R., Ma, S., Shimamoto, T., High-velocity frictional properties of Alpine Fault rocks: Mechanical data, microstructural analysis, and implications for rupture propagation, Journal of Structural Geology (2017), doi: 10.1016/j.jsg.2017.02.003 This is a PDF file of an unedited manuscript that has been accepted for publication As a service to our customers we are providing this early version of the manuscript The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain ACCEPTED MANUSCRIPT High-velocity frictional properties of Alpine Fault rocks: mechanical data, microstructural analysis, and implications for rupture propagation Carolyn Boulton1,2*, Lu Yao3, Daniel R Faulkner2, John Townend4, Virginia G Toy5, Rupert Sutherland6,4, Shengli Ma3, Toshihiko Shimamoto3 Department of Geological Sciences, University of Canterbury, Christchurch, New Zealand School of Environmental Sciences, University of Liverpool, Liverpool, United Kingdom 10 11 Department of Geology, University of Otago, Dunedin, New Zealand 12 GNS Science, Lower Hutt, New Zealand 13 *Corresponding author: carolyn.boulton@liverpool.ac.uk RI PT State Key Laboratory of Earthquake Dynamics, Institute of Geology, China Earthquake Administration, Beijing, China AC C EP TE D M AN U 14 SC School of Geography, Environmental, and Earth Sciences, Victoria University of Wellington, Wellington, New Zealand High-velocity Alpine Fault friction ACCEPTED MANUSCRIPT Abstract 16 The Alpine Fault in New Zealand is a major plate-bounding structure that typically 17 slips in ~M8 earthquakes every c 330 years To investigate the near-surface, high- 18 velocity frictional behavior of surface- and borehole-derived Alpine Fault gouges and 19 cataclasites, twenty-one rotary shear experiments were conducted at MPa normal 20 stress and m/s equivalent slip velocity under both room-dry and water-saturated 21 (wet) conditions In the room-dry experiments, the peak friction coefficient (µp=τp/σn) 22 of Alpine Fault cataclasites and fault gouges was consistently high (mean 23 µp=0.67±0.07) In the wet experiments, the fault gouge peak friction coefficients were 24 lower (mean µp=0.20±0.12) than the cataclasite peak friction coefficients (mean 25 µp=0.64±0.04) All fault rocks exhibited very low steady-state friction coefficients 26 (µss) 27 µss=0.09±0.04) Of all the experiments performed, six experiments conducted on wet 28 smectite-bearing principal slip zone (PSZ) fault gouges yielded the lowest peak 29 friction coefficients (µp=0.10-0.20), the lowest steady-state friction coefficients 30 (µss=0.03-0.09), and, commonly, the lowest specific fracture energy values (EG=0.01- 31 0.69 MJ/m2) Microstructures produced during room-dry and wet experiments on a 32 smectite-bearing PSZ fault gouge were compared with microstructures in the same 33 material recovered from the Deep Fault Drilling Project (DFDP-1) drill cores The 34 near-absence of localized shear bands with a strong crystallographic preferred 35 orientation in the natural samples most resembles microstructures formed during wet 36 experiments Mechanical data and microstructural observations suggest that Alpine 37 Fault ruptures propagate preferentially through water-saturated smectite-bearing fault 38 gouges that exhibit low peak and steady-state friction coefficients experiments mean µss=0.16±0.05; wet experiments mean AC C EP TE D (room-dry M AN U SC RI PT 15 High-velocity Alpine Fault friction ACCEPTED MANUSCRIPT 39 40 41 Keywords: Alpine Fault, high-velocity friction, fracture energy, rupture propagation, microstructures, shear bands 42 Introduction RI PT 43 The Alpine Fault, South Island, New Zealand is a long-lived crustal-scale 45 continental transform fault that has accommodated at least 460 km of cumulative 46 displacement in the past c 45 Myr (Wellman, 1953; Sutherland et al., 2000) 47 Paleoseismological records indicate that the Alpine Fault produces quasi-periodic 48 large-magnitude (M~8) earthquakes that propagate along-strike for 300-600 km 49 (Wells and Goff, 2007; Sutherland et al., 2007; Berryman et al., 2012a) Single-event 50 strike-slip and dip-slip surface displacements on the Alpine Fault are 7.5-9 m and c 51 m, respectively (Barth et al., 2013) Boulton et al (2012) and Barth et al (2013) 52 measured the frictional strength and stability of smectitic principal slip zone (PSZ) 53 gouges from well-studied localities spanning c 220 km along strike of the central and 54 southern Alpine Fault They concluded that the velocity-strengthening frictional 55 properties of surface-outcrop PSZ gouges tested while fluid-saturated at room 56 temperature and low sliding velocities (v0.1 m/s) has been observed 64 repeatedly since the first rotary shear experiments by Tsutsumi and Shimamoto 65 (1997) (for reviews, see Wibberley et al., 2008; Di Toro et al., 2011; Niemeijer et al., 66 2012) A wide range of dynamic weakening mechanisms has been proposed to 67 explain this effect, including: melt lubrication (e.g., Hirose and Shimamoto, 2005a; 68 Nielsen et al., 2008; Niemeijer et al., 2011), silica gel lubrication (Goldsby and Tullis, 69 2002, Di Toro et al., 2004), flash heating (Rice, 2006; Beeler et al., 2008; Goldsby 70 and Tullis, 2011), powder lubrication (e.g., Han et al., 2010; Reches and Lockner, 71 2010; Chang et al., 2012), fluid film lubrication (Brodsky and Kanomori, 2001; Ferri 72 et al., 2011), and thermal pressurization (e.g., Sibson, 1973; Lachenbruch, 1980, 73 Wibberley and Shimamoto, 2005; Rice, 2006; Sulem et al., 2007; Tanikawa and 74 Shimamoto, 2009; Faulkner et al., 2011) or thermochemical pressurization (Brantut et 75 al., 2010; Chen et al., 2013; Platt et al., 2015) To what extent hanging wall, principal 76 slip zone (PSZ), and footwall Alpine Fault rocks undergo high-velocity weakening 77 remains untested EP TE D M AN U SC RI PT 62 The present study documents the results of room-dry and water-saturated high- 79 velocity, low-normal stress (v=1 m/s, σn=1 MPa) friction experiments conducted on 80 Alpine Fault gouge and cataclasite samples collected from the surface at Gaunt Creek 81 and Hokuri Creek, and from shallow depth from the Deep Fault Drilling Program 82 (DFDP-1) at Gaunt Creek (Figure 1) A focus of these experiments is to quantify the 83 peak coefficient of friction (µp), as this value represents the yield strength and thus a 84 barrier to rupture propagation An additional aim is to quantify the steady-state 85 coefficient of friction (µss) at high velocity as well as the slip-weakening distance (dw) AC C 78 High-velocity Alpine Fault friction ACCEPTED MANUSCRIPT over which µss is reached Finally, microstructures produced during six experiments 87 with varying velocity histories and pore-fluid conditions are compared with 88 microstructures formed in naturally occurring smectitic PSZ fault gouges By doing 89 so, we test: (1) the effect pore fluids have on microstructural evolution during high- 90 velocity sliding; (2) the effect decelerating slip and simulated afterslip have on 91 recovered experimental microstructures, and (3) the degree to which natural 92 microstructures resemble those produced during experimental deformation Our 93 results allow us to conclude that small variations in sliding velocity following a high- 94 velocity slip event have little effect on microstructures recovered and that natural 95 microstructures resemble those formed during wet high-velocity friction experiments 96 Fault rock descriptions 98 2.1 Analytical methods TE D 97 M AN U SC RI PT 86 Samples were collected from unoriented drill core retrieved during the first 100 phase of the Deep Fault Drilling Project (DFDP-1) at Gaunt Creek (hereafter GC) 101 (Figure 1) An additional sample of PSZ gouge was collected from a nearby outcrop 102 (the GC scarp outcrop of Boulton et al., 2012) All sample depths reported from 103 DFDP-1B are adjusted by +0.20 m from borehole lithological logs following 104 Townend et al (2013) Saponite-rich gouge collected from a 12 m-wide PSZ at 105 Hokuri Creek (HkC PSZ) on the southern Alpine Fault was also tested The gouge 106 mineralogy, microstructure, and low-velocity frictional and hydrological properties of 107 the HkC PSZ gouge were described in detail by Barth et al (2013) With the 108 exception of the DFDP-1B 144.04 m brown gouge, all samples were gently AC C EP 99 High-velocity Alpine Fault friction ACCEPTED MANUSCRIPT disaggregated using mortar and pestle, and the powdered material was passed through 110 a 100# sieve to obtain a 90% matrix grains