See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/270688689

Fault zone permeability structure evolution in

basalts

Article in Geology · December 2012

DOI: 10.1130/G33508.1

CITATIONS READS

22 115

4 authors, including:

Richard Walker R. E. Holdsworth

University of Leicester Durham University
27 PUBLICATIONS 203 CITATIONS 218 PUBLICATIONS 5,827 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Sills as a record of horizontal tectonic compression View project

PhD: Laboratory earthquakes View project

All content following this page was uploaded by Richard Walker on 24 May 2016.

The user has requested enhancement of the downloaded file.

 Geology, published online on 19 October 2012 as doi:10.1130/G33508.1

Fault zone permeability structure evolution in basalts

Richard J. Walker1,2, Robert E. Holdsworth2, Peter J. Armitage3, and Daniel R. Faulkner3
1
School of Earth and Ocean Sciences, Cardiff University, Park Place, Cardiff CF10 3AT, UK
2
Department of Earth Sciences, Durham University, Science Labs, Durham DH1 3LE, UK
3
Rock Deformation Laboratory, University of Liverpool, Liverpool L69 3BX, UK

ABSTRACT tics of such lava-hosted faults because they will

A combination of field, microstructural, and experimental static permeability character- likely influence fluid flow in the subsurface.
ization is used to determine fault permeability structure evolution in upper crustal basalt- Fluid flow in an upper-crustal brittle fault
hosted fault zones in the Faroe Islands. The faults comprise low-strain fracture networks to zone is dependent on its architecture and fault
high-strain breccias that form tabular volumes around a principal slip zone hosting gouge or rock assemblage permeability (e.g., Faulkner
cataclasite. Samples representative of these fault zone components are used for static experi- and Rutter, 2001). Models for clastic sedimen-
mental permeability measurement. Results indicate that within the appropriate effective pres- tary sequences and crystalline basement rocks
sure (depth) range (10–90 MPa; ~0.3–3.0 km), basalt-hosted faults evolve from relatively low- typically refer to a low-permeability fault core
permeability (<10−17 m2) structures with <1 m displacement to relatively high-permeability surrounded by a higher-permeability damage
(>10−17 m2) structures with ≥1 m displacement. Sample analyses reveal that static permeability zone, relative to the intermediate permeability
is controlled by the development of: (1) fault-parallel clay growth (decreasing permeability), of the undeformed host rock (e.g., Caine et al.,
and (2) porous zeolite vein connectivity due to hydrofracture (increasing permeability). Fault- 1996; Evans et al., 1997). In this paper, we detail
parallel permeability is increased relative to the host rock, while fault-normal permeability the permeability structure of shallow (0–3 km
is low throughout fault rock evolution. This configuration will tend to promote across-fault depth) basalt-hosted fault zones using examples
compartmentalization and along-fault fluid flow, facilitating migration between relatively of widely distributed brittle fracturing in the
high-permeability horizons (e.g., vesicular flow-unit tops and siliciclastic horizons), bypassing Faroe Islands Basalt Group, northeastern Atlan-
the bulk of the stratigraphy. tic margin. We use experimental permeability
measurements to quantify fault rock static per-
INTRODUCTION eral recent onshore and offshore studies in the meability in basalts. Qualitative field and micro-
Understanding subsurface fluid flow and Faroe-Shetland Basin (e.g., Moy and Imber, structural characterization of the same fault rock
sealing potential in basaltic sequences is 2009; Walker et al., 2011a, 2011b) have shown assemblages are used to rationalize measured
becoming increasingly important with the ris- that faults cut part or all of the Paleocene lava permeability and to infer dynamic permeability
ing economic significance of intra- and subvol- sequence together with the rocks in the under- associated with fault events. The results indicate
canic hydrocarbon plays (Schutter, 2003), con- lying sedimentary basins. It is important, there- that permeability during faulting is important in
ventional and enhanced geothermal systems, fore, to determine the permeability characteris- fluid migration, and that at stages in fault evolu-
and the need to find suitable geological carbon
sequestration or storage sites (Hawkins, 2004; cooling low strain
Oelkers and Cole, 2008). The efficacy of geo- A low strain
inclusions
joints zone (e.g., B) C zeolite & calcite
mineralization
logical CO2 storage is dependent on the reten- N chaotic
Greenland 70°N
breccia
tion time, the dynamic reservoir stability (i.e.,
the risk of leakage), and/or the mineral stor-
Faroe
age potential of the host rock (Gislason et al., Iceland Islands
principal
2010). Sealing potential in basaltic sequences slip host
60°N high strain mosaic
is important to intra- and sub-basaltic hydro- Atlantic
surface zone rock basalt breccia
(e.g., D) clasts mosaic clast
carbon plays, particularly in passive-margin Ocean
UK 0 250 mi
(e.g., C) cooling
breccia
20°W joints/fractures 20 mm
continental basins that are active targets for 10°W 0° 0 500 km clasts

hydrocarbon exploration. In volcanic geother- B mosaic D

crackle
mal systems, the methods used for geothermal breccia breccia
altered
exploration and resource development are clasts
basalt slip
highly dependent on conductive versus con- clasts surface
vective heat loss, and so fault permeability is
critically important.
Faults are a key component in the sealing
potential of any subsurface storage system, yet mosaic
crackle breccia
little is known about the hydraulic structure of breccia 100 mm 20 mm
basalt-hosted faults. Here we focus on part of
the European North Atlantic margin (Fig. 1A; Figure 1. A: Overview of a dip-slip normal fault in a columnar-jointed basalt unit (after
Appendix DR1 in the GSA Data Repository1), Walker et al., 2012), showing idealized strain distribution. Map shows the location of the
Faroe Islands. B: Low-strain-zone crackle- and mosaic-type fault breccias that accom-
which is covered by variable thickness of modate millimeter to centimeter displacements. C: High-strain (fault core) mineralized
basaltic lavas and volcaniclastic rocks related mosaic and chaotic breccias that accommodate ~5 m displacement. D: Principal slip zone
to the North Atlantic Igneous Province. Sev- chaotic breccia and cataclasites that accommodate >10 m displacement.

1
GSA Data Repository item 2013014, Appendix DR1 (location maps for the study fault zones, Faroe Islands, North Atlantic European continental margin) and
Appendix DR2 (experimental method description, including table of results and graphs for all data), is available online at www.geosociety.org/pubs/ft2013.htm, or on
request from [email protected] or Documents Secretary, GSA, P.O. Box 9140, Boulder, CO 80301, USA.

GEOLOGY, January 2013; v. 41; no. 1; p. 1–4; Data Repository item 2013014 | doi:10.1130/G33508.1 | Published online XX Month 2012
GEOLOGY | January
© 2012 Geological 2013
Society | www.gsapubs.org
of America. For permission to copy, contact Copyright Permissions, GSA, or [email protected]. 1
 Geology, published online on 19 October 2012 as doi:10.1130/G33508.1

tion, low-strain zones can act as barriers to fluid ervoir pressures (0.3–3.0 km) using the tran- anisotropy ratio (kp/kN) at 90 MPa effective pres-
flow whereas high-strain zones act as conduits. sient pulse decay (TPD) technique (Brace et al., sure ranges from 11 to 43, with the maximum
1968); a detailed methodology is provided by anisotropy in the lower-strain zone.
FAULT ZONE ARCHITECTURE Armitage et al. (2011). Argon was used as the The studied basaltic host rocks are generally
Faults in the Faroe Islands cut volcanic and permeant, and all experiments were conducted at low permeability, within the range of 10−20 to
volcaniclastic strata of the Paleocene-age Faroe a maintained room temperature of ~22 °C. The 10−18 m2, with the exception of the compound
Islands Basalt Group (59–56 Ma; Passey and resolution of the pressure measurements is ± 0.3 lava unit sample (L01; Fig. 2). The two sample
Jolley, 2009), and are representative of faulting MPa, and permeability can be measured accu- suites (aphyric basalt lava unit and aphyric to pla-
at depths between 1 and 3 km that developed rately within a range of 10−16 to 10−23 m2 using gioclase-phyric basalt dike) show a varied evolu-
before, during, and immediately after North our setup for the TPD technique. tion pathway from host rock to evolved fault rock
Atlantic break-up in the early Eocene (Walker et (Fig. 2). In the lava sample suite, permeability ini-
al., 2011a). Basaltic host units display joints and STATIC PERMEABILITY RESULTS tially increases by about two orders of magnitude
microfractures formed during cooling, which The bulk permeability structure of a fault (from host to crackle breccia), but then decreases
range in length from <1 mm up to ~30 m (i.e., zone is a composite of the properties of the by a comparable amount into the higher-strain
the thickness of the host unit). Fractures and architectural elements and their spatial arrange- mosaic breccias. By contrast, the dike sample
joints of this type increase in density toward the ment in three dimensions (e.g., Caine and For- suite shows an initial decrease in permeability
top of lava units and toward the margins of dikes. ster, 1999). Permeability measurements of the from host rock to low-strain fault rocks, fol-
Fault-related damage represents an increase in key fault zone components provide an indica- lowed by a near-reciprocal increase toward the
fracture density relative to the host lithology. Slip tion of in situ permeability. On the assumption fault core. That permeability increase continues
is accommodated on a surface or zone of catacla- that low-strain fault rock components represent in the dike suite into the fault core chaotic brec-
site or gouge, defining the principal slip surface early stages of high-strain zones, these data can cia, which records the highest permeabilities of
or zone (PSS or PSZ, respectively; Figs. 1A and be used to infer the evolution of static perme- this study (Fig. 2). A permeability increase is
1D). These are contained within the fault core, ability through time. also recorded from the lava unit mosaic breccia
which is defined here as the zone that accommo- Pressure cycling within the study range (i.e., (IB07) into the PSS cataclasite samples (IB13)
dates the majority of shear displacement and can 10–90 MPa effective pressure) was used here (Fig. 2). Overall the sample suites show lower
be characterized by mosaic and chaotic breccias to close any stress-relief microfractures (see permeability in low-strain zones, and higher per-
(Figs. 1A and 1C) (breccia definitions following Appendix DR2 for details). Our data reveal meability in high-strain zones.
Woodcock et al., 2006). Surrounding the fault that at a simulated reservoir depth of ~3 km (90
core is a zone of damage, which is characterized MPa), permeability spans more than four orders FAULT ROCK ASSEMBLAGES
by crackle to mosaic breccias (Figs. 1A and 1B). of magnitude, ranging from 1.3 × 10−20 m2 (L03; Microstructural characterization of the host-
Calcite and zeolite mineralization are ubiqui- Fig. 2) to 1.4 × 10−16 m2 (L05; Fig. 2). For sam- and fault-rock assemblages is important because
tous in syn-magmatic Faroe Islands fault zones, ples with fault-parallel (kp) and fault-normal it gives insight into the controls on the static
indicating that they acted as conduits for the data (kN), kN ranges from 1.5 × 10−20 m2 (IB07b; permeability laboratory measurements. Fur-
passage of hydrous fluids. Early-formed veins Fig. 2) to 2.8 × 10−18 m2 (IB13b; Fig. 2), and kp thermore it gives a qualitative indication of the
are filled with zeolite that grows inwards from ranges from 6.5 × 10−19 m2 (IB07a; Fig. 2) to likely evolution in synfaulting dynamic perme-
the fracture walls as hemiradial fibers, indicat- 3.0 × 10−17 m2 (IB13a; Fig. 2). The permeability ability with increasing deformation.
ing mineral growth into fluid-filled cavities. In
thicker veins (>1–2 mm thick), calcite is the
dominant mineral fill. In most fault zones, zeo- 10-14 Host Peripheral damage Fault core
lites are also observed lining the euhedral ter-
Effective pressure (MPa)
minations of individual calcite crystals. Textural 10-15 10 30 50 70 90
relationships suggest that zeolite and calcite IB13a basalt
L01 (kP )
mineralization were accompanied by the gen- 10-16 L05 cong.

eration of authigenic clays at all stages of fault volc.

Log permeability (m2)

L02 S02 L04 sandst.

zone evolution (Walker et al., 2012). 10-17
comp.
IB13b lava
METHODS 10-18
S05
(kN )
L03 IB07a fault normal (kN )
Host-rock and fault-zone assemblages were S01 (kP )
aphyric
KR2 simple
sampled in transects across representative fault 10-19
lava fault parallel (kP )
zones in the Faroe Islands to characterize fault
10-20 IB07b aphyric-
zone heterogeneity. The experimental static per- (kN ) pl.-phyric
meability measurements detailed below have dike
10-21 host crackle mosaic chaotic cat./gou.
sample volumes on the scale of cubic centime- no displacement < 1 m displacement > 1m displacement
ters. Since mesoscopic open fractures have not
been sampled explicitly, measured permeabili- Figure 2. Measured permeability data for host-rock and fault-rock assemblages at effective
ties may provide only a lower bound to the bulk pressures ranging from 10 to 90 MPa (static fluid pressure of 10 MPa). Fault-parallel (k p) and
fault-normal (kN) values are shown for anisotropic samples (IB07 and IB13). Host rock and
fault zone permeability. Cores were taken to chaotic breccias do not exhibit a clear fabric: kp and kN values are assumed equal within the
measure fault-parallel and fault-normal perme- experimental uncertainty. Within the sub–100 MPa confining pressure range, the high perme-
ability, but in most cases, the fault-parallel sam- ability of the dike fault core sample (L05) is at the limits of the transient pulse decay (TPD)
ples broke along fractures and veins and could technique using our experimental setup; successful analysis at higher effective pressures
(110–190 MPa; Appendix DR2 [see footnote 1]) and down-pressure extrapolation indicate
not be loaded into the high-pressure fluid-flow that fault core permeability ranges from ~2.5 × 10–15 m2 at 10 MPa, to 1.4 × 10–16 m2 at 90 MPa.
apparatus. Static permeability was measured cong.—conglomerate; volc.—volcanic; sandst.—sandstone; comp.—compound; pl.—pla-
experimentally under simulated subsurface res- gioclase; cat.—cataclasite.

2 www.gsapubs.org | January 2013 | GEOLOGY

 Geology, published online on 19 October 2012 as doi:10.1130/G33508.1

Sample suite host rocks are generally fine a cataclasite matrix formed of the same compo- sis, and vein dissection. The presence of perva-
crystalline basalt dominated by plagioclase nents. Established vein networks are cut by, and sive mineral vein networks and authigenic clays
and pyroxene, and display very minor mineral incorporated into, the cataclasites. within fault zones also indicates significant
alteration to clays (Fig. 3A). This clay develop- Thin sections of measured samples show a transient permeability development during rock
ment appears to be authigenic, occurring along strong correlation between clay development, fracture episodes (i.e., dynamic permeability).
microfractures and crystal boundaries. Longer section-plane vein connectivity, and static per- Although the host rocks from the sample suites
and wider fractures, where present, are clay- meability results. Where low-strain fault rocks are low-porosity fine crystalline basalts, micro-
lined, and no zeolite or calcite mineralization display connected zeolite veins, permeability is fractures related to cooling and/or distributed
is observed (Fig. 3A). higher than for samples with segmented veins. strain demonstrably facilitated transient low fluid-
Low-strain (<1 m displacement) fault rocks In the most permeable, high-strain zones, clays flux, as evidenced by the occurrence of authi-
(Fig. 3B) show increased alteration relative to are absent or dissected by zeolite and/or calcite genic clays. Depending on fluid chemistry (i.e.,
the host, with discrete zones of clay minerals (veins or continuous clasts of veins). CO2 concentration), pH, and temperature, altera-
developed immediately adjacent to, and parallel tion of the basaltic mineralogy to clays within the
with, fractures and veins (Fig. 3B). In samples IMPLICATIONS FOR PERMEABILITY study depth range (i.e., 0–3 km) can occur in a
L03 and IB07, zeolite veins are clay-lined and STRUCTURE EVOLUTION matter of years (Gysi and Stefánsson, 2008), i.e.,
poorly connected in the section plane, whereas The measured experimental static permeabil- at a rate comparable to aseismic fault creep. Dur-
in samples S02 and L04, zeolite veins are inter- ities, when combined with field and microstruc- ing initial meso- to macro-scale rock fracturing,
connected. Blue-stained zeolite veins in thin tural characterization of the fault rocks, show fluid flow is increased locally, leading to the pre-
section (Figs. 3B and 3C) indicate that these that shallow faults in basalt generally develop cipitation of zeolite and calcite as veins. In these
veins maintain porosity. through stages of low-strain, lower-permeability low-strain (<1 m displacement) fault rocks, veins
High-strain (fault core) rocks that have accom- assemblages to high-strain, higher-permeability and fractures may remain locally statically per-
modated meter-scale or greater displacements assemblages. This contrasts with most clastic- meable, but are lined by low-permeability clays
(Fig. 3C) show intense zeolite and calcite veining or crystalline-hosted fault zone models, where and are poorly interconnected (Fig. 3B). The
that forms a connected network. Authigenic clays high-strain fault cores are generally thought to development of clays acts to decrease permeabil-
developed during host alteration appear to be exhibit lower permeabilities due to grain com- ity relative to the host rocks (Figs. 3A, 3B, and 4).
absent in the crystalline groundmass, with zeo- minution effects (e.g., Caine et al., 1996; Evans Such lowered permeabilities could help promote
lites taking their place. PSS cataclasite samples et al., 1997). Microstructural evidence here indi- pore fluid trapping, ultimately resulting in pres-
that have accommodated ~30 m displacement cates that static permeability evolution in basalt- sure elevation and hydrofracture.
(Fig. 3D) comprise chaotic, generally matrix- hosted faults relates to the interplay between As fault displacement accumulates, zeolite
supported basalt, zeolite, and calcite clasts set in clay sealing, zeolite vein connectivity, catacla- and calcite veins link to form throughgoing,

A: host (no disp.) B: peripheral damage (< 1 m disp.) C: fault core (> 1 m disp.) D: mature fault core (>10 m disp.)
clay development clay replacement clay fabric re-development
0.5 mm 0.5 mm 0.5 mm 0.25 mm 0.5 mm

calcite
clay fabric clay fabric
groundmass

porous
zeolite clay fabric
fill

0.5 mm 0.5 mm 0.5 mm calcite 0.5 mm 1.0 mm

fault / fracture / joint network

clay fabric

clay fabric clay fabric crack-seal

zeolite
texture clay fabric

clay fabric
zeolite

0.25 mm 0.5 mm clay fabric 0.5 mm 1.0 mm

clay fabric
zeolite
clay fabric
vein network

no mineral veins zeolite

in host rock

calcite
calcite
porous
zeolite
calcite fill zeolite

vein development vein linkage vein network destruction

Figure 3. Plane-polarized light photomicrographs of host- and fault-rock assemblages used for experimental permeability measurement
(Fig. 2). Samples were prepared using blue-stained resin to show porosity. See text for description and discussion. A: Host rocks. B: Crackle
(left) and mosaic (right) breccias. C: Fault core chaotic breccia. D: Principal slip zone comprising chaotic breccia and cataclasite and/or
gouge. disp.—displacement.

GEOLOGY | January 2013 | www.gsapubs.org 3

 Geology, published online on 19 October 2012 as doi:10.1130/G33508.1

H., Sigfusson, B., Broecker, W.S., Matter, J.M.,

1: fault parallel
Stute, M., Axelsson, G., and Fridriksson, T.,
2010, Mineral sequestration of carbon dioxide
in basalt: A pre-injection overview of the Carb-
Fix project: International Journal of Greenhouse
log k

Gas Control, v. 4, p. 537–545, doi:10.1016/j

.ijggc.2009.11.013.
Gysi, A.P., and Stefánsson, A., 2008, Numerical mod-
eling of CO2-water-basalt interaction: Mineral-
2: fault normal
ogical Magazine, v. 72, p. 55–59, doi:10.1180
Time
/minmag.2008.072.1.55.
fault linkage stage core development stage PSZ cataclasis Hawkins, D.G., 2004, No exit: Thinking about leak-
(< 1 m displacement) (> 1 m displacement) (10s of m displacement)
age from geologic carbon storage sites: Energy,
v. 29, p. 1571–1578, doi:10.1016/j.energy.2004
.03.059.
Moy, D.J., and Imber, J., 2009, A critical analysis of
1
the structure and tectonic significance of rift-
oblique lineaments (‘transfer zones’) in the
2
Mesozoic–Cenozoic succession of the Faroe–
Shetland Basin, NE Atlantic margin: Journal
Figure 4. Summary model for the evolution of fault-parallel and fault-normal permeability (k), of the Geological Society, v. 166, p. 831–844,
based on experimental data combined with microstructural characterization of fault zone as- doi:10.1144/0016-76492009-010.
semblages. Black lines indicate fractures. Light gray to black zones show distribution of low Oelkers, E.H., and Cole, D.R., 2008, Carbon diox-
to high strain zones during fault development. PSZ—principal slip zone. ide sequestration. A solution to a global prob-
lem: Elements, v. 4, p. 305–310, doi:10.2113
/gselements.4.5.305.
Passey, S.R., and Jolley, D.W., 2009, A revised
transient high-permeability pathways. Zeolite ACKNOWLEDGMENTS lithostratigraphic nomenclature for the Palaeo-
veins retain porosity following crystal growth This study was funded by Statoil U.K. Ltd. and gene Faroe Islands Basalt Group, NE Atlantic
(Fig. 3B and 3C), maintaining high permeability the SINDRI Group. The authors thank D. Ellis, J. Ocean: Earth and Environmental Science Trans-
Imber, P. Cowie, and three anonymous reviewers for actions of the Royal Society of Edinburgh, v. 99,
after faulting. If these veins are interconnected, constructive discussions and comments on earlier ver- p. 127–158, doi:10.1017/S1755691009008044.
fault rock static permeability will remain high; sions of this manuscript. Schutter, S.R., 2003, Hydrocarbon occurrence and
if they are not connected, static permeability exploration in and around igneous rocks, in
will be low (e.g., Figs. 2 and 3). REFERENCES CITED Petford, N., and McCaffrey, K.J.W., eds., Hy-
Larger-displacement (>1 m) fault core sam- Armitage, P.J., Faulkner, D.R., Worden, R.H., Aplin, drocarbons in Crystalline Rocks: Geological
A.C., Butcher, A.R., and Iliffe, J., 2011, Experi- Society of London Special Publication 214,
ples principally comprise interconnected zeo- mental measurement of, and controls on, perme- p. 7–33, doi:10.1144/GSL.SP.2003.214.01.02.
lite-calcite mineral fills, with only minor clay ability and permeability anisotropy of caprocks Walker, R.J., Holdsworth, R.E., Imber, J., and El-
content (Fig. 3C); hence static permeability is from the CO2 storage project at the Krechba lis, D., 2011a, Onshore evidence for progres-
high. Further slip leads to static permeability Field, Algeria: Journal of Geophysical Research, sive changes in rifting directions during con-
v. 116, B12208, doi:10.1029/2011JB008385. tinental break-up in the NE Atlantic: Journal
reduction in the fault core through cataclasis, of the Geological Society, v. 168, p. 27–48,
Brace, W.F., Walsh, J.B., and Frangos, W.T., 1968,
clay gouge development, and dismemberment Permeability of granite under high pres- doi:10.1144/0016-76492010-021.
of existing zeolite/calcite veins. Nevertheless, sure: Journal of Geophysical Research, v. 73, Walker, R.J., Holdsworth, R.E., Imber, J., and El-
the static permeability of the PSS (Figs. 2 and p. 2225–2236, doi:10.1029/JB073i006p02225. lis, D., 2011b, The development of cavities
3D) remains more than an order of magnitude Caine, J.S., and Forster, C.B., 1999, Fault zone archi- and clastic infills along fault-related fractures
tecture and fluid flow: Insights from field data in Tertiary basalts on the NE Atlantic Margin:
higher (both kp and kN) compared to low-strain Journal of Structural Geology, v. 33, p. 92–106,
and numerical modeling, in Haneberg W.C., et
fault rocks, despite fragmentation of the rela- al., eds., Faults and Subsurface Flow in Shal- doi:10.1016/j.jsg.2010.12.001.
tively high-permeability veins and develop- low Crust: American Geophysical Union Geo- Walker, R.J., Holdsworth, R.E., Imber, J., and Ellis,
ment of a fine-grained matrix (cataclasite and/ physical Monograph 113, p. 101–127. D., 2012, Faults, fault rocks and fractures in
Caine, J.S., Evans, J.P., and Forster, C.B., 1996, Fault flood basalt provinces: A case study from the
or gouge). Inspection of hand specimens sug-
zone architecture and permeability structure: Faroe Islands, NE Atlantic Margin: Geological
gests that this could be due to out-of-section- Geology, v. 24, p. 1025–1028, doi:10.1130/0091 Society of America Bulletin, v. 124, p. 1382–
plane connectivity of permeable vein frag- -7613(1996)024<1025:FZAAPS>2.3.CO;2. 1393, doi:10.1130/B30512.1.
ments (Fig. 3D). Evans, J.P., Forster, C.B., and Goddard, J.V., 1997, Woodcock, N.H., Omma, J.E., and Dickson, J.A.D.,
The high-permeability–high-strain-zone, low- Permeability of fault-related rocks, and impli- 2006, Chaotic breccia along the Dent Fault,
cations for hydraulic structure of fault zones: NW England: Implosion or collapse of a fault
permeability–low-strain-zone model presented void?: Journal of the Geological Society, v. 163,
Journal of Structural Geology, v. 19, p. 1393–
here promotes lateral compartmentalization and 1404, doi:10.1016/S0191-8141(97)00057-6. p. 431–446, doi:10.1144/0016-764905-067.
along-fault fluid migration. However, because the Faulkner, D.R., and Rutter, E., 2001, Can the mainte-
low-strain zone forms a barrier to flow, fluids are nance of overpressured fluids in large strike-slip
able to migrate only from and to relatively high- fault zones explain their apparent weakness?: Manuscript received 10 April 2012
Geology, v. 29, p. 503–506, doi:10.1130/0091 Revised manuscript received 26 June 2012
permeability horizons (e.g., lava unit vesicular -7613(2001)029<0503:CTMOOF>2.0.CO;2. Manuscript accepted 12 July 2012
flow tops), bypassing the bulk of the lava flow Gislason, S.R., Wolff-Boenisch, D., Stefansson, A.,
stratigraphy via an ‘insulated’ fault core. Oelkers, E.H., Gunnlaugsson, E., Sigurdardottir, Printed in USA

4 www.gsapubs.org | January 2013 | GEOLOGY

View publication stats