Fractures, Fluid Flow and Mineralization

Prévia do material em texto

Fractures, Fluid Flow and Mineralization
Geological Society Special Publications
Series Editors
A. J. FLEET
R. E. HOLDSWORTH
A. C. MORTON
M. S. STOKER
It is recommended that reference to all or part of this book should be made in one of the following
ways:
MCCAFFREY, K. J. W., LONERGAN, L. & WILKINSON, J. J. (eds) 1999. Fractures, Fluid Flow and
Mineralization. Geological Society, London, Special Publications, 155.
CLOKE, I. R., CRAIG, J. & BLUNDELL, D. J. 1999. Structural controls on hydrocarbon and mineral
deposits within the Kutai Basin, East Kalimantan. In: MCCAFFREY, K. J. W., LONERGAN. L. &
WILKINSON, J. J. (eds) Fractures, Fluid Flow and Mineralization. Geological Society, London, Special
Publications, 155, 213-232.
GEOLOGICAL SOCIETY SPECIAL PUBLICATION NO. 155
Fractures, Fluid Flow and Mineralization
EDITED BY
KEN MCCAFFREY
School of Geological Sciences,
Kingston University, Surrey, UK
LIDIA LONERGAN & JAMIE WILKINSON
T. H. Huxley School of the Environment, Earth Sciences and Engineering
Imperial College of Science, Technology and Medicine, London, UK
1999
Published by
The Geological Society
London
THE GEOLOGICAL SOCIETY
The Society was founded in 1807 as The Geological Society of London and is the oldest geological society in the
world. It received its Royal Charter in 1825 for the purpose of 'investigating the mineral structure of the Earth'.
The Society is Britain's national society for geology with a membership of around 8500. It has countrywide
coverage and approximately 1500 members reside overseas. The Society is responsible for all aspects of the
geological sciences including professional matters. The Society has its own publishing house, which produces
the Society's international journals, books and maps, and which acts as the European distributor for publications
of the American Association of Petroleum Geologists. SEPM and the Geological Society of America.
Fellowship is open to those holding a recognized honours degree in geology or cognate subject and who have at
least two years' relevant postgraduate experience, or who have not less than six years' relevant experience in
geology or a cognate subject. A Fellow who has not less than five years' relevant postgraduate experience in the
practice of geology may apply for validation and, subject to approval, may be able to use the designatory letters
C Geol (Chartered Geologist).
Further information about the Society is available from the Membership Manager, The Geological Society.
Burlington House. Piccadilly, London W1V OJU, UK. The Society is a Registered Chanty. No. 210161.
Published by The Geological Society from:
The Geological Society Publishing House
Unit 7, Brassmill Enterprise Centre
Brassmill Lane
Bath BA13JN
UK
(Orders: Tel. 01225445046
Fax 01225 442836)
First published 1999
Reprinted 2001
The publishers make no representation, express or
implied, with regard to the accuracy of the information
contained in this book and cannot accept any legal
responsibility for any errors or omissions that may be
made.
c The Geological Society of London 1999. All rights
reserved. No reproduction, copy or transmission of
this publication may be made without written per-
mission. No paragraph of this publication may be
reproduced, copied or transmitted save with the pro-
visions of the Copyright Licensing Agency, 90 Totten-
ham Court Road, London W1P 9HE. Users registered
with the Copyright Clearance Center. 27 Congress
Street, Salem, MA 01970, USA: the item-fee code for
this publication is 0305-8719 99 $15.00.
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the
British Library.
ISBN 1-86239-034-7
ISSN 0305-8719
Distributors
USA
AAPG Bookstore
PO Box 979
Tulsa
OK 74101-0979
USA
(Orders: Tel. (918) 584-2555
Fax (918) 560-2652)
Australia
Australian Mineral Foundation
63 Conyngham Street
Glenside
South Austral ia 5065
Australia
(Orders: Tel. (08) 379-0444
Fax (08) 379-4634)
Affil iated East-West Press PVT Ltd
G-l 16 Ansari Road
New Delhi 110 002
India
(Orders: Tel. ( 1 1 ) 327-9113
F a x ( 1 1 ) 326-0538)
Japan
Kanda Book Trading Co.
Cityhouse Tama 204
Tsurumaki 1-3-10
Tama-Shi
Tokyo 0206-0034
Japan
(Orders: Tel. (0423) 57-7650
Fax (0423) 57-7651)
Typeset by Aarontype Ltd, Unit 47. Easton Busines
Centre, Felix Road', Bristol BS50HE. UK
Printed by Cambridge University Press.
Cambridge. UK
Contents
Dave Johnston: an appreciation and bibliography vii
LONERGAN, L., WILKINSON, J. J. & MCCAFFREY, K. J. W. Fractures, fluid flow and 1
mineralization: an introduction
Fracture populations
ROBERTS, S., SANDERSON, D. J. & GUMIEL, P. Fractal analysis and percolation properties 7
of veins
STOWELL, J. F. W., WATSON, A. P. & HUDSON, N. F. C. Geometry and population 17
systematics of a quartz vein set, Holy Island Anglesey, North Wales
GILLESPIE, P. A., JOHNSTON, D. J., LORIGA, M. A., MCCAFFREY, K. J. W., WALSH, J. J. & 35
WATTERSON, J. Influence of layering on vein systematics in line samples
LORIGA, M. A. Scaling systematics of vein size: an example from the Guanajuato mining 57
district (Central Mexico)
Fluid flow and fracture systems
SANDERSON, D. J. & ZHANG, X. Critical stress localization of flow associated with 69
deformation of well-fractured rock masses, with implications for mineral deposits
JONES, M. A., PRINGLE, A. B., FULTON, I. M. & O'NEILL, S. Discrete fracture network 83
modelling applied to groundwater resource exploitation in southwest Ireland
CONNOLLY, P. T. & COSGROVE, J. W. Prediction of static and dynamic fluid pathways within 105
and around dilational jogs
Structural controls on mineralization
Cox, S. F. Deformational controls on the dynamics of fluid flow in mesothermal gold 123
systems
BLENKINSOP, T. G. & SANDERSON, D. J. Are gold deposits in the crust fractals? A study of 141
gold mines in the Zimbabwean craton
JOLLEY, S. J., HENDERSON, I. H. C., BARNICOAT, A. C. & Fox, N. P. C. Thrust-fracture 153
network and hydrothermal gold mineralization: Witwatersrand basin, South Africa
ROSSETTI, P. & COLOMBO, F. Adularia-sericite gold deposits of Marmato (Caldas, 167
Colombia): field and petrographical data
BRANQUET, Y., CHEILLETZ, A., GIULIANI, G., LAUMONIER, B. & BLANCO, O. Fluidized 183
hydrothermal breccia in dilatant faults during thrusting: the Colombian emerald deposits
BROWN, M. A. N., JOLLY, R. J. H., STONE, W. & COWARD, M. P. Nickel ore troughs in 197
Archaean volcanic rocks, Kambalda, Western Australia: indicators of early extension
CLOKE, I. R., CRAIG, J. & BLUNDELL, D. J. Structural controls on hydrocarbon and mineral 213
deposits within the Kutai Basin, East Kalimantan
Irish Zn/Pb deposits: structure and fluid flow
HITZMAN, M. W. Extensional faults that localize Irish syndiagenetic Zn-Pb deposits and 233
their reactivation during Variscan compression
CONTENTS vi
EVERETT, C. E., WILKINSON, J. J. & RYE, D. M. Fracture-controlled fluid flow in the Lower 247
Palaeozoic basement rocks of Ireland: implications for the genesis of Irish-type Zn-Pb
deposits
LEWIS, H. & COUPLES, G. D. Carboniferous basin evolution of central Ireland - simulation 277
of structural controls on mineralization
SEVASTOPULO, G. D. & REDMOND, P. Age of mineralization of carbonate-hosted, base metal 303
deposits in the Rathdowney Trend, Ireland
O'REILLY, B. M., READMAN, P. W. & MURPHY, T. Gravity lineaments and Carboniferous- 313
hosted base metal deposits of the Irish Midlands
Index 323
Dave Johnston: an appreciation and bibliography
On 2 October 1995, Dave Johnston disappeared
while working in the field at Annagh Head,
Belmullet, County Mayo, Ireland. He is believed
to have been washed from the shoreline by a
freak wave. Dave graduated in 1980 with a first-
class honours degree in Geology from Trinity
College Dublin and he returnedthere in 1984
having been appointed as a lecturer in the
Department of Geology. In the intervening
years, Dave carried out postgraduate research
at Monash University, Melbourne and gained a
PhD for his work on the structural controls of
Uranium deposits in the Rum Jungle area,
Northern territories. His subsequent research
focused on the structural controls of the precious
and base-metal deposits and the underlying
tectonics and structure of Ireland. In recent
years he published a number of significant papers
on these topics and, in particular, he was at the
forefront in the application of fractal and chaos
theory to geological phenomena and the quanti-
fication of mineral deposits in particular.
Dave was an enthusiastic teacher and inspired
many of the students and researchers who he
taught or worked with. Indeed, many former
students now work in a diverse range of geology-
related industries. He firmly believed that aca-
demia could interact and collaborate more
closely with industry to their mutual benefit.
Thus, he was elected as a council member for
dthe Irish Association for Economic Geology.
In his other role as President of the Irish
Geological Association he was keen to bring
geology to a wider public audience. Dave talked
freely to students and colleagues and liked
nothing better than discussing geology over a
pint after a day in the field. He was a deeply
caring individual who made time to help anyone
who needed it. Throughout, he remained closely
attached to his home and family in North
Dublin and considered himself to be very for-
dtunate to be able to do what he loved best,
geological research based in the magnificent
surroundings of Trinity College.
Dave will be remembered as a lively and
colourful character who enjoyed life to the full,
whether it be playing rugby, scuba diving, doing
fieldwork or travelling. For many years he led the
singing at the late-night 'Friends of the Irish'
party at the Tectonics Studies Group annual
meeting. He always had a bad joke or a strange
story to tell, often it was the many slightly bizarre
incidents that always seemed to happen to him.
These usually involved the series of clapped-out
vehicles that he always seemed to own. There was
the time his car rolled down a cliff in Donegal,
the time his car's engine went on fire, but because
he was late for a meeting at the Lisheen deposit,
he threw some water on it and drove on! Anyone
who knew or met Dave has a similar story to
recount of his various exploits.
This volume on 'Fractures, Fluid Flow and
Mineralization' is a fitting tribute to his memory.
As George Sevastopulo so appropriately said at
the conference held in Dublin in November 1997
to commemorate Dave's life, 'Dave would have
loved to have been here'. He is greatly missed by
his former students, many colleagues and friends
around the world.
K. J. W. McCaffrey
School of Geological Sciences,
Kingston University, Kingston-upon-Thames,
Surrey KT1 2EE, UK
BIBLIOGRAPHY
Bibliography
WALL, V. J., ETHERIDGE, M. A., Cox, S. F. &
JOHNSTON, J. D. 1984. Regional metamorphic
ore deposits - structural and chemical controls.
In: MARSHALL, B. & GILLIGAN, L. B. (eds)
Mechanical and Chemical Mobilization of Metal-
liferous Mineralization. Geological Society of
Australia Short Course, 33-40.
JOHNSTON, J. D. 1985. Interpretation of refolding and
asymmetric folds using vergence concepts in drill-
core. Journal of Structural Geology, 7, 317-326.
JOHNSTON, J. D. 1988. Structural controls of gold
mineralization in Western Australia. Irish Associa-
tion for Economic Geology Annual Review 1988, 74.
SANDERS, I. S. & JOHNSTON, J. D. 1989. The Tor-
ridonian Stac Fada Member; an extrusion of
fluidised peperite? Transactions of the Royal
Society of Edinburgh: Earth Sciences, 80, 1-4.
PALMER, D., JOHNSTON, J. D., DOOLEY, T. & MAGUIRE,
K. 1989. The Silurian of Clew Bay, Ireland: part
of the Midland Valley of Scotland? Journal of
the Geological Society, London, 146, 385-388.
McELDUFF, B. & JOHNSTON, J. D. 1989. Bedrock
occurrence of Platinum Group Metals (PGMS) in
Ireland. Irish Association for Economic Geology
Annual Review 1989, 111.
JOHNSTON, J. D. 1990. The Untamed Earth. Technol-
ogy Ireland, 55-58.
SANDERS, I. S. & JOHNSTON, J. D. 1990. Reply to The
Torridonian Stac Fada Member: a discussion'.
Transactions of the Royal Society of Edinburgh:
Earth Sciences, 81, 249-250.
MURPHY, F. C, ANDERSON, T. B. DALY, J. S.,
GALLAGHER, V., GRAHAM, J. R., HARPER,
D. A. T., JOHNSTON, J. D., KENNAN, P. S., KEN-
NEDY, M. J., LONG, C. B., MORRIS, J. H., O'KEEFE,
W. G , PARKES, M., RYAN, P. D., SLOAN, R. J.,
STILLMAN, C. J., TIETZSCH-TYLER, D., TODD, S. P.
& RAFTER, J. P. 1991. An appraisal of Caledo-
nian suspect terranes in Ireland. Irish Journal of
Earth Sciences, 11, 11-41.
VAUGHAN, A. M. P. & JOHNSTON, J. D. 1992.
Structural constraints on closure geometry
across the lapetus Suture in eastern Ireland.
Journal of the Geological Societv, London, 149,
65-74.
JOHNSTON, J. D. 1992. The fractal geometry of vein
systems: the potential for ore reserve calculations.
In: BOWDEN, A. A., EARLS, G., O'CONNOR, P. G.
& PYNE, J. F. (eds) The Irish Minerals Industry
1980-1990. Irish Association for Economic
Geology, Dublin, 105-117.
JOHNSTON, J. D. 1993. Ice wedge casts in the Dalradian
of south Donegal: evidence for subaerial expo-
sure of the Boulder Bed. Irish Journal of Earth
Sciences, 12, 13-26.
JOHNSTON, J. D. 1993. Three-dimensional geometries
of veins and their relationship to folds; examples
from the Carboniferous of eastern Ireland. Irish
Journal of Earth Sciences, 12, 47-63.
MCCAFFREY, K. J. W., JOHNSTON, J. D. & FEELY. M.
1993. Use of fractal statistics in the analysis of
Mo-Cu mineralisation at Mace Head. County
Galway. Irish Journal of Earth Sciences. 12,
139-148.
JOHNSTON, J. D., TAIT, J. A., OLIVER. G. J. H. &
MURPHY, F. C. 1993. Evidence for a Caledonian
orogeny in Poland. Transactions of the Royal
Society of Edinburgh: Earth Sciences, 85, 131-142.
JOHNSTON, J. D., MCCAFFREY, K. J. W., LORIGA, M. A..
WATTERSON, J., WALSH, J. J. & GILLESPIE, P. A.
1994. A Manual Describing Recording, Analysis
and Prediction of Vein and Related Fracture
Distribution. MIRO, Lichfield.
FITZGERALD, G., FEELY, M., JOHNSTON, J. D.. CLAY-
TON, G., FITZGERALD, L. J. & SEVASTOPULO, G. D.
1994. The Variscan thermal history of west Clare.
Ireland. Geological Magazine, 131. 545-558.
JOHNSTON, J. D. 1995. Major northwest-directed
Caledonian thrusting and folding in Precambrian
rocks, northwest Mayo, Ireland. Geological Mag-
azine, 132. 91-112.
JOHNSTON, J. D. 1995. Pseudomorphs after ikaite in a
glaciomarine sequence in the Dalradian of Done-
gal, Ireland. Scottish Journal of Geology. 31. 3-9.
JOHNSTON, J. D. & PHILLIPS, W. E. A. 1995. Terrane
amalgamation in the Clew Bay region, west of
Ireland. Geological Magazine. 132. 797-806.
JOHNSTON, J. D. 1995. A review of the pre-Devonian
rocks of Ireland. ///: ANDERSON. K., ASHTON. J..
EARLS. G., HITZMAN, M. & TEAR, S. (eds) Irish
Carbonate-hosted Zn-Pb Deposits. Society of Eco-
nomic Geologists Guidebook Series 21. 209-217.
JOHNSTON, J. D. 1995. Variscan deformation in Ire-
land. In: ANDERSON, K.. ASHTON. J.. EARLS. G..
HITZMAN, M. & TEAR, S. (eds) Irish Carbonate-
hosted Zn-Pb Deposits. Society of Economic
Geologists Guidebook Series, 21, 111-113.
MCCAFFREY, K. J. W. & JOHNSTON. J. D. 1996.
Fractal analysis of a mineralised vein deposit:
Curraghinalt gold deposit. County Tyrone.
Mineralium Deposita. 31. 52-58.
JOHNSTON, J. D. & MCCAFFREY, K. J. W. 1996.
Fractal geometries of vein systems and the
variation of scaling relationships with mechan-
isms. Journal of Structural Geology. 18. 349-358.
WILKINSON, J. J. & JOHNSTON. J. D. 1996. Pressure
fluctuations, phase separation, and gold precipi-
tation during seismic fracture propagation.Geol-
ogy, 24, 395-398.
JOHNSTON, J. D., COLLER, D., MILLAR, G. & CRITCH-
LEY, M. F. 1996. Basement structural controls on
Carboniferous-hosted base metal mineral depos-
its in Ireland. ///: STROGEN, P.. SOMERVILLE. I. D.
& JONES, G. LL. (eds) Recent Advances in Lower
Carboniferous Geology. Geological Society. Lon-
don, Special Publications, 107. 1-21.
JOHNSTON, J. D. 1997. Localization of mid-crustal
thrust ramps by metadolerite sheets in the
Dalradian of northwest Ireland. Geological
Magazine. 134, 199 212.
Compiled by P. N. Wyse Jackson, Department of Geology, Trinity College, Dublin 2.
Fractures, fluid flow and mineralization: an introduction
LIDIA LONERGAN,1 JAMIE WILKINSON1 & KEN McCAFFREY2
1
 T. H. Huxley School of Environment, Earth Sciences & Engineering, Imperial College of
Science, Technology and Medicine, London SW7 2BP, UK
School of Geological Sciences, Kingston University, Kingston-upon-Thames,
Surrey KT1 2EE, UK
The historical context
Prior to Hubbert and Rubey's classic paper
(Hubbert & Rubey 1959), which specifically
recognized the role of high fluid pressures in
lowering the shear stress required to move and
emplace large thrust sheets, geologists had
tended to ignore the importance of fluids in
crustal deformation (Fyfe et al. 1978). Much has
changed in the intervening 40 years. It is now
accepted that not only do fluids enable deforma-
tion but that the converse is also true, i.e. that
faulting can cause fluid migration. The grow-
ing interest in the relationship between fluid
flow and deforming rock in a range of geological
environments has led to knowledge on the
topic becoming increasingly specialized. This is
reflected by the number of major books pub-
lished in the last decade spanning subject areas
more traditionally considered the realm of struc-
tural geologists, metamorphic or sedimentary
petrologists and hydrogeologists. Notable exam-
ples include: Fluids in Subduct ion Zones (Tarney
et al. 1991); The Role of Fluids in Crustal Pro-
cesses (National Research Council, US Geophy-
sics Study Committee 1990); Geofluids (Parnell
1994), focusing on the flow of fluid in sedi-
mentary basins; Rock Fractures and Fluid Flow
(National Research Council, US Committee
on Fracture Characterization and Fluid Flow
1996); and Fluid Flow and Transport in Rocks
(Jamtveit & Yardley 1997) which addresses the
integration of the physics and chemistry of fluid
flow through crustal rocks.
Data collated from the major geoscience
reference database, GeoRef, shows an exponen-
tial growth in the number of papers published
since 1975 that include 'faults/fracturing' and
'fluid' in their title or among their keywords (Fig.
la). This growth, culminating in a publication
rate in excess of 400 papers per annum, reflects
the explosion in research interest in the subject
area in the last 10-15 years. In the period from
1975 to the early 1980s, most of these publica-
tions, typically of the order of 40 per year,
were concerned with ore deposits or mineral-
ization (Fig 1b). This is not surprising given
that many modern ideas concerning the flow of
fluids in fractures stem from observations made
as a result of human exploitation of mineral
resources.
The origin of many modern concepts may be
traced to works such as Waldemar Lindgren's
classic text Geology of Mineral Deposits (Lind-
gren 1913) in which he presented the first classifi-
cation and interpretation of mineral deposits
within a genetic context. It was recognized, even
then, that 'the great majority of mineral deposits
were formed where the path of [solutions] was
prescribed by openings in the rocks other than
those of ordinary pore space... These openings
of super-capillary size... are chiefly found in the
zone of fracture' (considered to be generally
<6km in the crust).
By the late 1980s and early 1990s papers were
no longer limited to mineralization studies but
had spread across a number of disciplines, princi-
pally rock mechanics, structural geology, meta-
morphic petrology, hydrology, water resources
and petroleum geology. Much of this recent
research has benefited from stimuli in three
main areas:
• an increased subsurface fluid database arising
from drilling for economic resources (hydro-
carbon, geothermal and mineral indust-
ries), as well as ODP and deep crust drilling
programs;
• the increased computing power of the last 20
years allowing more complex numerical
simulations to be performed;
• the impetus from groundwater resource man-
agement and nuclear waste disposal sectors.
Whilst endeavour in these fields has led to
significant progress along a number of avenues,
interdisciplinary papers and research colloquia
LONERGAN, L., WILKINSON, J. J. & MCCAFFREY, K. J. W. 1999. Fractures, fluid flow and mineralization:
an introduction. In: MCCAFFREY, K. J. W., LONERGAN, L. & WILKINSON, J. J. (eds) Fractures, Fluid Flow and
Mineralization. Geological Society, London, Special Publications, 155, 1-6.
L. LONERGAN ET AL.
Fig. 1. (a) Total number of references with 'fault/fracture' and 'fluid' in the title, abstract or keywords from 1975
to 1997 inclusive (data from GeoRef database). Data have been normalized for total number of publications in the
database for each year. Search terms used: '(fracture* or fault*) and fluid*, (b) Data in (a) shown on non-
normalized scale (•) and numbers of those publications which were about ore or mineralization ( )- Search
term used was '[(fracture* or fault*) and fluid*] and (ore or mineral*)'.
crossing the boundaries of the traditional subject
disciplines remain relatively rare. It seemed that
now, some 15 years on, was a timely moment to
reassess the subject of fluid flow and fracturing
in the context of mineralization, and to explore
how advances within different research commu-
nities could be applied to the understanding of
the formation of economic resources. It was also
felt that this was an appropriate way to celebrate
the life and work of Dave Johnston whose
interests were very much in this field. With these
aims in mind three communities of geoscien-
tists were invited to a meeting in Dublin in
November 1997: structural geologists, geoscien-
tists, with interests in fluid flow, and economic
geologists. The papers collected in this volume
evolved from those presented at the meeting and
are intended to give a state-of-the-art review of
current ideas on the interrelationships between
fracturing and fluid flow, particularly in relation
to the genesis of hydrothermal mineral deposits.
Permeability, fracture networks
and mineralization
An issue that has long been recognized as
fundamental in fluid flow research is how rock
permeability is defined and how to quantify it
adequately for fractured rock systems. Indeed,
in the introduction of Fluid Flow and Transport
in Rocks (Jamtveit & Yardley 1997) it is empha-
sized that rock permeability is the key parameter
in any model of fluid flow and transport in
2
FRACTURES, FLUID FLOW AND MINERALIZATION
rocks. This is particularly true for hydrothermal
mineral deposits that form as a result of focused
flow of a large volume of fluid. As many of the
papers in this volume illustrate (indicated in
bold), such mineralization is usually structurally
controlled, and hence the need to understand the
role of faulting and fracturing in enhancing rock
permeability becomes paramount.
The papers included in this volume reflect
how knowledge in this area has moved on sig-
nificantly since Lindgren's text (Lindgren 1913).
One of the most important new concepts, high-
lighted by the landmark paper of Sibson et al.
(1975), was the idea that faults and fracture
systems were not 'passive' fluid conduits but
could be active in transporting fluid through
rock during deformation. Research in neotec-
tonic areas (e.g. Muir Wood & King 1993)
confirmed, with observationaldata, that active
faults do move large volumes of fluid through
the crust. Now much research effort is focused
on how faulting alters the hydrogeological prop-
erties of the rock itself, by modifying the frac-
ture pore volume as well as the permeability of
the fault and adjacent areas (e.g. Knipe 1993;
Matthai & Roberts 1997).
The coupling between deformation processes
and rock permeability through linked fracture
systems is one of the main themes that recurs in
various sections in this volume. Some of this
work has its origins in the 'fractal revolution'
that has occurred in structural geology in the
last ten years or so. This approach has proved
effective in quantifying the geometry and scaling
characteristics of fracture systems, and has led
to an increased understanding of how fault
systems evolve and grow. Localization of
deformation in fractures and shear zones leads
to fluid flow localization. When active faults and
shear zones link to form percolation networks,
the large scale flow systems essential for miner-
alization can be established (e.g. Cox, Roberts
et al., Sanderson & Zhang). The integration of
advances from 'traditional' structural geology
with developments in percolation theory will
aid both the understanding of the genesis of
hydrothermal mineral deposits and improve the
ability to explore for these resources.
Vein populations
The first section in the volume addresses vein
systems. Veins are a key natural data set
preserving a visible and permanent record of
fluid flow and mineral transport in the rock
mass. Evidence from field studies shows that
various aspects of veins, especially their thick-
nesses, have fractal properties (for example, the
paper by Stowell et al. on veins hosted in
Precambrian rocks of Anglesey, Wales). Roberts
et al. deduce that the measured fractal dimen-
sion (or D-value) is a useful descriptive para-
meter and can potentially be used to predict the
connectivity of vein systems. From field data
and models Roberts et al. conclude that vein
systems with power-law thickness distributions
initiate due to the opening of isolated fractures,
but at this stage only locally derived fluids enter
the fractures. When the isolated vein systems
become connected they retain power-law thick-
ness distributions but with characteristically
lower D-values. Deformation is localized and
fluid transport over increased length scales can
occur, leading to mineral precipitation given
suitable physical and chemical conditions. Thus,
thickness distributions of mineralized vein sys-
tems would be predicted to have low D-values.
From a study of the vein system of the Guana-
juato epithermal Ag-Au deposit in Mexico,
Loriga also provides data which show that vein
thickness populations can be described by a
power-law distribution over four orders of mag-
nitude, inferring a common growth process for
the veins in the deposit. Like Roberts et al. and
Gillespie et al., she concludes that veins with
power-law distributions will be more prone to
mineralization because they form larger con-
nected fracture networks allowing large scale
fluid transport. Gillespie et al. identify funda-
mental differences between stratabound vein
arrays (arrays in layered rocks, where the veins
are restricted to individual mechanical units)
and non-stratabound vein arrays based on
studies of vein arrays at ten localities in the
British Isles, France and Mexico. The strata-
bound vein arrays have regular spacings con-
trolled by the layer thickness and do not show a
power-law thickness distribution. In compari-
son, non-stratabound veins are clustered and
exhibit a power law thickness distribution, with
D-values that appear to be independent of
lithological variation.
Fluid flow and fracture systems
In the second section of the book the specifics of
fluid flow in fracture systems is addressed from a
number of different perspectives. Sanderson &
Zhang investigate fluid flow and deformation
in fractured rock using numerical models, in
which mechanical and hydraulic behaviour are
coupled. They show that, at a critical stress
state, initially diffuse flow though fracture net-
works changes to a highly localized flow. The
3
L. LONERGAN ET AL.
overall hydraulic conductivity within and across
the fracture networks increases suddenly when
the critical stress state is reached. At the critical
state, the variation in vertical flow rates through
different parts of the network shows a fractal
distribution. Jones et al provide an example of a
sophisticated 3D model of a fractured ground-
water reservoir in Carboniferous limestones in
southwest Ireland, which explicitly represents
fractures using data from boreholes to con-
strain the distribution of the fracture popula-
tions. This contribution illustrates the sort of
complex modelling and simulation that are more
common in the hydrological and nuclear waste
sectors, but which have great potential for
mineralization or hydrocarbon studies. Connolly
& Cosgrove address the importance of stress
distribution as a controlling factor in influenc-
ing fluid migration in tectonically active upper
crustal regimes. They use photoelastic experi-
ments and theory of brittle failure to obtain pre-
dictions of stresses, fractures and fluid pathways
around dilation jogs during stick-slip faulting.
They investigate the case of applying a constant
load to faults that are not propagating (static)
and then go on to consider fluid migration path-
ways associated with dynamic faulting events.
In their example of a dilational jog that starts
out with no overlap and evolves to an overlap-
ping fault system, fluid is initially driven away
from the jog but at later growth increments is
focused into the jog at high flow rates.
Structural controls on mineralization
The third section of the volume is devoted to
mineral deposits. A collection of seven papers
covers a variety of topics with the general theme
of structural controls of mineral deposit for-
mation, including some new thoughts on old
themes. A provocative paper by Cox re-examines
existing ideas on the genesis of mesothermal gold
deposits, applying new concepts in percolation
theory to their formation. Arguments are pre-
sented for the critical percolation state of these
systems, providing a natural analogue for the
models described by Sanderson & Zhang. Gold
mineralization of this type is considered on a
regional scale by Blenkinsop & Sanderson who
demonstrate that deposits in the Zimbabwean
craton have a fractal distribution with a D-value
of c. 1. If it is assumed that this D-value can be
extended to other areas, then the authors predict
that companies can improve their exploration
strategies by comparing observed deposit dis-
tributions with fractal dusts. An important new
data set is presented by Jolley et al., who discuss
the major control of lithology on the geometry of
structures formed during thrusting in the extra-
ordinarily gold-rich Witwatersrand Basin in
South Africa. Critical evidence is presented
which shows that deformation was coincident
with, and controlled the distribution of, hydro-
thermal fluid flow and gold mineralization. This
is key information in relation to the ongoing
controversy surrounding the genesis of Witwa-
tersrand gold and strongly supports a hydro-
thermal origin for these deposits, long regarded
by many workers to be alluvial placers. New
thoughts on an old theme are presented by Brown
et al., who provide a novel interpretation of the
famous Kambalda nickel deposits. Structural
data are used to infer that the komatiitic flows
hosting mineralization were constrained within
early extensional graben structures, similar to the
fissure swarms observed in Iceland at the present
day. This interpretation has considerable sig-
nificance with regard to exploitation of these
deposits.
Descriptions of vein textures areuseful in
interpreting the competing processes of fracture
opening and mineral precipitation, a competition
which ultimately governs the temporal evolution
of permeability in dynamic systems. Vein data
and fluid inclusions - the only direct record of
palaeofluid properties - are used in two other
contributions. Rossetti & Colombo present field
and petrographic data from the little known
adularia-sericite-type epithermal gold deposits
of Marmato, Colombia. Mineralization, hosted
within a Neogene volcanic centre, is controlled
by fractures associated with a regional fault
system. Variations in fluid properties with depth
are observed, indicating that gold precipitation
occurred in response to competing processes
of boiling and dilution in the vein system.
Branquet et al. consider the development of
fluidized hydrothermal breccias in dilatant zones
during thrusting to be the major control of
mineralization in the unusual emerald deposits
in the Eastern Cordillera of the same country.
The final paper in this section, by Cloke et al..
integrates seismic, gravity and structural data
to develop a model for the Kutai Basin in
Indonesia that shows how basement structure
controlled basin evolution and the location of
volcanism and hydrothermal activity responsible
for major gold deposits. It is shown how a
regional understanding of basin evolution and
fluid pathways is important for mineral and
hydrocarbon exploration. This contribution
illustrates how economic geologists, whether
their quest be for ore or hydrocarbons, can
benefit from modern concepts of basin analysis
coupled with effective use of regional geophysi-
cal data sets.
4
FRACTURES, FLUID FLOW AND MINERALIZATION
Irish base metal deposits
At the November 1997 conference in Dublin we
were fortunate to be able to bring together a
unique group of researchers working on the
Carboniferous base metal deposits of Ireland for
a lively discussion covering many aspects of the
genesis of Europe's premier Zn-Pb deposits.
A selection of these papers is presented in the
final section of this volume as a unified case
study where readers can see aspects of the differ-
ent approaches documented in previous sections
applied.
There has been exciting recent progress in
understanding the genesis of the Irish base metal
deposits with Hitzman, Everett et al., Lewis &
Couples and O'Reilly et al. discussing various
aspects of the importance of structural reactiva-
tion in basement rocks for controlling fluid
flow and mineralization. Hitzman discusses the
nature of the extensional fault systems which
localize the deposits and considers the evi-
dence for, and implications of, their inversion
during subsequent Variscan compression. This
represents a development of some of the ideas
presented by Dave Johnston (Johnston et al.
1996) in one of his last papers. One of the
currently popular models for mineralization in
Ireland proposes that the deposits formed as a
result of long distance flow of fluids through a
sandstone aquifer, driven by topographic head
from the rising Variscan mountains to the south
of the Irish Midlands Basin. This interpretation
is challenged by Everett et al., who present new
evidence that indicates that fluid flow in
fractured basement rocks was an important
process in the genesis of the Zn-Pb deposits.
This is a theme also addressed by O'Reilly et al.,
who use reprocessed gravity data to show how
the deposits are strongly localized by Caledo-
nian basement faults and that intersections with
north-northeast and northwest trending struc-
tures appear to play an important control on
localizing fluid flow and mineralization. Lewis &
Couples also contribute to the debate on fluid
flow within and beneath the sedimentary basins
of the Irish Midlands by constructing 2D
coupled fluid and heat flow simulations of the
two competing models (topo-graphically driven
regional flow and deposit specific density driven
convection). Their paper also shows the power
of sequential 3D basin reconstructions in aiding
the understanding of the tectonic evolution of
the Irish Midlands during the Carboniferous.
Finally, a controversial interpretation of the
timing of mineralization with respect to the age
of host rocks is presented by Sevastopulo &
Redmond. Their work implies that younger host
rocks than previously considered may be pro-
spective for Zn mineralization in Ireland.
A multidisciplinary problem
The integration of all available techniques is the
hallmark of current research in the field of fluid
flow and mineralization, and many of the papers
presented in this volume exemplify the use of
multidisciplinary approaches. The geochemistry
of fluids trapped within veins (e.g. Everett et al.),
geophysical remote sensing such as gravity or
seismic methods (O'Reilly et al. and Cloke et al.)
and numerical modelling (e. g. Jones et al.,
Lewis & Couples and Sanderson & Zhang)
all have their part to play alongside more
classical field geology in helping to solve out-
standing complex problems. One of the main
aims of this volume is to promote the cross-
fertilization of ideas between geoscientists inter-
ested in either the passage of fluids through the
Earth's crust or the genesis of mineral deposits.
Such interdisciplinary collaboration is essential
if further advances are to be made in this
research area.
This volume arose from a joint meeting of the Tectonic
Studies Group, Mineral Deposits Studies Group, Irish
Association for Economic Geology and Irish Geolo-
gical Association on 'Structural Controls and Genesis
of Economic Resources: Mineral and Hydrocarbon
Deposits' which was held in November 1997 in Trinity
College, Dublin. We thank Rio Tinto Mining &
Exploration, BHP Minerals, Billiton International
Development pic, CSA Ltd, ERA-Maptec and the
IAEG for their generous sponsorship of the meeting
which allowed us to bring together nearly 200
delegates from all around the world, to celebrate and
commemorate the life of Dave Johnston with a
stimulating scientific meeting and field trip to the
Irish base metal deposits. Behind the scenes, the
tireless logistical support given by Patrick Wyse
Jackson, Neil Kearney, Frank Hendron, Declan
Burke, Stephen Donnolly, JefT Lord, Mags Duncan,
Mary Foody, Joann Layng, Lynn Pink and Dick
Campain ensured a smoothly run, successful con-
ference. Garth Earls, Mike Boland and the staff at
Tara Mines (Navan), Minorco Services Ireland (Lish-
een) and Arcon Mines (Galmoy) are acknowledged for
their assistance in running the field trip. We thank all
contributors to the meeting for presenting the science
that is contained in this volume and for creating an
occasion that Dave himself would have truly enjoyed.
The following referees donated their time and
expertise for reviewing the manuscripts, often under
tight time constrains, and helped us bring this volume
to fruition: David Alderton, John Ashton, Chris Bean,
Tom Blenkinsop, Adrian Boyce, Olivier Bour, Joe
Cartwright, Mike Coward, Patience Cowie, Stephen
Cox, Ian Davidson, Garth Earls, Stuart Egan, Chris
5
L. LONERGAN ET AL.
Elders, Terry Engelder, Martin Feely, Kerry Galla-
gher, Paul Gillespie, Alan Herbert, Richard Herring-
ton, Murray Hitzman, Bob Holdsworth, Rob
Hunsdale, David James, Gawen Jenkin, Steve Jolley,
Richard Jolly, Richard Lisle, Iain Main, Ken McClay,
John McCloskey, Eric Nelson, Noelle Odling, Carl
Renshaw, Stephen Roberts, Dave Sanderson, George
Sevastopulo, Robin Shail, Rick Sibson, Ian Somer-
ville, Cees Swager, John Walsh, Nicky White, James
Wood, Robert Zimmerman and three other anon-
ymous referees.
References
FYFE, W. S., PRICE, N. J. & THOMPSON, A. B. 1978.
Fluids in the Earth's Crust. Elsevier.
GEOREF (CD-ROM) 1975-6/1998. American Geolo-
gical Institute, SilverPlatter.
HUBBERT, M. K. & RUBEY, W. W. 1959. Role of fluid
pressure in mechanics of overthrustfaulting. Geo-
logical Society of America Bulletin, 70, 115-166.
JAMTVEIT, B. & YARDLEY, B. 1997. Fluid Flow and
Transport in Rocks. Chapman & Hall.
JOHNSTON, J. D., COLLER, D., MILLAR, G. & CRITCH-
LEY, M. F. 1996. Basement structural controls on
Carboniferous-hosted base metal mineral deposits
in Ireland. In: STROGEN, P., SOMERVILLE, I. D. &
JONES, G. LI. (eds) Recent Advances in Lower Car-
boniferous Geology. Geological Society, London,
Special Publications, 107, 1-21.
KNIPE, R. J. 1993. The influence of fault zone pro-
cesses on fluid flow and diagenesis. In: HORBURY,
E. D. & ROBINSON, A. G. (eds) Diagenesis and
Basin Development. AAPG Studies in Geology.
36, 135-154.
LINDGREN, W. 1913. Geology of Mineral Deposits.
McGraw-Hill.
MATTHAI , S. K. & ROBERTS, S. G. 1997. Trans-
ient versus continuous fluid flow in seismically
active faults: an investigation by electric analogue
and numerical methods. In: JAMTVEIT. B. &
YARDLEY, B. (eds) Fluid Flow and Transport in
Rocks. Chapman & Hall, 263-295.
MUIR WOOD, R. & KING, G. C. P. 1993. Hydrological
signatures of earthquake strain. Journal of Geo-
physical Research, 98, 22 035-22 068.
NATIONAL RESEARCH COUNCIL. US COMMITTEE ON
FRACTURE CHARACTERIZATION AND FLUID
FLOW. 1996. Rock Fractures and Fluid Flow.
Contemporary Understanding and Applications.
National Academy Press.
NATIONAL RESEARCH COUNCIL. US GEOPHYSICS
STUDY COMMITTEE. 1990. The Role of Fluids in
Crustal Processes. National Academy Press.
PARNELL, J. (ed.) 1994. Geofluids: Origin, Migration
and Evolution of Fluids in Sedimetary Basins.
Geological Society, London. Special Publications.
78.
SIBSON, R. H., MOORE. J. M. & RANKIN, A. 1975.
Seismic pumping - a hydrothermal fluid trans-
port mechanism. Journal of the Geological Societv,
London, 131, 653-659.
TARNEY, J., PICKERING, K. T., KNIPE, R. J. & DEWEY.
J. F. 1991. The Behaviour and Influence of Fluids in
Suhduction Zones. Royal Societv. London.
6
Fractal analysis and percolation properties of veins
S. ROBERTS,1 D. J. SANDERSON13 & P. GUMIEL2
1
 School of Ocean and Earth Science, Southampton University,
Southampton Oceanography Centre, Southampton SO 14 3ZH, UK
Instituto Tecnologico Geominera de Espana, Rios Rosas 23, 28003 Madrid, Spain
3Present address: T. H. Huxley School of Environment, Earth Sciences of Engineering,
Imperial College of Science, Technology and Medicine, Prince Consort Road,
London SW7 2BP, UK
Abstract: Systematic sampling of vein systems often reveals a power-law distribution of the
number of veins (N) to the vein thickness (t) of the form N = Ct-D . A comparison of vein
thickness distributions from different geological settings establishes systematic variations
between vein populations, with lower D-values characterizing well-connected vein systems.
Low D-value vein systems are reported from mineral deposits and from veins developed
within and around fault zones, often at constant vein frequency. Simple models are
described to explain the field observations which suggest that the observed variations in vein
thickness distributions are directly related to the connectivity of the sampled vein network.
Mineralized vein systems are an economically
important manifestation of the flow of hydro-
thermal fluids through the Earth's crust. Major
ore deposits of Au, Cu, Sn, W, Pb and Zn are all
found within the confines of vein systems. This
paper describes physical methods of investigat-
ing the development and connectivity of hydro-
thermal vein systems and examines the increased
range of information which can be acquired.
Field data are reported from mineralized and
barren vein systems, and theoretical models
designed to describe their development. It is
argued that the methods described provide new
and useful information in the quest to under-
stand, explore and ultimately exploit mineralized
fracture networks.
Vein thickness distributions
Measurements of vein thickness along linear
traverses, ideally normal to the trend of the
veins, can often be made alongside more typical
vein descriptions of orientation, architecture,
age relationships, mineral content and fibre
orientations. In such instances, the thickness,
position and orientation of all veins exceeding
some minimum thickness, e.g. >l mm, are
recorded along the traverses. The data can be
compared with normal, log-normal, power-law
and negative exponential distributions based on
the shapes of the cumulative frequency curves.
Typically, cumulative frequency plots of vein
thickness conform to power-law distributions, of
the form
N= Ct-D (1)
where N is the number of veins with thickness
>t\ C represents the frequency of veins > unit
size; the exponent D is often termed the fractal
dimension of the distribution. A plot of log N v.
log t will be a straight line for a power-law
distribution, with slope —D (Fig. 1). Some of the
departure from the power-law distribution can
be attributed to two types of truncation: under-
sampling of the thinnest veins (usually <2 mm);
the absence of larger veins, either due to some
finite limit of vein opening or the low probability
of intersects on a sample traverse of short finite
length (Barton & Zoback 1990; Pickering et al
1996); these effects can be noted to varying
degrees in many of the plots given in this paper.
Vein thickness distributions have been widely
studied during the past few years (Sanderson
el al 1994; Clark et al 1995; Johnston &
McCaffrey 1996; McCaffrey & Johnston 1996;
Roberts et al 1998) and have generally been
found to conform to power-law distributions.
Data from two regions of Central Iberia are
outlined below to illustrate this approach.
Quartz veins, Central Portugal
Metamorphic quartz veins were investigated
from Precambrian Complejo Esquisto Grauva-
quico of the Castello Branco region of Central
ROBERTS, S., SANDERSON, D. J. & GUMIEL, P. 1999. Fractal analysis and percolation properties of veins.
In: MCCAFFREY, K. J. W., LONERGAN, L. & WILKINSON, J. J. (eds) Fractures, Fluid Flow and Mineralization.
Geological Society, London, Special Publications, 155, 7-16.
S. ROBERTS, D. J. SANDERSON & P. GUMIEL
Fig. 1. Cumulative frequency plot of vein thickness (t)
from two transects in the Castello Branco area,
Portugal. The data conform to power-law distributions
and Locality 3 shows a lower exponent (D-value)
reflecting the development of thicker veins. Both
data sets show evidence of truncation of data in the
1-3 mm region.
Portugal (Murphy & Roberts 1997) (Fig. 1). The
Complejo Esquisto Grauvaquico is a thick
sequence of metagreywackes, comprising inter-
bedded shales and silts, all metamorphosed to
lower greenschist facies. Well-exposed road
sections, typically >50m in length, contain
syn- and post-kinematic quartz veining of
variable character and intensity, typically orien-
tated N140°E, and ranging in thickness from
1 to 420mm. A series of line traverses were
completed, CB1-CB12, and a summary of the
data is given in Table 1. Two contrasting
examples of the data sets, CB3 and CB12, are
plotted on Fig. 1. Both data sets have similar
frequencies of veining > l m m (0.5-lm"1) but
contrast in their power-law exponents (D = 0.15
for CB3 and D= 1.3 for CB12). This reflects the
presence of thicker veins within the CB3 section,
with the majority of data falling between 10 and
80mm thickness, whereas for CB12 the majority
of the data fall between 2 and 10mm thickness.
Thus, the line samples plotted as log-log
cumulative frequency plots provide information
about the vein density for a given thickness at
each site, and the extent to which the system is
dominated by large or small veins. The latter
property is reflected in the exponent (D-value) of
the distribution, with low D-values reflecting an
increased presence of thick veins. Furthermore,
such an approach readily allows different sites tobe compared.
Ore-Mineralized system, La Codosera,
Western Spain
Analysis of ore-mineralized vein systems have
also revealed power-law vein thickness distribu-
tions (Sanderson et al. 1994; McCaffrey &
Johnston 1996; Roberts et al. 1998). In parti-
cular, a study involving the systematic measure-
ment of thickness data was introduced as part of
the routine drill core logging procedure in an
exploration programme of the La Codosera area
of western Spain (Roberts et al. 1991; Sanderson
et al. 1994). Measured vein thickness distribu-
tions generally conformed to power-law distri-
butions (Fig. 2). A comparison of the borehole
data shows a decrease in the exponent through
boreholes #25, #28 and #27 (Fig. 2), which
reflects an increase in vein thickness and vein
density, particularly at 10mm thickness, evident
within each hole. Notably, the lower D-value is
Table 1. Summary of line transect data from the Castillo Branco area of Portugal
Sample
CB1
CB2
CB3
CB4
CB5
CB6
CB7
CB8
CB9
CB10
CB11
CB12
n
94
72
46
45
71
49
12
38
72
36
94
79
Traverse
length
(m)
90
80
90
60
60
60
60
60
60
60
60
90
Thickness
Min.
2
2
3
2
2
2
2
2
2
2
2
1
(mm)
Max.
420
500
500
480
220
170
60
110
460
85
195
50
Thickness
(mm/m)
30.00
26.30
24.70
34.40
45.40
21.40
3.05
13.40
33.30
13.30
36.20
4.00
Density
(m-1)
1.04
0.90
0.51
0.75
1.18
0.82
0.20
0.63
1.20
0.60
1.57
0.88
D
0.80
0.60
0.75
0.60
0.70
0.77
0.73
0.75
0.75
0.70
0.95
1.30
8
FRACTAL ANALYSIS AND PERCOLATION PROPERTIES OF VEINS
Fig. 2. Vein thickness plots from the La Codosera area, southwest Spain. Number of veins per metre of core (N)
with thickness >t plotted v. thickness (t) from three different drill holes. Drill hole #21 shows the highest Au
assays and the lowest power-law exponent [after Sanderson et al. (1994)].
also broadly correlated with increasing amounts
of gold, reported from the assays returned from
each hole, which is not simply related to the total
vein thickness (Sanderson et al. 1994, fig. 2).
Vein thickness distributions and the
development of fracture networks
The characteristics of the power-law vein thick-
ness distributions observed, in particular the
D-values, must reflect the mechanisms respon-
sible for the formation and evolution of the vein
system. Thus, models produced to explain the
development of the vein systems should ideally
result in the generation of power-law distribu-
tions of vein thickness as opposed to normal,
log-normal, etc. This section considers a theore-
tical model and a detailed analysis of veins
developed within the damage zones of faults.
The results provide insights into how these vein
thickness distributions arise and their potential
significance.
A theoretical model
A simple conceptual model for the develop-
ment of a vein system is shown in Fig. 3 in which
an initial set of small fractures grows to pro-
duce a linked fracture network (Roberts et al.
1998). Zhang & Sanderson (1994) have discussed
network evolution in terms of the random addi-
tion of fractures with a power-law length dis-
tribution. They recognize this as a percolation
phenomenon (e.g. Stauffer & Aharony 1992),
suggesting that a system of fractures will event-
ually connect to form an infinite cluster at a
certain fracture density, which can be character-
ized by its fractal dimension. The set of fractures
within the infinite cluster, which provide path-
ways across the network, is termed the backbone
and this provides a path for fluid flow and
localized deformation.
An & Sammis (1996) developed a cellular
automaton model to simulate the growth of a
network of faults, which employed a set of
simple rules based on fracture mechanics, to
Fig. 3. Conceptual model of growth of a vein system: (a) initial, small, isolated fractures; (b) growth proportional
to length produces local linkage; (c) linkage to form an infinite cluster. Note that the opening (vein thickness)
is concentrated on the backbone of the infinite cluster. After Roberts et al. (1998).
9
10 S. ROBERTS, D. J. SANDERSON & P. GUMIEL
control initiation, growth and linkage of frac-
tures. Based on these studies a conceptual model
has been developed (Fig. 3) in which the
following rules are incorporated schematically:
• an initial power-law distribution of fracture
length with no further nucleation (Nur 1982;
Segall & Pollard 1980) (Fig. 3a); fractures are
located randomly and with random devia-
tion from a single mean orientation;
• thickness (opening) is proportional to frac-
ture length (e.g. Vermilye & Scholz 1995);
• change in length (AL) occurs where
ALocZ/, where k is a constant dependent
on the mechanisms controlling growth: fol-
lowing An & Sammis (1996), a value of k = 1
is used which develops geometries similar to
those observed in nature;
• linkage occurs between two fractures of
length LI and L2 when the distance between
their tips is <0.1 (L1 +L2).
Thus, it is possible to model the development of
an infinite cluster by growth rather than addition
of fractures.
At the percolation threshold, a cluster
of infinite length suddenly develops, which
localizes both deformation and fluid flow
(Zhang & Sanderson 1994) (Fig. 3c). As thick-
ness is proportional to length, this will also
produce a sudden opening of fractures located
on the backbone of this cluster, probably due to
the reduced elastic bending stresses required to
open the long arrays of linked fractures. Thus,
veins which form part of the infinite cluster will
have a wider range of thicknesses than those
forming isolated veins or clusters, and this
will be reflected in a lower fractal dimension (D).
For line samples through such vein systems,
the isolated veins will be dominated by small
length and thickness (D > 1), whereas the
connected systems will be dominated by larger
veins (D < 1).
Using stochastic models, Clark et al. (1995)
suggest that power-law distributions of vein
thicknesses can arise if the growth of the veins
and the associated network involves limited initi-
ation of new fractures and incremental growth in
proportion to the existing vein length. Adapting
these ideas and incorporating the fracture
linkage model (Fig. 3) allows vein thickness
distributions to be interpreted in the following
simple model. Pairs of cracks are randomly
sampled from an initial power-law distribution
of lengths and thicknesses (D = 1). Each pair is
linked by summing the lengths and thicknesses:
this assumes tip to tip linkage but the model
could be modified to incorporate other types of
Fig. 4. Results of simple model for vein thickness
distribution. An initial power-law distribution of
lengths and thickness (D = 1) is randomly sampled
and pairs of cracks linked by summing their lengths
and thicknesses. The resulting distribution for linking
of 50 and 75% of cracks, with growths of two and
four times, respectively, shows lower D-values and a
departure from ideal power-laws (after Roberts
et al. 1998).
linkage. The process is repeated until a given
proportion of cracks are linked and all cracks
have grown in proportion to their length. The
resulting distribution is shown in Fig. 4 for
linking of 0, 50 and 75% of cracks. The resulting
distributions have lower D-values (D < 1) and
show departure from ideal power-laws, particu-
larly for the larger cracks. The models also
produce patterns of vein thickness distribution
similar to those seen in field studies, cf. Fig. 4
with Figs 1 and 2.
Vein development in damage zones of faults
The above models predict the localized open-
ing of fractures, promoting enhanced fluid
flow, when connected fracture systems develop.
One situation where the transition from uncon-
nected to connected vein systems can be
examined is in the vicinity of fault zones where
regionalveins (usually unconnected) often
intensify and become connected in the damage
zones. In this section the relationship between
vein thickness distribution and localization of
deformation is examined from three traverses
across damage zones associated with exten-
sional faults.
FRACTAL ANALYSIS AND PERCOLATION PROPERTIES OF VEINS 11
Kilve, Somerset. An 8m long traverse across
three east-west trending normal fault zones
(FZ1-FZ3) of displacement 0.4, 0.1 and <0.1 m,
respectively, was examined within a single lime-
stone unit in the Lower Lias limestones and
shales to the east of Kilve Pill (Fig. 5). A total of
3.8m of the traverse was within damage zones
of faults, with the remaining 4.2 m lying outside
these zones. The limestone layer contains many
thin calcite veins with displacements <1 mm. All
veins that were observable (probably of thick-
ness >0.2 mm) were recorded, but the sample is
only considered to be complete for veins c. >0.5
mm in thickness, the average density at this
thickness being c. 10 m"1. The veins in the fault
zones penetrate the 0.25m thick limestone bed
and connect with veins in the underlying and
overlying shales, the remainder being confined
to the limestone.
The frequency of veins does not change within
the fault zones (Fig. 5a) but their opening, as
shown by steeper cumulative thickness slopes,
does. Log-log plots (Fig. 5b) suggest an approx-
imate power-law distribution, but the power-
law exponent (D-value) within the damage zones
is 0.5, compared with a value of c. 1.5 out-
side (Table 2). The significant result from this
study is that the thickness distributions are
markedly different for veins inside and outside
the fault damage zones. These vein distribu-
tions are interpreted as resulting from the local-
ized opening of fractures within the limestone
layers due to displacement and strain within
the fault zones.
Fig. 5. (a) Cumulative vein thickness and vein frequency developed over three small fault zones within an 8 m
traverse line at Kilve, Somerset. Note the increase in vein thickness but not vein frequency through the fault
zones, (b) Cumulative frequency thickness distributions. O, Data collected outside the recognized damage zones
of the faults (D = 1.55); • , data collected from within the damage zones of the faults (D = 0.5).
S. ROBERTS, D. J. SANDERSON & P. GUMIEL
Table 2. Summary of data from the damage zones of faults
Sample
Kilve
FZ
E
Millook
FZ
E
Wick
FZ
E
n
64
95
66
102
113
52
Traverse
length
(m)
3.8
4.2
1.6
6.9
45
60
Thickness
Min
0.2
0.2
0.2
0.2
1
1
(mm)
Max
100.0
5.5
210
8
163
11
Veins
(%)
16.37
1.10
40.01
1.68
2.45
0.15
D
0.5
1.5
0.7
1.1
0.8
1.5
Millook, Cornwall. A low displacement,
normal-dextral fault (trending northeast-south-
west) is exposed in a sandstone-shale sequence
(Crackington Formation, Namurian) on the
foreshore at Millook. Two traverses of 4 and
4.5 m (total 8.5 m) were made in a sandstone unit
across the fault zone, which has a damage zone
(c. 1m wide). A total of 1.6m of traverse was
measured within the damage zone, which com-
prises a network of connected quartz veins up to
210mm in thickness. Outside this damage zone
the veins are <8mm thick and occur as isolated
veins, or in short arrays, of overlapping vein
segments. The sandstone layer contains many
thin veins (displacements <l mm). Although
attempts were made to measure all observable
veins (thickness >0.2mm), the sample is only
considered to be complete for veins >1 mm. The
average density at this thickness is c. 7m"1 out-
side the damage zone, increasing to c. 30m"1
within the zone. Log-log plots (Fig. 6a) sug-
gest an approximate power-law distribution, but
the power-law exponent (D-value) within the
damage zones is 0.7, cf. c. 1 outside (Table 2).
South Head, Wick, north Scotland. A section
through two small normal faults (trending
north-northwest-south-southweast), with down-
throws of c. 1 m and 0.5m to the west, occurs in
Middle Devonian rocks (Caithness Flagstones)
at South Head, Wick. Extension is associated
with a series of N160°E calcite and Fe-carbonate
veins. Three traverses, taken approximately
normal to the faults, were made; a 45 m traverse
through the two faults and the intervening
rocks, and two 30m traverses in the same units
away from the faults (Table 2). All veins with
thickness >1 mm were measured. In the faulted
traverse the veins are up to 100mm thick
and form a well-connected network with thicker
carbonate areas forming pull-aparts or jogs
between more planar veins. Away from faults.
Fig. 6. (a) , Data from outside the damage zones (D = 1.1); • . data from within the damage zones of faults
(D = 0.7); both observed at Millook, Cornwall, (b) Similar data from South Head, Wick, north Scotland.
, Data from outside the damage zones of the faults (D = 1.5); • . data from within the damage zones of the
faults (D = 0.8).
12
FRACTAL ANALYSIS AND PERCOLATION PROPERTIES OF VEINS 13
veins are usually <10mm thick (Table 2) and
occur as isolated veins and side-stepping arrays.
Locally, veins are arranged in en echelon arrays,
some of which extend from the tips of recogni-
sable faults. The average density of veining is
usually < 1 m"1 away from faults but increases
to c. lOm"1 within the fault zones, averaging
2.5m"1 in the traverse through the two faults.
Log-log plots (Fig. 6b) suggest an approximate
power-law distribution with D c. 1.5 away from
the faults, decreasing to 0.8 within the faulted
traverse and, locally, lower still in the fault
zones themselves.
Vein thickness distributions and grade
Gold mineralization in quartz veins
Analysis of veins from the La Codosera area
of southwest Spain shows that boreholes sam-
pled which contained elevated gold grades
(>300ppb) display low D-values (fitted to the
bulk of the data) compared to boreholes which
consistently provided assays of <300ppb (San-
derson et al 1994) (Fig. 2). By comparison,
borehole and adit data from the Curraghinault
gold deposit (McCaffrey & Johnston 1996) have
consistently lower exponents (Fig. 7).
These data from mineral deposits are consis-
tent with the models and observations of the
damage zones of faults. They predict that
natural vein networks with low D-values char-
acterize well-connected systems and are more
prospective. Connectivity facilitates the ingress
of 'externally' derived fluid and permits a
significant increase in the fluid flux through
such systems. For example, at La Codosera, a
nitrogen anomaly was observed for fluid inclu-
sions within quartz samples from assay intervals
with enhanced gold values (Dee & Roberts
1993). This was attributed to the trapping of
fluids derived from metamorphic devolatization
reactions at depth within the host shear zone.
This observation is consistent with recent work
by Pettke & Diamond (1997) who used Sr
isotope signatures to trace fluid evolution and
suggested that fluids derived from deep within
the crust are an important component of
mineralized shear zones at Brusson, northwest
Alps. At La Codosera, the gold mineralization is
associated with an unmixing aqueocarbonic
fluid. Recently, Wilkinson & Johnston (1996)
demonstrated that unmixing and gold precipita-
tion can be closely associated with the devel-
opment of vein systems and, in particular, the
linking of veins. These data tend to suggest that
the mechanical development and the geochem-
ical evolution of the vein systems is intimately
linked.
Sn-W Mineralization
Similar data sets have been acquired from a
series of Sn-W veins within the Iberian Penin-
sula (Roberts et al. 1998). Sn-W mineralization
is associated with intrusions of syn- to late
kinematic Hercynian granites, typically occur-
ring as quartz veins and stockworks which cut
the granites and local countryrocks. The veins
sampled were oriented northeast-southwest and
dip towards the southeast. The vein-fill is
predominantly quartz with no apparent fibres
or development of crack-seal textures. The
deposits are small with economic extraction of
ore occurring only at the La Parilla mine
[c. 6000 tonnes of Sn (0.2 wt%) and WO3
(0.05wt%)]. Barren vein sets of similar age,
but with no visible sulphides, were studied from
Fig. 7. (a) Vein thickness distribution from Curraghinalt gold deposit, after McCaffrey & Johnston (1996).
Data shown are typical for adit and drill-hole analysis, (b) Comparative data from a borehole at La Codosera;
this sample has lower gold grades but an increased vein density compared to Curraghinalt.
14 S. ROBERTS, D. J. SANDERSON & P. GUMIEL
Table 3. Summary of data from the Sn-W mineralization
Sample
La Parrilla, 1
La Parrilla, 2
A. Montanches
El Trasquilon
Albala
Trujillo, 1
Trujillo, 2
Trujillo, 3
Trujillo, 4
Trujillo, 5
Trujillo, 6
Trujillo, 7
Trujillo, combined
n
37
112
74
60
40
9
23
16
17
10
27
23
125
Traverse
length
(m)
110.0
100.0
37.0
34.0
36.0
11.5
39.0
16.0
30.0
6.0
9.0
25.0
136.5
Thickness (mm)
Min.
16.0
1.0
2.5
6.0
1.0
4.0
.0
.0
.0
.0
.0
.0
1.0
Max.
250
400
136
300
48
12
15
15
30
17
22
30
30
Thickness
- (m mm)
26.30
42.30
36.60
79.80
14.60
5.70
3.10
4.30
3.70
5.50
16.20
5.36
5.00
Density D
(m-1)
0.34 0.7
1.12 0.7
2.00 1.0
1.76 0.9
1.11 0.7
0.78 1.3
0.59 .2
1.00 .2
0.57 .4
1.67
3.00 .4
0.92 .5
0.92 .6
the Trujillo granite, which elsewhere hosts Sn-W
mineralization. The data are shown in Table 3
and plotted on log-log frequency distribu-
tion plots (Fig. 8).
Data from the mineralized sites at La Parrilla,
Arroyomolinos de Montanches, Albala and
El Trasquilon (Table 3) show a similar intensity
of vein development (one vein >1 m m m 1 ) , with
vein thicknesses ranging from 2.0 to 2.5 orders
thickness (mm)
Fig. 8. Log-log plot of the number of veins per metre
(N) v. thickness (t) for selected samples of veins from
the Caceres district: Trujillo, unmineralized veins;
Albala, partially mineralized quartz veins; La Parrilla,
quartz-scheelite veins from worked ore deposit [after
Roberts et al. (1998)].
of magnitude. At all four localities the bulk of
the data conform to a power-law distribution,
with a tendency for the thicker veins to depart
from the power-law distribution (Fig. 8). This
may in part be attributed to truncated samples,
but probably reflects an approach to some upper
limit of vein opening. The former mining locality
of La Parrilla shows the lowest D-value of 0.7,
with vein thicknesses ranging between 1 and
400mm. Thus, thickness distributions from the
variably mineralized sites contrast with those
from the unmineralized sites within the Trujillo
granite, by showing lower D-values, indicating a
greater influence of thicker veins and breaks
within the vein thickness distributions.
The data obtained for the Sn-W mineralized
veins are consistent with the models outlined
above and with models for Sn-W mineralization
which attempt to explain the development of the
mineralization through the tapping of a fluid
reservoir (Pollard & Taylor 1986). Moreover,
recent geochemical modelling (Heinrich 1995)
suggests that for Sn-bearing vein systems to
develop, fluid flow must be focused within a
restricted part of the fracture network in order
to prevent reaction with wall rocks which would
otherwise lead to Sn deposition at a much earlier
stage. The development of a critical cluster of
connected fractures is required to tap the fluid
reservoir in an efficient way (Roberts et al. 1998).
The focusing of the fluid flow on the backbone
of the critical cluster (see Stauffer & Aharony
1992) provides the necessary channelling of the
fluid to prevent significant fluid-rock reaction at
an early stage. The vein thickness data suggest
that fractal analysis of Sn-W mineralized vein
systems can provide information to help inter-
pret the formation of Sn-W ore bodies.
Sn-W veins, Caceres
FRACTAL ANALYSIS AND PERCOLATION PROPERTIES OF VEINS 15
Thickness distributions and alteration
Models predicting increasing fracture connectiv-
ity and enhanced fluid flow suggest that an
increase in the alteration may well be observed
at sites with well-connected fracture networks;
assuming that an impermeable armour of vein
material does not form along the walls of the
fracture. A geochemical traverse was completed
alongside vein thickness measurements for a
60m road section at Castello Branco, Portugal.
Using alteration indices, such as relative deple-
tion/enrichment of major elements (SiO2, Na2O,
A12O3, K2O and TiO2), mass balance calcula-
tions (isocon diagrams), XRD analysis and ele-
mental analysis (C and N), two distinct zones of
enhanced alteration were recognized within the
slates. These zones tended to show a loss of Si,
Na and Ca and an increase in K with the
accompanying development of a two-mica plus
chlorite assemblage. The vein thickness data
were then divided into two subsamples accord-
ing to the presence or absence of local alteration
(Fig. 9).
Overall, the distributions of vein thicknesses
in the Castello Branco section suggest the
presence of a well connected system. Closer
inspection of the altered zones shows a distribu-
tion with an increased vein density and/or
thickness compared to the overall section. This
observation would appear to suggest that the
more pervasive alteration is related to rock
volumes with greater fracture permeability, due
to increased density and/or fracture opening
(aperture) (see Sanderson & Zhang 1999). The
Fig. 9. Vein thickness distributions from a line
transect in the Castello Branco area, Portugal. Data
are subdivided according to recognized alteration of
the host rock. The areas of alteration are characterized
by a vein population with an increased vein density.
data are certainly consistent with the more
intense alteration related to interaction with an
externally derived fluid and a high fluid flux
(fluid-rock interaction) through the zone.
Summary and conclusions
The fractal and percolation properties of veins
provide an important insight into the role of
fracture connectivity in the formation of vein
systems and vein-style ore deposits. Measure-
ment of the fractal dimension of vein thick-
ness distributions provides a powerful descrip-
tive parameter and a potential predictor of
connected vein systems. Observation of natural
systems and models suggests that vein sys-
tems initiate due to opening of isolated frac-
tures. At this stage, only locally derived fluid
enters fractures, with negligible fluid migra-
tion. As veins grow, a percolation threshold is
reached at which point a connected vein system
develops through the rock mass. Data from
mineral deposits and damage zones of faults
indicates that connected vein networks are char-
acterized by power-law distributions and by
lower D-values than unmineralized networks.
The development of the connected vein net-
work localizes deformation and fluid flow. This
allows fluid to be transported over increased
lengths and increases the likelihood of external
fluids entering the system. The transported fluids
produce vein-hosted ore deposits where physi-
cal and chemical conditions are suitable for
mineral precipitation. Thus, thickness distribu-
tions of mineralized vein systems should show
low D-values, recognition of which can serve as
a guide to prospecting in such systems.
This work was carried out as part of a grant (GR9/
1485) funded by NERC. Joe Cartwright and Paul
Gillespie are thanked for detailed and constructive
reviews of earlier versions of the manuscript.
References
AN, L.-J. & SAMMIS, C. G. 1996. A cellular automaton
for the development of crustalshear zones.
Tectonophysics, 253, 247-270.
BARTON, C. A. & ZOBACH, M. D. 1990. Self-similar
distribution of macroscopic fractures at depth in
crystalline rock in the Cajon Pass scientific drill
hole. In: BARTON, N. & STEPHANSSON, O. (eds)
Rock Joints. Balkema, Rotterdam, 163-170.
CLARK, M. B., BRANTLEY, S. L. & FISHER, D. M.
1995. Power-law vein-thickness distributions and
positive feedback in vein growth. Geology, 23,
975-978.
16 S. ROBERTS, D. J. SANDERSON & P. GUMIEL
DEE, S. J. & ROBERTS, S. 1993. Late kinematic gold
mineralization and the role of anomalous nitro-
gen; an example from the La Codosera area, SW
Spain. Mineralogical Magazine, 57, 437-450.
HEINRICH, C. A. 1995. Geochemical evolution and
hydrothermal mineral deposition in Sn (-W-Base
Metal) and other granite related ore systems:
some conclusions from Australian examples: In:
THOMPSON, J. F. H. (ed.) Magmas, Fluids and Ore
Deposits. Mineral Association of Canada Short
Course Series, 23, 203-220.
JOHNSTON, J. D. & MCCAFFREY, K. J. W. 1996.
Fractal properties of vein systems and the
variation of scaling relationships with mechanism.
Journal of Structural Geology, 18, 349-358.
MCCAFFREY, K. J. W. & JOHNSTON, J. D. 1996.
Fractal analysis of mineralized vein deposit:
Curraghinalt gold deposit, County Tyrone.
Mineralium Deposita, 31, 52-58.
MURPHY, P. J. & ROBERTS, S. 1997. Evolution of a
metamorphic fluid and its role in lode gold
mineralization in the Central Iberian Zone.
Mineralium Deposita, 32, 459-474.
NUR, A. 1982. The origin of tensile fracture linea-
ments. Journal of Structural Geology, 4, 31-40.
PETTKE, T. & DIAMOND, L. W. 1997. Oligocene gold
quartz veins at Brusson, NW Alps: Sr isotopes
trace the source of ore-bearing fluid to over 10-km
depth. Economic Geology, 92, 389-406.
PICKERING, G., BULL, J. M. & SANDERSON, D. J. 1996.
Scaling of fault displacements and implications
for estimation of sub-seismic strain: In: BUCHA-
NAN, P. G. & NIEUWLAND, D. A. (eds) Modern
Developments in Structural Interpretation, Valida-
tion and Modelling. Geological Society, Lodnon,
Special Publications, 99, 11-26.
POLLARD, P. J. & TAYLOR, R. G. 1986. Progressive
evolution of alteration and tin mineralization:
controls by interstitial permeability and fracture-
related tapping of magmatic fluid reservoirs in tin
granites. Economic Geology, 81, 1795-1800.
ROBERTS, S.. SANDERSON, D. J. & GUMIEL, P. 1998.
Fractal analysis of Sn-W mineralization from
Central Iberia: Insights into the role of fracture
connectivity in the formation of an ore deposit.
Economic Geology, 93, 360-365.
DEE, S. J. & GUMIEL. P. 1991. Tectonic
setting and fluid evolution of auriferous quartz
veins from the La Codosera area. Western Spain.
Economic Geology, 86, 1012-1022.
SANDERSON, D. J. & ZHANG, Z. 1999. Critical stress
localization of flow associated with deformation
of well-fractured rock masses, with implications
for mineral deposits. This volume.
, ROBERTS, S. & GUMIEL, P. 1994. A fractal
relationship between vein thickness and gold
grade in drill core from La Codosera. Spain.
Economic Geologv, 89. 168-173.
SEGALL, P. & POLLARD, D. D. 1980. Mechanics of
discontinuous faults. Journal of Geophysical
Research, 85. 4337-4350.
STAUFFER, D. & AHARONY, A. 1992. Introduction to
Percolation Theory. Taylor & Francis. Bristol.
VERMILYE, J. M. & SCHOLZ, C. H. 1995. Relation
between vein length and aperture. Journal of
Structural Geology, 17, 423-434.
WILKINSON. J. J. & JOHNSTON, J. D. 1996. Pressure
fluctuations, phase separation, and gold precipita-
tion during seismic fracture propagation. Geology*
24, 395-398.
ZHANG, X. & SANDERSON, D. J. 1994. Fractal struc-
ture and deformation of fractured rock masses:
In: KRUHL, J. H. (ed.) Fractals and Dynamic Sys-
tems in Geosciences. Springer, Berlin.
Geometry and population systematics of a quartz vein set,
Holy Island, Anglesey, North Wales
JULIA F. W. STOWELL, ADRIAN P. WATSON & NEIL F. C. HUDSON
Division of Earth Sciences, School of Environmental and Applied Sciences,
University of Derby, Kedleston Road, Derby DE22 1GB, UK
Abstract: A steeply dipping set of quartz-chlorite-muscovite-biotite veins was emplaced, at
depths of at least 14km and temperatures >400°C, into Mona Complex metasedimentary
rocks of Holy Island, after the D4 event. Vein trends range from northeast-southwest to
north-northwest-south-southeast. Vein offshoots, consistently oriented c. 10-35° anticlock-
wise relative to the main vein, are common. Quartz fibres within different veins show a range
of orientations, from northwest-southeast to east-west, and are interpreted as tracking vein
opening directions. Vein and fibre orientations are integrated into a four-stage model for
vein emplacement. At each stage, new main veins either open extensionally, with fibres
subnormal to their boundaries, or by hybrid extension-shear. Hybrid main veins commonly
have extensional offsoots. Extensional main veins are orientated progressively further
anticlockwise with time from a dominantly northeast-southwest trend to a dominantly
north-south trend, but hybrid main veins have out-of-sequence orientations and probably
utilized pre-existing fractures. The dominance of extensional and hybrid extension-shear
vein opening indicates that the differential stress was small and pore fluid pressure was high.
A study of length-thickness relationships has revealed that this vein set is self-affine with
vein widening progressing more quickly than elongation during growth. Linear transect
analyses show that the vein spacings have fractal characteristics.
Vein arrays are a ubiquitous feature of regional
metamorphic terrains and are an important
indication that large quantities of fluids are
channelled through the crust during tectonic
events (Walther & Orville 1982). Recent studies
of vein arrays have largely focused on three
aspects: the geometry of the array (e.g. Sander-
son et al 1994; Johnston & McCaffrey 1996;
Smith 1996); the origin of the fluid from which
the vein was precipitated (e.g. Boiron et al. 1996;
Haggerty & Bottrell 1997); the fluid-rock inter-
action at scales ranging from subgrain (Cathe-
lineau et al. 1993) to whole terrains (Cox 1993).
Ultimately, integration of information from
these different fields of investigation will enhance
the understanding of the development and
growth of veins and mechanisms of fluid flow
through the crust. This study concentrates on
the geometry and population systematics of a
vein array on Holy Island, Anglesey, North
Wales. Preliminary observations on the fluid
composition and conditions of emplacement are
also presented.
Geological setting
Rocks from the New Harbour and South Stack
Groups of the Monian Supergroup (Shackleton
1975) are exposed on the southwest coast of
Holy Island (Fig. 1), Anglesey. Both groups
comprise a series of greenschist facies metasedi-
ments, deposited as turbidite fans (Phillips
199la), with the New Harbour Group being
intruded by a suite of serpentinites prior to the
onset of deformation (Maltman 1977). Four
main phases of deformation have been recog-
nized in the New Harbour Group (Hudson &
Stowell 1997) and the South Stack Group
(Phillips 19916), although correlation between
them is problematical (Hudson & Stowell 1997).
Veins are common throughout both groups
and have been divided into five sets, V1-V5
(Table 1), on the basis of the timing of their
emplacement relative to the established struc-
tural history. This paper examines the geometry
and systematics of the V3 vein set.
V3 veins
V3 veins are common throughout southwest
Holy Island and are ideal for a study of the
development and growth of vein systems. The V3
veins developed post-D4 (Table 1), post-dating
all the major deformation events, and their
observed orientations reflect their emplacement
orientations. In addition, V3 veins commonly