Journal of African Earth Sciences 102 (2015) 70–85
Contents lists available at ScienceDirect
Journal of African Earth Sciences
journal homepage: www.elsevier.com/locate/jafrearsci
Fracturing of doleritic intrusions and associated contact zones:
Implications for fluid flow in volcanic basins
Kim Senger a,b,c,⇑, Simon J. Buckley a, Luc Chevallier d, Åke Fagereng e,1, Olivier Galland f, Tobias H. Kurz a,
Kei Ogata b,2, Sverre Planke g,h, Jan Tveranger a
a
Centre for Integrated Petroleum Research, Uni Research, Allégaten 41, 5007 Bergen, Norway
Department of Arctic Geology, University Centre in Svalbard, P.O. Box 156, 9171 Longyearbyen, Norway
Department of Earth Science, University of Bergen, Allégaten 41, 5007 Bergen, Norway
d
Council for Geosciences, 3 Oos Street, 7535 Bellville, South Africa
e
Department of Geological Sciences, University of Cape Town, 13 University Avenue, 7701 Rondebosch, South Africa
f
Physics of Geological Processes (PGP), Department of Geosciences, University of Oslo, Postboks 1047, Blindern, 0316 Oslo, Norway
g
Centre for Earth Evolution and Dynamics (CEED), University of Oslo, Postboks 1028, Blindern, 0315 Oslo, Norway
h
Volcanic Basin Petroleum Research AS, Oslo Innovation Center, Gaustadalléen 21, 0349 Oslo, Norway
b
c
a r t i c l e
i n f o
Article history:
Received 31 January 2014
Received in revised form 29 October 2014
Accepted 31 October 2014
Available online 21 November 2014
Keywords:
Karoo
Igneous intrusions
Fluid flow
Fracturing
Permeability
Dolerite
a b s t r a c t
Igneous intrusions act as both carriers and barriers to subsurface fluid flow and are therefore expected to
significantly influence the distribution and migration of groundwater and hydrocarbons in volcanic
basins. Given the low matrix permeability of igneous rocks, the effective permeability in- and around
intrusions is intimately linked to the characteristics of their associated fracture networks. Natural fracturing is caused by numerous processes including magma cooling, thermal contraction, magma emplacement and mechanical disturbance of the host rock. Fracturing may be locally enhanced along
intrusion–host rock interfaces, at dyke–sill junctions, or at the base of curving sills, thereby potentially
enhancing permeability associated with these features. In order to improve our understanding of fractures associated with intrusive bodies emplaced in sedimentary host rocks, we have investigated a series
of outcrops from the Karoo Basin of the Eastern Cape province of South Africa, where the siliciclastic
Burgersdorp Formation has been intruded by various intrusions (thin dykes, mid-sized sheet intrusions
and thick sills) belonging to the Karoo dolerite. We present a quantified analysis of fracturing in- and
around these igneous intrusions based on five outcrops at three individual study sites, utilizing a combination of field data, high-resolution lidar virtual outcrop models and image processing. Our results show
a significant difference between the three sites in terms of fracture orientation. The observed differences
can be attributed to contrasting intrusion geometries, outcrop geometry (for lidar data) and tectonic setting. Two main fracture sets were identified in the dolerite at two of the sites, oriented parallel and perpendicular to the contact respectively. Fracture spacing was consistent between the three sites, and
exhibits a higher degree of variation in the dolerites compared to the host rock. At one of the study sites,
fracture frequency in the surrounding host rock increases slightly toward the intrusion at approximately
3 m from the contact. We conclude by presenting a conceptual fluid flow model, showing permeability
enhancement and a high potential for fluid flow-channeling along the intrusion–host rock interfaces.
Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction
⇑ Corresponding author at: Electromagnetic Geoservices ASA, Dronning Mauds gt.
15, 0250 Oslo, Norway. Tel.: +47 95291592.
E-mail addresses: ksenger@emgs.com, senger.kim@gmail.com (K. Senger).
1
Present address: School of Earth & Ocean Sciences, Cardiff University, Park Place,
CF10 3AT Cardiff, United Kingdom.
2
Present address: Dipartimento di Fisica e Scienze della Terra ‘Macedonio Melloni’,
Università degli Studi di Parma, Campus Universitario – Parco Area delle Scienze 157/
A, I-43124 Parma, Italy.
http://dx.doi.org/10.1016/j.jafrearsci.2014.10.019
1464-343X/Ó 2014 Elsevier Ltd. All rights reserved.
The matrix of crystalline igneous rocks is typically tight, with
sub-milliDarcy permeability (e.g., Sruoga et al., 2004) and primary
porosity commonly less than 0.5–1% (e.g., Petford, 2003; van Wyk,
1963). Fractured igneous aquifers may nonetheless be considered
as potential groundwater reservoirs (e.g., Gustafson and Krásný,
1994; Woodford and Chevallier, 2002) due to the presence of
fracture networks, a characteristic feature of virtually all igneous
71
Ecca Grp
Stormberg Grp
B
East London
Port Elizabeth
Main Karoo
Basin boundary
100km
B
E
L. Elliot
Burgersdorp
Katberg
Nonesi’s neck
L
Permian
Hillcrest
Golden Valley
10km
E
L
D
Ecca
Bonkola
C
C
U. Elliot
Molteno
M
D
Karoo dolerites
L
Triassic
Cape Fold Belt
Cape Town
Drakensberg
Prince Albert
Dwyka
Beaufort Grp
E
Fm
V
V V V
V
V
VV
V V V
VV
V
Gp
V
Dwyka Grp
Stormberg
Drakensberg Grp
Beaufort
VV V V V
A
Jurassic Prd
K. Senger et al. / Journal of African Earth Sciences 102 (2015) 70–85
E
Bonkola
GV_W_1
GV_E_2
GV_E_3
GV_W_2
GV_E_4
GV_E_6
Nonesi’s neck
GV_E_1
GV_W_3
GV_E_8
GV_E_7
Hillcrest
3km
3km
Fig. 1. Location and geological setting of the study area. (A) Location of study area (black rectangle) superimposed on a simplified geological map of southern Africa, after
Smith (1990) and Svensen et al. (2012). (B) Satellite image of the study area, illustrating the sills (dark red polygons) and dykes (bright red lines). Note the evident exposure of
the saucer-shaped intrusions and the study area locations (yellow pins). (C) Close-up of the Golden Valley saucer-shaped sill, and the locations of the photo-mosaics
(photographed rim segments and photograph location) presented in this study. (D) Close-up of the Queenstown area, with the three lidar scan sites marked. Refer to Table 1
for details. (E) Simplified stratigraphic column of the Karoo Supergroup, after Tankard et al. (2009). The star at 183 Ma indicates the timing of the Karoo dolerite emplacement
(Svensen et al., 2012). All satellite images from Google Earth. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this
article.)
rocks. This dual nature of the permeability of igneous intrusions
causes them to influence reservoir fluid flow in an apparently
non-predictable manner, sometimes forming fully sealing, impermeable barriers (e.g., Gurba and Weber, 2001; Thomaz Filho
et al., 2008), at other times high-permeability pathways (e.g.,
Mège and Rango, 2010; Sankaran et al., 2005), and sometimes both
(e.g., Rateau et al., 2013; Stearns, 1942).
The economic and societal consequences of improved understanding of fractured igneous aquifers is significant, with underexplored fractured igneous reservoirs, such as in the Argentinian
72
K. Senger et al. / Journal of African Earth Sciences 102 (2015) 70–85
Table 1
List of studied localities. The locations are given in the WGS84 datum.
Study site
Latitude
Longitude
Bonkola N
31°470 18.7700 S
26°570 14.7000 E
0
00
0
00
Scan size
(width * height)
Igneous features
Intrusion contact (dip dir/dip
angle)
Outcrop
orientation
Host rock
36 * 5 m
2 m thick dyke
110/75
057/237
Burgersdorp
Fm.
Burgersdorp
Fm.
Burgersdorp
Fm.
Burgersdorp
Fm.
Burgersdorp
Fm.
Burgersdorp
Fm.
Bonkola S
31°47 18.77 S
26°57 14.70 E
47 * 12 m
2 m thick dyke
119/63
057/237
Nonesi’s
neck
Hillcrest N
31°500 28.7000 S
26°590 24.5400 E
69 * 14 m
324/74
024 (curving)
31°530 25.2300 S
26°480 9.9600 E
110 * 13 m
256/36
293/113
0
5 m thick inclined
intrusion
>80 m wide inclined
sheet
>100 m wide inclined
sheet
18 * 10 km saucershaped sill
Poor exposure
293/113
N/A
Variable
0
00
00
Hillcrest S
31°53 25.23 S
26°48 9.96 E
150 * 10 m
Golden
Valley
31°550 28.7000 S
26°170 5.8000 E
No scan
Neuquén Basin, hosting commercial quantities of hydrocarbons
(e.g., Bermúdez and Delpino, 2008; Gudmundsson and Løtveit,
2012; Rodriguez Monreal et al., 2009; Schutter, 2003; Witte
et al., 2012). Applied aspects of permeability associated with intrusions can be best illustrated by the regional-scale exploration for
groundwater in the semi-arid Karoo Basin, which has focussed on
identifying and mapping fracture zones yielding high flow
(Chevallier et al., 2001; van Wyk, 1963). Extensive drilling and
A Nonesi’s Neck
injection tests (12 wells) across an inclined dolerite sheet at
Qoqodala in the Eastern Cape province confirmed this fact, highlighting how the igneous body itself forms a barrier to fluid flow,
whereas the intensely fractured lower dolerite–host rock interface
channels groundwater flow along the contact (Chevallier et al.,
2004). The influence of fractures on groundwater yields is undisputed, but many studies rely on fracture information from boreholes and regional lineament analyses from remote sensing. The
Lidar
Photograph
on
si
tru . 5 m
n
i a
c
intrusion
Que
ens
tow
n
R39
NND_
6
B Hillcrest N
01
C
Lidar
Photograph
Photograph
blast marks
interpreted
fractures
Lidar
measuring tape
scanline
‘virtual’
scanline
Fig. 2. Outcrop photographs and screenshots of the virtual outcrops illustrating the level of detail and applicability for fracture measurements. (A) Nonesi’s neck. Car (circled)
and intrusion (ca. 5 m thick) for scale. (B) Side-view of the Hillcrest N locality, illustrating the structural details of the virtual outcrop model. Field notebook (circled) and
measuring tape for scale. (C) Detail of a blast mark at Hillcrest N. Measuring tape for scale.
73
K. Senger et al. / Journal of African Earth Sciences 102 (2015) 70–85
Table 2
Summary of scanlines (Lidar, fieldwork and photo-mosaics) and structural stations, with corresponding fracture characteristics. Refer to Table 3 for additional details on virtual
profiles.
ID
Data set
Lithology
Length (m)
Number of fractures
Fractures/meter (avg)
Scanlines
BD_01
BD_01L
HC_01
HC_01L
NND_01
NND_01P
NND_02
Fieldwork
Lidar
Fieldwork
Lidar
Fieldwork
Photo-mosaic
Fieldwork
Mixed
Mixed
Mixed
Mixed
Mixed
Mixed
Mixed
10
10.4
20
20
24
23.4
10
67
49
124
61
158
79
37
6.7
4.7
6.2
3.1
6.6
3.4
3.7
Structural stations
Bonkola S
Bonkola S
Bonkola N
Bonkola N
Hillcrest S
Hillcrest S
Hillcrest N
Hillcrest N
Nonesi’s Neck
Nonesi’s Neck
Lidar
Lidar
Lidar
Lidar
Lidar
Lidar
Lidar
Lidar
Lidar
Lidar
Host rock
Dolerite
Host rock
Dolerite
Host rock
Dolerite
Host rock
Dolerite
Host rock
Dolerite
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
176
60
158
50
308
532
55
473
134
101
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
number of detailed field-based studies of fractures in igneous rocks
is relatively limited (e.g., Delaney et al., 1986; McCaffrey et al.,
2003).
Our study focusses on the fracture patterns in- and around doleritic intrusions within the Karoo Basin and presents new outcropscale fracture data from three sites in the Eastern Cape Province of
South Africa. We acquired new data, combining both state-of-theart digital techniques, in particular lidar scanning (Buckley et al.,
2008), and traditional field methods. Our objectives are two-fold.
Firstly, we illustrate, assess and verify various methods of collecting fracture spacing and orientation data from igneous intrusions
at different scales. Secondly, we apply this data set to investigate
the fracture patterns at intrusion–host rock contacts, presenting
a conceptual model for fracture-channeled permeability at intrusion–host rock contact zones.
2. Fracturing in igneous rocks
The complexity of fracturing within and around igneous intrusions is partly related to the distinct formation mechanisms developing three main types of fractures: (1) emplacement-induced
fractures in the host-rock, (2) post-emplacement fractures within
the intrusion (i.e., cooling joints), and (3) non-intrusion related
fractures (i.e., tectonic fractures) in both the intrusion and country
rock.
The first set of fractures occurs in the country rock (referred to as
‘syn-emplacement fractures’, SEF for short, in this contribution, and
as ‘inflation fractures’ by Kattenhorn and Schaefer (2008)), and is
induced by the emplacement of the intrusion. Because the magma
is overpressured when it is emplaced (e.g., Lister and Kerr, 1991;
Rubin, 1995), it triggers substantial damage in the country rock at
both local (e.g., Abdelmalak et al., 2012; Delaney and Pollard,
1981; Delaney et al., 1986; Galland and Scheibert, 2013; Meriaux
et al., 1999; Schofield et al., 2012a) and regional scale (e.g.,
Jackson and Pollard, 1990; Rubin and Pollard, 1988). Abdelmalak
et al. (2012) show how the deformation pattern reflects the mechanism of emplacement of the magma intrusions. Quantitative
detailed mapping of fracture patterns around intrusions, however,
have rarely been carried out, and a proper understanding of the
fracture pattern induced by magma intrusions in the country rock
is lacking. The second set of fractures (referred to as ‘post-emplacement fractures’, ‘PEF’ for short, and as ‘cooling fractures’ by
Kattenhorn and Schaefer (2008)) forms within the intrusion and
is induced by post-emplacement processes, such as thermal contraction during cooling (e.g., Hetényi et al., 2012). The third set of
fractures (referred to as ‘tectonic fractures’, TF for short) forms
due to tectonic forces unrelated to the intrusion emplacement. Tectonic fractures may be present both in the intrusion (if tectonism
post-dated the magma emplacement) and in the host rock (both
pre- and post-emplacement). These three sets of fractures (i.e.,
‘SEF’, ‘PEF’, and ‘TF’) are juxtaposed, and the way they interact and
are connected is unknown.
3. Geological setting
All studied localities exhibit dolerites of the Karoo large igneous
province (e.g., Svensen et al., 2012 and references therein)
emplaced into the Permo-Triassic siliciclastic Beaufort Group,
which is part of the regional Karoo Supergroup (Fig. 1). Tectonically, the Karoo Basin is traditionally interpreted as a retro-arc
foreland basin bounded by the Cape Fold Belt to the south (e.g.,
Johnson et al., 1996). The depositional history of the Karoo Supergroup is well covered in the published literature (e.g., Smith, 1990;
Tankard et al., 2009) and only briefly summarized here. Deposition
commenced during the Permo-Carboniferous with glacio-marine
sedimentation (Dwyka Group), followed by shallow marine (Lower
Ecca Group), deltaic (Upper Ecca Group) and fluvial/lacustrine
(Beaufort Group) deposition (Smith, 1990). The Burgersdorp Formation, comprising the uppermost unit of the Beaufort Group
and forming the host rock at the studied localities, consists of grayish-red and greenish-gray mudstone with subordinate fine-grained
sandstone units (Hiller and Stavrakis, 1984). The top of the Burgersdorp Formation marks the base of the Stormberg Group, comprising, from the base upwards, perennial braid-plain deposits
(Molteno Formation) overlain by continental red beds (Elliot Formation) and aeolian and playa deposits (Clarens Formation;
Smith, 1990). The rapid emplacement of magma at ca. 183 Ma
led to the formation of an extensive (ca. 367 000 km3 of magma)
sill complex (Svensen et al., 2012). At that time, the associated lava
complex, the Drakensberg Group, likely exceeded the 140 000 km2
remnant exposed today (Bristow and Saggerson, 1983; Smith,
1990). This volcanism effectively terminated deposition of the
Karoo Supergroup and the resistant dolerites and basalts have
protected the underlying sedimentary succession from
denudation since the Early Jurassic. It thus offers a rare glimpse
into a relatively undisturbed ancient intrusion–host rock complex
74
K. Senger et al. / Journal of African Earth Sciences 102 (2015) 70–85
Table 3
Summary of virtual profiles with associated fracture parameters.
Profile ID
Locality
Vertical/horizontal
Lithology
Length (m)
High-resolution profiles
B-S-1
Bonkola South
B-S-2
Bonkola South
B-S-3
Bonkola South
B-S-4
Bonkola South
B-S-5
Bonkola South
B-S-6
Bonkola South
B-N-1
Bonkola North
B-N-2
Bonkola North
B-N-3
Bonkola North
B-N-4
Bonkola North
B-N-A
Bonkola North
B-N-B
Bonkola North
B-N-C
Bonkola North
HC-N-01
Hillcrest North
HC-N-02
Hillcrest North
HC-N-03
Hillcrest North
HC-N-04
Hillcrest North
HC-N-05
Hillcrest North
HC-N-06
Hillcrest North
HC-N-A
Hillcrest North
HC-N-B
Hillcrest North
HC-N-C
Hillcrest North
HC-N-D
Hillcrest North
HC-N-E
Hillcrest North
HC-N-F
Hillcrest North
HC-N-G
Hillcrest North
HC-N-H
Hillcrest North
HC-N-I
Hillcrest North
H-S-A
Hillcrest South
H-S-B
Hillcrest South
H-S-C
Hillcrest South
Horizontal
Horizontal
Horizontal
Horizontal
Diagonal
Diagonal
Horizontal
Horizontal
Horizontal
Horizontal
Horizontal
Horizontal
Horizontal
Horizontal
Horizontal
Horizontal
Horizontal
Horizontal
Horizontal
Vertical
Vertical
Vertical
Vertical
Vertical
Vertical
Vertical
Vertical
Vertical
Horizontal
Horizontal
Horizontal
Mixed
Mixed
Mixed
Mixed
Mixed
Mixed
Mixed
Mixed
Mixed
Mixed
Host rock
Host rock
Host rock
Dolerite
Dolerite
Dolerite
Dolerite
Dolerite
Dolerite
Dolerite
Dolerite
Dolerite
Dolerite
Dolerite
Dolerite
Dolerite
Dolerite
Dolerite
Host rock
Host rock
Host rock
22.1
22.1
22.1
22.1
12.5
13.5
12.1
12.1
12.1
11.8
14.7
14.7
14.6
20.8
20.5
20.4
20.4
20.3
20.4
11.7
11.4
11.5
11.7
11.8
11.8
11.8
11.5
11.6
19.5
29.5
29.7
51
49
53
45
31
39
44
50
63
50
20
24
25
25
24
25
19
18
12
4
4
11
11
9
6
8
9
9
34
74
62
2.3
2.2
2.4
2.0
2.5
2.9
3.6
4.1
5.2
4.2
1.4
1.6
1.7
1.2
1.2
1.2
0.9
0.9
0.6
0.3
0.4
1.0
0.9
0.8
0.5
0.7
0.8
0.8
1.7
2.5
2.1
Regional-scale profiles
GV_E_2
Golden Valley
GV_E_3
Golden Valley
GV_E_4
Golden Valley
GV_E_6
Golden Valley
GV_E_7
Golden Valley
GV_E_8
Golden Valley
GV_W_1a
Golden Valley
GV_W_1b
Golden Valley
GV_W_2
Golden Valley
GV_W_3
Golden Valley
Mary
Eastern Cape
Graaf Reinet
Eastern Cape
R364
Eastern Cape
R364_2
Eastern Cape
Horizontal
Horizontal
Horizontal
Horizontal
Horizontal
Horizontal
Horizontal
Horizontal
Horizontal
Horizontal
Horizontal
Horizontal
Horizontal
Horizontal
Dolerite
Dolerite
Dolerite
Dolerite
Dolerite
Dolerite
Dolerite
Dolerite
Dolerite
Dolerite
Dolerite
Dolerite
Dolerite
Dolerite
700
140
170
250
340
190
310
80
260
360
140
190
70
70
205
30
47
62
53
36
93
19
80
135
39
69
27
42
0.3
0.2
0.3
0.2
0.2
0.2
0.3
0.2
0.3
0.4
0.3
0.4
0.4
0.6
1845
908
0.5
1.8
Total (all):
Total (high-resolution only):
3782.8
512.8
Number of fractures
Fractures/meter (avg)
and its associated fracturing patterns. The Karoo igneous rocks
were studied with respect to regional geochemistry across southern Africa (Bristow and Saggerson, 1983), and have been linked
to synchronous igneous complexes in Antarctica (Ferrar) and South
America (Paraná; Cox, 1992). The regional geometry of the dolerites and associated fracturing was particularly well studied by
the South African Water Resource Commission in conjunction with
groundwater exploration and management (e.g., Chevallier et al.,
2001; Chevallier and Woodford, 1999; Woodford and Chevallier,
2002).
geometries include a thick (>100 m thick) inclined dolerite sheet
(Hillcrest), a thin (ca. 2 m thick) dyke (Bonkola), and a moderately-sized (ca. 5 m thick) inclined intrusion (Nonesi’s Neck). The
road-cut exposures at both Hillcrest and Bonkola facilitated lidar
acquisition at both sides of the road. Blast marks, and related fractures associated with road construction can be clearly identified
both on lidar and field data and were not included in the analyses.
The well-studied Golden Valley sill (e.g., Polteau et al., 2008;
Schofield et al., 2010) was chosen as a site for the large-scale fracturing study.
4. Methods and study sites
4.2. Lidar scanning
4.1. Study sites
Terrestrial laser scanning (also known as lidar) is a well-established 3D measurement technique that can be used for collecting
large amounts of digital, topographic data for building spatially
complete and precise models of geological outcrops. It is particularly used for studying outcrop analogues of reservoirs targeted
by the hydrocarbon industry (e.g., Enge et al., 2007). In this paper,
The sites for this study were selected based on two criteria:
They should display a range of intrusion geometries within a similar geologic setting, and all should thus be located within the
Burgersdorp Formation (Table 1 and Fig. 1). The observed igneous
75
K. Senger et al. / Journal of African Earth Sciences 102 (2015) 70–85
LiDAR
Fieldwork
Fracture frequency diagrams
14
BD_01
Numberoffractures
Bonkola N
BD_01L
Continuous
12
Discontinuous
10
Lidar profile
8
6
4
2
0
1
3
2
4
5
6
7
8
9
10
Distancealongscanline (m)
Continuous
12
Numberoffractures
Hillcrest S
HC_01L
HC_01
Discontinuous
10
Lidar profile
8
6
4
2
0
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15 16 17 18 19 20
Distancealongscanline (m)
Nonesi’s Neck
intrusion
20
Bed-confined
18
Continuous
16
Discontinuous
14
ImageJ
12
10
8
6
4
Numberoffractures
22
Through-going
2
0
1
2
3
4
5
6
7
8
9
10 11
12 13 14 15 16 17 18 19 20
21 22 23 24 25
Distancealongscanline (m)
Fig. 3. Comparison of field data and digital data sources (lidar and photo-mosaics) from selected localities. Stereoplots are shown with an equal-area lower hemisphere
projection, with fractures displayed as contoured poles to planes. The fracture frequency diagrams show the field data (vertical bars, subdivided by fracture type) compared to
the digitally interpreted scanlines (red line). See Table 2 for details on the various data sets. (For interpretation of the references to color in this figure legend, the reader is
referred to the web version of this article.)
the method has been applied as described by Buckley et al. (2008),
with a Riegl LMS-Z420i terrestrial lidar scanning system employed
for data acquisition. Data gaps within the lidar point cloud were
kept to a minimum by positioning the instrument at multiple locations. The lidar instrument was equipped with a Nikon D200
(10.2 MP) camera, which allows automatic registration of calibrated digital photographs within the lidar coordinate system.
The main processing steps involve the registration of several
high-density lidar point clouds in a common coordinate system,
point cloud reduction and point cloud triangulation to create a
meshed 3D surface. The meshed surface is then textured with
the photographs, resulting in photorealistic 3D outcrop models
used to conduct fracture measurements. While the 3D information
of the lidar data allows structural measurements, the lithology
(e.g., contact between dolerite and host rock) can be easily recognized from the photographs, which would be difficult from the
lidar point cloud alone. Fracture characterization was conducted
manually using the in-house software LIME. At such close range
(maximum 30 m from scanner to outcrop), the virtual outcrop
model provided nearly as high resolution as traditional photography (less than one centimeter), as illustrated by the blast mark
and measuring tape in Fig. 2C.
4.3. Photo-based interpretation
Due to limited accessibility and scanning range restrictions
imposed by the size of the outcrop and the unavailability of a helicopter-mounted lidar scanner (e.g., Rittersbacher et al., 2013), the
outer rims of the Golden Valley saucer-shaped intrusion were photographed at a distance of 0.5–3 km with a Nikon D90 digital camera (12.9 MP) with a telephoto lens (focal length of 200 mm). The
resultant images were stitched together as a conventional flattened
photo-mosaic, fractures were interpreted manually and subsequently processed using the ImageJ image processing software
76
K. Senger et al. / Journal of African Earth Sciences 102 (2015) 70–85
fractures in host rock
A Nonesi’s Neck
Bonkola S
Hillcrest S
Bonkola N
All
Hillcrest N
1
2
n=176
n=134
n=308
n=158
n=55
n=831
Bonkola S
Bonkola N
Hillcrest S
n=60
n=50
n=532
All
Hillcrest N
fractures in dolerite
B Nonesi’s Neck
n=101
n=473
n=1216
0-10
10-20
20-30
30-40
40-50
50-60
60-70
70-80
80-90
n=59
n=50
n=53
n=49
n=51
n=67
n=55
n=100
n=61
n=40
n=36
n=39
n=60
90-100m
Hillcrest S
C
n=60
n=31
Hillcrest N
n=57
n=59
n=78
Fig. 4. Structural data derived from the virtual outcrop models. See Table 2 for details on the data sets. (A) Fracture orientations mapped in sedimentary rocks at the three
study sites. The rose diagram shows the strike of all the fractures in the host rock. The master jointing sets from the Western Karoo published by Woodford and Chevallier
(2002) are shown for comparison: (1) Master joints with a predominant NNW-trend, (2) E–W jointing set particularly prominent in the Graaff-Reinet area. (B) Fractures in
dolerite compared to the main dyke orientation and the outcrop orientation. The rose diagram shows the strike of all the fractures in dolerite. (C) Orientation of fractures in
the Hillcrest igneous bodies, grouped in 10 m intervals away from the contact zone. Both the northern and southern outcrops are shown for comparison. All stereoplots are
illustrated using an equal-area, lower hemisphere projection, with a Schmidt grid. The main dyke orientation (solid blue line) and the outcrop orientation (green stippled line)
are shown where applicable. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
(Abràmoff et al., 2004) to provide ‘virtual scanlines’. The resolution
and accuracy of this method is much lower than for terrestrial lidar
scanning, but it is nonetheless very useful for determining whether
the overall fracturing frequency changes around a single largescale intrusion. Furthermore, in order to ensure a robust comparison during subsequent analysis, the same method was also successfully applied for determining fracture spacing on manually
interpreted high-resolution images generated from the virtual outcrop models.
4.4. Field work
Traditional field methods were employed at the scan localities
in order to ground-truth the digital data. The line-intersection
method (i.e., scanlines; Singhal and Gupta, 2010) was used to characterize the fractures (e.g., fracture orientation, fracture infill)
along 1D transects at Hillcrest, Bonkola and Nonesi’s neck. Fractures in dolerites were classified as either continuous (main fractures extending across most of the outcrop) or discontinuous
(shorter fractures often abutting against other fractures). Fracture
orientation (strike and dip) was measured using both a standard
geological compass and a digital GeoClino clinometer, both of
which were controlled for unwanted magnetic interference from
the dolerites. A magnetic declination correction of 24.1° west
was applied prior to data analysis (NOAA, 2013).
5. Results
Results from scanlines and structural stations are compiled in
Table 2, while virtual profiles from all localities are listed in
Table 3.
5.1. Field versus digital data
An initial comparison of the lidar-derived virtual outcrop models (VOMs) and digital photographs highlight the differences in resolution (Fig. 2). The most significant advantage of VOM compared
to traditional 2D photo-mosaics is the three-dimensional rendering of the outcrop which in many cases allows direct measurement
of fracture orientations on the model. The uncertainty attached to
using VOM is related to scan-resolution, scanning angle and
77
K. Senger et al. / Journal of African Earth Sciences 102 (2015) 70–85
A
B
13
12
number of fractures per 10m
11
GV_W_1
GV_E_2
GV_E_3
GV_W_2
GV_E_4
GV_E_6
GV_W_3
GV_E_7
GV_E_8
10
9
8
7
6
5
4
3
2
1
100
80
60
10
40
D
R364-2
R364
Graaf Reinet
Mary
GV_E_8
GV_E_7
GV_E_6
GV_E_4
GV_E_3
GV_E_2
GV_W_1b
12
ln(Y) = -2.3574 * ln(X) + 6.3036
8
20
frequency
fracture spacing (m)
C
GV_W_3
5 km
GV_W_2
GV_W_1a
0
6
4
2
10
8
6
4
0
GV_W_3
GV_W_2
GV_W_1b
GV_W_1a
GV_E_8
GV_E_7
GV_E_4
GV_E_2
2
1
0.8 1
2
4
6
8 10
20
40
fracture spacing (m)
Fig. 5. Regional study of the outer rims of the Golden Valley sill, illustrating fracture frequency in different segments of the saucer-shaped intrusion. (A) Location of the
studied segments on the eastern and western limbs of the sill. (B) Whisker plots (minimum, lower quartile, mean, upper quartile, maximum) illustrating the variable fracture
frequency in 10 m intervals at the Golden Valley ‘virtual scanlines’ as well as control locations. (C) Whisker plots illustrating the individual fracture spacing at selected ‘virtual
scanlines’ at Golden Valley. (D) Log–log plot of all the fracture spacing measurements at the sites shown in (C), grouped into 0.25 m wide bins. The data suggests a power-law
distribution in the 3–30 m interval.
influences introduced during processing, such as point cloud decimation or smoothing (Buckley et al., 2008).
Both VOM and photo-mosaic data were checked using traditional field techniques at all localities (scanlines, Fig. 3). The orientations show some discrepancy, particularly at Bonkola N, where
orientations are hampered by the overall dip of the VOM toward
the road. Depending on whether fracture planes are measured
along a line (i.e., virtual scanline) or across the whole outcrop,
the well-exposed road-facing fracture surfaces may thus be overrepresented. Orientation of field data, on the other hand, may be
somewhat hampered by the highly magnetic dolerite, which may
influence measurements made using a magnetic compass, even
though the absence of this effect was checked in the field.
Fracture frequency data collected using ‘real’ and ‘virtual’ scanlines show good agreement. More fractures, particularly the
shorter discontinuous fractures, were identified using conventional
field techniques compared to both VOMs (Bonkola N, Hillcrest S)
and photo-mosaics (Nonesi’s Neck). Qualitatively, the overall pattern of higher fracture frequency within the intrusions is evident
in all data sets. However, the VOMs, photo-mosaics and field data
each sample a specific range of fracture sizes due to their differing
resolutions, and quantitative comparison of the results from the
different methods should be treated with caution.
5.2. Fracture orientation
Analysis of fracture orientation was primarily carried out using
the virtual outcrop models, with control data provided by a limited
number of traditional scanlines (Table 2). Orientations are presented as poles to planes in Fig. 4, illustrating the dominant orientation of fractures in the sedimentary host rock (Fig. 4A), fractures
in the dolerite (Fig. 4B) and fractures in the dolerite at specific
intervals from the intrusion contact (Fig. 4C).
An interesting observation is that the fractures are typically
aligned parallel to the outcrop orientation. This is counter-intuitive
for traditional scanlines, which tend to be biased toward fractures
oriented perpendicular to the outcrop orientation, since these are
more likely to intersect the outcrop-parallel scanlines. However,
on virtual outcrop models fractures aligned sub-parallel with the
outcrop are rendered more clearly on the 3D image, and consequently more easily identified during interpretation. Bias correction (e.g., Lato et al., 2010) can be applied, but small fractures
C
HC-N-I
HC-N-H
HC-N-F
HC-N-E
HC-N-D
HC-N-C
HC-N-A
A
HC-N-G
K. Senger et al. / Journal of African Earth Sciences 102 (2015) 70–85
HC-N-B
78
intrusion contact
10 fractures
HC-N-06
HC-N-05
HC-N-04
HC-N-03
HC-N-02
HC-N-01
intrusion
contact
D
E
G
F
H
I
_0
6
_N
C
C
H
H
_N
_0
5
_0
4
_N
_0
3
C
_N
C
H
H
_N
C
C
_N
10
5
_0
2
_0
1
D
H
HC-N-06
4
3
2
1
C
B
H
B
A
0
2
4
6
8
10
12
14
16
18
2
1
fracture spacing (m)
3
HC_N_I
HC_N_H
0
HC_N_G
20
15
10
5
0
4
HC_N_F
HC-N-01
4
3
2
1
0
5
vertical profiles
HC_N_E
20
15
10
5
0
HC_N_D
HC-N-02
4
3
2
1
1
HC_N_C
20
15
10
5
2
HC_N_B
HC-N-03
4
3
2
1
3
HC_N_A
20
15
10
5
4
cumulative number of fractures
fractures per metre
HC-N-04
4
3
2
1
horizontal profiles
20
15
10
5
fracture spacing (m)
5
HC-N-05
4
3
2
1
20
Distance (m)
Fig. 6. Fracturing pattern within 20 m of the contact zone at Hillcrest N. (A) High-resolution screen capture of the virtual outcrop model, with interpreted fractures (black
lines) and profile locations (yellow stippled lines). (B) Histograms and cumulative frequency plots of the six horizontal profiles HC-N-01 to HC-N-06 away from the contact.
(C) Cumulating frequency plots of the eight vertical profiles HC-N-A to HC-N-I at varying distance from the contact. (D) Whisker plots (minimum, lower quartile, mean, upper
quartile, maximum) illustrating the fracture spacing in the horizontal and vertical profiles. (For interpretation of the references to color in this figure legend, the reader is
referred to the web version of this article.)
aligned perpendicular to the outcrop will often remain undersampled, due to the line-of-sight effect and smoothing during lidar
processing.
Fractures observed in the sedimentary rocks display a wide
range of orientations (Fig. 7A), but mainly align along an E–W and
a NW–SE axis. As stated above, there is a clear bias due to
individual outcrop orientation, which in turn influences statistics
of the complete dataset. This can be illustrated by the Hillcrest S
(n = 308) locality which dominates the orientation plot as it
contains 37% of all fractures analyzed in the sedimentary host rock.
The Hillcrest S and Hillcrest N localities also illustrate the effect of
analyzing outcrops on opposite sides of road-cuts. As expected,
fractures predominantly dip to the north (toward the road)
at Hillcrest S, and to the south (also toward the road) at Hillcrest N.
Fracture orientation data from host rocks and intrusions exhibit
significant differences, as shown in Fig. 4. While fractures in the
host rock are likely related to regionally extensive joint patterns,
fractures in the dolerite are aligned mainly along a NW–SE trend
and a sub-ordinate E–W trend. This is remarkably similar to the
regional dyke orientation, but the orientations at the Hillcrest site
(supplying the majority of measurements) also align to the outcrop
orientation. This may mask the natural fracturing pattern.
At Hillcrest, the fracture orientation was sampled along 10 m
wide segments to investigate potential orientation changes toward
79
K. Senger et al. / Journal of African Earth Sciences 102 (2015) 70–85
A
B-N-C
Intrusion
B-N-B
B-N-A
B-N-C
4
3
2
1
4
3
2
1
B-N-B
4
3
2
1
B-N-A
0
20
15
10
5
20
15
10
5
Intrusion
20
15
10
5
1
2
3
4
5
6
7
8
9
10
11
12
13
cumulative number of fractures
fractures per metre
1m
Distance(m)
B
H-S-C
H-S-B
Intrusion
H-S-A
4
3
2
1
4
3
2
1
H-S-C
60
40
20
H-S-B
60
40
20
H-S-A
60
40
20
0
1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
cumulative number of fractures
4
3
2
1
Intrusion contact
fractures per metre
1m
Distance(m)
Fig. 7. Fracture frequency in sedimentary layers away from intrusion contacts. Fractures in dolerite are not represented in this figure. (A) Fracture frequency in three
sandstone layers within 15 m of the thin dyke at Bonkola N. The fracture frequency is displayed both as a histogram (binned fractures/meter, scale on left) and cumulative
number of fractures toward the contact (scale on right). (B) Fracture frequency in three sandstone layers within 30 m of the inclined dolerite sheet at Hillcrest S.
the intrusion contact at 0 m (Fig. 4C). The stereoplots on both sides
of the road (Hillcrest S and Hillcrest N) both indicate a broader
spread in fracture orientations 0–10 m from the contact zone. This
may suggest enhanced structural complexity at the contact zone.
However, similarly variable fracture orientations are also evident
at some distance from the contact zone (particularly in the
30–40 m and 50–70 m intervals) that may be related to
intrusion-internal structural complexities. In the segment closest
to the contact at Hillcrest N, there appears to be a minor overlap
between the fracture orientation and the orientation of the main
dyke (blue diamond illustrated in Fig. 4C). This overlap is not
clearly evident in the inner part of the intrusion. However, the
large scale and uncertain overall geometry of the inclined sheet
exposed at Hillcrest may generate complex external geometries
with irregular contacts as well as correspondingly complex internal fracturing patterns.
5.3. Fracture spacing
The fracture spacing, directly linked to overall permeability in
crystalline rocks with low or no matrix porosity, was investigated
both at the regional and local scale. On the regional scale, high-resolution photo-mosaics were acquired at both limbs of the Golden
Valley saucer-shaped intrusion (Fig. 5A). The analyzed data, plotted
as whisker-plots together with some ‘control sills’ in the vicinity
(Fig. 5B), show a similar fracture density spanning from 1 to 4
80
K. Senger et al. / Journal of African Earth Sciences 102 (2015) 70–85
B-N-4
B-N-3
B-N-2
B-N-1
1m
fractures per metre
40
20
B-N-3
6
4
2
40
20
B-N-2
6
4
2
40
20
B-N-1
6
4
2
0
host rock
40
intrusion
20
cumulative number of fractures
B-N-4
6
4
2
0
0
1
2
3
4
5
6
7
8
9
10
11
12
Distance (m)
Fig. 8. Fracturing patterns at Bonkola N. The analyzed fracture frequency profiles (histograms and cumulative frequency) are shown at the same scale below.
fractures per 10 m (mean) in all the photo-mosaics. The fracture
spacing (Fig. 5C) was approximately 3 m (mean), though it can
be longer locally (e.g., profile GV_E_7, ca. 5 m). Maximum fracture
spacing was generally below 15 m, though in some profiles poor
exposure has resulted in significant maxima. A log–log plot of fracture spacing versus frequency (Fig. 5D) illustrates a typical powerlaw distribution above approximately 3 m, considered to represent
the lower separation limit on this regional scale.
On the local scale, the dolerite-internal fracturing was extensively analyzed within 20 m of the contact at the Hillcrest N locality. Six horizontal virtual profiles and nine vertical profiles are
illustrated in Fig. 6. The horizontal profiles (Fig. 6B) clearly illustrate a fracture swarm aligned roughly perpendicular to the contact, and intersecting the lowermost profile (HC-N-01) at 9 m.
The 2 m thick fracture swarm generates the highest fracture frequency (up to 4 fractures per meter; f/m) in all profiles, and was
also well imaged on two vertical profiles (HC-N-C and HC-N-D).
Apart from this fracture swarm, no obvious fracture frequency
increase is apparent toward the contact on either horizontal or vertical profiles at this site.
Fracturing was also investigated in the sedimentary host rocks
around the intrusion, using the thin dyke at Bonkola N and thick
inclined sheet at Hillcrest S as case studies (Fig. 7). At Bonkola N
(Fig. 8), three horizontal virtual profiles along three distinct sandstone beds all illustrated enhanced fracturing within 2–4 m of the
intrusion contact. This was most obvious from the cumulative fracture plots, but even the binned fracture frequency was typically
(except for profile B-N-C, see Fig. 7A) highest in this interval (up
to 4 f/m). Away from the intrusion, background fracturing was typically 0–2 f/m in all three layers.
At Hillcrest S, the background fracturing, measured up to 25 m
away from the contact, was somewhat higher (1–4 f/m; Fig. 7B).
This is likely related to the strengthening of the host rock due to
contact metamorphism associated with the emplacement of this
relatively large-scale intrusion (Haave, 2005). Closer to the contact,
fracture frequency appeared similar to background fracturing. On
the two uppermost profiles (H-S-B and H-S-C), which follow sandstone bodies presumably wedged up by the intrusion, the cumulative frequency curve was nonetheless somewhat steeper within
the first 3 m away from the intrusion.
Fracture frequency in virtual profiles across both sedimentary
rocks and the intrusion were investigated using the Bonkola dyke
as a case study (Figs. 8 and 9). At Bonkola N, four virtual profiles
across an interpreted high-resolution screen capture of the virtual outcrop model revealed a complex fracture network, particularly in the host rock. In the intrusion itself, fractures were
typically aligned both perpendicular and parallel to the contact.
Fracturing appeared to be minimal within the intrusion (profile
B-N-3), though this may be an artifact due to poor fracture exposure in the dyke, which is partly covered by vegetation. The pervasive fracturing of the host rock, particularly in the shale
between profiles B-N-1 and B-N-2, may be related to surficial
weathering given that the fractures do not display any distinct
orientation.
On the other side of the road, at Bonkola S, enhanced fracturing
is evident within the intrusion, as are clearer, continuous, fractures
within the host rock (Fig. 9). The dipping intrusion emphasizes the
two main fracture sets within the dyke, namely parallel and perpendicular to the intrusion contact. Both of these sets contrast with
the predominantly vertical fracture exposures in the country rock.
A limited number of fractures in the host rock are also aligned
81
K. Senger et al. / Journal of African Earth Sciences 102 (2015) 70–85
parallel to the intrusion contact (highlighted by arrows in Fig. 9),
but all occur within 1 m of the intrusion.
Comparing the fracture spacing in dolerites and host rocks
illustrates the broad range of fracture spacing in dolerites,
particularly when measured along vertical profiles (Fig. 10).
The range of fracture spacing was well-constrained for both host
rock and mixed lithology profiles, with the upper to lower
quartile falling between 0.1 and 1 m on all profiles. In dolerites,
however, the fracture spacing spanned from 0.01 m (min)
to 4 m (max), and the mean spacing varied between 0.3 and
1.5 m. Scanlines, due to their higher resolution, displayed a
consistent mean fracture spacing of approximately 0.1–0.2 m,
corresponding well with the mixed lithology profiles at
Bonkola N.
2m
B
-S
-6
5
S-
B-
B-S-4
B-S-3
B-S-2
B-S-1
B-S-4
3
1
B-S-3
fractures per metre
3
40
30
20
10
2
1
B-S-2
3
40
30
20
10
2
1
B-S-1
3
40
30
20
10
0
2
1
0
0
2
3
4
5
6
7
8
9
Distance(m)
B-S-5
6
f/m
1
13
14
B-S-6
15
16
17
18
19
20
21
cumulative number of fractures
40
30
20
10
2
22
20 fractures
4
2
0
0
1
2
3
host rock
4
5
6
7
intrusion
8
9
1
2
3
4
5
6
7
8
9
10
11
12
Distance(m)
Fig. 9. Fracturing patterns at Bonkola S. The analyzed fracture frequency profiles (histograms and cumulative frequency) are shown at the same scale below.
82
K. Senger et al. / Journal of African Earth Sciences 102 (2015) 70–85
10
fracture spacing (m)
Dolerite
Host rock
Mixed lithology
Scanlines
1
0.1
0.01
NND-02
HC-01
NND-01
B-N-4
BD-01
B-N-3
B-N-2
B-N-1
B-S-6
B-S-5
B-S-4
B-S-3
B-S-2
B-S-1
H_S_C
H_S_B
H_S_A
B_N_C
B_N_B
B_N_A
HC_N_I
HC_N_H
horizontal profiles
HC_N_G
HC_N_F
HC_N_E
HC_N_D
HC_N_C
HC_N_B
HC_N_A
vertical profiles
HC_N_06
HC_N_05
HC_N_04
HC_N_03
HC_N_02
HC_N_01
horizontal profiles
Fig. 10. Summary of fracture spacing from a variety of high-resolution profiles and scanlines, subdivided by the dominant lithology. The whisker-plots illustrate the
minimum, lower quartile, mean, upper quartile and maximum fracture spacing at each profile. Note the logarithmic scale. Refer to Table 3 for details.
6. Discussion
6.1. Heterogeneity of fracture networks and tectonic setting
The most striking result of this study relates to the pervasive
heterogeneity of the study sites. Even within the same outcrop
face, profiles do not reflect a consistent pattern of fracturing
both in the host rock and the intrusion. Furthermore, the
primary fracture orientations are significantly different at the
three sites, being heavily biased by the outcrop orientation. This
complex fracturing pattern may be related to tectonic pre- and
post-intrusion heterogeneities within the host rock (‘TF’), juxtaposed with syn-emplacement fractures in the host rock (‘SEF’)
and post-emplacement fractures in the intrusion (‘PEF’). A good
example is illustrated in the profiles at Bonkola S, where the
intrusion-related fracturing was masked by the pervasive fracturing in the host rock, particularly on the north side of the
intrusion. In order to simplify the complex fracturing patterns,
we discuss the three main fracture sets introduced previously:
(1) syn-emplacement fractures in host rock, (2) post-emplacement fractures in the intrusions and (3) tectonic fractures in
both intrusion and host rock.
6.2. Syn-emplacement fracturing in host rock
The emplacement of intrusions has mechanical consequences
for the surrounding host rock, as exemplified, for example, by the
jacked-up country rock at Nonesi’s Neck. Theoretically, the
increased deformation associated with magma emplacement
should lead to enhanced fracturing of the host rock in the vicinity
of the intrusions. We have shown limited increase (doubling of the
background fracture frequency within one meter of the intrusion,
but only in one profile) in host rock fracturing toward the intrusion
contacts, most developed at the Bonkola N site. The lack of a clear
increase in fracture frequency toward the intrusion is thought to be
masked by the underlying host rock heterogeneity, and, to some
extent, also the resolution limit of the virtual outcrop models,
which clearly underestimate the frequency of short discontinuous
fractures.
On the regional scale within the Karoo Basin, dyke-related fracturing was extensively studied due to its direct influence on
groundwater potential. Woodford and Chevallier (2002) suggested
that fractures related to the emplacement of dykes are present
within 5–15 m from the dyke, forming a structurally complex zone
of enhanced fracturing.
6.3. Post-emplacement fracturing patterns in intrusions
Our study illustrates that the fractures within the same intrusion are typically oriented perpendicular and parallel to the contact, best exemplified at Bonkola N (Fig. 9). These fractures are
thus interpreted as related to thermal contraction during cooling,
with the cooling front propagating from the relatively cool host
rock interface. This is in accordance with our observations, as
well as numerical modeling by Kattenhorn and Schaefer (2008)
who predicted four types of fractures: column-bounding, column-normal, entablature (all of which are cooling fractures,
‘PEF’), and inflation fractures (‘SEF’). This observation is also in
good agreement with the analyses of Bermúdez and Delpino
(2008).
Fractures related to magma cooling will clearly also be controlled by the intrusion geometry and the magma emplacement
chronology and direction. Schofield et al. (2012b), using both outcrops and 3D seismic data, describe the development of magma
lobes within a single sill complex. Proximal to the magma source,
these lobes coalesce into a single sheet. In distal settings, however,
the same sill will form numerous magma lobes. Needless to say,
the fracturing pattern is expected to be different in a large-sized
sheet compared to smaller, more localized and structurally complex magma lobes.
It is also expected that the fracturing pattern will be different in
a large-sized sheet compared to smaller, more localized and
83
K. Senger et al. / Journal of African Earth Sciences 102 (2015) 70–85
Positive topographic
relief
Negative topographic
relief
Groundwater
well
Terrain surface
HOST ROCK
Shale
Fracture Set 1:
syn-emplacement (SEF);
parallel to contact
within zone 3
Increasing due to weathering
and reduced lithostatic pressure
DOLERITE DYKE
ct
ur
es
Sandstone
or
tio
n
of
o
pe
n
Fracture Set 3:
background tectonic
deformation (TF);
not aligned
to intrusion
Pr
op
4
fra
Fracture Set 2:
post-emplacement (PEF);
a: perpendicular to contact
b: parallel to contact)
Sandstone
3
not to scale
2
Shale
Most complex fracture
pattern in the central part
of zone 1 and entire zone 2
HOST ROCK
Decreasing due to
increased lithostatic pressure
1
2
43
Conglomerate
Sandstone
Zone 4:
due to low-grade contact metamorphism)
Zone 3: contact aureole of host sediments “Hornfels zone”, calcite veining,
alteration of petrophysical properties due to contact metamorphism
Zone 2: chilled margin of dolerite dyke
(increased fracturing => permeability pathway?)
Zone 1:
Fig. 11. Conceptual model of fracturing along a hypothetical inclined intrusion, representative of both sills and dykes. The fractures are subdivided into three main sets: (1)
syn-emplacement fractures within the country rock formed during magma emplacement, (2) post-emplacement fractures within the intrusion formed primarily by cooling
mechanisms, and (3) tectonic fractures formed by processes unrelated to the intrusions. At depth (i.e., below the weathering depth), many fractures are closed and the
intrusion primarily acts as a barrier. At shallower levels, where pre-existing fractures may be opened due to weathering, the intrusion may act as a carrier, particularly along
the structurally complex intrusion–host rock interface (zones 2 and 3).
structurally complex magma lobes, as observed by for example
Witte et al. (2012). Nevertheless, it is not trivial to interpret the differences of fracture patterns observed on the studied outcrops in
terms of intrusion thickness only. Indeed, both the size and the
shapes of each studied intrusion are different, such that it is challenging to discriminate between the effects of the intrusion shape
from those of the intrusion thickness on the observed fracture patterns, in particular based on three outcrops only. More systematic
measurements are thus necessary.
6.4. Tectonic fracturing
Tectonic fractures (including compaction and exhumation
fractures) are present at all studied localities, as evidenced by the
background fracturing pattern in the host rock away from the
intrusion contacts. They may both pre- and post-date the intrusive
activity. Tectonic fractures may also be present in the intrusions
themselves, provided that the tectonic event post-dates the intrusive event. Slicken-sided surfaces along fractures within intrusions
84
K. Senger et al. / Journal of African Earth Sciences 102 (2015) 70–85
are a clear indicator of post-intrusive tectonism (e.g., Ogata et al.,
2014).
6.5. Conceptual model of fracturing at intrusion–host rock contact
zones
On the basis of our field observations and the current state-ofknowledge, we propose a conceptual model of fracturing in- and
around a hypothetical intrusion, with focus on the contact zones
(Fig. 11). This structural model should be considered together with
studies on aureole processes in different host rock lithologies (e.g.,
Haave, 2005), which will influence the size of the hornfels zone
(zone 3). The model assumes a symmetric fracturing pattern at
the base and top of the intrusion, and is thus applicable for both sills
and dykes, but not for lava flows. The schematic zonation ranges
from the inner part of the intrusion (zone 1), through the highly
fractured chilled margin (zone 2) and the baked margin of the host
rock (zone 3) to the structurally unaffected country rock (zone 4).
Clearly, the nature of the magmatism (e.g., intrusion geometry, melt
temperature, emplacement duration and direction) and the host
rock properties (e.g., matrix permeability and porosity, pre-existing
fracture network, temperature, fluid presence) will superimpose
heterogeneities onto this conceptual framework. In our model, we
simplify the fracturing pattern into the three main sets: (1) synemplacement fractures in the host rock, (2) post-emplacement fractures in the intrusion, and (3) tectonic fractures in both host rock
and intrusion. Many individual fractures in fracture sets 1 and 2
are thought to be aligned parallel to the intrusion contact. Fractures
perpendicular to the contact within zone 1 reflect the development
of columnar joints.
The permeability through the fracture network will, in addition
to the obvious fracture network distribution identified in our model,
be a function of the fracture connectivity and the fracture aperture.
Cooling fractures perpendicular to the contact are often viewed as
primary permeability pathways (e.g., Rateau et al., 2013), and the
multi-sided pervasive jointing exposes much of the intrusion to percolating fluids. However, as pointed out by Petford (2003), the
geometry of columnar joints lacks the fracture connectivity
required for high levels of permeability. In essence, the columnar
joints may locally provide permeability pathways across the intrusion, provided that they are open, but permeability is thought to be
higher parallel to the intrusion contacts. This hypothesis was confirmed by hydraulic testing of the Palisades sill in New Jersey
(Matter et al., 2006) and an inclined doleritic sheet at Qoqodala in
the Eastern Cape province (Chevallier et al., 2004), both of which
revealed increased permeability along the fracture system located
at the host rock–intrusion interface.
The fracture aperture is a critical parameter directly related to
the flow rate (Singhal and Gupta, 2010). As illustrated in our conceptual model, fracture aperture, and thereby fracture permeability, is
likely to decrease with depth due to various factors, including
decreasing weathering, an increasing lithostatic pressure and
increasing thermal expansion of rocks (Singhal and Gupta, 2010).
This phenomenon of decreasing permeability with depth across
individual dykes is well-known and clearly illustrated by the influence of dykes on the water table at different depths, corresponding
to the weathering depth (Perrin et al., 2011). Nonetheless, the determination of fracture aperture at outcrops is hampered by uncertainty, and proxies such as packer tests are used to estimate the
fracture aperture in various crystalline rocks to less than 400 lm
(Cook, 2003 and references therein). Goldberg and Burgdorff
(2005) used an acoustic televiewer at the Palisades Sill to illustrate
very open (cm-scale) fractures, though it is unclear whether these
relate to individual fractures or fracture corridors. This indicates
that additional work focussed on fracture apertures is required to
constrain the applicability of the conceptual model with depth.
To summarize, the fracturing in the contact zone is particularly
important since it drives the majority of the successful groundwater boreholes (van Wyk, 1963; Woodford and Chevallier, 2002),
and the proposed model (Fig. 11) may provide a framework on
which to conduct additional research.
7. Conclusions
On the basis of our field observations of fracturing patterns in
and around igneous intrusions within the Eastern Cape we infer
the following:
– The studied Karoo dolerite intrudes the Burgersdorp Formation
host rock at three study sites. The dolerite exhibits different
forms, namely a thin dyke (Bonkola), a medium-sized inclined
intrusion (Nonesi’s Neck) and a thick inclined sheet (Hillcrest).
– Fractures are classified into three groups based on the likely
mode of formation, namely (1) syn-emplacement fractures in
host rocks, (2) post-emplacement fractures in dolerite and (3)
tectonic fractures in both host rock and dolerite.
– The primary fracture orientations are significantly different at
the three sites, being heavily biased by the outcrop orientation.
– Fracture spacing is consistent between the three sites, and is
shown to be more variable and irregular within the dolerite
compared to the host rock.
– At one of the sites (Bonkola N), fracturing in the surrounding
host rock increases qualitatively toward the intrusion within
approximately 3 m of the contact.
– Based on our field observations and the current state-of-the-art,
we propose a conceptual model for fluid flow along intrusion–
host rock interfaces. The four-zoned model spans from the unaffected country rock to the central part of igneous intrusions, and
predicts that fluid flow will most likely be focussed along the
intrusion–host rock contact zones.
Acknowledgements
This work was partly financed by the Research Council of Norway, through the CLIMIT program (‘Geological input to carbon
storage’ project) and through its Centres of Excellence funding
scheme (project number 223272). Fieldwork was funded by a
World Universities Network researcher mobility grant and a Meltzerfondet stipend, both to K.S. Lilith Kuckero assisted during the
field work. The Council for Geosciences kindly provided digital geological maps of the study area. OpenStereo (Grohmann and
Campanha, 2010) was used for plotting stereoplots, Google provided an academic license of Google Earth Pro and Riegl GmbH provided ongoing software support. We thank the participants of the
LASI V workshop, held in Port Elizabeth 29–30 October 2012, for
fruitful discussions. Finally, we sincerely appreciate the constructive comments from one anonymous reviewer and the associate
editor Jean-Paul Liégeois.
References
Abdelmalak, M.M., Mourgues, R., Galland, O., Bureau, D., 2012. Fracture mode
analysis and related surface deformation during dyke intrusion: results from 2D
experimental modelling. Earth Planet. Sci. Lett. 359–360, 93–105. http://
dx.doi.org/10.1016/j.epsl.2012.10.008.
Abràmoff, M.D., Magalhães, P.J., Ram, S.J., 2004. Image processing with ImageJ.
Biophotonics Int. 11 (7), 36–42.
Bermúdez, A., Delpino, D.H., 2008. Concentric and radial joint systems within basic
sills and their associated porosity enhancement, Neuquén Basin, Argentina. In:
Thomson, K., Petford, N. (Eds.), Structure and Emplacement of High-Level
Magmatic Systems. Geological Society of London Special Publication #302, pp.
185–198. http://dx.doi.org/10.1144/sp302.13.
K. Senger et al. / Journal of African Earth Sciences 102 (2015) 70–85
Bristow, J., Saggerson, E., 1983. A general account of Karoo vulcanicity in southern
Africa. Geol. Rundsch. 72 (3), 1015–1059. http://dx.doi.org/10.1007/bf01848353.
Buckley, S.J., Howell, J.A., Enge, H.D., Kurz, T.H., 2008. Terrestrial laser scanning in
geology: data acquisition, processing and accuracy considerations. J. Geol. Soc.
165 (3), 625–638. http://dx.doi.org/10.1144/0016-76492007-100.
Chevallier, L., Woodford, A., 1999. Morpho-tectonics and mechanism of
emplacement of the dolerite rings and sills of the western Karoo, South
Africa. S. Afr. J. Geol. 102 (1), 43–54.
Chevallier, L., Goedhart, M., Woodford, A.C., 2001. The Influences of Dolerite Sill and
Ring Complexes on the Occurrence of Groundwater in Karoo Fractured Aquifers:
A Morpho-tectonic Approach. Water Resource Commission Reports. WRC
Report No. 937/1/01: 165.
Chevallier, L., Gibson, L.A., Nhleko, L.O., Woodford, A.C., Nomquphu, W., Kippie, I.,
2004. Hydrogeology of Fractured-rock Aquifers and related Ecosystems within
the Qoqodala Dolerite Ring and Sill Complex, Great Kei Catchment, Eastern
Cape. Water Resource Commission Reports, WRC Report No. 1238/1/04: 150.
Cook, P., 2003. A guide to regional groundwater flow in fractured rock aquifers.
CSIRO Land and Water, Adelaide, Australia, 115 pp.
Cox, K.G., 1992. Karoo Igneous Activity, and the Early Stages of the Break-up of
Gondwanaland, vol. 68(1). Geological Society, London, Special Publications, pp.
137–148. http://dx.doi.org/10.1144/gsl.sp.1992.068.01.09.
Delaney, P.T., Pollard, D.D., 1981. Deformation of Host Rocks and Flow of Magma
during Growth of Minette Dikes and Breccia-bearing Intrusions near Ship Rock,
New Mexico. U.S. Geological Survey Professional Paper #1202, Washington,
USA, 61 pp.
Delaney, P.T., Pollard, D.D., Ziony, J.I., McKee, E.H., 1986. Field relations between dikes
and joints: emplacement processes and paleostress analysis. J. Geophys. Res.:
Solid Earth 91 (B5), 4920–4938. http://dx.doi.org/10.1029/JB091iB05p04920.
Enge, H.D., Buckley, S.J., Rotevatn, A., Howell, J.A., 2007. From outcrop to reservoir
simulation model: workflow and procedures. Geosphere 3 (6), 469–490. http://
dx.doi.org/10.1130/ges00099.1.
Galland, O., Scheibert, J., 2013. Analytical model of surface uplift above
axisymmetric flat-lying magma intrusions: implications for sill emplacement
and geodesy. J. Volcanol. Geoth. Res. 253, 114–130. http://dx.doi.org/10.1016/
j.jvolgeores.2012.12.006.
Goldberg, D., Burgdorff, K., 2005. Natural fracturing and petrophysical properties of the
Palisades dolerite sill. In: Harvey, P.K., Brewer, T.S., Pezard, P.A., Petrov, V.A. (Eds.),
Petrophysical Properties of Crystalline Rocks. Geological Society, Special
Publications, London, pp. 25–36. http://dx.doi.org/10.1144/gsl.sp.2005.240.01.03.
Grohmann, C.H., Campanha, G.A., 2010. OpenStereo: Open Source, Cross-Platform
Software for Structural Geology Analysis. AGU Fall Meeting, 13–17 December
2010, San Francisco, USA. <http://www.igc.usp.br/index.php?id=openstereo>
Gudmundsson, A., Løtveit, I.F., 2012. Sills as fractured hydrocarbon reservoirs:
examples and models. In: Spence, G.H., Redfern, J., Aguilera, R., Bevan, T.G.,
Cosgrove, J.W., Couples, G.D., Daniel, J.M. (Eds.), Advances in the Study of
Fractured Reservoirs. Geological Society of London Special Publication #374.
Geological Society of London, London, p. 21. http://dx.doi.org/10.1144/SP374.5.
Gurba, L.W., Weber, C.R., 2001. Effects of igneous intrusions on coalbed methane
potential, Gunnedah Basin, Australia. Int. J. Coal Geol. 46 (2–4), 113–131. http://
dx.doi.org/10.1016/S0166-5162(01)00020-9.
Gustafson, G., Krásný, J., 1994. Crystalline rock aquifers: their occurrence, use and
importance. Appl. Hydrogeol. 2 (2), 64–75. http://dx.doi.org/10.1007/
s100400050051.
Haave, C., 2005. Metamorphic and Petrophysical Effects of Sill Intrusions in
Sedimentary Strata: The Karoo Basin, South Africa. Unpublished MSc Thesis,
University of Oslo, Norway Thesis, 168 pp.
Hetényi, G., Taisne, B., Garel, F., Médard, É., Bosshard, S., Mattsson, H., 2012. Scales of
columnar jointing in igneous rocks: field measurements and controlling factors.
Bull. Volcanol. 74 (2), 457–482. http://dx.doi.org/10.1007/s00445-011-0534-4.
Hiller, N., Stavrakis, N., 1984. Permo-Triassic fluvial systems in the southeastern
Karoo Basin, South Africa. Palaeogeogr. Palaeoclimatol. Palaeoecol. 45 (1), 1–21.
http://dx.doi.org/10.1016/0031-0182(84)90106-8.
Jackson, M.D., Pollard, D.D., 1990. Flexure and faulting of sedimentary host rocks
during growth of igneous domes, Henry Mountains, Utah. J. Struct. Geol. 12 (2),
185–206. http://dx.doi.org/10.1016/0191-8141(90)90004-I.
Johnson, M.R., Van Vuuren, C.J., Hegenberger, W.F., Key, R., Show, U., 1996.
Stratigraphy of the Karoo Supergroup in southern Africa: an overview. J. Afr.
Earth Sci. 23 (1), 3–15. http://dx.doi.org/10.1016/S0899-5362(96)00048-6.
Kattenhorn, S.A., Schaefer, C.J., 2008. Thermal–mechanical modeling of cooling history
and fracture development in inflationary basalt lava flows. J. Volcanol. Geoth. Res.
170 (3–4), 181–197. http://dx.doi.org/10.1016/j.jvolgeores.2007. 10.002.
Lato, M., Diederichs, M., Hutchinson, D.J., 2010. Bias correction for view-limited
lidar scanning of rock outcrops for structural characterization. Rock Mech. Rock
Eng. 43 (5), 615–628. http://dx.doi.org/10.1007/s00603-010-0086-5.
Lister, J.R., Kerr, R.C., 1991. Fluid-mechanical models of crack propagation and their
application to magma transport in dykes. J. Geophys. Res. 96 (B6), 10049–
10077. http://dx.doi.org/10.1029/91JB00600.
Matter, J., Goldberg, D., Morin, R., Stute, M., 2006. Contact zone permeability at
intrusion boundaries: new results from hydraulic testing and geophysical
logging in the Newark Rift Basin, New York, USA. Hydrogeol. J. 14 (5), 689–699.
http://dx.doi.org/10.1007/s10040-005-0456-3.
McCaffrey, K.J.W., Sleight, J.M., Pugliese, S., Holdsworth, R.E., 2003. Fracture
formation and evolution in crystalline rocks: insights from attribute analysis.
In: Petford, N., McCaffrey, K.J.W. (Eds.), Hydrocarbons in Crystalline Rocks.
Geological Society, Special Publications, London, pp. 109–124. http://dx.doi.org/
10.1144/gsl.sp.2003.214.01.07.
85
Mège, D., Rango, T., 2010. Permanent groundwater storage in basaltic dyke fractures
and termite mound viability. J. Afr. Earth Sci. 57 (1–2), 127–142. http://
dx.doi.org/10.1016/j.jafrearsci.2009.07.014.
Meriaux, C., Lister, J.R., Lyakhovsky, V., Agnon, A., 1999. Dyke propagation with
distributed damage of the host rock. Earth Planet. Sci. Lett. 165 (2), 177–185.
NOAA, 2013. NOAA Magnetic Field Calculators. <http://www.ngdc.noaa.gov/
geomag-web/> (accessed 13.08.13).
Ogata, K., Senger, K., Braathen, A., Tveranger, J., Olaussen, S., 2014. Regional fracture
patterns of relevance for fluid flow in the Longyearbyen CO2 Lab reservoircaprock succession. Svalbard. Norw. J. Geol., in press.
Perrin, J., Ahmed, S., Hunkeler, D., 2011. The effects of geological heterogeneities
and piezometric fluctuations on groundwater flow and chemistry in a hard-rock
aquifer, southern India. Hydrogeol. J. 19 (6), 1189–1201. http://dx.doi.org/
10.1007/s10040-011-0745-y.
Petford, N., 2003. Controls on primary porosity and permeability development in
igneous rocks. In: Petford, N., McCaffrey, K.J.W. (Eds.), Hydrocarbons in
Crystalline Rocks. Geological Society, Special Publications, London, pp. 93–
107. http://dx.doi.org/10.1144/gsl.sp.2003.214.01.06.
Polteau, S., Ferre, E.C., Planke, S., Neumann, E.R., Chevallier, L., 2008. How are saucershaped sills emplaced? Constraints from the Golden Valley Sill, South Africa. J.
Geophys. Res. 113 (B12104), 1–13. http://dx.doi.org/10.1029/2008JB005620.
Rateau, R., Schofield, N., Smith, M., 2013. The potential role of igneous intrusions on
hydrocarbon migration, West of Shetland. Petrol. Geosci. 19 (3), 259–272.
http://dx.doi.org/10.1144/petgeo2012-035.
Rittersbacher, A., Buckley, S.J., Howell, J.A., Hampson, G.J., Vallet, J., 2013.
Helicopter-based laser scanning: a method for quantitative analysis of largescale sedimentary architecture. In: Martinius, A.W., Howell, J.A., Good, T. (Eds.),
Sediment-body Geometry and Heterogeneity: Analogue Studies for Modelling
the Subsurface. Geological Society of London Special Publication #387, London,
pp. 1–18. http://dx.doi.org/10.1144/SP387.3.
Rodriguez Monreal, F., Villar, H.J., Baudino, R., Delpino, D., Zencich, S., 2009.
Modeling an atypical petroleum system: a case study of hydrocarbon
generation, migration and accumulation related to igneous intrusions in the
Neuquen Basin, Argentina. Mar. Petrol. Geol. 26 (4), 590–605. http://dx.doi.org/
10.1016/j.marpetgeo.2009.01.005.
Rubin, A.M., 1995. Propagation of magma-filled cracks. Annu. Rev. Earth Planet. Sci.
23, 287–336. http://dx.doi.org/10.1146/annurev.ea.23.050195.001443.
Rubin, A.M., Pollard, D.D., 1988. Dike-induced faulting in rift zones of Iceland and
Afar. Geology 16 (5), 413–417. http://dx.doi.org/10.1130/0091-7613(1988)
016<0413:difirz>2.3.co;2.
Sankaran, S., Rangarajan, R., Dhar, R., 2005. Delineation of hydraulic connectivity
across a dolerite dyke through hydrogeological, geophysical and tracer studies—
a case study. Environ. Geol. 48 (4), 411–419. http://dx.doi.org/10.1007/s00254005-1230-5.
Schofield, N., Stevenson, C., Reston, T., 2010. Magma fingers and host rock
fluidization in the emplacement of sills. Geology 38 (1), 63–66. http://
dx.doi.org/10.1130/g30142.1.
Schofield, N., Brown, D.J., Magee, C., Stevenson, C.T., 2012a. Sill morphology and
comparison of brittle and non-brittle emplacement mechanisms. J. Geol. Soc.
169 (2), 127–141. http://dx.doi.org/10.1144/0016-76492011-078.
Schofield, N., Heaton, L., Holford, S.P., Archer, S.G., Jackson, C.A.-L., Jolley, D.W.,
2012b. Seismic imaging of ‘broken bridges’: linking seismic to outcrop-scale
investigations of intrusive magma lobes. J. Geol. Soc. 169 (4), 421–426. http://
dx.doi.org/10.1144/0016-76492011-150.
Schutter, S.R., 2003. Hydrocarbon occurrence and exploration in and around
igneous rocks. In: Petford, N., McCaffrey, K.J.W. (Eds.), Hydrocarbons in
Crystalline Rocks. Geological Society, Special Publications, London, pp. 7–33.
http://dx.doi.org/10.1144/gsl.sp.2003.214.01.02.
Singhal, B.B.S., Gupta, R.P., 2010. Applied Hydrogeology of Fractured Rocks.
Springer, Dordrecht, 382 pp.
Smith, R.M.H., 1990. A review of stratigraphy and sedimentary environments of the
Karoo Basin of South Africa. J. Afr. Earth Sci. (and the Middle East) 10 (1–2),
117–137. http://dx.doi.org/10.1016/0899-5362(90)90050-o.
Sruoga, P., Rubinstein, N., Hinterwimmer, G., 2004. Porosity and permeability in volcanic
rocks: a case study on the Serie Tobifera, South Patagonia, Argentina. J. Volcanol.
Geoth. Res. 132 (1), 31–43. http://dx.doi.org/10.1016/s0377-0273(03)00419-0.
Stearns, H.T., 1942. Hydrology of volcanic terranes. In: Meinzer, O.E. (Ed.),
Hydrology. Dover Publishing Inc., New York, pp. 678–703.
Svensen, H., Corfu, F., Polteau, S., Hammer, Ø., Planke, S., 2012. Rapid magma
emplacement in the Karoo Large Igneous Province. Earth Planet. Sci. Lett. 325–
326, 1–9. http://dx.doi.org/10.1016/j.epsl.2012.01.015.
Tankard, A., Welsink, H., Aukes, P., Newton, R., Stettler, E., 2009. Tectonic evolution
of the Cape and Karoo basins of South Africa. Mar. Pet. Geol. 26 (8), 1379–1412.
http://dx.doi.org/10.1016/j.marpetgeo.2009.01.022.
Thomaz Filho, A., Mizusaki, A.M.P., Antonioli, L., 2008. Magmatism and petroleum
exploration in the Brazilian Paleozoic basins. Mar. Pet. Geol. 25 (2), 143–151.
http://dx.doi.org/10.1016/j.marpetgeo.2007.07.006.
van Wyk, W.L., 1963. Ground-water Studies in Northern Natal, Zululand and
Surrounding Areas, Geological Survey of South Africa, Memoir 52. Republic of
South Africa, Pretoria, 145 pp.
Witte, J., Bonora, M., Carbone, C., Oncken, O., 2012. Fracture evolution in oilproducing sills of the Rio Grande Valley, northern Neuquén Basin, Argentina.
AAPG Bull. 96 (7), 1253–1277. http://dx.doi.org/10.1306/10181110152.
Woodford, A., Chevallier, L., 2002. Hydrogeology of the Main Karoo Basin: Current
Knowledge and Future Research Needs. Water Resource Commission, WRC
Report No. TT 179/02, 506 pp.