Content uploaded by Young-Seog Kim
Author content
All content in this area was uploaded by Young-Seog Kim on Jul 26, 2022
Content may be subject to copyright.
Article
Vol. 26, No. 1, p. 79−93, February 2022
https://doi.org/10.1007/s12303-021-0018-2
pISSN 1226-4806 eISSN 1598-7477
Geosciences Journal
GJ
Descriptive classification of dyke morphologies based
on similarity to fracture geometries
Seok-Jun Yang
1
and Young-Seog Kim
2
*
1
Mineral Resources Research Division, Korea Institute of Geoscience and Mineral Resources, Daejeon 34132, Republic of Korea
2
Department of Earth & Environmental Sciences, Pukyong National University, Busan 48513, Republic of Korea
ABSTRACT: Fractures (including faults and joints) are important pathways for magma in the upper crust. Two theories of dyke intru-
sion are generally accepted: dykes open pre-existing fractures or dykes inject propagating new fractures at their tips. Thus, there is a
close interrelationship between dyke intrusion and fractures. We present a field-based study focused on geometric patterns of intru-
sions, because detailed descriptions and analyses of exposed dykes can provide useful complementary information regarding the roles
of fractures as dyke formation pathways. Most non-planar dyke examples that we analyzed are from four areas (Geoje Island, Kori, Uljin, and
Suncheon) of Korea. Some dykes show morphological similarities to fractures, such as branching, bending, and splaying. Therefore,
we classified the intrusion patterns based on the equivalent terminology for fractures (e.g., tip, wall, and linkage), which is useful for describ-
ing dykes at the outcrop scale and for research into inter-relationships between dykes and fractures. This study improves our under-
standing for the role of fractures in guiding dyke emplacement.
Key words: dyke intrusion pattern, classification, fractures, shallow-level intrusion, dyke emplacement mechanism
Manuscript received November 9, 2020; Manuscript accepted May 28, 2021
1. INTRODUCTION
Dykes rarely form simple planar structures, but instead show
a variety of intrusion patterns, especially within the shallow crust
(e.g., Hoek, 1991; Jolly and Sanderson, 1995; Kattenhorn and
Watkeys, 1995; Ryan, 1995; Platten, 2000; Babiker and Gudmundss on,
2004; Gudmundsson, 2005; Clemente et al., 2007; Khodayar
and Franzson, 2007; Paquet et al., 2007; Goulty and Schofield,
2008; Mathieu et al., 2008; Rivalta et al., 2015; Kavanagh et al.,
2017). These different intrusion patterns provide information
regarding related structures and the mechanisms of magma
intrusion (e.g., Baer et al., 1994; Yang et al., 2008; Edwards et al.,
2017; Tibaldi and Bonali, 2017; Walker et al., 2017).
Two end-member mechanisms for dyke intrusion have been
described (e.g., Martínez-Poza et al., 2014). The first mechanism
is based on the hypothesis that it is easier for magma to intrude
along pre-existing fractures than to generate new fractures (e.g.,
Delaney et al., 1986; Spacapan et al., 2016). If the magma
overpressure, which is the difference between the magmatic and
lithostatic pressures, is greater than the tectonic compressive
stress acting perpendicular to the pre-existing fractures, magma
will intrude along pre-existing fractures (e.g., Tibaldi and Bonali,
2017). However, this process appears to conflict with hydraulic
fracturing mechanisms (e.g., Baer et al., 1994), and numerous
studies have suggested that fracturing-related sheet intrusions
(sills and dykes) can develop in host rocks (Pollard, 1973; Spence
and Turcotte, 1985; Rogers and Bird, 1987; Lister and Kerr, 1991;
Baer et al., 1994). In such instances, the magma pressure would
exceed the minimum effective normal stress acting on the host
rock at the dyke tip (e.g., Gudmundsson, 1995, 2005). An impor tant
commonality between these two theories is that most magma
intrusions occur along fractures, which may be newly formed or
pre-existing. Thus, the mechanism of shallow magma intrusion is
strongly influenced by fracture mechanics, geometries, and their
distribution characteristics in a specific area.
A dyke is a magma material that fills the open space of a
fracture formed through displacement of both walls (e.g., Hoek,
1991), and therefore the dyke intrusion pattern preserves the
stress-strain relationship and provides insight into paleostress
regimes. This study aims to analyze and classify dykes based on
*Corresponding author:
Young-Seog Kim
Department of Earth & Environmental Sciences, Pukyong National
University, 45, Yongso-ro, Nam-Gu, Busan 48513, Republic of Korea
Tel: +82-10-2993-7909, Fax: +82-51-629-6623, E-mail: ysk7909@pknu.ac.kr
©
The Association of Korean Geoscience Societies and Springer 2022
80 Seok-Jun Yang and Young-Seog Kim
https://doi.org/10.1007/s12303-021-0018-2
https://www.springer.com/journal/12303
Fig. 1. (a) Simplified location map of the Gyeongsang Basin in southeast Korea formed by subduction of Pacific plates (modified from
Kamata, 1998). (b) Geological map around the Gyeongsang Basin showing the locations of study sites. Cretaceous granitoids and sed-
imentary rocks were intruded with dykes during the Late Cretaceous to the Early Tertiary (modified from Kang and Paik, 2013; Jin et al.,
2018).
Descriptive classification of dyke morphologies 81
https://www.springer.com/journal/12303
https://doi.org/10.1007/s12303-021-0018-2
associated fracture geometry to investigate the role of fractures
in the formation of dyke intrusion patterns. To investigate the
roles of fractures, examples showing clear relationship between
fracture and dyke were collected from four regions in Korea
(Geoje Island, Kori, Uljin, and Suncheon) and post-intrusion
fractures were separated based on cross-cutting relationships.
The results show that a variety of non-straight intrusion
patterns can be developed due to complex fracture systems. To
classify intrusion patterns in terms of fracture geometry,
we employed a fault damage zone classification concept (e.g.,
Kim et al., 2004). Geometric and kinematic analyses of dyke
intrusion patterns based on these concepts can clarify the
processes of local changes in dyke direction, dyke geometry,
and opening direction. This study emphasizes the importance
of analyzing fractures surrounding a dyke in the field and
analyzing dyke intrusion pattern in terms of fracture geometry.
This descriptive analysis can help the kinematic and dynamic
interpretations of intruded dykes.
2. METHODS
2.1. Location and Geological Background of Outcrop
Data
The Gyeongsang Basin, located in southeast Korea, comprises
Cretaceous non-marine sedimentary, volcanic, and pyroclastic
rocks and Late Cretaceous to Early Tertiary felsic rocks (Fig. 1;
Chang, 1975; Chang et al., 2003; Hwang et al., 2008; Kang and
Paik, 2013). These rocks are intruded with Tertiary dykes of
various compositions (Son et al., 2007). Tectonically, the basin
is Cretaceous continental back-arc basin formed by subduction
of Pacific plate beneath the Eurasia plate (Chough and Sohn,
2010; Fig. 1a), thus igneous activities took place actively and
various intrusion patterns are exposed along the coast line of
the southeast Korea. In addition, although the Suncheon area is
located outside of the Gyeongsang Basin, a mafic dyke with a
specific pattern in the Jurassic granite was identified in that
area and included in this classification study.
We analyzed examples of specific dyke intrusion patterns to
infer the roles of fractures associated with dyke intrusion. Although
the shape and direction of the exposed dyke plane can lead to
confusion regarding dyke morphology (i.e., irregular shapes),
the dyke intrusion patterns observed in this study indicate a clear
relationship between fractures and intrusion patterns. Photographs
and sketches were recorded for specific intrusion patterns, and
these data were analyzed geometrically. The deformation history
of all outcrops was established based on the cross-cutting
relationships between dykes and fractures to identify fractures
related to intrusion.
2.2. Basic Classification Concepts of Descriptive
Dyke Intrusion Pattern
Dykes can be characterized by a combination of the fracture
system and the dilation vector (e.g., Hoek, 1991). The dilation
vector can be estimated from the displacement of fracture walls.
Therefore, detailed descriptive analysis of fracture systems related
to dykes must be performed to understand the roles of fractures
in the formation of specific intrusion patterns.
A fracture can provide either a straight or non-straight conduit
for magma. Most dyke intrusions accompany tectonic events,
and thus the propagation or reactivation of fracture systems during
the intrusion event can generate various geometries depending
on the stress state. Non-straight conduits can occur in the following
situations: simultaneous intrusion into more than two directions
of fracture sets, oblique openings to the fracture walls, fractures
with asperities, rotation of local stress, and differences in fracture
geometries depending on the location of the parent fault or
fracture (i.e., tip, wall, and linking zone). These factors may affect
the fluid flow through conduits individually or in combination.
Therefore, the proper analysis and integrated interpretation for a
dyke system is very important to understand the dyke forming
mechanism and the characteristics of fluid flow.
The fault damage zone classification concept (e.g., Kim et al.,
2004) applied in this classification study allows the kinematic
analysis of the fault system based on the fracture geometries at
each location (i.e., tip, wall, and linking zone). Geometric analysis
and classification of dyke intrusion patterns based on this concept
can support the kinematic analysis of the fracture system during
dyke intrusion and help to clarify the overall dyke system. Therefore,
we applied a similar classification concept and terminology
(Fig. 2) used for fault description to explain the roles of fractures
at each dyke location.
Fig. 2. Schematic diagram showing the three classes (array, linkage,
and segment patterns) and two sub-classes (wall and tip patterns) of
dyke classification based on the exposed part of a dyke or dykes.
82 Seok-Jun Yang and Young-Seog Kim
https://doi.org/10.1007/s12303-021-0018-2
https://www.springer.com/journal/12303
3. CLASSES AND TYPES
The classification system used in this study is divided into
three classes based on observations in two-dimensional outcrops:
the array pattern describes the distribution through the whole dyke,
the linkage pattern shows the relationships between neighboring
dykes, and the segment pattern represents individual dyke geometry
(Fig. 2). The class of segment pattern is subdivided into wall
pattern and tip pattern to describe different parts of the exposed
dyke. All patterns are divided into straight and non-straight types.
3.1. Major Classes
In the classification system (Fig. 3), three major classes (array,
linkage, and segment patterns) are expressed with white letters
in blue ellipses and numbered with Roman numerals sequentially
from the largest scale.
3.1.1. Array pattern
In this study, an array defines the distribution characteristics
of several neighboring dykes (more than two dyke segments).
This pattern can be subdivided into straight and non-straight
types according to the distribution of dyke segments. The straight
type may develop through intrusion into a systematically well-
developed fracture system in one direction, while the non-straight
type occurs with intrusions into more than two fracture systems
or into a radial or complex fracture system.
3.1.2. Linkage pattern
In this study, the term linkage is used to classify the geometry
between two dyke segments. This pattern can be classified into
straight and non-straight types based on the geometry of the
two segments at their linkage. The non-straight type can show
various linkage types related to stress trajectory changes, oblique
openings, and existing fractures between neighboring dyke
Fig. 3. Descriptive classification chart for dyke intrusion patterns; dyke types are categorized into three main patterns and the segment pat-
tern is subdivided into two sub-patterns. Straight and non-straight types were determined based on the presence of specific intrusion pat-
terns. Each lowercase letter followed by a parenthesis indicates the source of a specific dyke intrusion pattern.
Descriptive classification of dyke morphologies 83
https://www.springer.com/journal/12303
https://doi.org/10.1007/s12303-021-0018-2
segments, whereas the straight type forms when no such effect
is present.
In particular, it is important to distinguish linkages from
structures such as bridges (segments separated by host rock),
offsets (apparent displacements), and steps (bends) that may
form during dyke propagation.
3.1.3. Segment pattern
In this study, the term segment is used to classify geometric
shapes observed within one dyke segment. This pattern is
subdivided into two sub-patterns of wall and tip according to
the portion of the dyke under consideration. Sub-patterns are
represented by white letters and green rounded squares, and are
numbered III A and III B.
The wall is defined as the main body of a dyke segment (i.e.,
no dyke tip is involved). This pattern can be largely classified
into straight and non-straight types based on the geometry of
the segment wall. The straight type intrudes along a straight frac ture
set, while the non-straight type generally intrudes along curved
fractures or fracture sets with more than two orientation planes.
In the non-straight type, the opening direction can be inferred
through fracture restoration using matching pairs on the two
dyke walls.
The tip is defined as the marginal part or end of a dyke segment
on the 2-D plane. Similar to a fracture, a dyke has the greatest
change in stress at the tip, which can lead to the formation of many
irregular shapes. This pattern can be classified into straight and
non-straight types based on the geometry of the segment tip.
No change in direction is observed in tapered and blunt-ended
shapes, whereas direction changes are observed in deflected,
curved, and splayed shapes. Furthermore, deflected, curved, and
splayed shapes can be caused by preexisting fractures with other
directions (major and secondary fractures) or local change of
the stress trajectory.
3.2. Types
All patterns can be subdivided into straight and non-straight
types based on the presence of characteristic intrusion features
(i.e., changes in intrusion direction). Types are represented by
white letters and navy squares. These types are described in this
study, as well as previous studies with clear relationships between
the fracture and dyke intrusion pattern, to demonstrate how
variable intrusion patterns are controlled by fractures. Each
example is numbered with alphabets, which represent the sources
of the examples listed bottom of the classification system.
Details of the types described in other studies are listed in Table
1; some types are described in both this study and previous
studies. New examples can be added to each type through
further research.
In this classification system (Fig. 3), the types are classified
based on the following criteria (i.e., specific intrusion geometries):
(1) Non-straight type of Array pattern: existence of curved
segments (b) or intersecting dykes (c). (2) Straight and Non-straight
type of Linkage pattern: connected (e, g) or non-connected (d, f). (3)
Non-straight type of Wall pattern: deflected in one direction (i),
deflected in more than two directions (j), systematically deflected (k)
or branched (l). (4) Straight type of Tip pattern: tapered (m), blunted
(n) or splitted (o). (5) Non-straight type of Tip pattern: deflected
in one direction (p), deflected in more than two directions (q)
or splay (r).
Tab l e 1. Specific intrusion types from other studies used in the present classification
Classification in
this study (Fig. 3) Intrusion pattern Controlling factor Reference
Array
pattern
a) En-echelon array Pre-existing joint set fig. 3a of Healy et al. (2018)
c) Intersecting dyke array Pre-existing dyke fig. 12a of Hoek (1991)
Linkage
pattern
d) Right-stepped offset dyke Pre-existing joint network and
regional stress field fig . 5a of Martinez-Poza and Druguet (2016)
f) Unconnected curved tips at the linkage zone Shear zone fig. 5a of Clemente et al. (2007)
g) Dyke segments connected by branched dykes Two-direction fracture set fig. 2 of Hoek (1991)
Wall
pattern
j) Curved fold Pre-existing fracture sets fig. 11 of Tibaldi et al. (2008)
k) Zigzag-shaped Newly formed fractures fig. 11 of Hoek (1991)
l) Branched Main fault plane and secondary fractures fig. 2a of Spacapan et al. (2016)
Tip patte rn
m) Tapered tip Fractures in the fault zone fig. 3c of Dering et al. (2019)
n) Blunt-ended High-angle discontinuity of the magma
propagation direction fig. 2 of Kattenhorn and Watkeys (1995)
o) Blunt-ended Pre-existing joint set fig. 3a of Healy et al. (2018)
q) Deflected tip High-angle discontinuity of the magma
propagation direction fig. 6 of Gudmundsson (2011)
r) Split tip Pre-existing main fracture and
a secondary fracture plane fig. 6a of Edwards et al. (2017)
84 Seok-Jun Yang and Young-Seog Kim
https://doi.org/10.1007/s12303-021-0018-2
https://www.springer.com/journal/12303
4. OUTCROP DATA
Analysis of dykes showing characteristic intrusion patterns
and their surrounding fractures was performed, and the history
of deformation at each outcrop was established based on mutual
cross-cutting relationships.
It remains controversial whether dyke formation uses magma-
driven fracturing or pre-existing fractures (e.g., Hoek, 1991) as
its intrusion mechanism. To distinguish these possibilities in the
field, understanding of the regional fracture pattern and analysis
of the fracture tip pattern are ne cessary; in the hydraulic mag matic
fracturing model, dyke tips tend to be irregular, whereas in the
pre-existing fracture-fill model, the dyke tips are straight and
wedge-shaped (e.g., Ghodke et al., 2018). However, the main purpos e
of this study is to study the role of fractures in determining the
geometry of the dyke by characterizing the fracture system that
controls the direction of the dyke, rather than distinguishing pre-
existing or newly fractures. Thus, stage I was defined as dyke
intrusion event, and fracture sets related to dyke intrusion were
classified based on their deformation history.
Bussell’s method (1989), which is commonly used to determine
opening direction, involves matching pairs of features on the
two walls of a dyke. However, most examples introduced in this
study show insufficient numbers of matching pairs because of
their limited dimensions or exposures. Thus, opening direction
was assumed based on dyke geometry; if the dyke is straight on
the observed scale, the opening direction is perpendicular to
the dyke wall, whereas if the dyke is deflected with constant
thickness, the opening direction is the direction bisecting the
bending fracture (i.e., the preferred orientation to open both
fractures; Yang et al., 2008). The assumed opening direction must
be modified using Bussell’s method to avoid contamination of data.
4.1. Array Pattern
4.1.1. Straight type: example 1
A straight dyke array was identified in a granite outcrop in the
Kori area (Figs. 4a and b). The maximum thickness of these
acidic dyke segments (dipping < 30°) is approximately 20 cm,
and they are cross-cut by both densely spaced NE-SW-striking
(Set-b) and NW-SE-striking fractures (Set-c).
Based on the relationship between the dykes and fracture
systems (Set-a to -c), the deformation history of this outcrop
can be divided into three stages (Fig. 4e). Stage I: dyke intrusion
controlled by N-S-striking fractures (Set-a). Set-a is anastomosed
by post-fracture systems and thus has low capacity for extension.
Stage II: Densely spaced Set-b fractures cross-cut the dyke array
with a minor right-lateral horizontal offset. Stage III: Set-c is
interpreted as the final deformation event, evidenced by slightly
curved tips of Set-c fractures that are abutted to Set-b and
portions of Set-b cross-cutting Set-c.
N-S-striking fractures (Set-a) parallel to the dyke are distr ibuted
around the dyke segments and do not cross-cut the dykes,
indicating a syn- or pre-dyking fracture set. An E-W-trending
opening direction can be assumed from the N-S striking dyke
segment arrangement, which is reasonably constant without any
change in direction at the overlapping parts of each segments.
Thus, in this outcrop, the systematically oriented fractures (Set-
a) and opening direction subp erpendicular to Set-a are interpreted
as important controlling factors.
4.1.2. Non-straight type: example 2
A non-straight array of three dyke segments (Segment-a, -b,
and -c) with slightly different thicknesses and shapes was
identified in a granite gneiss from the Uljin area (Figs. 4c and d).
The maximum thickness of these andesitic dyke segments is
approximately 30 cm, and they are cross-cut by densely developed
fractures of the host rock, which can be divided into three sets:
Set-a (NNE-SSW-striking fractures), Set-b (NW-SE-striking
fractures), and Set-c (E-W-striking fractures). NNE-SSW- and
NW-SE- striking quartz veins are present in both dykes and the
host rock.
Based on the relationships among the dykes, fracture systems
(Set-a to -c), and veins, the deformation history of this outcrop
can be divided into three stages (Fig. 4f ). Stage I: dyke intrusion
controlled by Set-a and Set-b. Stage II: Quartz veins were
developed along the Set-a and Set-b on both the host rock and
dykes, indicating reactivation of both sets. Stage III: Set-c cross-
cuts the dyke segments and is anastomosed by Set-a and Set-b,
indicating that Set-c is the final deformation event.
On the observed outcrop scale, Segment-a is planar, whereas
Segment-b and Segment-c have zigzag shapes. In particular,
Segment-c has an obvious zigzag shape bounded by Set-a and
Set-b, and two matching pairs were observed; this segment
indicates a WNW-ESE-trending opening direction (black dashed
double-headed arrows in Fig. 4d). Thus, in this outcrop, the
intersecting fractures (Set-a and -b) and opening oblique to
both sets are interpreted as important controlling factors.
4.2. Linkage Pattern
4.2.1. Straight type: example 1
A set of dyke segments with straight walls and rectangular-
shaped tips was identified in a granite outcrop in the Kori area
(Figs. 5a and b). This segmented mafic dyke is up to 2 m thick
and is surrounded by fracture sets oriented parallel to the dyke
(Set-a) and at a high angle (Set-b) to it. The N-S-striking fractures
can be divided into two unconnected groups: joints within the
Descriptive classification of dyke morphologies 85
https://www.springer.com/journal/12303
https://doi.org/10.1007/s12303-021-0018-2
dyke that are generally at a high angle to the dyke walls, and
fractures outside the dyke that do not penetrate the dyke (Set-b).
The deformation history, based on the relationships between
the dykes and fracture systems (Set-a to -c) in this outcrop, can
be divided into two stages (Fig. 5e). Stage I: dyke intrusion
controlled by Set-a and b. Stage II: Set-a fractures were reactivated,
and right-lateral movement is inferred from the geometry of the
linkage zone.
The extensive WNW-ESE-striking Set-a is oriented parallel to
the dyke segments. Comparatively sharp contact to the host
rock of the two segments and a small-scale bending shape of the
northern segment (blue dotted circle in Fig. 5b) support the
interpretation that the dyke intrusion is controlled by the fracture
sets, Set-a and Set-b. A NNE-SSW-trending opening direction
is assumed based on the general dyke direction. This direction is
parallel to Set-b, indicating an unfavorable direction of magma
Fig. 4. Examples of dyke arrays. Representative fractures of each fracture set are highlighted with thick red lines. (a and b) Straight acidic dyke
array in a granite from the Kori area (35°19'53''N/129°17'44''E). (c and d) Non-straight dyke array in a granite gneiss from the Uljin area
(36°47'36'' N/129°27'46''E). Black dotted arrows connect matching pairs on the two dyke walls, indicating opening direction. (e and f) Defor-
mation history based on the relationships between dyke segments and surrounding fractures for (a) and (c), respectively.
86 Seok-Jun Yang and Young-Seog Kim
https://doi.org/10.1007/s12303-021-0018-2
https://www.springer.com/journal/12303
Fig. 5. Examples of linkage patterns. Representative fractures of each fracture set are highlighted with thick red lines. (a and b) Straight link-
ing dyke segments in a granite from the Kori area (35°19'53''N/129°17'44''E). (c and d) Non-straight (between Segment-a and -b) and straight
(between Segment-b and -c) linking dyke segments from a sedimentary rock in Geoje Island (modified from Yang et al., 2008; 34°44'18''N/
128°39'34'' E). The black dashed arrow connects matching pairs on the two dyke walls, which indicate the opening direction. (e and f) Defor-
mation histories based on the relationships between dyke segments and surrounding fractures for (a) and (d), respectively.
Descriptive classification of dyke morphologies 87
https://www.springer.com/journal/12303
https://doi.org/10.1007/s12303-021-0018-2
propagation. This finding supports the interpretation that the
tetragonal tip of the dyke was formed because Set-b fractures
acted as barriers. Thus, in this outcrop, the roles of fractures can
be divided into pathway (Set-a) and barrier (Set-b) functions;
these functions can be differentiated based on opening direction
and are both important controlling factors.
4.2.2. Non-straight type: example 2
A set of andesitic dyke segments with linking geometries
similar to fractures was identified in sedimentary rocks in Geoje
Island (Figs. 5c and d). The maximum thickness of these dyke
segments is approximately 15 cm, and the segments are surr ounded
by E-W-striking fractures (Set-a) and NE-SW-striking fractures
(Set-b).
For observational convenience, the basaltic dyke array was
arbitrarily separated as follows: Segment-a to -c, and the linking
dyke. Segment-a and -b are linked by a NE-SW-striking linking
dyke, while no linking dyke is present between Segment-b and -
c. The E-W-striking dyke segments with blunt (i.e., abruptly
terminating tip at the eastern tip of Segment-b) and offshoot
(i.e., an unconnected small dyke segment near the dyke tip at
the western tip of Segment-c) tips cross-cut the N-S-striking
acidic dyke, indicating an intrusive sequence. E-W-striking
fractures (Set-a) running parallel to the dyke direction and NE-
SW-striking fractures (Set-b) parallel to the linking dyke are
present around the dykes.
The deformation history based on the relationship between
the dykes and fracture systems (Set-a and -b) in this outcrop can
be divided into two stages (Fig. 5f). Stage I: N-S-striking acidic
dyke intrusion. Stage II: mafic dyke intrusion controlled by Set-
a and -b.
Set-a fractures are parallel to the dyke direction, rarely
penetrate the dyke, and are widely distributed in the host rock.
Set-b fractures run parallel to the linking dyke, rarely penetrate
the dyke, and are concentrated between the acidic dyke and the
linking dyke. The thickness of the E-W-striking dyke segments
is reasonably constant, while the thickness of the linking dyke is
more variable. An opening direction of NW-SE was estimated
from the bent region formed through deflection from Segment-
c to the linking dyke (black dashed double-headed arrow in Fig.
5d). Thus, in this outcrop, intersecting fractures (Set-a and -b)
and the oblique openings of both sets are interpreted as
important controlling factors.
4.3. Wall Pattern
4.3.1. Non-straight type: example 1
A curved fold-shaped basaltic dyke was identified in a granite
outcrop in Geoje Island. The maximum thickness of the dyke is
approximately 8 cm (Figs. 6a and b), and it is surrounded by
NW-SE-striking fractures (Set-a), E-W-striking fractures (Set-
b), and N-S-striking fractures (Set-c).
Based on the relationship between the dyke and fracture
systems (Set-a to -c), the deformation history of this outcrop
can be divided into three stages (Fig. 6e). Stage I: dyke intrusion
controlled by Set-a and -b. Stage II: Set-b fractures were
reactivated indicating left-lateral movement, evidenced by a
linking fracture in the dyke. Stage III: Set-c fractures cross-cut
the dyke and the Set-b fractures (some Set-c fractures are attached
to Set-b fractures with curved tips) with minor right-lateral
displacement.
The Set-a fractures run parallel to the dyke direction, but do
not penetrate the dyke, and are rarely present around the dyke;
Set-b fractures run parallel to the bending direction of the dyke,
penetrating the dyke, and are concentrated at the bend in the
dyke. However, there is no distinct deformation on the dyke
caused by shearing or folding related to the movement of Set-b.
A NNE-SSW-trending opening direction in which both Set-a
and Set-b are opened equally was assumed based on the constant
thickness of the dyke. Thus, in this outcrop, intersecting fractures
(Set-a and -b) and the oblique openings of both sets are interpreted
as important controlling factors.
4.3.2. Non-straight type: example 2
A kink-shaped mafic dyke was identified in a granite outcrop
in the Suncheon area (Figs. 6c and d). The maximum thickness
of this mafic dyke is approximately 20 cm, and it is surrounded
by N-S- (Set-a), NE-SW- (Set-b), and WNW-ESE-striking (Set-
c) fractures.
Based on the relationship between the dyke and fracture
systems (Set-a to -c), the deformation history of this outcrop
can be divided into two stages (Fig. 6f). Stage I: dyke intrusion
controlled by Set-a and -b. Stage II: Set-c fractures cross-cut the
entire structure.
Set-a runs parallel to the general dyke direction, while Set-b is
parallel to the kink direction of the dyke and occurs densely
around the dyke. These two fracture sets do not penetrate (or
displace) the dyke. A NW-SE-trending opening direction in
which both Set-a and Set-b are opened equally was assumed
based on the constant thickness of the dyke. Thus, in this outcrop,
intersecting fractures (Set-a and -b) and the oblique openings of
both sets are important controlling factors.
4.4. Tip Pattern
4.4.1. Non-straight type: example 1
A straight basaltic dyke with a curved tip was identified in a
granite outcrop in Geoje Island (Figs. 7a and b). The maximum
88 Seok-Jun Yang and Young-Seog Kim
https://doi.org/10.1007/s12303-021-0018-2
https://www.springer.com/journal/12303
thickness of this mafic dyke is approximately 40 cm (decreasing
toward the tip of the dyke, then deflecting into a curved tip) and
it is surrounded by E-W- (Set-a), ENE-WSW- (Set-b), and
NNW-SSE-striking (Set-c) fractures.
Based on the relationships between the dyke and fracture
systems (Set-a to -c), the deformation history of this outcrop
can be divided into three stages (Fig. 7e). Stage I: dyke intrusion
controlled by Set-a and its secondary fractures. Stage II: Set-b
fractures cross-cut the dyke and affected to some fractures of
Set-a and its secondary fractures. Mode I fractures at the tip of
Set-b indicate left-lateral movement. Stage III: Set-c fractures
cross-cut the entire structure and are attached with curved tips
Fig. 6. Examples of wall patterns. Representative fractures of each fracture set are highlighted with thick red lines. (a and b) Non-straight dyke
wall in a granite outcrop in Geoje Island (34°42'45'' N/128°37'42''E). (c and d) Non-straight dyke wall in a granite in the Suncheon area
(35°3'42'' N/127°33'49''E). (e and f) Deformation histories based on the relationships between the dyke segments and surrounding structures
for (b) and (d), respectively.
Descriptive classification of dyke morphologies 89
https://www.springer.com/journal/12303
https://doi.org/10.1007/s12303-021-0018-2
Fig. 7. Examples of tip patterns. Representative fractures of each fracture set are highlighted with thick red lines. (a and b) Non-straight dyke
tip observed in a granite in Geoje Island (34°48'6'' N/128°41'56''E). (c and d) Non-straight dyke tip in a granite in the Kori area (35°19'53''N/
129°17'44'' E). The black dotted arrow connects matching pairs on the two dyke walls, indicating the opening direction. (e and f) Deformation
histories based on the relationships between dyke segments and surrounding structures for (a) and (c), respectively.
90 Seok-Jun Yang and Young-Seog Kim
https://doi.org/10.1007/s12303-021-0018-2
https://www.springer.com/journal/12303
to some Set-b fractures.
The curved tip may be controlled by pre-existing secondary
fractures or rotation of local stresses occurring at the fracture
tip. These structures are difficult to interpret accurately from
this outcrop alone, but Set-a clearly served as the main conduit
for magma.
4.4.2. Non-straight type: example 2
A curved fold-shaped and branched mafic dyke (main dyke)
with a splayed tip was identified in a granite outcrop in the Kori
area. The maximum thickness of the dyke is approximately 2 m
(Figs. 7c and d), and it is surrounded by WNW-ESE- (Set-a),
NE-SW- (Set-b), and NNW-SSE-striking (Set-c) fractures.
Based on the relationships between the dykes and fracture
systems (Set-a to -c), the deformation history of this outcrop
can be divided into three stages (Fig. 7f ). Stage I: dyke intrusion
controlled by Set-a and Set-b. Stage II: Set-a and Set-b fractures
were reactivated and showed left-lateral and right-lateral movement,
respectively, as evidenced by minor displacements of the split
dyke. The occurrence of splayed dykes displaced by approximately
2 cm across NW-SE-striking fractures indicates that these fractures
were reactivated. Stage III: Set-c fractures cross-cut all structures
and attached to some Set-a and Set-b fractures.
Set-a and Set-b fractures do not penetrate the dyke and run
parallel to the direction of the branched dyke (similar to a
horse-tail fracture tip) and the split dyke, respectively. A NNW-
SSE-trending opening direction was estimated from matched
pairs. Figure 5c shows small NE-SW-striking splayed dykelets
present only on the northern wall of the WNW-ESE-striking long
thin dykelet. This asymmetric development of dykelets may
indicate local openings related to left-lateral shearing along the
WNW-ESE-striking fractures. Thus, in this outcrop, intersecting
fractures (Set-a and -b) and the oblique openings of both sets
are interpreted as important controlling factors.
5. DISCUSSION
5.1. Possible Kinematic and Dynamic Implications of
Geometric Classification
The geometries and distribution characteristics of planar
igneous bodies such as dykes and sills have been used to reco nstruc t
paleostress conditions and to characterize emplacement mechanism s,
melt source locations, and tectonic and magmatic processes (e.g.,
Magee et al., 2019). Thus, the interpretation of structures related
to dyke intrusion provides information regarding stress changes
and stress states, because dyke emplacement and propagation
provide clues to crustal evolution from the scale of fractures to
fragmentation of supercontinents (e.g., Ghodke et al., 2018). In
particular, because dyke propagation is conceptually similar to
faulting in shallow-level crust (e.g., Dering et al., 2019), the accurate
analysis of the relationship between a dyke and the surrounding
fracture (or fault) system provides useful information concerning
deformation events related to dyke intrusion.
Recently, numerous studies for dykes have been conducted to
describe dyke propagation mechanisms (e.g., Walker et al., 2017;
Healy et al., 2018), tectonic events related to intrusion (Pallister,
2010; Ghodke et al., 2018), and dyke intrusion-induced faulting
(e.g., Dering et al., 2019, Trippanera, 2019). Most of these studies
have been based on the detailed analyses of dyke geometries,
which is an essential component of kinematic and dynamic
analyses of dykes (e.g., Magee et al., 2019).
The dyke in Figure 5c shows different linking patterns depending
on the stepping direction. This asymmetric development of
linking dykes may be due to the presence of an acidic dyke and
absence of Set-b fractures between Segment-b and Segment-c.
However, left-lateral shear acting on Set-a to Set-c resulting from
the oblique opening is another possible mechanism for the
observed asymmetry. In this stress state, a mode I fracture set
may develop preferentially at the left-stepping linkage zone,
similar to the interaction between Segment-a and -b in the fault
damage zone conceptual model (Kim et al., 2004).
The dyke illustrated in Figure 7c clearly shows multiple
geometries, including a bent shape at the wall and splayed shape
at the tip. Estimation of opening direction from the bent wall
generated oblique openings in both directions at the fracture
wall and tip; in particular, left-lateral shear acting on fracture
Set-a resulted in an asymmetrical splayed tip. This tip geometry
is similar to horsetail structures at the fault tip damage zone
(Kim et al., 2004), indicating that this dyke intruded into mode I
fractures at the tip damage zone.
Geometrical classification of dyke in terms of fracture system
is helpful to the kinematic analysis of the dyke system (Fig. 8).
Some specific geometries within linking and tip damage zones
could be used as good shear sense indicators during faulting.
Some associated fracture geometries observed in dykes here
(Figs. 5c and 7c) clearly indicate shear sense along the fracture
planes. Furthermore, based on detailed geometric and kinematic
analyses of fracture (dyke) geometry, it is possible to restore the
fracture map of the pre-intrusion. The fracture map can be helpful
to the analysis of dynamics associated with dyke intrusion;
applied stress (σ
n
,
σ
s
), strain of the host rock, and scaling factor
of the displacement (mm to km) of the original fracture walls.
5.2. Future Directions for Geometric Analysis of
Dyke Intrusion Patterns
The behavior of fractures in the subsurface is an active field of
Descriptive classification of dyke morphologies 91
https://www.springer.com/journal/12303
https://doi.org/10.1007/s12303-021-0018-2
research in structural geology and fracture mechanics. Some
studies have aimed to lin k dyke emplacement to fracture me chani cs
(e.g., Jolly and Sanderson, 1995; Zhang et al., 2007). In the present
study, we collected detailed field data to investigate the geometric
nature of exposed dykes, with the aim of elucidating the
characteristics of dyke intrusions patterns and the roles of
fractures as conduits. Although dykes can be precisely observed
and analyzed in the field, integrating such field data with the
current mechanical model of dyke intrusion remains difficult
(Rivalta et al., 2015). To overcome this limitation, it is necessary
to establish a dyke propagation model based on analysis of the
dynamic shape of the dyke from a three-dimensional perspective
(Rivalta et al., 2015). However, because it is generally difficult to
observe three-dimensional dykes, accurate geometric analysis
using fractures rather than dykes can help to better explain
kinematic and dynamic processes of dyke intrusion.
Many factors can affect dyke development during shallow-
level intrusion, and such intrusions can result in various intrusion
patterns. The final shape and geometry of a dyke is determined
by the rate of magma supply, composition of the magma, host
rock properties, type of stress conditions, and tectonic setting
(e.g., Ghodke et al., 2018). Thus, further research into intrusion
patterns should be based on three-dimensional intrusion patterns
and should involve the interpretation of multiple possible
formation mechanisms.
6. CONCLUSIONS
Various dyke intrusion patterns have been analyzed and
classified in terms of fracture patterns, leading to the definition
of three main classes (array, linkage, and segment) and two sub-
classes (wall and tip) that can be used to describe the exposed
part of a dyke system or the location around a dyke. Each class is
subdivided into two types based on the straightness of specific
dyke geometries.
Variable non-straight types were reported in this study,
indicating that various non-straight conduits can be provided
during dyke intrusion by interactions between opening direction
and fracture sets (or fracture geometries). Therefore, the proper
analysis of dyke intrusion patterns in terms of fractures is necessar y
to avoid misinterpretation for dyke intrusion systems with non-
straight dyke intrusion patterns.
This newly proposed classification system will support better
systematic description of intrusions, especially for the interpretation
of the intrusion kinematics and mechanisms. Although the
classification system does not include all types of dykes. It may
prove useful for expanding knowledge regarding the relationship
between the dyke and surrounding fractures in the field.
ACKNOWLEDGMENTS
This work was supported by the Korea Institute of Geoscience
and Mineral Resources (KIGAM) Basic Research Project
“Development of precise exploration technology for energy
storage minerals (V) existing in Korea and the resources estimation
(20-3211)” funded by the Ministry of Science and ICT of Korea.
The manuscript benefited from the careful reviews of two
anonymous reviewers and the editorial staff of Geosciences
Journal. We thank Prof. D. J. Sanderson for constructive review
on an early version of this manuscript.
REFERENCES
Babiker, M. and Gudmundsson, A., 2004, Geometry, structure and
emplacement of mafic dykes in the Red Sea Hills, Sudan. Journal of
African Earth Sciences, 38, 279–292.
Baer, G., Beyth, M., and Reches, Z., 1994, Dykes emplaced into fractured
basement, Timna Igneous Complex, Israel. Journal of Geophysical
Research, 99, 24039–24050.
Bussell, M.A., 1989, A simple method for the determination of the dila-
tion direction of intrusive sheets. Journal of Structural Geology, 11,
679–687.
Chang, K.H., 1975, Cretaceous stratigraphy of southeast Korea. The Jour-
Fig. 8. Conceptual models for dyke intrusion patterns based on the fault damage zone concept. (a) Linking dyke of Figure 5d might be
related with the extensional overstep zone. (b) Split dyke of Figure 7d might be related with the extensional tip damage zone.
92 Seok-Jun Yang and Young-Seog Kim
https://doi.org/10.1007/s12303-021-0018-2
https://www.springer.com/journal/12303
nal of the Geological Society of Korea, 11, 1–23.
Chang, K.H., Suzuki, K., Park, S.-O., Ishida, K., and Uno, K., 2003,
Recent advances in the Cretaceous stratigraphy of Korea. Journal of
Asian Earth Sciences, 21, 937–948.
Chough, S.K. and Sohn, Y.K., 2010, Tectonic and sedimentary evolu-
tion of a Cretaceous continental arc-backarc system in the Korean
peninsula: new view. Earth-Science Reviews, 101, 225–249.
Clemente, C.S., Amorós E.B., and Crespo, M.G., 2007, Dyke intrusion
under shear stress: effects on magnetic and vesicle fabrics in dykes
from rift zones of Tenerife (Canary Islands). Journal of Structural
Geology, 29, 1931–1942.
Delaney, P.T., Pollard, D.D., Ziony, J.I., and McKee, E.H., 1986, Field
relations between dikes and joints: emplacement processes and
paleostress analysis. Journal of Geophysical Research Solid Earth:
Solid Earth, 91, 4920–4938. https://doi.org/10.1029/JB091iB05p04920
Dering, G.M., Micklethwaite, S., Cruden, A.R., Barnes, S.J., and Fiorentini,
M.L., 2019, Evidence for dyke-parallel shear during syn-intrusion
fracturing. Earth and Planetary Science Letters, 507, 119–130.
Edwards, P., Choi, J.-H., and Kim, Y.-S., 2017, Variations in thickness of
fault-controlled dikes and its implications: a case study of Gosung
area, South-East Korea. Island Arc, 26, e12181. https://doi.org/10.1111/
iar.12181
Ghodke, S.S., Rathna, K., Kokandakar, G.J., Nagaraju, B., More, L.B.,
Bhosle, M.V., Kumar, K.V., 2018, Emplacement and growth of alkaline
dikes: insights from the shonkinite dikes (Elchuru alkaline complex,
SE India). Journal of Structural Geology, 117, 219–236.
Goulty, N. and Schofield, N., 2008, Implications of simple flexure the-
ory for the formation of saucer-shaped sills. Journal of Structural
Geology, 30, 812–817.
Gudmundsson, A., 1995, Infrastructure and mechanics of volcanic sys-
tems in Iceland. Journal of volcanology and Geothermal Research,
64, 1–22.
Gudmundsson, A., 2005, The effects of layering and local stresses in
composite volcanoes on dyke emplacement and volcanic hazard.
Comptes Rendus Geoscience, 337, 1216–1222.
Gudmundsson, A., 2011, Deflection of dykes into sills at discontinu-
ities and magma-chamber formation. Tectonophysics, 500, 50–64.
Healy, D., Rizzo, R., Duffy, M., Farrell, N.J.C., Hole, M.J., and Muir-
head, D., 2018, Field evidence for the lateral emplacement of igne-
ous dykes: implications for 3D mechanical models and the plumbing
beneath fissure eruptions. Volcanica, 1, 85–105. https://doi.org/
10.30909/vol.01.02.85105
Hoek, J.D., 1991, A classification of dyke-fracture geometry with exam-
ples from Precambrian dyke swarms in the Vestfold Hills, Antarc-
tica. Geologische Rundschau, 80, 233–248.
Hwang, B.H., Son, M., Yang, K., Yoon, J., and Ernst, W.G., 2008, Tectonic
evolution of the Gyeongsang Basin, southeastern Korea from 140 Ma
to the present, based on a strike-slip and block-rotation tectonic
model. International Geology Review, 50, 343–363.
Jin, K., Kim, Y.-S., Yang, S.-J., Choi, J.-H., and Kim, K.-O., 2018, Defor-
mation history and characteristics of the Ilgwang Fault in south-
east Korea. Geosciences Journal, 22, 209–226.
Jolly, R.J.H. and Sanderson, D.J., 1995, Variation in the form and distri-
bution of dykes in the Mull swarm, Scotland. Journal of Structural
Geology, 17, 1543–1557.
Kamata, H., 1998, Quaternary volcanic front at the junction of the
South-west Japan Arc and the Ryukyu Arc. Journal of Asian Earth
Sciences, 16, 67–75.
Kang, H.-C. and Paik, I.S., 2013, Review on the geological ages of the
formations in the Gyeongsang Basin, Korea. Journal of the Geolog-
ical Society of Korea, 49, 17–29. (in Korean with English abstract)
Kattenhorn, S.A. and Watkeys, M.K., 1995, Blunt-ended dyke segments.
Journal of Structural Geology, 17, 1535–1542.
Kavanagh, J.L., Rogers, B.D., Boutelier, D., and Cruden, A.R., 2017,
Controls on sill and dyke-sill hybrid geometry and propagation in
the crust: the role of fracture toughness. Tectonophysics, 698, 109–120.
Khodayar, M. and Franzson, H., 2007, Fracture pattern of Thjórsárda-
lur central volcano with respect to rift-jump and a migrating trans-
form zone in South Iceland. Journal of Structural Geology, 29, 898–
912.
Kim, Y.-S., Peacock, D.C.P., and Sanderson, D.J., 2004, Fault damage
zones. Journal of Structural Geology, 26, 503–517.
Lister, J.R. and Kerr, R.C., 1991, Fluid-mechanical models of crack
propagation and their application to magma transport in dykes.
Journal of Geophysical Research, 96, 10049–10077.
Magee, C., Muirhead, J., Schofield, N., Walker, R., Galland, O., Holford,
S., Spacapan, J., Jackson, C., and McCarthy, W., 2019, Structural
signatures of igneous sheet intrusion propagation. Journal of Struc-
tural Geology, 125, 148–154.
Martínez-Poza, A.I. and Druguet, E., 2016, Structure and tectonic set-
ting of the SE Sardinia mafic dyke swarm. Insights for the stress
state during magma emplacement in the upper crust. Journal of
Geodynamics, 101, 170–185.
Martínez-Poza, A.I., Druguet, E., Castaño, L.M., and Carreras, J., 2014,
Dyke intrusion into a pre-existing joint network: the Aiguablava
lamprophyre dyke swarm (Catalan Coastal Ranges). Tectonophys-
ics, 630, 75–90.
Mathieu, L., De Vries, B.V.W., Holohan, E.P., and Troll, V.R., 2008,
Dykes, cups, saucers and sills: analogue experiments on magma intru-
sion into brittle rocks. Earth and Planetary Science Letters, 271, 1–13.
Pallister, J.S., McCausland, W.A., Jónsson, S., Lu, J., Zahran, H.M., Had-
idy, S.E., Aburukbah, A., Stewart, I.C.F., Lundgren, P.R., White,
R.A., and Moufti, M.R.H., 2010, Broad accommodation of rift-related
extension recorded by dyke intrusion in Saudi Arabia. Nature Geo-
science, 3, 705–712.
Paquet, F., Dauteuil, O., Hallot, E., and Moreau, F., 2007, Tectonics and
magma dynamics coupling in a dyke swarm of Iceland. Journal of
Structural Geology, 29, 1477–1493.
Platten, I.M., 2000, Incremental dilation of magma filled fractures: evi-
dence from dykes on the Isle of Skye, Scotland. Journal of Struc-
tural Geology, 22, 1153–1164.
Pollard, D.D., 1973, Derivation and evaluation of a mechanical model
for sheet intrusions. Tectonophysics, 19, 233–269.
Rivalta, E., Taisne, B., Bunger, A.P., and Katz, R.F., 2015, A review of
mechanical models of dike propagation: schools of thought, results
and future directions. Tectonophysics, 638, 1–42.
Rogers, R.D. and Bird, D.K., 1987, Fracture propagation associated with
dyke emplacement at the Skaergaard intrusion, East Greenland.
Journal of Structural Geology, 9, 71–86.
Ryan, B., 1995, Morphological features of multigeneration basic dykes
Descriptive classification of dyke morphologies 93
https://www.springer.com/journal/12303
https://doi.org/10.1007/s12303-021-0018-2
near Nain, Labrador: clues to original emplacement mechanisms
and subsequent deformation. Precambrian Research, 75, 91–118.
Son, M., Kim, J.-S., Hwang, B.-H., Lee, I.-H., Kim, J., Song, C.W., and
Kim, I.-S., 2007, Paleogene dyke swarms in the eastern Geoje Island,
Korea: their absolute ages and tectonic implications. The Journal of
the Petrological Society of Korea, 16, 82–99. (in Korean with English
abstract)
Spacapan, J.B., Galland, O., Leanza, H.A., and Planke, S., 2016, Control
of strike-slip fault on dyke emplacement and morphology. Journal
of the Geological Society, 173, 573–576.
Spence, D.A. and Turcotte, D.L., 1985, Magma-driven propagation of
cracks. Journal of Geophysical Research: Solid Earth, 90, 575–580.
https://doi.org/10.1029/JB090iB01p00575
Tibaldi, A. and Bonali, F.L., 2017, Intra-arc and back-arc volcano-tec-
tonics: magma pathways at Holocene Alaska-Aleutian volcanoes.
Earth-Science Reviews, 167, 1–26.
Tibaldi, A., Vezzoli, L., Pasquaré, F.A., and Rust, D., 2008, Strike-slip fault
tectonics and the emplacement of sheet-laccolith systems: The Thver-
fell case study (SW Iceland). Journal of Structural Geology, 30,
274–290.
Trippanera, D., Ruch, J., Passone, L., and Jónsson, S., 2019, Structural
mapping of dike-induced faulting in Harrat Lunayyir (Saudi Ara-
bia) by using high resolution drone imagery. Frontiers in Earth Sci-
ence, 7, 168. https://doi.org/10.3389/feart.2019.00168
Walker, R.J., Branney, M.J., and Norry, M.J., 2017, Dike propagation
and magma flow in a glassy rhyolite dike: a structural and kine-
matic analysis. Geological Society of America Bulletin, 129, 594–606.
Yang, S.-J., Jin, K., and Kim, Y.-S., 2008, Paleostress conditions based on
dyke intrusion patterns and deformation histories in Geo-je island,
SE Korea. Journal of the Geological Society of Korea, 44, 747–764.
(in Korean with English abstract)
Zhang, X., Jeffrey, R.G., and Thiercelin, M., 2007, Deflection and prop-
agation of fluid-driven fractures at frictional bedding interfaces: a
numerical investigation. Journal of Structural Geology, 29, 396–410.
Publisher’s Note Springer Nature remains neutral with regard to jurisdic-
tional claims in published maps and institutional affiliations.
