Anda di halaman 1dari 51

Journal of Volcanology and Geothermal Research 298 (2015) 85135

Contents lists available at ScienceDirect

Journal of Volcanology and Geothermal Research


journal homepage: www.elsevier.com/locate/jvolgeores

Review

Structure of volcano plumbing systems: A review of


multi-parametric effects
Alessandro Tibaldi
Department of Earth and Environmental Sciences, University of Milan-Bicocca, Piazza della Scienza 4, 20126 Milan, Italy

a r t i c l e

i n f o

Article history:
Received 31 December 2014
Accepted 21 March 2015
Available online 11 April 2015
Keywords:
Plumbing system
Dyke
Sill
Volcano
Stress
Magma chamber

a b s t r a c t
Magma is transported and stored in the crust mostly through networks of planar structures (intrusive sheets),
ranging from vertical dykes to inclined sheets and horizontal sills, and magma chambers, which make up the
plumbing system of volcanoes. This study presents an overview of plumbing systems imaged at different depths
and geodynamic settings, in order to contribute to assessing the factors that control their geometry. Data were
derived from personal eld surveys and through the analysis of publications; observations include local lithology
and tectonics of the host rock with special reference to local fault kinematics and related stress tensor, regional
tectonics (general kinematics and far-eld stress tensors), geology and shape of the volcano, topographic settings, and structural and petrochemical characteristics of the plumbing system. Information from active volcanoes and eroded extinct volcanoes is discussed; the shallow plumbing system of active volcanoes has been
reconstructed by combining available geophysical data with eld information derived from outcropping sheets,
morphometric analyses of pyroclastic cones, and the orientation and location of eruptive ssures. The study of
eroded volcanoes enabled to assess the plumbing system geometry at deeper levels in the core of the edice
or underneath the volcano-substratum interface. Key sites are presented in extensional, transcurrent and contractional tectonic settings from North and South-America, Iceland, the Southern Tyrrhenian Sea and Africa.
The types of sheet arrangements illustrated include swarms of parallel dykes, diverging rift patterns, centrallyinclined sheets, ring and radial dykes, circum-lateral collapse sheets, sills, and mixed members. This review
shows that intrusive sheet emplacement at a volcano depends upon the combination of several local and regional
factors, some of which are difcult to be constrained. While much progress has been made, it is still very challenging to forecast the likely paths and geometry of sheet propagation and emplacement during volcanic unrest
events.
2015 Elsevier B.V. All rights reserved.

Contents
1.
2.
3.
4.

5.
6.

7.
8.
9.

Introduction . . . . . . . . . . . . . . . . . . . . . . . .
Denition of a magma plumbing system . . . . . . . . . . .
Propagation and arrest of intrusive sheets . . . . . . . . . .
Deep plumbing system . . . . . . . . . . . . . . . . . . .
4.1.
Felsic magmas . . . . . . . . . . . . . . . . . . .
4.2.
Mac magmas . . . . . . . . . . . . . . . . . . .
Shallow plumbing system . . . . . . . . . . . . . . . . .
Plumbing system in the interior of volcanoes . . . . . . . . .
6.1.
Tectonic vs. magmatic components . . . . . . . . . .
6.2.
Intrusive sheets and orientation of tectonic stress tensor
6.2.1.
Vertical 1 . . . . . . . . . . . . . . . . .
6.2.2.
Vertical 3 . . . . . . . . . . . . . . . . .
6.2.3.
Vertical 2 . . . . . . . . . . . . . . . . .
6.3.
Topographic inuence . . . . . . . . . . . . . . . .
Plumbing systems and surface deformation . . . . . . . . . .
Plumbing systems at calderas . . . . . . . . . . . . . . . .
Conclusions . . . . . . . . . . . . . . . . . . . . . . . .

http://dx.doi.org/10.1016/j.jvolgeores.2015.03.023
0377-0273/ 2015 Elsevier B.V. All rights reserved.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.

86
87
87
91
91
92
95
107
108
110
110
110
113
114
119
120
126

86

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction
This paper addresses the fast-growing eld of studies on the structure of magma plumbing systems below and inside volcanoes, the
sites where processes like magma transfer, storage and evolution
occur. A magma plumbing system can be dened as a network of conduits along which magma moves, interconnected with chambers
where magma accumulates. A plumbing system may be connected
to the surface, at least during transient times, to which magma is
transported producing eruptions. During transfer from the deep uid
source and subsequent storage in the crust, magmas are subject to a series of processes that lead to their differentiation. From the shallow
chambers to the surface, magmas are subject to further processes that
eventually dictate the type, intensity and duration of eruption. All
these processes are inuenced by a series of parameters that include,
for example, depths of differentiation, amount of wall-rock assimilation,
rates and timescales of magma generation, and times of storage (Annen
and Zellmer, 2008). Magma storage and ascent are above all tightly
linked to the structure and state of stress of the crust (Chaussard
and Amelung, 2014), but in turn magma intrusions might control
plate boundary evolution (Acocella, 2014). As a consequence, the reconstruction of the structure and geometry of the plumbing system is of
paramount importance for understanding how the subvolcanic engine
works.
The assessment of volcanic hazard is also dependant on the comprehension of the structure of magma plumbing systems. Processes acting
at open conduit volcanoes, and leading to paroxistic explosions, have
been recently addressed taking into consideration the structure of the
shallower conduits. Chouet et al. (1997), for example, studied the
wave elds of tremors and explosions at Stromboli Volcano, Italy, demonstrating that the source of this phenomenon is localised beneath
the summit crater in the shallower part of the plumbing system at
depths b 200 m. The model for degassing and explosion occurrence is
consistent with a vertical, NESW-striking crack-like conduit. This geometry fully coincides with the eld observations carried out at
Stromboli's plumbing system, which crops out in the more dissected
parts of the volcano (Pasquar et al., 1993; Tibaldi, 1996, 2001). Later,
more detailed geophysical studies on the active conduit (Chouet et al.,
2008) and eld data on Holocene conduits (Tibaldi, 2003; Corazzato
et al., 2008) put further constraints on conduit geometry suggesting a
dip towards NE.
Also the evaluation of the areas most prone to the opening of new
vents and eruptive ssures is intimately linked to the understanding
of the structure of plumbing systems, especially at volcano-tectonic
rift zones on volcano slopes (e.g. Bonali et al., 2011). Mac magma, in
fact, is normally supplied to the surface along planar and mostly
steeply-dipping intrusive sheets that may group to form dyke swarms
(Dieterich, 1988; Carracedo, 1994; Moore et al., 1994; Walter and
Schmincke, 2002), and eventually volcano-tectonic rift zones formed
by hundreds of such parallel dykes (Fiske and Jackson, 1972; Walker,
1999). These rift zones can be studied at the surface by analysing the
orientation and location of eruptive ssures, vents and the morphometric characteristics of pyroclastic cones (Tibaldi, 1995), as well as by
interferometric methods (Massonnet and Sigmundsson, 2000). At
the same time, much information can be obtained by studying in the
eld the eroded parts of the rift zones where sheet intrusions are exposed. Field studies of thousands of sheet intrusions also show that
most sheets become arrested on their way to the surface, and that unrest commonly does not lead to an eruption (Fig. 1) (Gudmundsson
and Brenner, 2005). The comprehension of this phenomenon depends
on a better understanding of the physical conditions for the injection

128
128

of sheet intrusions from the source magma chamber and their propagation to the surface, which again requires knowledge of the structure of
the plumbing system and of the host rock.
Classically, studies of plumbing systems focused on igneous processes that are elucidated by sampling of exposed rocks and laboratory analyses, more recently integrated with experimental approaches. The new
eld of studies of volcanotectonics devoted to reconstructing the
structure of plumbing systems uses eld data, analogue modelling and
numerical modelling. The experimental approaches are rapidly growing
in number and quality, but it is fundamental to anchor and validate their
results with eld truth. In recent years, exposed plumbing systems have
been studied in sufcient detail with modern techniques of structural
geology in order to get information aimed at a better understanding of
the entire process from crustal magma storage to eruption at the
surface.
The present paper contributes to the knowledge of magma plumbing systems by reviewing the most relevant literature and integrating
it with eld data mostly collected by the author. The focus is on the analysis of the structure of plumbing systems in the uppermost crust and inside volcanoes, where more data obtained from eld evidence and
geophysics are available. This is integrated with a summary of the literature on the deeper part of plumbing system above the melt generation
zone. These data provide a backdrop for understanding the entire
magma plumbing system and the processes that take place within it.
Through a series of examples, from deeply eroded volcanoes to the surface of active volcanoes, I describe the various parameters that control
the geometry of plumbing systems; these in fact are sensitive to multiple factors that frequently work together to dictate the nal conguration of the conduit array. Intact cones and volcanoes that experienced
lateral failure or caldera collapse are taken into consideration.
One of the most complex conditioning factors is represented by the
tectonic settings; the tectonic inuence on magma migration and volcanism has received a lot of attention recently, but several issues are still
open and controversial. As an example, for decades volcanism and regional extensional tectonics have been thought to be tightly linked, as

Fig. 1. A magma plumbing system is represented as a network of vertical, inclined and horizontal conduits that channel magma towards the surface, and a series of chambers where
magma can be stored. Main nomenclature is shown. Not to scale.

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

this type of stress state favours magma upwelling along vertical fractures perpendicular to the regional least principal stress (3) that is horizontal (Anderson, 1951; Cas and Wright, 1987; Watanabe et al., 1999).
For arc volcanism occurring at convergent margins, Nakamura (1977)
stated that the overall tectonics of the arcs should be strike-slip (with
3 and greatest principal stress, 1, both horizontal) instead of compressional reverse (3 vertical). Strike-slip tectonics would allow magma to
ascend through vertical dykes parallel to the direction of 1 (Nakamura
and Uyeda, 1980). By contrast, a pure contractional tectonic environment, with reverse or transpressional faulting, is usually regarded as a
highly unfavourable setting for volcanism (Glazner, 1991; Hamilton,
1995; Watanabe et al., 1999), where only intrusive emplacement is expected (Cas and Wright, 1987), although in reality local magma stress
can give rise to eruption also in this setting. In the present paper I investigate also these issues through examples that encompass different
geodynamic settings including compressional and extensional regimes
with normal, strike-slip and reverse fault kinematics. Plumbing systems
form a major component of diverging plate margins and orogenic belts
at converging margins, and can be highly varied both spatially and temporally, which has led to a number of controversies about their architecture and evolution.
The paper is structured in order to present a review and analysis of
many existing data and models, by introducing magma plumbing systems and their settings, and the possible inuencing parameters at different depths. First we will investigate the deeper level from the uid
source zone, then the upper crust zone, and nally the interior of volcanoes and their surface.
2. Denition of a magma plumbing system
Volcanoes are underlain by magmatic plumbing systems that transport magma from the Earth's mantle and crust towards the surface. A
large fraction of this magma never reaches the surface and becomes solidied in the Earth's interior as magmatic intrusions, providing clues to
the past geometry of magma pathways. As globally assumed in the
Earth Sciences, the behaviour of a geological process in the past can be
a key to forecast the future development of the same process. Thus
the reconstruction of the recent plumbing system below a volcano can
help to assess the present magma paths at an active volcano, but, as
we will see later on, plumbing systems are extremely dynamic and
changes can occur at a fast rate.
A plumbing system can be represented as a network of vertical or inclined conduits that channel magma towards the surface, and sills and
chambers where magma can be stored (Fig. 1). At a certain time, one
or more of these members can be partially or totally molten. Further
magma injection can be supplied from a deeper source in the mantle,
or from one or more active magma chambers in the crust. Magma chambers located at different levels in the plumbing system can act as a sink
for magma from the deeper reservoir, and as a source for magma injections into the surrounding crust (Gudmundsson, 2012).
Magma conduits have been represented for decades with a cylindrical shape and a circular section. However, direct observations of
eruptions along fractures, ground deformation measures, and the distribution of seismicity associated with magma intrusions provide evidence
for the role of dykes in mac, intermediate and felsic magma transport
(Pollard et al., 1983; Rubin and Pollard, 1987; Peltier et al., 2005;
Yamaoka et al., 2005; Aloisi et al., 2006; Mattia et al., 2007). Conduits
have also been studied by eld evidence in eroded volcanoes where tabular sheets compose the bulk of the plumbing system (Gudmundsson,
1987, 1988, 1990; Tibaldi, 2001; Gudmundsson, 2002; Corazzato et al.,
2006; Pasquar and Tibaldi, 2007; Tibaldi et al., 2008a,b,c, 2009). In
the case of andesitic volcanism, some authors still suggest that eruptions are more focused and the conduits are more cylindrical (Zellmer
and Annen, 2008), as explained through progressive melting of host
rocks (Quareni et al., 2001) and in terms of the sharp increase in
magma viscosity close to the surface due to decompression, degassing

87

and crystallisation. At caldera structures also felsic magmas are normally emplaced as ring dykes, although this concept has been argued by
Legros et al. (2000) who suggest that magma ascent at caldera ringssures is less favoured than at more restricted conduits.
These data indicate that the most of the magma is generally
transported towards the surface through conduits that can be represented by tabular intrusions (sheets) with planar, curved, en-chelon,
or more complex geometries, but that invariably have a very high length
(and depth) to thickness ratio. A magmatic sheet is thus a crystalline
rock that formed in a crack, primarily caused by the magma itself, in a
pre-existing rock body. In the eld, intrusive sheets are classically divided into three end-members based on their attitude and the relations
with the country rocks: vertical to steep-dipping dykes, inclined sheets,
and sills, formed when magma is injected between rock layers forming a
horizontal or gently-dipping sheet. In origin, dykes have been described
as tabular bodies that intrude normally or obliquely to the bedding
plane of the host rock, to be called also discordant intrusion, while
the term sill indicated a sub-parallel or concordant intrusion
(Billings, 1972; Best, 1982; Hall, 1987). Other authors, like Bates and
Jackson (1987) adopted a terminology consistent with the intrusion attitude and referred to dykes as vertical tabular intrusions and to sills as
horizontal ones. Hatayama et al. (1980) dene a dyke as a vertical intrusion, a sheet as a horizontal intrusion, and a sill as a horizontal and concordant intrusion. The denition of these parts of a plumbing system as
based on their relations with the bedding of the host rock is misleading,
also because it implies that dykes have to produce cracks in order to
propagate across the bedding, whereas sill intrusion is easier along
pre-existent weakness zones like bedding discontinuities. Although
some authors consider that magma lling already existing fractures is
an important dyke intrusion mechanism (e.g. Delaney et al., 1986;
Bear et al., 1994; Delaney and Gartner, 1997; Valentine and Krogh,
2006), it is important to note that not always do sills follow preexisting discontinuities and not always do dykes propagate along selfgenerated fractures. I here encourage to use a terminology that does
not bear reference to the relations with fractures and is based on inherent geometric attributes: dykes are those tabular intrusions that have
a dip 76, inclined sheets have a dip N 10 and b 76, and sills have a
dip 10. This terminology is essentially descriptive of the present sheet
attitude and does not take into account their possible tilting due to postintrusion deformation. Post-sheet regional tilting must be retrodeformed in order to get the original sheet architecture.
Magma chambers can be totally or partially molten, and in the latter
case a large part may consist of a crystal mush that behaves as poroelastic
(Maale and Scheie, 1982; McKenzie, 1984; Gudmundsson, 1987; Marsh,
1989; Sinton and Detrick, 1992; Marsh, 2000). A totally or partially molten magma chamber can be recognised by geophysical methods due to
its size that usually is much larger than sheet intrusions and to its physical properties that differ from those of the host rock. Also, a completely
solidied, buried magma chamber usually has different properties than
the host rocks, which enable to recognise it. Solidied magma chambers
can have dimension spanning from tens of km3 (plutons) to small intrusions with size b 1 km3, which have been named differently in the past, in
relation to their shape. This can be from spherical to strongly ellipsoid or
tabular, and this geometric feature is very important since the shape of an
active magma chamber dictates the orientation of the stress exerted on
the host rock by magma overpressure (Gudmundsson, 2006, 2012).
The more complex plumbing systems can be composed of a plexus of interconnected sills, inclined sheets, dykes and multiple magma chambers
(Hildreth, 1981; Lahr et al., 1994; Donoghue et al., 1995; La Delfa et al.,
2001; Preston, 2001; Dawson et al., 2004; Marsh, 2004; Sanchez et al.,
2004; Cartwright and Hansen, 2006).
3. Propagation and arrest of intrusive sheets
Magma propagation occurs along planes of weakness that are already present in the host rock, or along hydrofractures generated by

88

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

internal uid pressure. Hydrofractures are primarily extension fractures


whose propagation can be modelled as mode I cracks (extension normal
to the fracture wall) (Bahat, 1980; Gautneb et al., 1989; Gudmundsson
et al., 2001). Many studies have been devoted to understanding
hydrofractures in general and signicant progress has been made on
dyke propagation in particular. Earlier papers proposed that a sufciently large buoyant force can cause fractures to propagate upwards
(Weertman, 1971; Secor and Pollard, 1975), whereas other studies did
not consider buoyancy (Spence and Sharp, 1985). From numerical
modelling, Spence et al. (1987) and Lister (1990) obtained a solution
of fracture surface displacement that results in the formation of a dyke
with a bulbous head and a constant width tail. Lister and Kerr (1991)
showed that magma moves along dykes under a local balance between
buoyancy forces and viscous pressure drop. Rubin (1993) and Chen
et al. (2007) modelled dyke propagation mainly based on a given source
pressure (see also Rubin, 1995a and references therein, and Roper and
Lister, 2005). Buoyancy and source pressure have been modelled by
Meriaux and Jaupart (1998) and Bonafede and Rivalta (1999). Analogue
modelling on magma-fracture propagation has been carried out for example by Maale (1987), Takada (1990), Menand and Tait (2002), Ito
and Martel (2002) and Tibaldi et al. (2014).
These papers suggest that buoyancy forces or magma overpressure
can contribute to the propagation of hydrofractures to the surface,

although in most cases the fracture and the intrusive sheet are arrested
inside the rock succession. The resulting magma pressure induces stress
concentration at the tip that, in turn, creates the fracture. Numerical
models and eld data suggest that the magma front should lag behind
the hydrofracture tip during its propagation, and thus there should be
a dry part of the fracture (Warpinski, 1985; Bonafede and Olivieri,
1995; Garagash and Detournay, 2000). Sometimes in the eld it is possible to observe the fracture at the sheet tip that still hosts magma with
a considerably minor thickness, in the order of a few mmcm (e.g.
Fig. 2A). Ahead of the tip, a hydrofracture opens if it has a suitable orientation with respect to the surrounding stress eld in the host rock, and if
the tensile stresses exceed the rock strength. If the rock succession is not
homogeneous and contains discontinuities and layers with different
mechanical properties, the tensile stress generated at the fracture tip
cannot be large enough to overcome the resistance to fracture propagation and the sheet can become arrested (e.g. Fig. 2B). Intrusive sheets
can stop when they meet: (1) discontinuities; (2) stress barriers; or
(3) rock layers with strongly contrasting Young's moduli. The opening
of discontinuities (1) with a different orientation from the propagating
sheet, can dissipate stresses; when a horizontal discontinuity, like a
bedding plane, is encountered at the contact ahead of a hydrofracture
tip, the hydrofracture may be unable to propagate through the horizontal, open discontinuity (Gudmundsson and Brenner, 2001, 2005;

Fig. 2. A. Photo (left) and sketch (right) of a sheet showing a sudden decrease in thickness in correspondence of the hydrofracture propagated at its tip and initial intrusion along the fracture; note that only a few centimetres of the fracture's length are dry (Skye Island, UK). B. Photo (left) and sketch (right) of a dyke arrested at the contact with a discontinuity (bedding) and
a layer with different mechanical properties (a thick lava ow) (Reunion Island).

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

Gudmundsson, 2002). (2) Stress barriers are layers with local stresses
unfavourable for intrusion propagation, such as layers that contain
stresses in excess of up to 510 MPa with respect to the adjacent
rocks (Gretener, 1969; Gudmundsson, 1986, 1990; Parsons et al.,
1992). Regarding point (3), ascending dykes can stop at the contact of
bedding layers with different mechanical properties, such as in the example of Fig. 2B where a dyke is intruded into a succession of thin
(b20 cm) lava ows and dominant breccia layers and then stops at

89

the contact with the thick (~ 2 m) lava ow. Experiments also show
that soft layers can be more effective in arresting hydraulic fractures
than stiff layers (Charlez, 1997; Yew, 1997), as can be seen in the example of Fig. 3A where an inclined sheet gradually splits into minor ngers
and nally becomes arrested within a soft layer media made of poorly
consolidated breccia deposits.
The termination of an intrusive sheet can have different shapes that
represent the diverse settings for the stoping of the fracture propagation

Fig. 3. Photos (left side) and sketches (right side) of: A. an inclined sheet that gradually splits into minor ngers and nally becomes arrested within poorly consolidated breccia deposits
(northwest coast, Sao Miguel, Azores Archipelago), since soft layers can be more effective at arresting hydraulic fractures than stiff layers (Charlez, 1997; Yew, 1997); B. a gently-dipping
sheet that terminates with a series of secondary intrusions (asymmetric ngers) that depart upward (Stromboli volcano, Italy); and C. an arrested steeply-dipping sheet with an upward
symmetric fanning arrangement (Ross Lake, Washington, USA).

90

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

(Gudmundsson, 2003). The sheet can maintain a unique main plane


with a bulbous tip shape, or it can show sets of ngers that depart upward (Fig. 3B) or in a fanning arrangement (Fig. 3C). The propagation
of ngers uses more mechanical energy on the processes of ow, dilation, and deformation than the propagation by a continuous sheet geometry (Pollard et al., 1975) and thus can explain the arrest. Finger
formation may occur when local inhomogeneities in the rock mass
along the wall of the sheet tip are located ahead of the bulk of the intrusion, and the magma may accelerate at the inhomogeneity as it nds
less resistance there. Alternatively, ngers can form where there are
perturbations in the cooling distribution in the magma sheet and
lobes with higher temperatures advance as the others have a greater
viscosity (Wylie et al., 1999). A sheet can also change its geometry
into an en-chelon pattern expressed by individual ngers that
are oblique, in section view, with respect to the main sheet plane
(Fig. 4A). Sheet en-chelon segmentation commonly occurs in response
to stress eld rotation at an angle to the principal stress directions, frequently found approaching the surface, as also experimentally obtained
by Tibaldi et al. (2014) (Fig. 4B). En-chelon sheet ngers can show systematic step-overs between segments owing to tangential stress on
the sheet walls (Pollard, 1987). Segmentation may also reect dyke

propagation through heterogeneous geological media with local deviations of the stress eld, and in this case the segments' step direction is
non-systematic (Daniels et al., 2012). In layered host rocks, overlapping
sheet segments may tend to link together by folding and thinning of the
rock between the overlapping segments (bridge) and by thickening of
the sheet, by a propagating crack that is usually straight. In isotropic
host rocks, the termination of the single sheet segment tends to bend towards the adjacent segment, breaking the bridge (Fig. 4C); this occurs
because there is a mechanical interaction between the adjacent crack
tips, with the stress eld at one tip altering the stress of the adjacent
propagating tip by inducing curved cracks (Nicholson and Pollard,
1985). At overlapping sheet segments, a local temperature decrease at
one sheet segment might hinder further magma propagation here and
coalescence at the adjoining segment, with the possible effect of dispersion of the magma overpressure and stoping of the intrusion propagation. However, if magma ux is high enough, the original bridge can
be completely cut across, resulting in the two original sheet segments
now linked (Fig. 4D) (bridge xenoliths of Rickwood, 1990). Ination
can proceed until a complete coalescence has been reached and the continuous planar sheet conguration has been restored (Hutton, 2009;
Schoeld et al., 2012).

Fig. 4. A. Photo of en-chelon ngers in section view departing from a single intrusive sheet (Skye Island, UK). B. Photo of an analogue model, physically scaled, of a sheet intrusion showing
the upward protrusion of en-chelon ngers approaching the free topographic surface; white and black boxes have 1 cm side. C. Photo (left side) and sketch (right side) of two partially
overlapping sheet segments whose tips bend towards the adjacent segment, typical of isotropic host rocks (Skye Island, UK). D. Photo (left side) and sketch (right side) of a bridge xenoliths
isolated by two partially overlapping sheet segments linked together (Skye Island, UK).
B. After Tibaldi et al. (2014).

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

4. Deep plumbing system


Information on the deeper part of magma plumbing systems mostly
comes from geophysical and petrological research. Outcrops of
the deeper portions of plumbing systems are extremely rare and
are not representative of the actual root zone, where melts start to ascend. Much research has been carried out on both felsic and mac
root zones, resulting in different models also depending on magma
geochemistry.
4.1. Felsic magmas
In granitic systems, a classical view is that percentages within the
range 3050% of melt fraction may allow for the mobilisation of zones
hundreds of metres in lateral extent (Wickham, 1987). For example,
in the Trois Seigneurs massif in the Pyrenees (France), metamorphic
rocks pass gradationally into migmatitic biotitesillimanite gneisses
and then into a biotite granitequartz diorite body (Wickham, 1987).
Within the metamorphic sequence there are ponds of granite with diameters ranging from l m to l km across. Melting at 670700 C, 3.0
4.0 kbar under water-rich conditions was followed by segregation and
movement of melt into fractures. When large enough, these bodies intruded the overlying metasediments, but were soon halted as they became water-saturated. The zone of granite ponds corresponds to the
zone of partial melting, segregation and collection (a) in the division
of plumbing systems of Atherton (1993), above which there are the
transport zone (b) and the emplacement zone (c), where freezing occurs (Fig. 5). In zone (a), with increasing melt proportion there is a
change of the mechanical properties of the rock melt that changes
from linearly-elastic to elasticplastic and nally to viscous; this is
reected in the ow behaviour that therefore will vary with melt fraction, and thus in the formation of the deeper part of plumbing systems
(Shaw, 1965; Murase and McBirney, 1973; Shaw, 1980; Wickham,
1987). In the case of low melt content, small batches of magma will be
able to rise and will soon stop. In the case of large melt fraction, such
as for example in the case of high thermal gradients (80100 C/km),
large magma batches can ascend giving rise to huge plutons, especially
if associated with contemporaneous rifting or local extension related to
pull-apart structures or releasing bends within transcurrent shear zones
and the consequent horizontal regional ductile extension (Atherton,
1993). Similarly, also Zellmer (2008, 2009) based on global correlations
between eruptive style, surface heat ux and convergence rates at different volcanic arcs, infers that the rate of melt production in the mantle
wedge ultimately controls the deep dynamics of magma transfer.
From zone (a) of melt production, uids should nd their pathway
to move upwards through zone (b). Models of uid transport in the lithosphere propose diapiric ascent or uid ow through networks of

Fig. 5. Subdivision of a granitic plumbing system modied after Atherton (1993) for Cordilleran setting, with: (a) zone of partial melting, segregation and collection; (b) transport
zone; and (c) emplacement zone where freezing occurs.

91

interconnected static microfractures (porosity networks). For the


lower crust, magma diapirism is thought to be a common mechanism
of magma transport (Whitehead and Luther, 1975; White and
Chappell, 1977; Pitcher, 1979; Hildreth, 1981; Marsh, 1982). In the
deeper section of plumbing systems, in fact, it is considered that the
thermal and pressure conditions are such as to allow for the ductile
ow of the host rock around the rising magma mass, whereas at
upper levels (middle to upper crust) this is regarded as thermally and
mechanically unrealistic. Moreover, the diapiric ascent model is also
controversial since numerical modelling indicates that magma solidication should occur due to the slow ascent (Clemens and Mawer,
1992). The model of ow through porosity networks is accepted, for example, in sediments at low conning pressures, but the model is
questioned for geopressured zones since it cannot explain all aspects
of the observed large-scale uid ow (Nunn, 1996). The third model
of propagation relies on uid-lled fractures and is considered valid
throughout the lithosphere. If diapirs rise in the lower crust, their upward pressure can induce fracturing in the roof rocks, where uidlled fractures can propagate much faster than diapirs, thus overcoming
the problem of time-dependant solidication.
Tectonics plays a major role not only in contributing to the creation
of heat ux conditions, but also in governing the weakness zones that
may favour magma upwelling. Several examples, mainly based on
eld data, indicate that melt upwelling in the deeper parts of plumbing
systems was channelled along transcurrent zones (Mitjavila et al., 1997;
Olivier et al., 1999; Rossetti et al., 2000; Rosenberg, 2004; Weinberg
et al., 2004; Marcotte et al., 2005). This is mainly related to two facts:
(1) transcurrent faults are vertical shear zones that extend downward
to lower crustal levels, whereas reverse and thrust faults and normal
faults are less steep and tend to acquire a horizontal geometry with
depth (Fig. 6) (Davis et al., 1996) and (2) magma can enter ductile extensional jogs within shear zones as envisaged for example by Aydin
et al. (1990), D'Lemos et al. (1992) and Chiarabba et al. (2004a). Recently it has been suggested that even transpressional tectonics can be an efcient mechanism for moving magma through the lithosphere (Saint
Blanquat et al., 1998), although magma upwelling under transpression
might result in the movement of only a small volume of magma to the
surface (Marcotte et al., 2005), and nally an in-depth review by
Tibaldi et al. (2010) shows several examples of volcanism along
transpressional, transcurrent and transtensional fault zones. For felsic
magmas it is also interesting to note a close temporal relation between
peak anatexis and regional strike-slip displacements, suggesting that
strike-slip motions may initiate along zones that were thermally softened by anatexis and that regional tectonics and granite formation
may be inextricably linked processes (D'Lemos et al., 1992). Based on
this, the shape of granitic intrusions should be represented by elongated
vertical bodies with a horizontal major axis from parallel to sub-parallel
to the strike of a wrench zone (e.g. Farris et al., 2006). Notwithstanding
this, some authors indicate that granitic plutons are mostly thin with
aspect ratios (thickness to width) up to 1:15 and a at shape (e.g.
Atherton, 1993). Geophysical and structural studies showed that some
plutons are sill-like, low aspect-ratio, tabular bodies (Bridgwater et al.,
1974; Lefort, 1981; Cruden, 1998; Petford et al., 2000), whereas
magma chambers and plutons are often represented as spherical bodies
(e.g. Pitcher, 1979).
The importance of major, deep crustal fracturing in triggering felsic
magma melting and assisting magma upwelling has been put forward
by many authors (e.g. Rittmann, 1962). They suggest that the deep
plumbing system of granitic magmas is dominated by intrusions that
frequently use vertical and horizontal extensional features linked to
strike-slip faulting as in the British Tertiary Province and Irish
Caledonides (Leake, 1990), at the Coastal Batholith of Peru (Pitcher,
1978), at the Main Donegal granite (Hutton, 1982), at the Peruvian
Cordillera Batholith (Petford and Atherton, 1992), in Scotland
(Watson, 1984), in southern Greenland (Harrison et al., 1990), and at
the Boulder and Tobacco Root Batholiths (Montana, USA, Schmidt

92

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

Fig. 6. Deep magma plumbing system is conditioned also by the geometry of tectonic weakness zones that can channel magma ow. Transcurrent faults are vertical shear zones that extend
downward at lower crustal levels and can capture ascending magmas. On the contrary, reverse faults with their typical ramp and at geometry tend to stop dykes or divert dykes into sills.
Finally, normal faults are shallower-dipping than strike-slip faults and tend to acquire a horizontal geometry with depth where they exert a lower inuence on ascending magmas.

et al., 1990). Most authors thus agree that zone (b) (Fig. 5) of plumbing
system of granitic magmas should be characterised by sheet-like complexes of intrusions that give rise to the conduit system below plutons,
the latter also being controlled by shear zones. This corresponds to the
mechanism of transport and emplacement in the middle crust.
In zone c of Atherton (1993) storage processes occur and most
magma is emplaced in the form of plutons. Storage includes a series of
mechanisms that can be summarised by two end-members, although
it is reasonable to consider that a spectrum of intermediate possibilities
exists. One end-member is represented by a plutonic body that results
from a single, connected and fairly closed batch of magma that freezes
while rising through the crust. This mechanism can also be viewed as
a diapir that has been disconnected from its source (Paterson et al.,
2010). The other end-member is represented by a plutonic body that
is a frozen part of a former complex plumbing system that remains active for an extended duration. In this case the plumbing system is
huge, evolves over time, is used during several magma upwelling
events, and remobilises material from older pulses or from host rock
(Paterson et al., 2010). Pluton growth can be further complicated
by other processes such as: magma pulses moving back down the
magma pathway during rise of other pulses; loss of magma from the
plutonic system by volcanic eruptions; reheating of crystal mush
zones and consequent continuous movements in a magma conduit; internal differentiation processes. A long-lasting scientic discussion revolved around the question if magmas more commonly ascend by
continuous upwelling or by pulse-like ow. Abundant eld evidence
suggests that the latter mechanism might be more common than previously thought; several sheeted dyke and sill complexes indicate that
plutonic bodies have been constructed through magma pulsing
(Pitcher and Berger, 1972; Hardee, 1982; Hutton, 1982; Lagarde et al.,
1990; Paterson and Vernon, 1995; McNulty et al., 1996; Vigneresse
and Bouchez, 1997; Paterson and Miller, 1998; Wiebe and Collins,
1998; Johnson et al., 1999; Miller and Paterson, 2001; Miller and
Miller, 2002; Memeti et al., 2010).
In the middle-upper crust, dyking is hence considered as the most
consistent magma transfer mechanism (Clemens and Mawer, 1992;
Clemens et al., 1997; Petford et al., 2000). Granite intrusion along ductile
extensional shear zones is considered the most suitable process for upwelling and storage and this solves the room problem posed by pluton
emplacement by diapirism (Hutton et al., 1990). A model that instead
extends diapirism also to the middle crust is the one proposed by
Miller and Paterson (1999); these authors suggested that felsic
magma can rise and be emplaced as visco-elastic diapirs through a series of multiple magma batches within regional tectonic deformation,
accompanied by downward movements of the host rocks through multiple processes including brittle deformation and stoping (Cabello et al.,
2006; Farris et al., 2006; Zak and Paterson, 2006). Recently, diapirism
has been extended up to the middleupper crust by del Potro et al.
(2013) who, based on the inversion of high-resolution gravity data

beneath the Bolivia Altiplano, suggest that partially molten granitic bodies ascend by way of a diapiric mechanism through the hot ductile crust.
4.2. Mac magmas
In the case of mac magma, the deep plumbing system is considered
to be governed by a higher mobility than in the case of a felsic magma at
parity of other conditions. Mac magma is transported through cracks
in the lithosphere that give rise to swarms of dykes. These dykes can
be connected with deep to shallow magma chambers, or they may directly convey magma from the mantle towards the surface without a
shallow reservoir, as derived for example from the very high magma
supply rate of 0.7 km3/yr during the 1959 Kilauea Iki eruption
(Takada, 1999) respect to the values of 0.02 to 0.18 km3/yr during
19521983 (Dvorak and Dzurisin, 1993). Nevertheless, as for felsic
magmas, mac magmas mostly also stop rising in the crust. Although
direct observations of exhumed deep intrusions are rare, some outcrops
indicate that mac magma bodies emplaced within or at the base of the
lower crust, result from the accretion of successive magma pulses. For
example, a deep section of a mac plumbing system has been exhumed
after orogenic uplift and erosion at the Ivrea mac body in the Italian
Alps, where the evidence of individual pulses of magma can be directly
observed (Quick et al., 1994).
An explanation for the deep magma accumulation along a horizontal
zone was initially proposed by Ryan (1987) and Lister and Kerr (1991)
at the level of neutral buoyancy (LNB). Here it was retained that magma
density equals the density of the host rock and buoyancy becomes null,
favouring sill emplacement. This hypothesis was revised in more recent
years based on the fact that several dykes cut through the LNB and even
through rock successions with a lower density than magma. Sills in fact
can emplace due to other mechanical and physical conditions that will
be discussed later. If the magma supply rate is low, sills can freeze,
whereas if the magma supply rate is high, they may combine and result
in the formation of a magma chamber (Gudmundsson, 1990, 1995).
Once formed, the magma chamber can further stop other ascending
dykes, thus growing in size. Below continental crust, mac magmas
can be trapped near the Moho or within the lower crust. Below oceanic
island volcanoes, it has been suggested that basaltic to gabbroid rocks
are denser than basaltic magmas and thus LNB conditions are not
reached within the lower crust. In this condition, a LNB can be present
a few kilometres below the volcano or even inside the volcanic edice,
giving rise to very shallow magma chambers (Klgel et al., 2005).
Deep tabular horizontal sills have also been individuated by means of
seismic proles: layering in the lower crust has been interpreted as
due to the presence of sills (Fuchs et al., 1987; Wenzel and Brun,
1991; Franke, 1992). Exposure of very deep lithospheric section at the
Oman Ophiolite shows that magma-induced fracturing and intrusion
propagation are common processes that occur even at the Moho level
(Nicolas et al., 1994).

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

The occurrence of mac intrusions in the lower crust has also been
modelled by numerical simulations by several authors (Petford and
Gallagher, 2001; Annen and Sparks, 2002; Dufek and Bergantz, 2005;
Annen et al., 2006a), who showed that if the composition of the crust
in contact with the mac intrusions is amphibolitic, partial melting of
the crust is very limited. This on one side implies that large quantities
of silicic melts can be generated by incomplete crystallisation of the intruded mac magma (Dufek and Bergantz, 2005; Annen et al., 2006b),
and on the other side that a deeper plumbing system should preserve
its original architecture given by tabular intrusions, because the contacts with the host rocks are subject to limited modications. Dahm
(2000) modelled the possible deep plumbing system at convergent
margins. He found that uid-lled fractures can propagate through
the inhomogeneous stress eld of the asthenospheric mantle wedge
above a subduction zone. The magma paths depend on the dip angle
of the subducting slab, the distance from the wedge corner, the subduction velocity, the mantle viscosity, and the apparent buoyancy force of
dykes. In particular, dykes propagate upwards parallel to the slab if
the angle of subduction is N 45, whereas dykes are bended in subhorizontal curved sills, and thus magma is trapped without reaching
the lithospheric layer, in the case of shallow subduction angles (Dahm,
2000).
More information comes from recent geophysical studies that allow
imaging of the entire crust down to the source region of the magma in
the uppermost mantle, as for example at the deep plumbing system of
Mt Etna volcano; Chiarabba et al. (2004b) show the structure beneath
Mt Etna down to a depth of 24 km through tomographic inversions of
P- and S-wave arrival times from local earthquakes (Fig. 7). Tomography shows the lower plumbing system made by a main narrow conduit
zone, 46 km long, beneath the central part of the volcano between 9
and 18 km of depth, and a wide region of low Vp in the uppermost mantle (below 34 km of depth) associated with the magma source region.
The upper plumbing system is characterised by an intrusive complex
below the southern craters and the Valle del Bove, and another complex

93

extending between 3 and 9 km of depth and 810 km in length beneath


the central and eastern sectors of the volcano.
At Kilauea Volcano (Hawaii) several studies have been carried out on
the plumbing system in the last 45 years. Wright (1971) suggested a
vertical plumbing system located below the summit zone of Kilauea
with a shallow magma chamber at a few km depth, a storage zone at
2025 km of depth, and a region of partial melting at 65 km of depth.
These features account for the high pressure crystal fractionation
emphasised by Clague (1987) and the xenolith populations in the
lavas that require a magma reservoir at the base of the oceanic crust.
The presence of a magma chamber at a depth N 16 km has been suggested also by Frey et al. (1990) based on geochemical data. One of
the rst and most complete models of the Kilauea plumbing system
was presented by Ryan (1988), who outlined the presence of a deep
main conduit zone located down to a depth of 60 km below the summit
zone of the volcano (Fig. 8). This conduit is elongated and probably is
composed of a series of vertical and sub-vertical dykes which are linked
to a long feeding system located in the uppermost kilometres below the
volcanic rift. Based on microgravimetry data, it appears that the dense
central core of the volcano extends south of the summit, implying a
structural continuity between the Southwest Rift Zone and the East
Rift Zone (Kauahikaua et al., 2000). The surface of the dense core
beneath the East Rift Zone is steeper to the south than to the north,
suggesting southward migration of more recent eruptive activity
(Kauahikaua, 1993). Microgravimetry also suggests possible connections of the plumbing system at depths of 1014 km between different
volcanoes of the Island of Hawaii. These connections reach a depth compatible with the oceanic crust and their existence would suggest that
magma may have exploited linear zones of weakness in the oceanic
plate before the conduits became separated, or that island-building volcanoes can originate along the rift zones of older volcanoes (Kauahikaua
et al., 2000). The main conduits of Kilauea are punctuated by a series of
at bodies that represent various magma chambers highlighted by
aseismic zones. The maximum volume of the aseismic zone is 40 km3

Fig. 7. The deep plumbing system of Mt Etna (Italy) has been revealed by tomographic inversions of P- and S-wave arrival times from local earthquakes down to 24 km depth. Two tomographic sections (oriented SSENNW and SWNE) show the presence of a high velocity body (HVB) interpreted as a main solidied intrusive body. Magma intrusions prevalently affect
the western and north-western border of the HVB, following NNWSSE and NESW structures.
After Chiarabba et al. (2004b), reproduced under license number 3490690796656 of Oct 16, 2014.

94

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

Fig. 8. One of the rst and most complete models of the Kilauea plumbing system was presented by Ryan (1988), who outlined the presence of a deep main conduit zone located down to a
depth of 60 km below the summit zone of the volcano. This elongated conduit is composed of a series of vertical and sub-vertical dykes which are linked to a long feeding system located in
the uppermost kilometres below the volcanic rift.
Reproduced under license number 3490680301921 of Oct 16, 2014.

(Klein et al., 1987); however, it must be noted that an aseismic zone


may include a crystal-mush zone and a hot, ductile region surrounding
the magma reservoir. More recent studies showed that rapid variations
of 206Pb/204Pb isotopes on a time scale of years to decades require shortterm changes in the pathway that melt takes from the source zone to
surface (Greene et al., 2013, and references therein). This suggests
that the plumbing system is more complicated than expected and that
it might be made of interconnected conduits with various geometries.
Moreover, the integration of previous geophysical and geochemical
data with new ndings suggests the presence of a deeper magma chamber at about 1012 km depth (Lin et al., 2014), another at 35 km depth
(Fiske and Kinoshita, 1969; Klein et al., 1987), and a shallow chamber at
about 1 km depth (Cervelli and Miklius, 2003; Johnson et al., 2010).
As opposed to the geometry of the plumbing systems of Mt Etna
and Kilauea volcanoes, geophysical studies at the ShinmoeDake volcano, Japan, indicate the presence of a magma feeding system with a general oblique dip (Fig. 9A). Magnetotelluric data from Aizawa et al.
(2014) combined with petrographic data show the presence of a deep
basalticandesitic magma chamber located in an offset position from

this volcano at a depth N 10 km. Evidence from the 2011 eruption suggests that magma migrated towards the volcano along inclined pathways. These pathways coincide with a conductor zone located through
resistivity analyses. This type of deep to medium-depth magmatic system implies a component of lateral transport of magma and contributes
to the recent discussion on magma chambers that are in an offset position from the volcanoes they are connected with. Another recent example comes from Ishizuka et al. (2008) who found geophysical and
petrological evidence of long-distance lateral horizontal magma transport below the submarine/subaerial volcanic chain of the northern Izu
arc (Fig. 9B). Lava outpoured in different parts of the chain resulted to
come from the same basaltic primary magma, and this suggests that
magma was laterally transported for at least 20 km in the middle to
lower crust (1020 km deep here). Later on, the same magmas experienced crystal fractionation and accumulation at a shallow magma
chamber with further transport at distance b 5 km. Ishizuka et al.
(2008) also suggested that the long-distance magma transport has
been controlled by a regional extensional stress regime, while the
shorter (b5 km) distance transport has been controlled by a local stress

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

95

Fig. 9. Models of oblique and horizontal deep magma plumbing systems. A. Geophysical studies at the ShinmoeDake volcano, Japan indicate the presence of a magma feeding system with
a general oblique dip. Magnetotelluric data from Aizawa et al. (2014) combined with petrographic data show the presence of a deep basalticandesitic magma chamber located in an offset
position from this volcano at a depth N 10 km. Evidence from the 2011 eruption suggests that the magma migrated towards the volcano along inclined pathways. B. Ishizuka et al. (2008)
found evidence of long-distance lateral horizontal magma transport below the submarine/subaerial volcanic chain of the northern Izu arc: lava emitted in different parts of the chain resulted to derive from the same basaltic primary magma; this enabled to conclude that magma was laterally transported for at least 20 km in the middle to lower crust (1020 km deep).
Later on, the same magmas experienced crystal fractionation and accumulation at shallow magma chamber with further transport at distance b5 km.

regime linked to the load of the volcanic chain. Similarly, signicant


lateral magma migration, in the order of ~ 15 km, has been found at
El Hierro Island (Canary Archipelago) during the JulyOctober 2011
seismicvolcanic crisis (Gonzlez et al., 2013). On October 2011, a submarine ssure eruption occurred at a distance of 15 km from the initial
earthquake loci. Geodetic analysis reveals two distinct shallow magma
reservoirs, at 9.5 4.0 km and 4.5 2.0 km, vertically offset from
one another and linked by vertical conduits.
5. Shallow plumbing system
The shallow part of plumbing systems, that is to say the rst
kilometres below volcanoes, is better constrained than the deeper part
due to the larger accessibility of outcrops at eroded volcanoes and to
more feasible geophysical studies. The plumbing system can go from
very simple dyke complex, made of tens to thousands parallel vertical
dykes that link the magma chamber to the uppermost volcano conduit,
to a very complex intrusive plexus made of dykes, sills and inclined
sheets, which interconnect multiple shallow magma chambers. For a
long time it has been believed that the simpler kind of plumbing system
without shallow chambers can be found at oceanic island volcanoes, but
more recent observations indicate the opposite (Amelung and Day,
2002; Klgel et al., 2005; Gonzlez et al., 2013). The possible absence
of shallow magma chambers has been interpreted as due to local processes that can modify the structure of a volcano such as catastrophic
ank collapses (Amelung and Day, 2002; Tibaldi, 2004; Tibaldi et al.,
2008a, 2008b, 2008c; Manconi et al., 2009) or due to poor knowledge
of the subvolcanic system.
At El Hierro Island (Canary archipelago), geodetic data indicate the
presence of a shallow magma reservoir at a depth of 4.5 2.0 km beneath the volcano ank of the southern rift (Gonzalez et al., 2013), although a deeper chamber has been suggested by Becerril et al. (2013).
At the Island of Hawaii, beneath Kilauea Volcano there are two main
magma storage zones at shallow level, at ~ 1 km and ~ 3 km depth
(Cervelli and Miklius, 2003; Baker and Amelung, 2012). At Piton de la
Fournaise, two active magma chambers have been detected at ~ 2.5
and ~7 km of depth (Peltier et al., 2009). The Fernandina Island volcano
(Galpagos Islands) has two levels of magma storage at depths of ~ 1
and ~ 5 km (Chadwick et al., 2011; Bagnardi and Amelung, 2012).
Other volcanoes of the Galpagos Islands have a single shallow crustal
magma reservoir, detected geodetically, such as at Alcedo, Cerro Azul,
Darwin, Sierra Negra, and Wolf volcanoes. The existence of shallow
magma chambers has been attributed to the existence of a high melt
supply rate from the deep mantle by Clague and Dixon (2000).
Gonzlez et al. (2013) suggest that at the Canary Islands the high melt
supply rate is not enough for the existence of shallow magma reservoir,
because the volcanoes of this archipelago have the lowest magma

supply rates for oceanic island volcanoes (~ 0.0002 km/yr, Amelung


and Day, 2002). Hildner et al. (2012) proposed that the age of the underlying lithosphere could control the presence of one or more chambers within the upper part of the plumbing system, because the older
oceanic lithosphere is much cooler and denser and tends to sink due
to advanced thermal contraction together with a deepening of the
magma chambers. Gonzlez et al. (2013) similarly suggested that
the main controlling factor is the thermomechanical structure of the
crust, which in turn depends on the deep mantle magma supply rate
and on the age of the oceanic lithosphere. Fialko and Rubin (1998)
and Grandin et al. (2012) claimed that the presence of double shallow
magma chambers is linked to the presence of two horizons of neutral
buoyancy, where magma accumulation occurs preferentially and its lateral migration is facilitated, giving rise to eruptions along rift zones. The
deeper horizon associated with olivine-rich cumulates exists, for example, at a depth of 610 km for Hawaii and at ~9.5 km at El Hierro, and a
shallower one at 24 km at Hawaii and ~ 4.5 km at El Hierro (Ryan,
1988; Gonzlez et al., 2013).
Shallow magma chambers have been detected also at several eroded
volcanoes for example in the British Tertiary Igneous Province (United
Kingdom and Ireland) and in Iceland (Walker, 1958, 1960; Bald et al.,
1971; Walker, 1974, 1975; Fridleifsson, 1977; Gudmundsson, 1990,
1995, 2002; Klausen, 2004, 2006; Tibaldi and Pasquar, 2008; Tibaldi
et al., 2011). These chambers may host magma with the same pressure
as the host rock and in this case their presence does not perturb sensibly
the surrounding stress eld. However, the presence of magma is linked
to high temperatures and a different rheology than the host rock, conditions that can perturb the propagation of dykes from below (Karlstrom
et al., 2010). Chambers that instead are signicantly overpressured with
respect to the stress in the host rock can perturb the stress eld in the
surrounding region; below the chamber there may be a rock volume inside which the stress produced by magma forces is large enough to perturb the trajectories of rising dykes, focusing them towards the magma
chamber (Fig. 10A) (Karlstrom et al., 2009), although, as far as I know,
there are no eld examples of such focusing below eroded magma
chambers.
On the contrary, there are plenty of examples of the structure of
plumbing systems immediately above shallow magma chambers, especially in the case of mac magmas. These plumbing systems are made of
swarms of inclined sheets that have a typical arrangement consisting in
variable dips that tend to converge on a central area. The denition of
inclined sheet was rst used by Harker (1904) for intrusions at Skye,
while the term cone sheet was later introduced by Bailey et al.
(1924) to describe the occurrence of sheets arranged in a conical
swarm geometry around a main intrusion. Diverse models have been
introduced to explain the formation of cone sheets, which have been
widely studied due to their importance in magma transport, also

96

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

Fig. 10. Chambers that are signicantly overpressured with respect to the stresses in the host rock can perturb the stress eld in the surrounding region. A. Below the chamber there may be
a rock volume inside which stresses produced by magma force are large enough to perturb the trajectories of rising dykes, focusing them towards the magma chamber. B. Boundary element results showing a pressurised spherical magma chamber with an internal magmatic excess pressure of 5 MPa as the only loading; the trajectories of the greatest principal stress
depart at the right angle from the chamber walls, creating a radial pattern.
A. Redrawn after Karlstrom et al. (2009). B. Redrawn after Gudmundsson (1998).

laterally from the central zone of a volcano; cone sheets are crucial also
in terms of the possibility of locating the magma chamber based on their
geometry and of the fact that the piecemeal intrusion of hundreds of inclined sheets induces a major upward deformation of the host rock (Le
Bas, 1971). Bailey et al. (1924) suggested that any increase in the pressure of a magma reservoir at depth would superimpose a tensional regime on the host rock and favour the formation of inward-dipping
fractures, the intrusion of which produces cone sheets (Fig. 11A). A decrease in the magma chamber pressure instead may promote instability
in the roof rock resulting in subsidence along outward-dipping or subvertical fractures, the intrusion of which produces ring dykes
(Fig. 11B) (Bailey et al., 1924). Anderson (1936) considered that ring
dykes should be emplaced along shear fractures that dip outward in
the range of 6070 becoming shallower upward. Robson and Barr
(1964) and Durrance (1967) put forward the importance of shear
planes in guiding the emplacement of cone sheets, although now it is
recognised that most sheets occupy extensional fractures. The model
of Phillips (1974) suggests that cone sheets are restricted to a concentric
swarm departing from the shoulders of a magma chamber and following shear fractures, and thus they are missing above the centre of the
magma chamber, something usually different in the eld (Fig. 11C).
Shear fractures form as rotational strain develops along inwarddipping strips that delimit roof uplifts relative to the immobile side
regions. Phillips (1974) also suggested that cone sheets could extend
upward in sub-vertical shear fractures with a bowl-like geometry
(Fig. 12) or they may curve into sub-horizontal sills upwards with a
trumpet-shape (Fig. 12). In the present paper I prefer to use the term
centrally-inclined sheet swarm because it is more descriptive and
does not bear reference to the previous cone sheet models.

Within a centrally-inclined sheet swarm, the sheet dip may vary


from one sheet to another and also at the same sheet. Anyway, it is
worth noting that outcrops of centrally-inclined sheet swarms are usually limited in lateral/vertical extent and different geometries of inclined
sheets can be found; accordingly, different models are proposed to
explain them. These models put forward: (1) concave-downward
(trumpet-shaped) sheets with increasing dip closer to the magmatic
source (Fig. 12A) (Phillips, 1974); in this model, sheets are missing in
the central part; (2) concave-upward (bowl-shaped) sheets with decreasing sheet dip with depth from a pressurised magma chamber
(Fig. 12B) (Phillips, 1974); also here, sheets are missing in the central
part; (3) radial planar sheets from a spherical magma chamber
(Fig. 12C) (Chadwick and Dieterich, 1995; Gudmundsson, 1998);
and (4) planar parallel to sub-parallel sheets originated from a lobate
(sill-like) magma chamber (Fig. 12D) (Gudmundsson, 1998; Tibaldi
et al., 2011; Bistacchi et al., 2012).
In westernmost Iceland, along the southern coast of the Snaefellsnes
Peninsula that is essentially made up of TertiaryQuaternary basalts,
there are two major intrusions, each surrounded by a centrallyinclined sheet swarm. The intrusions correspond to the Midhyrna gabbro and Lysuskard granophyre (Upton and Wright, 1961) and both intruded Tertiary basaltic lava ows. The Midhyrna and Lysuskard
intrusions are surrounded and intruded by two centrally-dipping
sheet swarms (Fig. 13A) (Tibaldi et al., 2013); these sheets show no
gradual variation in dip with distance from the focus area, are rectilinear
in section view, and intrude with the same geometry the main intrusive
bodies as well as the layered Tertiary lavas (Fig. 13B). The diameter of
both sheet swarms is about 12 km, the average sheet thickness is
0.63 m, and the average dip is 28 (Tibaldi et al., 2013), lower than

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

97

Fig. 11. Models of cone sheet and ring dyke emplacement in a vertical section view: A. an expanding magma chamber induces a radial 1 (continuous lines) and concentric 3 (segmented
lines) parallel to the vertical section view, cone sheets are emplaced along tension fractures (diverging arrows). B. Magma chamber deation induces shear zones at the periphery along
which ring dykes may intrude. C. An expanding magma chamber produces tensional fractures (diverging arrow) along which cone sheets may intrude with a downward-concave shape,
and/or shear zones with rotational strain along which cone sheets may intrude with an upward-concave shape. Note that emplacement of cone sheets is not foreseen above the central part
of the magma chamber.
A. and B. Modied after Bailey et al. (1924) and Anderson (1936). C. Modied after Phillips (1974).

other sheet swarms in Iceland whose average dip is 34. Farther south
there is the Thverfell centrally-inclined sheet swarm that was emplaced
mostly within almost isotropic hyaloclastite deposits (Pasquar and
Tibaldi, 2007; Tibaldi et al., 2008c) (Fig. 14). Also in this case the sheets
do not show a systematic dip decrease outwards from the focus zone.
The Cuillin Igneous Complex in the Island of Skye (UK) represents
the deep core of a Tertiary, large basaltic volcano. It is composed of
coarse-grained layered and unlayered basic and ultrabasic huge intrusive bodies that are in turn intruded by centrally-inclined sheets. The

overlying succession that forms the original volcano, up to a thickness


of 2 km, has been eroded. Bell et al. (1994) stated that centrallyinclined sheet magmas were generated by large degrees of partial melting of a depleted mantle source, combined with assimilation of Lewisian
gneisses, and fractional crystallisation that occurred in a magma chamber subjacent to the nal site of emplacement and undergoing periodic
replenishment. Also in this case, the about one thousand sheets measured do not show a systematic variation of dip angle with distance
from the focus zone (Tibaldi et al., 2011) as can be seen in Fig. 15. It is

98

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

Fig. 12. Possible geometries of centrally-inclined sheet swarms resulting from internal excess magma pressure: A. radial planar sheets from a spherical magma chamber; B. concave-upward (bowl-shaped) sheets from a spherical magma chamber; C. concave-downward (trumpet-shaped) sheets from a sill-shaped magma chamber; and D. planar parallel sheets from a
laccolith-like chamber.
A. After Chadwick and Dieterich (1995) and Gudmundsson (1998). B. After Chadwick and Dieterich (1995) and Gudmundsson (1998). C. After Phillips (1974) and Chadwick and Dieterich
(1995). D. After Bistacchi et al. (2012).

particularly worth noting the similarity of the average dip of the inclined sheets near the centre of the swarm with respect to those
cropping out at the periphery. This is also quantied by the distribution
of sheet dip measured along four transects across the whole swarm
(Fig. 16). Also at the Tejeda complex, in the island of Gran Canaria,
there is a centrally-inclined sheet swarm with homogeneous dip angles
measured across the whole swarm (Fig. 17). Schirnick et al. (1999)

interpreted this architecture of the centrally-inclined sheets as the effect


of magma forces exerted by an ellipsoidic magma chamber with a at
upper surface. Also seismic observations (e.g., Barker and Malone,
1991; White et al., 2008) and geodetic inversions from active volcanic
areas (e.g., Newman et al., 2006) suggest an ellipsoidal shape for
magma chambers located at a few kilometres depth. At Ardnamurchan
(NW Scotland) there is a centrally-inclined sheet swarm with a

Fig. 13. A. Photo of the western centrally-dipping sheet swarm of the LysuskardMidhyrna intrusions. B. Graph of sheet dip angle vs. distance from the centre of each centrally-dipping
sheet swarm: box and circles represent the data of the two swarms; note that these sheets show no gradual variation in dip attitude with distance from the focus area, are rectilinear
in section view, and intrude with the same geometry the main intrusive bodies as well as the layered Tertiary lavas. Box shows the location.
A. Modied after Tibaldi et al. (2013).

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

99

Fig. 14. A. Photo of the western part of Thverfell (Iceland) centrally-inclined sheet swarm: note the constant dip angle of the sheets. B. Photo and C. sketch of part of the sheet swarm where
it is possible to observe that some of the intrusions cross the boundary between lava ows and the underlying hyaloclastite breccias. Box shows the location.

diameter of 18 km (Richey and Thomas, 1930; Emeleus, 2009). This example has been recently used by Magee et al. (2012) to question the formation of this sheet pattern by proposing an alternative model. These
authors claimed that the Ardnamurchan sheet swarm may be linked
to laterally propagating regional dykes, sourced from laterally adjacent
magmatic systems, which are deected around a central complex by
stress eld interference.
Since the geometry and location of centrally-inclined sheet
swarms is considered as mainly dependant from the geometry of the
overpressured magma chamber, it is important to quantify the maximum distance the sheets can reach. In fact, if the stress exerted by the
magma chamber overpressure on the surrounding rock guides the development of these inclined sheets, we can estimate the distance by
which this stress dominates. I have collected most of the available
data on well exposed centrally-inclined sheet swarms in order to evaluate their width (Fig. 18A). Here it is possible to see that most swarms
have diameters ranging from a minimum of 5 km to a maximum of
20 km, with an average value of 12.3 km. In order to quantify also
the typical thickness of centrally-inclined sheets, I collected data
from different locations in Iceland and in the British Isles and
summarised them in the graph of Fig. 18B. In this graph it is possible
to observe the distribution of dip angles and thickness for a population
of 2340 sheets, with the larger number of intrusions with dip angles in
the range of 2065 and thickness of 0.21.8 m (largest thickness up
to 1012 m). It has been shown that in some centrally-inclined sheet
swarms, shallower-dipping sheets are thinner than steeper-dipping
sheets (Gudmundsson, 2003), which is not consistent with the data presented above. This might be due to the fact that it has been carefully
avoided to include in the sheet population of Fig. 18B any steeplyinclined sheets that might in reality represent regional dykes, which is
to say to intrusions that are not related to the central magma chamber.
Regional dykes, in fact, are much thicker in average than inclined sheets
linked to local magma chambers: In Iceland, for example, regional dykes

vary in thickness from a few centimetres to about 60 m, with a mean


thickness of 26 m (Gudmundsson et al., in press). The data of
Fig. 18B indicate that the average dip angle is 41 and the average thickness is 0.93 m. The average inclined sheet dip angle in the British Isles
has been found at ~45 (Billings, 1972; Hills, 1972), and at Gran Canaria
at ~41 (Schirnick et al., 1999), and at some individual sheet swarms in
Iceland at 34 (Gudmundsson et al. 1996). The average sheet thickness
of 0.93 m found from the data of Fig. 18B is larger than the one found
in some individual swarm in Iceland, which is around 0.1 m
(Gudmundsson et al., 1996) but identical to the average thickness of
the 1128 sheets in the Thverartindur Central Volcano (Klausen, 2004).
These eld examples indicate the effects of overpressured magma
chambers on the overlying host rock. These data can be compared
with numerical modelling that takes into account the variation in the
stress eld around an overpressured magma chamber. This overpressure can develop through a series of different causes that include injection of new magma from below into the chamber, volatile exsolution,
melting or solidication of wall rocks, and magmatic differentiation
(Tait et al., 1989; Folch and Mart, 1998; Annen and Sparks, 2002). The
overpressure considered large enough to produce fracture in the chamber walls and propagation of intrusive sheets depends on the criterion
of rock rupture; Sartoris et al. (1990) considered a value of 110 MPa
large enough to exceed the tensile strength of the surrounding rocks
under a tensile mechanism of rupture, a value adopted also by other authors (e.g. Atkinson and Meredith, 1987; Gudmundsson, 1988). Other
authors suggest that this is likely an underestimation (e.g. Karlstrom
et al., 2009) and proposed greater values (10100 MPa) based on thermal considerations of long-distance dyke propagation and on the composition and tectonic settings of the chamber (Rubin, 1995b; Jellinek
and DePaolo, 2003).
A pressurised magma chamber changes the overall stress pattern of
the host rock signicantly, as shown by numerical and analogue modelling (e.g. McLeod and Tait, 1999). In the case of a pressurised spherical

100

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

Fig. 15. Photos of the centrally-inclined sheets at the Cuillin Igneous Complex at the Isle of Skye (UK). Note that the sheets have the average dip angle both in the core of the swarm
(A) where the highest frequency of intrusion is observed, and in the peripheral part (C). B. Location of the photos.

magma chamber, boundary element results show in the case of internal magmatic excess pressure of 5 MPa as the only loading that the
trajectories of the greatest principal stress (1, ticks) depart at right
angle from the chamber walls, creating a radial pattern in section view
(Fig. 10B) (Gudmundsson, 1998). Based on the work of Anderson
(1951), it is commonly assumed that magma paths select the orientation of least resistance by opening along planes across which the
normal stress is the lowest. The consequence is that magma paths follow and trace trajectories perpendicular to 3 within the stress eld occurring before sheet propagation (Stevens, 1911; Anderson, 1936, 1938;
Od, 1957). This conclusion has been questioned by McKenzie et al.
(1992) and Meriaux and Lister (2002) who pointed out that several
theoretical works ignored the fact that dykes radically alter the surrounding stress eld as they propagate. Meriaux and Lister (2002)

concluded that neglecting the dyke effect gives in any case good qualitative agreement in the pattern of principal-stress trajectories, but substantially different estimates of the stresses. Based on the above, it can
be assumed that magma escaping from the chamber walls should follow
the direction of less resistance giving rise to intrusive sheets that propagate along planes that contain 1 and 2 (i.e. normal to 3). Fig. 10C
shows the boundary element results of the stress eld around a vertical
section of a spherical magma chamber subject to remote horizontal tensile stress of 5 MPa as the only loading. The trajectories of the 1 are
mostly vertical to sub-vertical suggesting that dykes departing from
the magma chamber have a vertical geometry at a short distance.
Fig. 10D presents the same setting as in Fig. 10A but with a at
magma chamber (laccolith-like shape). Although we are aware of the
limitation of considering only magma overpressure as the acting stress,

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

101

Fig. 16. Distribution of the sheet dip angles measured along four transects across the whole centrally-inclined sheet swarm of the Cuillin Complex (Island of Skye, UK). Note the similarity
between the average dip angle of the inclined sheets near the centre of the swarm and those cropping out at the periphery.
Redrawn after Tibaldi et al. (2011).

this example is useful to discern the contribution of the magma chamber force in dictating the intrusive sheets pattern. In this case in fact,
the trajectories of 1 above the magma chamber are mostly parallel to
each other, due to the at upper surface of the chamber, resulting in a
series of inclined sheets whose dip angle does not vary signicantly.
As a matter of fact, the host rock is subject to further stresses of different origins, including the lithostatic component and the tectonic
component of the regional far eld; Russo et al. (1996, 1997), Russo
and Giberti (2000), van Wyk de Vries and Matela (1998), Grosls
(2007) and Mart and Geyer (2009) present numerical models where
the gravitational body force and tectonic boundary stresses are considered together with the stresses transmitted by the magma chamber.
Bistacchi et al. (2012), in particular, focused on the possible trajectories
of intrusive sheets modelled under a total stress eld resulting from
three components of boundary forces (Ranalli, 1995): (1) the superposition of diverse magma overpressures in the magma chamber; (2) the
presence of different free topographic surfaces ranging from a at topography to a positive topography, simulating the load of a volcanic edice, or to a negative surface corresponding to a caldera; (3) regional
tectonics with diverse remote stresses, plus a body force (4) corresponding to gravitational acceleration. The analysis used a set of
quasi-static Finite Element Method mechanical models that showed
the conguration of the resulting stress eld (Fig. 19) under diverse
pressures in an oblate magma chamber. With negative values, which
correspond to a deating chamber, the trajectories of 1 dip outwards

with respect to the magma chamber. Assuming the case of intrusive


sheets that follow paths corresponding to surfaces perpendicular to
3 (thus containing 1 and 2), this case corresponds to the formation
of a ring dyke complex. When 3 is low, the dip of ring dykes at 2
3 km depth is 6580. Within a distance from the central axis of
about 11.2 diameters of the magma chamber, radial dykes are
favoured for intermediate overpressures, and centrally-inclined sheets
dominate in the axial domain for higher ( 10 MPa) overpressures.
The dip angles of centrally-inclined sheets are predicted in the
range 4565 for overpressure 20 MPa. In all cases, radial dykes
dominate beyond the 11.2 diameter limit. By simulating different
depths of magma chamber in the range 23.5 km, the only difference
is that inclined stress trajectories consistent with centrally-inclined
sheets appear for lower overpressures in the case of a shallower
magma chamber. By using different possible shapes of topography,
results of stress trajectories at 23 km depth do not vary sensibly,
apart when there is a caldera depression that produces variation in
the trajectories. At shallower depths, the topography inuence increases. The effect of an increasing gravitational force or of an increasing regional extensional tectonics is to steepen the sheets above the
magma chamber, whereas the effect of increasing magma overpressure (in the case of a chamber with a lobate shape) or increasing regional horizontal compression is to decrease the sheet dip angle.
In the eld, it has been frequently observed that centrally-inclined
sheets are associated with radial dykes; this can be clearly seen, for

Fig. 17. Geologicalstructural section of the Tejeda complex, on the island of Gran Canaria, with a centrally-inclined sheet swarm that shows homogeneous dip angles across the whole
swarm. The authors interpreted the geometry of the centrally-inclined sheets as the effect of the magma force exerted by an ellipsoidic magma chamber with a at upper surface.
Redrawn after Schirnic et al. (1999).

102

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

Fig. 18. A. Graph of the diameter of well exposed centrally-inclined sheet swarms; it is possible to note that most swarms have a diameter that ranges from a minimum of 5 km to a maximum of 20 km, with an average value of 12.3 km. B. Graph of the distribution of dip angles (X axis) and thickness (Y axis) of 2340 sheets surveyed at the centrally-inclined sheet systems of
Iceland and British Islands (Ardnamurchan). Note the larger number of intrusions with dip angles in the range of 2065 and thickness of 0.21.8 m.
B. Data after Magee et al. (2012).

example, at the Cuillin Complex centrally-inclined sheet swarm at Skye


Island (UK), where Tibaldi et al. (2011) recognised three zones with different characteristics: (1) a central zone in the core of the centrallyinclined sheet swarm where inclined sheets dominate; (2) a more
external zone where inclined sheets coexist with vertical dykes; and
(3) an outer zone where dykes dominate (Fig. 20). In zone 2) several
generations of dykes are interleaved with generations of inclined sheets,
suggesting quick changes in the state of stress and consequent geometry of intrusion. These eld data are consistent with the above described
numerical models in which the geometry of intrusions above a magma
chamber changes at a critical distance that depends on the state of stress
resulting from the combination of the various sources. In zone 1) the
paths of the intrusive sheets are controlled by the magma chamber
overpressure. In zone 2) the intrusive paths can rapidly change following pulsations of the magma chamber: with large ination the zone of
centrally-inclined sheets expand outwards, whereas with a decrease
of the overpressure, the magma chamber inuence decreases and
dykes develop. In zone 3) the regional background stress dominates
and intrusive sheets departing form the same magma chamber assume
a dyke-like geometry. Galland et al. (2014), based on analogue models,
suggested that the shift from cone sheets to dykes is controlled by the
magma dynamics expressed by the ratio between the viscous stresses
in the owing magma to the host rock strength; dykes form preferentially when their magma source is deep compared to its size, or when
magma inux rate (or viscosity) is low, and vice-versa for cone sheets.
Some analogue models show that these cone sheets can depart from
dykes, although eld data are necessary to conrm this hypothesis.
The importance of considering the possible effects of the velocity of
magma upwelling is reinforced by the fact that it is becoming clear
that very diverse values are possible for both mac and felsic magmas;

values from 10 4 to 10 2 m/s have been classically considered


(Huppert and Sparks, 1981; Sparks et al., 1984), although recent ndings show that values up to 1117 m/s for basaltic magmas are possible
(Lloyd et al., 2014), as well as for andesitic magmas from 10 3 to
102 m/s (Ruprecht and Plank, 2013) and up to 1 m/s even for rhyolitic
magmas (Castro and Dingwell, 2009).
All this description assumes that intrusive sheets follow planes of
fracturing that contain 1 and 2. Some eld evidence shows another
possibility: at the Isle of Skye it has been observed that some inclined
sheet intruded along shear planes, although the great majority followed
fractures with an opening mode I (Fig. 21). Lower angle shear planes
show normal motions, whereas steeper planes have reverse motions
(Bistacchi et al., 2012). These planes form a conjugate set bisected by
the dominant dilational sheets, and are thus consistent with each
other. The frequently found scatter in centrally-inclined sheet dip angles
can thus be explained as the result of intrusions along both mode I fracture planes and conjugate shear planes with a higher- or lower-thanaverage dip angle. The presence of reverse shear planes has been
found also by Schirnick et al. (1999) in Gran Canaria and by Mathieu
and van Wyk de Vries (2009) at the Mull volcano in Scotland.
The centrally-inclined sheets contribute to magma transport upward: the sheets located above the centre of the magma chamber may
be steeper and tend to transport magma vertically into the volcano,
whereas those located more externally tend to move magma laterally
with respect to the volcano, possibly contributing to lateral (eccentric)
eruptions at the volcano's foot. Anyway, only a small fraction of these
sheets is capable of bringing magma to the surface and most remain
trapped into the shallower crust. This might occur through bending or
stoping. As regards bending, inclined sheets and dykes can rotate
to a shallower dip angle and reach a sill-like conguration. An old

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

103

Fig. 19. Results of quasi-static Finite Element Method mechanical models showing the conguration of the stress eld around an oblate magma chamber that incorporates: (1) diverse
magma overpressures; (2) the presence of different free topographic surfaces ranging from a at topography to a positive topography, simulating the load of a volcanic edice, or to a
negative surface corresponding to a caldera; (3) regional tectonics with different remote stress values, plus a body force (4) corresponding to gravitational acceleration. Results are
shown on a vertical cross-sectional symmetry plane of the 2D axial-symmetric model. Each gure corresponds to an increasing magma chamber pressure. Vector symbols show orientation of 1 and 2; 3 is always perpendicular to the 12 plane, 2 axes plot as small dots in areas where they are perpendicular to the plane of the cross-section. Colour shows the values of
least compressive to tensional 3 stress axis; red indicates the negative values, related to the emplacement of magma in tensional ssures. (For interpretation of the references to colour in
this gure legend, the reader is referred to the web version of this article.)
Modied after Bistacchi et al. (2012).

104

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

Fig. 20. At the Cuillin Complex cone sheet system (Skye Island, UK), there are three zones with different characteristics: (1) a central zone in the core of the cone sheet where inclined
sheets dominate; (2) a more external zone where inclined sheets coexist with vertical dykes; and (3) an outer zone where dykes dominate.
Modied after Tibaldi et al. (2011).

explanation for the dyke-sill transition was the effect of neutral buoyancy forces (Corry, 1988). Nevertheless, much eld evidence shows that
sheets were injected through host rock successions dominantly composed of fragmented deposits, like hyaloclastites and breccias, the density of which is lower than magma density (Pasquar and Tibaldi, 2007;
Tibaldi and Pasquar, 2008; Tibaldi et al., 2008c), without any geometrical change, so neutral buoyancy cannot be the main explanation.
Similarly, Thomson (2007) suggested that the presence of lithological
contrasts, particularly ductile horizons such as overpressured shales,
may guide sill formation at any depth below the neutrally buoyancy
level. Another explanation may be the intersection of the intrusive
sheet with an already existing horizontal, freely slipping joint
(Weertman, 1980) or in a similar way, the intersection between a
weak bedding plane and a steep normal fault at shallow depths
(Gaffney et al., 2007). The upward propagation of dykes and inclined
sheets can be diverted also due to the generation of stress barriers, i.e.
layers with local stresses unfavourable for the intrusion propagation
(Gretener, 1969; Gudmundsson, 1986, 1990; Parsons et al., 1992;
Gudmundsson, 2011). The deviation from a dyke to a sill has also been
reproduced experimentally by Kavanagh et al. (2006) at the interface
between upper, rigid layers overlaying lower, weaker layers. Valentine
and Krogh (2006) explained dyke bending in terms of local stress rotation along 3-D variations on a normal fault plane. Finally, dykes can
bend into sills in the case of remote tectonic compression, as they adjust
their trajectory reaching a geometry parallel to the 12 plane

(Menand et al., 2010). Among all these possibilities, eld data suggest
that the further propagation of sheets is strongly inuenced by lithologic boundaries in the host rock. As an example, at the Thverfell eroded
volcano (SW Iceland), the centrally-inclined sheets bend and reach a
horizontal geometry as they approach the contact between a lava succession and the underlying hyaloclastites (Fig. 22AB). As another example, a dyke bends into a sill at a similar contact at the Stardalur
eroded volcano (SW Iceland) (Fig. 22C), and a similar conguration
can be seen also at Nysiros volcano (Greece), where a dyke bends into
a sill near a contact between hyaloclastites and lavas (Fig. 22D). This
eld evidence suggests that dykes and inclined sheets can be diverted
or stopped at contacts between layers with lithologies having very different stiffness (Gudmundsson, 1986; Kavanagh et al., 2006).
Changes in intrusion geometry can also derive from dykedyke interactions. Numerical and analogue simulations have been developed
to analyse how ascending dykes might interact, showing that their
self-induced stress eld may locally be more signicant than the regional stress eld (Khn and Dahm, 2008). Coeval ascending dykes may
converge towards regions of increased dyke densities and may lead to
the formation of sills; alternatively, they can change their dip and
magma paths by getting closer to each other, or they can merge crating
dykes with increasing magma ux (Ito and Martel, 2002; Canon-Tapia
and Merle, 2006), or they can have a higher probability of becoming
arrested (Jin and Johnson, 2008). The dykedyke interaction is greater
when the intrusion is more than some dyke lengths away from the

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

105

Fig. 21. Photos taken at the Cuillin cone sheet in the Skye Island (UK), showing examples of inclined sheets intruded along shear planes with reverse motion (R) and normal motion (N),
although most followed fractures with an opening mode I (extension normal to the fracture wall) (T). Lower angle shear planes show normal motions, whereas steeper planes have reverse motions, and both form a conjugate set bisected by the dominant dilational sheets (T), and are thus consistent with each other. The frequently found scatter in cone sheet dip angles
can thus be explained as the result of intrusions along both mode I fracture planes and conjugate shear planes with a higher- or lower-than-average dip angle.

feeding reservoir or the free surface, whereas the interaction is generally small when the horizontal tensional stress is large compared to the
compressional stresses induced by the emplacement of a single dyke
(Khn and Dahm, 2008).
Numerical and analogue models indicate that also the load exerted
by a volcano can inuence the trajectory of the magma path below.
Pinel and Jaupart (2004) found, by numerical modelling, that if a volcano is present, at small distances from the axis, conning stresses due to
the volcano load hinder vertical magma propagation. This produces horizontal dyke propagation and, in case the dyke reaches the surface, the
formation of a distal eruptive centre. They found also that, everything
else being equal, a decreasing magma supply rate or a decreasing
magma viscosity generate eruptive centres at increasing distances
from the focal area. Gaffney and Damjanac (2006) found a similar tendency of magma ow to be diverted away from a highland zone (like
a volcano) towards the lowland. By analogue modelling, Kervyn et al.
(2009) showed that the local loading stress eld, due to the presence
of a volcano, favours rising magma away from the volcano centre if a
central conduit is not established or is blocked. Experiments show that
the load compressive stress stops rising dykes as they approach the
cone stress eld especially in the case of dykes with limited overpressure and steep volcanic cones (Fig. 23). Dyke overpressure builds up
as intrusion continues and dykes extend laterally until their tips rise
vertically again, leading to eruptions. This might explain the presence
of volcanic centres located at the base of a volcano, whose distance
from the volcano's axis depends on volcano slope angle, magma overpressure and substratum thickness, whereas extensional processes favour magma propagation in the axial volcanic zone. Similar results

were obtained by Peltier et al. (2005) who, using seismic data, analysed
the direction and velocity of dyke propagation from a shallow reservoir
at a depth of 3 km at Piton de la Fournaise (Reunion Island) between
1998 and 2004. They found a change from vertical ascent to slower lateral propagation when the dyke reached the Dolomieu cone base. This
change was attributed to the presence of a fractured zone as well as to
the effect of volcano load. The nal geometry of a sheet is also a function
of the distribution of magma overpressure within the intrusion; dykes
with either uniform internal overpressure or with a simple viscous pressure drop propagate across much greater distances than the corresponding unpressurised cracks, and uniformly loaded dykes tend to
propagate with less curvature than those with a viscous pressure drop
(Meriaux and Lister, 2002).
Several eld examples show that crosscutting relationships between
generations of intrusive sheets may give clues to the parameters
that control the geometry of magma conduits. For example, as already
introduced before, at the eroded Thverfell volcano (Iceland) the
centrally-inclined sheets bend until they become horizontal at the contact between the lava succession and the underlying hyaloclastites
(Fig. 22AB). The resulting sills stacked on top of each other and created
a laccolith-like body, showing the difculty of magma uprising to the
surface (Tibaldi et al., 2008c). However, the stacked sills are crosscut
by a vertical dyke that affected all the exposed rock succession
(Fig. 24AB). A similar situation can be seen at another site of the
same eroded Thverfell volcano, where inclined sheets intruded the
lava succession and are in turn cut by a series of parallel, sub-vertical
dykes (Fig. 24CD). Both cases reect a change in the geometry of
the plumbing system, that rst favoured lateral magma transport

106

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

Fig. 22. Examples suggesting that dykes and inclined sheets can be diverted at contacts between layers with lithologies characterised by very different stiffness. Photo (A) and sketch (B) at
the Thverfell eroded volcano (SW Iceland), showing centrally-inclined sheets that bend attaining a horizontal geometry as they approach the contact between a lava succession and the
underlying hyaloclastites. At the Stardalur eroded volcano (SW Iceland) (C), a dyke bends into a sill at a similar contact. At the Nysiros volcano (Greece) (D), a similar conguration can be
seen where a dyke bends into a sill nearby a contact hyaloclastite/lava.

(horizontal or inclined) and then vertical magma upwelling. This can be


explained in terms of a change in the stress state within the host rock,
induced by a rotation of the 3 axis from vertical (sill case) or gentledipping (inclined sheet) to horizontal. This change is the effect of a decrease of the horizontal far eld stress, as supported by the fact that the
dykes that intruded the whole rock succession are perpendicular to the
orientation of the Icelandic Rift extension axis. If we consider that the
stacked sills and the centrally-inclined sheets of the Thverfell example
were located below the volcano, the successive dyke intrusions indicate
that extensional tectonics may favour upward magma propagation in
the axial volcanic zone, consistent with the modelling of Kervyn et al.
(2009). A similar situation was encountered at the Otoge Complex
(Japan), where eld evidence shows a centrally-inclined sheet system
crosscut by a swarm of more recent parallel dykes (Geshi, 2005). Petrographic analyses indicate that the centrally-inclined sheets have a less
fractionated character with little compositional variations, whereas

the dyke swarm has a fractionated character with wide compositional


variations. Geshi (2005) interpreted this compositional change by fractional crystallisation due to the decline in the supply of less fractionated
hot magma into the reservoir. The change in the geometry of the plumbing system from centrally-inclined sheets to dyke has thus been
interpreted as reecting a change from conditions with large overpressure and high magma supply, to conditions with pressure decrease in
the magma chamber and lower magma supply. During the decline of
magma replenishment into the reservoir, dykes reach an orientation
perpendicular to the tectonic 3 (Geshi, 2005).
Most of the above described papers assume that sheets intrude along
planes containing 1 and 2. However, a sheet could propagate oblique
to this plane if it invades a preexisting fracture that is misaligned
with respect to the principal stress directions; sheets that occupy
faults or preexisting joints are described in several studies (Johnson,
1961; Curtie and Ferguson, 1970; Gudmundsson, 1983; Delaney and

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

107

Fig. 23. Analogue models showed that the local loading stress eld, due to the presence of a volcano, favours magma ascent away from the volcano centre. Dyke overpressure builds up as
intrusion continues and dykes extend laterally until their tips rise vertically again, leading to eruption.
Modied after Kervyn et al. (2009).

Gartner, 1997). Ziv et al. (2000) presented a numerical model for sheet
intrusion into preexisting fractures that are oblique to the principal
stresses. They conclude that it appears to be very difcult for a sheet
to follow a preexisting fault unless one or more of the following conditions are met: (1) the fracture is nearly perpendicular to 3; (2) the resolved shear stress on the fracture is small compared to the excess
magma pressure; and (3) the effective ambient sheet-normal stress is
small compared to the rock tensile strength. The same authors suggest
that it may be quite difcult for sheets emerging from midcrustal to
lower crustal depths to follow faults that are oblique to 3. One of the
main observations is related to the fact that rocks are usually fractured
medium with possible pervasive discontinuities. A preexisting fault
with a different orientation from the 12 plane acting at the time of
sheet intrusion is usually surrounded and intersected by other faults
and joints of various lengths. Field data demonstrate in fact that
magma paths can change orientation at short distance when the rock
is densely fractured (e.g. Fig. 25). Thus the presence of these secondary
discontinuities makes it even more difcult for the magma to follow a
misaligned pre-existing fault (Ziv et al., 2000). In spite of this suggestion, a recent work by Garca et al. (2014) demonstrated, by seismological data and numerical modelling, that magma migrated several times

along very diverse paths during the seismo-volcanic crisis of 2011


2013 at El Hierro (Canary Islands). Magma intruded southward and
then northward along a NS-striking pre-existing fracture zone, and
successively migrated along an EW-striking fracture zone. The different intrusions were guided by a combination of two magma pressure
sources located at diverse position and depth, and the presence of the
discontinuities.
6. Plumbing system in the interior of volcanoes
As we look at the structural characteristics of plumbing systems
within volcanoes, we have to consider further parameters that play a
role in dictating the nal conguration. As we have seen in the previous
chapters, the structure of magma conduits is essentially governed by the
combination of preexisting structures and the total state of stress of the
host rock. The latter results from several possible inputs, including fareld tectonics, lithostatic loading and magma force, but as Earth's surface is approached, topography contributes more sensibly to the total
stress eld. In order to distinguish the various components of the stress
sources that contribute to the total stress eld inside a volcanic edice, I
will start analysing the tectonic component vs. the magma component.

108

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

Fig. 24. Examples of changes in the geometry of magma plumbing system at shallow level at the eroded Thverfell volcano (Iceland). A. Sketch and B. photo of stacked sills crosscut by a
vertical dyke that affected all the exposed rock succession. C. Sketch and D. photo of centrally-inclined sheets intruded into the lava succession and, in turn, cut by a series of parallel, subvertical dykes. Both settings reect a change in the geometry of the plumbing system that rst favoured lateral magma transport and then vertical magma upwelling. This can be explained
in terms of a change in the stress state in the host rock promoted by a rotation of the 3 axis from vertical (sill case) or gentle-dipping (inclined sheet) to horizontal.

6.1. Tectonic vs. magmatic components


Field data show that the geometry of magma plumbing system in
volcanoes changes in different tectonic settings and with the variation
of the tectonic stress magnitude. To introduce these concepts, I will
use the examples of Alicudi and Stromboli volcanoes (Aeolian Arc,
southern Tyrrhenian Sea, Italy, Fig. 26): both volcanoes have an
age b 100 ka BP, lie above the same substratum, have a similar size
(rising about 2.5 km from sea bottom), and are mainly of basaltic and
basalticandesitic composition; however, they have different patterns
of dykes (or eruptive ssures that represent their homologous at the
surface). The evolution of the summit, emergent part of Alicudi has
been characterised by several main episodes of cone growth alternating
with lateral volcano-tectonic collapses directed southward (Tibaldi
et al., 2005). During the growth stages, dyking occurred with a radial
pattern, which can be dened as a group of dykes that have a similar
frequency in every direction with regard to the axis of the volcanic edice (Fig. 26A). Several dyke generations, suggested also by some
crosscutting relationships, show the persistence of radial dyking that reects the stability over time of the plumbing system geometry. Oceanographic studies by Calanchi et al. (1995) show that no large faults are
present beneath Alicudi volcano. Also the growth of Stromboli volcano
was repeatedly interrupted by caldera collapses and lateral collapses
(Pasquar et al., 1993; Tibaldi, 1996, 2001). The plumbing system accompanying the various growth phases was strongly dominated by
dyking along a NE-trending weakness axial zone crossing the summit
of the volcano (Fig. 26B). Inside this zone, single dykes strike between
NNE and ENE. Relationships between the unconformity surfaces of the
various volcano growth phases and these dykes indicate that the NEtrending weakness zone persisted over the entire history of the volcano
(Tibaldi, 1996, 2003).
These two volcanoes can thus be considered examples of end members of the possible patterns of plumbing systems; all other conditions
being similar, Alicudi represents the end member where local magma

stress dominates in the volcano, whereas Stromboli reects the control


of far-eld tectonic stress. More in detail, Alicudi developed a radial pattern of dykes that has been caused by a dominating magma force along a
central main conduit zone. Magma exerts an isotropic force on the surrounding rocks that produces hydraulic fracturing followed by radial
dyking (Fig. 27A). In this case the magma ow direction should be horizontal or sub-horizontal and outward-directed. Another explanation
for the radial pattern refers to the distribution of the gravitational
stresses due to the load of the edice. The load controls the trajectories
of the 1 that become subparallel to the volcano slopes while 3 is tangential (Dieterich, 1988). A third explanation is that radial dykes may be
fed from a deeper level of the magma plumbing system at the basement
of the volcano (Geshi, 2008) (Fig. 27). In this third case, the source may
be located beneath the centre of the radial dyke structure where a
pressurised magma chamber with a vertically elongated spindle shape
will induce a zone of compression in which the radial dykes can be
emplaced (Chadwick and Dieterich, 1995; Gudmundsson, 1998);
magma ow should be oriented vertically or obliquely and upwarddirected. This process will be encouraged when the central conduit is
solidied (Porreca et al., 2006). Other examples of radial dykes include
Fernandina volcano at the Galpagos islands (Chadwick and Dieterich,
1995), Summer Coon volcano in Colorado (Poland et al., 2004, 2008),
Kliuchevskoi volcano in Kamchatka (Takada, 1997) and Komochi volcano in Japan (Geshi, 2008).
On the contrary, Stromboli shows a main swarm of rectilinear parallel dykes that crosscuts the summit zone of the volcano and has been active since 100 ka BP. Another two minor zones of dyking have been
active at 8520 ka BP (Corazzato et al., 2008) and b 13 ka BP (Tibaldi,
2001) and will be discussed later. Rectilinear dyke swarms and their
surface expression in the form of aligned volcanic centres and eruptive
ssures are dened as volcano-tectonic rift zones (VTRZ). VTRZ are
usually aligned with the regional tectonic stress eld, i.e. perpendicular
to the 3 and parallel to the horizontal greatest principal stress (Hmax),
which indicates a control on intrusion geometry exerted by far-eld

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

109

Fig. 25. A. Photo and B. sketch of a dyke intruded in a rock affected by several preexisting
fractures with diverse orientations (Asja Peninsula, west Iceland). Note the sharp change
in dyke geometry along the intrusion path.

stresses. Other examples are Askja and Kraa in Iceland (Gudmundsson


and Nilsen, 2006), Erta Ale, Ale Bagu, Dabbahu and Gabho in Afar
(Barberi and Varet, 1970; Wright et al., 2006), Tongariro in New
Zealand (Nairn et al., 1998), Nyiragongo and Nyamuragira in Congo
(Komorowski, 2002). The interpretation of the radial pattern at Alicudi
and of the VTRZ pattern at Stromboli in terms of the control exerted by
magma forces vs. tectonic forces, respectively, is consistent with the
geodynamic setting of the Aeolian Volcanic Arc (Fig. 26C). The western
part of the arc has poorly developed faults and in the late Quaternary
has been subject to low tectonic loading or to compression (Neri et al.,
2003; Goes et al., 2004; Argnani et al., 2007; Di Roberto et al., 2008).
This situation is suitable for magma stress dominating over tectonic
stress. The eastern part of the arc, instead, has been subject to regional
extension with a NWSE trend of 3, creating suitable conditions for
the development of the NESW rift at Stromboli.
Between these two end-members there may be intermediate settings where both magma pressure and tectonic stresses can interact,
giving rise to more complex magma path patterns within volcanoes. A
classical example is the dyke pattern at Spanish Peaks (Colorado,
USA), a site which has been studied for decades (Fig. 28A). Dykes are
arranged radially around the Spanish Peaks stock and have been
interpreted to result from radial stresses caused by the pressurised
stock (Od, 1957). At larger distances from the stock, dykes gradually
bend until they reach an average N80 strike. All more recent works
agree that the overall dyke orientations result from the combination

Fig. 26. Alicudi (A) and Stromboli (B) volcanoes (Aeolian Arc, southern Tyrrhenian Sea,
Italy) have an age b 100 ka BP, lie above the same substratum, have similar size and geochemical characteristics, but have different dyke patterns. Alicudi shows a radial dyking
pattern, whereas Stromboli has a main NE-striking rift zone active since 100 ka ago and
two minor zones of dyking (active at 6436 ka BP and b13 ka BP). The two end-member
dyke patterns reect the effect of a hydraulic force exerted by magma in the case of
Alicudi, and a control exerted by regional tectonics at Stromboli. C. Shows the tectonic settings of the two volcanoes where it is possible to observe that Stromboli lies in an area affected by extension normal to its rift zone.
Data of Alicudi and regional tectonics from Tibaldi et al. (2005); data of Stromboli from
Pasquar et al. (1993) and Tibaldi, 1996, 2001).

110

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

edice as at Askja, Iceland (Gudmundsson and Nilsen, 2006) and Erta


Ale, Tat Ali and several other volcanoes in the Afar region (Barberi and
Varet, 1970; Acocella and Neri, 2009). Sometimes the rift zone may extend beyond the base of the volcano with dykes reaching as far as tens of
km from the volcano perimeter, such as at Kraa in Iceland and at several volcanoes in the Afro-Arabian Rift (Gudmundsson, 1995; Ebinger
and Casey, 2001; Sigmundsson, 2006). This orientation of the regional
stress tensor with a vertical 1 gives way to swarms of vertical, parallel
dykes inside the volcano, a percentage of which are capable of crossing
the summit part of the cone or its anks, giving rise to aligned pyroclastic cones and ssure eruptions. This situation is the simplest from a
point of view of the structure of the shallower magma feeding system,
although it may be complicated by the interaction with topography, as
we will see in the next chapter. Another complication is represented
by the possible development of sills also within volcanoes under an extensional stress eld. In fact, sills have been observed, for example, in
extensional settings associated with active/recent volcanoes such as
Stromboli and Nysiros, and at eroded centres like in the Paiute Ridge
area (Nevada, USA) (Valentine and Krogh, 2006) and in the North Atlantic Igneous Province (Hansen, 2006). It is worth noting that in general
these sills have been observed at lithological boundaries (Fig. 22D), suggesting that their geometry has been controlled by differences in stiffness within the rock succession (Gudmundsson, 1986), as already
mentioned.

Fig. 27. Sketch of possible models of propagation of radial dykes in volcanoes. A. Magma
exerts an isotropic force along the central conduit walls that produces hydraulic fracturing
followed by radial dyking; this is favoured in the case of an open conduit. B. Radial dykes
may be fed from a deeper level of the magma plumbing system at the basement of the volcano and propagate upward; this is more favoured in the case of a closed conduit.
B. Modied after Geshi (2008).

of radially distributed stresses and remotely applied regional tectonic


stresses (Muller and Pollard, 1977; Smith, 1987), although Muller
(1986) observed that there are four diverse dyke populations indicating
a rotation of the remote tectonic stress eld. The remote stresses caused
the reorientation of the distal parts of dykes that reach a strike parallel
to Hmax. A similar pattern has also been obtained by Meriaux and
Lister (2002) by calculation of the trajectories of dyke paths following
mode I opening of fractures. When dykes intercept the volcano slopes,
they produce radial ssure eruptions and fractures in the central uppermost part of the edice, and parallel fractures, eruptive ssures and
aligned pyroclastic cones along two opposite volcano anks (Fig. 28B)
(Nakamura, 1977; Nakamura et al., 1977; Meriaux and Lister, 2002,
and references therein).
6.2. Intrusive sheets and orientation of tectonic stress tensor
6.2.1. Vertical 1
Regional tectonic stresses can propagate and dominate across a volcano with the result that they can guide the geometry of magma paths.
Where the active regional tectonics is characterised by a vertical 1, volcanoes are usually elongated perpendicularly to 3 and show a volcanotectonic rift zone crossing the summit part of the cone or the entire

6.2.2. Vertical 3
Volcanoes can also overlie zones of active compression with horizontal 1 and 2 (for a review see Tibaldi et al., 2010) and there are
some eld and experimental data that indicate which planes are the
prime candidates to host magma paths in such contractional settings.
Magma migration along pre-existing faults is consistent with several
cases found in different geodynamic settings and spanning from evidence at deep crustal level (e.g. Rosenberg, 2004), to near surface
level (e.g. Tibaldi et al., 2010). Below volcanoes, examples from different
contractional settings indicate that the control exerted by ramp and at
structures can occur after the main collision event or during active
thrust movements (syn-kinematic). Examples of the rst case are the
early Proterozoic Chilimanzi granites (Zimbabwe) that were emplaced
50 Ma after a continental collision event (Fig. 29A) (Dirks and Jelsma,
1998). These granites intruded the crust along sills coinciding with
sub-horizontal thrust and along inclined sheets corresponding to
ramp structures. In Nigeria, the Neoproterozoic Rahama monzogranite
pluton was emplaced 60 Ma after a collision event (Ferr et al., 1997,
2002). Magma was emplaced along reverse inclined shear zones
interpreted as thrust ramps with structures of the intrusive rocks suggesting some late syn-kinematic deformations (Fig. 29B) (Ferr et al.,
2012). The Miocene tonalites of the Hidaka Belt (Japan) were emplaced
along both ramp and at structures of a duplex system (Fig. 29C)
(Shimura, 1992; Toyoshima et al., 1994). Regarding examples of synkinematic intrusions, the Palaeozoic Wyangala granites (Australia)
were emplaced shortly after a continental collision event (Paterson
et al., 1990; Tobisch and Paterson, 1990). In particular for the Yarra pluton, the data of Tobisch and Paterson (1990) are consistent with magma
ascent along a ductile, reverse shear zone that was still moving during
magma cooling, and this attests to syn-kinematic emplacement
(Fig. 29D). Halliday et al. (1987) showed that alkaline magmas in NW
Scotland intruded along the Moine thrust zone and documented the
persistence of magmatism, over 30 Ma, in a belt shortened by thrusting.
Mart et al. (1992) found that the Cenozoic magmatism of the Valencia
trough (Spain) is characterised by a rst cycle of Early to Middle Miocene age with calc-alkaline rocks emplaced during thrust development
under a compressional regime. Mkweli et al. (1995) studied the Umlali
Thrust Zone between the granitegreenstone rocks of the Zimbabwe
craton and the granulite-facies rocks of the Limpopo Belt, demonstrating the syn- to post-kinematic intrusion of porphyritic granites.

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

111

Fig. 28. A. At Spanish Peaks (Colorado, USA) dykes (red lines) are arranged radially around the Spanish Peaks stock and have been interpreted as resulting from radial stresses caused by the
pressurised stock (Ode, 1957); at greater distances from the stock, dykes gradually bend and, in a distal position, assume a parallel strike that results from remotely applied regional tectonic stresses (Muller and Pollard, 1977; Smith, 1987). Black lines here represent the calculated trajectories of dyke paths following mode I opening of fractures. B. When dykes intercept
the volcano slopes, they produce radial ssure eruptions and fractures in the central uppermost part of the edice, and parallel fractures, eruptive ssures and aligned pyroclastic cones
along two opposite volcano anks (Nakamura, 1977; Nakamura et al., 1977). (For interpretation of the references to colour in this gure legend, the reader is referred to the web version of
this article.)
A. Modied after Meriaux and Lister (2002).

The analogue experiments carried out by Ferr et al. (2012) to simulate basic and felsic magma upwelling in a contractional setting show
that uids of different viscosities can move along ramp and at

structures, similarly to the previously described eld examples. They


conclude that thrust ramp structures provide an inclined pathway for
magma to ascend along pre-existing mechanical discontinuities. Ferr

Fig. 29. Examples of control exerted by ramp and at structures on magma intrusions. A. Early Proterozoic Chilimanzi granites (Zimbabwe) emplaced 50 Ma after a continental collision
event along ramp and at structures (Dirks and Jelsma, 1998). B. Neoproterozoic Rahama monzogranite pluton emplaced 60 Ma after a collision event along ramp structures (Ferr et al.,
1997, 2002, 2012). C. Miocene tonalites of the Hidaka Belt (Japan) emplaced along both ramp and at structures of a duplex system (Shimura, 1992; Toyoshima et al., 1994). D. Yarra pluton
of the Palaeozoic Wyangala granites (Australia) emplaced along a ductile, reverse shear zone that was still moving during magma cooling (Paterson et al., 1990; Tobisch and Paterson,
1990).

112

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

et al. (2012) also suggest that additional room for magma can be provided by local dilational jogs along the thrust plane. Similar experiments
carried out by Galland et al. (2007a) (Fig. 30A), show that uids can
move upwards along thrust planes at the leading edges of a plateau. In
the experiment, vertical tension fractures formed on the surface of the
plateau, in a direction almost parallel to the imposed shortening. This
situation has been attributed by Galland et al. (2007a) to superposition
of a local load, because of the uplifted area, on the regional stress eld.
At a shallow depth, the stress eld was mainly due to the load of the
uplifted area, so that the greatest stress was vertical, whereas at deeper
levels the stress was mainly due to regional compression, with a horizontal greatest stress. In any case, the vertical tension fractures did not
act as conduits in these experiments.
As regards an example of volcanism and contractional tectonics at
shallow level, at Iwate volcano (NE-Japan) the focal mechanism analysis
of the M = 6.1, 1998 earthquake indicates reverse faulting with the horizontal principal compressive stress axis in an EW direction (http://
hakone.eri.u-tokyo.ac.jp/vrc/erup/iwate.html). NS-striking reverse active faults are present in this area: The activity of one of these, the
Nishine-fault, was estimated to be about 0.7 m/1000 years by the
Japan Active-Fault Research Group. GPS surveys showed a steady and
continuous extension between southern and northern sites of the volcano in 1998 (Miura et al., 2000). Based on these results, it has been suggested that intrusion of an EW-striking dyke occurred at around 10 km
of depth below the summit of the volcano, thus resulting in a magma
path parallel to the direction of 1. At Tromen volcano (Chile, Galland
et al., 2007a), lying above the hanging wall of a coeval west-dipping

thrust fault, most of the eruptions have occurred in the upper part of
the edice. In the central part of the volcano, the magmatic conduits
are in the form of subvertical andesitic dykes, striking almost EW. On
the eastern and southern anks, EW trending alignments of volcanic
domes probably follow underlying fractures (Galland et al., 2007a,b).
Also in this case, dykes intruding the volcano are parallel to the direction
of 1. At Guagua Pichincha volcano, Ecuador (Legrand et al., 2002),
and Miyakejima volcano, Japan (Fujita et al., 2004), where the tectonic
settings are also compressional, seismic data suggest subhorizontal
magma transport at depth, followed by subvertical transport nearer
the surface. Several geophysical data have been collected also at
Mt Redoubt in Alaska, during the 198990 and 2009 eruptions; Lahr
et al. (1994) showed that the location of an earthquake swarm of the
198990 eruption is consistent with a narrow conduit steeply dipping
to the NE (i.e. a NWSE-striking dyke). The eruption of 2009 was immediately preceded by a period marked by volcano-tectonic events indicating a NESW-trending P-axis orientation corresponding to the
horizontal displacement of the conduit walls in the same direction
(Roman and Gardine, 2013), a nding that can be interpret as the possible effect of a NWSE dyke. These NWSE dykes are parallel to the direction of the horizontal, regional 1 (Nakamura, 1977; Nakamura et al.,
1977, 1980; Koehler et al., 2012). A plexus of dykes and sills has also
been recognised beneath the northern ank of Mt Redoubt at deeper
level (Benz et al., 1996). At the active Trident volcano (Alaska),
Wallmann et al. (1990), based on the presence of a set of NWSE ssures crossing the margins of the 1912 Novarupta crater, local platemotion vectors and regional stress orientation, suggest that a NWSE

Fig. 30. Examples of experiments where volcanoes in contractional settings have been modelled. In A, intrusion of magma in the shallower crust along thrust planes. In B, C and D, photos in
plan view (to the left) and fault traces (to the right) of experiments of volcanic cones lying above substrate reverse faults, dipping to the left, with different positions with respect to the
cone: B. cone summit located on the footwall block, C. cone summit located above the surface trace of the substrate fault, and D. cone summit located on the hanging-wall block; E. section
view of the deformation zone inside the cone of case C.
A. Modied after Galland et al. (2007a). E. Modied after Tibaldi (2008).

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

113

feeder dyke propagated from a reservoir beneath Trident volcano to the


Novarupta vent. Anyway, the contemporaneous formation of the Mt
Katmai caldera represents a possible magmatic connection between
the reservoir beneath Katmai and Trident; this connection has been
interpreted in terms of a propagating sill or the failure of septa between
NE-aligned batches of magma (Hildreth, 1987; Wallmann et al., 1990;
Lowenstern et al., 1991).
The experiments of Tibaldi (2008) were mostly aimed at simulating
deformation within a cone lying above a reverse fault, by varying the location of the fault with respect to the cone (Fig. 30BD). Fractures parallel to the contraction direction did not form, probably because the
reverse/thrust faults affecting the substrate are not arcuate in plan
view, as opposed to those simulated in the experiments by Galland
et al. (2007a). In Tibaldi's (2008) experiments, the propagation of the
rectilinear (in plan view) reverse faults from the substrate across the
volcanic cone results in the formation of a steep normal fault zone (F1
in Fig. 30C, D and E) and a reverse low angle fault (F2). This implies
the developing of a local extensional eld with a horizontal 3 parallel
to the general contraction direction in the summit part of the cone.
Summarising the above, all these data suggest that in contractional
tectonic settings, magma ascending can be partitioned between a
deeper transport by inclined sheets and a shallow transport by vertical
dykes. In the crust below a volcano, magma may migrate horizontally
along at structures and upward along ramp structures. Not necessarily
this magma migration may lead to eruptions, but in any case it emerges
that pre-existing contractional discontinuities in the crust my act as preferred magma pathways upwards, even during active compressional
deformation. Inside a volcano, magma upwelling may occur along the
steeper-dipping fault zone F1 in case the reverse fault propagates
through the cone and splits upwards. This magma migration along the
steeper fault is facilitated by the very local extension and is corroborated
by eld evidence of eruptive vents close to or along F1 zones, such as
at Trohunco caldera and Los CardosCentinela volcanic complex
(Argentina). In several other cases, dykes within the volcanic cone are
parallel to the regional 1, and also sill emplacement seems to be a frequent mechanism of intrusion.
6.2.3. Vertical 2
As an example of magma paths in strike-slip tectonic settings
(vertical 2), Pasquar and Tibaldi (2003) studied by eld and analogue
data several volcanoes of the Bicol Peninsula (Luzon, Philippines),
where transcurrent tectonics dominate; they concluded that at depth
magma most likely used the main NW-striking strike-slip regional
faults, whereas at shallower level, the dykes followed fractures subparallel to 1 and sub-perpendicular to 3. A similar change in dyke orientation with depth was also discovered in another transcurrent zone at
Galeras volcano (Colombia, Tibaldi and Romero, 2000). This volcano is
also characterised by the presence of a sector collapse towards the
east within which is the active crater zone (Fig. 31A). The orientation
of this collapse matches exactly the results obtained by means of
analogue modelling by Lagmay et al. (2000), who showed that sigmoidal deformation zones form on the volcano anks close to the underlying fault, along the right side of a right-lateral strike-slip fault, or along
the left side of a left-lateral strike-slip fault. If we compare their analogue model deformed above a right-lateral strike-slip fault, we observe
that the collapse of Galeras volcano occurred in the same position
and orientation (Fig. 31BC). This deformation zone above a strike-slip
fault should be the most prone to develop failures, intrusions and
eruptions.
Holohan et al. (2007), who analysed by analogue modelling the interactions between structures associated with regional tectonic strikeslip deformation and volcano-tectonic caldera subsidence, suggested a
similarity between the roof-dissecting Riedel shears and Y-shears
appearing in their models and the regional strike-slip faults that dissect
the central oors of the Negra Muerta (Riller et al., 2001; Ramelow et al.,
2006) and Hopong calderas, with indication of volcanoes developed

Fig. 31. A. DEM of Galeras volcano (Colombia) with location of sector collapse and main
right-lateral strike-slip faults affecting the cone; the sketch has been rotated for easier
comparison with the analogue model of BC. B. Photo and C. interpretative sketch of analogue modelling of a volcano located above a right-lateral strike-slip fault. Note the
matching between the Galeras example and the experiment.
A. Modied after Tibaldi and Romero (2000). BC. Modied after Lagmay et al. (2000).

inside pull-apart basins (Fig. 32A). According to Holohan et al. (2007),


these fault systems might be regarded as preferential pathways for
the ascent of magma and other uids before, during, or after caldera

114

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

that most dykes conformed to an en-chelon pattern and were affected


by a right-lateral shear component along the intrusion planes during
propagation.
Summing up the above, data suggest that at volcanoes lying in
transcurrent settings, magma ascending can be partitioned at plumbing
systems that have a different structure depending on depth. The deeper
magma transport below a volcano can be dominated by dyking along
the main, rectilinear segment of a strike-slip fault zone, or at associated
releasing bends. The shallow magma transport within a volcanic cone
can occur along dykes oriented parallel to 1 (T structures) or oblique
to 1 (Riedel shears).
6.3. Topographic inuence

Fig. 32. In strike-slip fault zones, volcanism can occur at pull-apart basins (A); at releasing
bend structures (B); directly along rectilinear strike-slip faults (C); and at the tips of the
main strike-slip faults (horsetail structures (D). The most common dyke orientation inside
the volcanoes located within these diverse fault settings is represented by red lines. Stratovolcanoes, shield volcanoes, pyroclastic cones and domes may occur at all the above
types of strike-slip fault structures, whereas calderas are preferentially located within
pull-apart basins. (For interpretation of the references to colour in this gure legend, the
reader is referred to the web version of this article.)

formation. Busby and Bassett (2007) document that the intrabasinal


lithofacies of the Santa Rita Glance Conglomerate record repeated intrusion and emission of small volumes of magma emplaced along a main
strike-slip fault, as frequently found (Bellier and Sebrier, 1994; Bellier
et al., 1999), and along releasing fault bends (sketch in Fig. 32B).
Marra (2001), studying the Mid-Pleistocene volcanic activity in the
Alban Hills (Central Italy), documented that a local, clockwise block rotation between parallel NS strike-slip faults might have generated local
crustal decompression, enabling volatile-free magma to rise from deep
reservoirs beneath the Alban Hills and feeding ssure lava ows.
Chiarabba et al. (2004a), on the basis of a shallow seismic tomography
of Vulcano Island (Aeolian Arc, Italy) observe that at a depth N 0.5 km,
the rise of magma is controlled by NWSE fractures associated with
the activity of the NWSE-striking, right-lateral strike-slip to obliqueslip, TindariLetojanni fault system (Fig. 32C) (Mazzuoli et al., 1995).
At shallower depth (i.e. b0.5 km), the plumbing system of the volcano
is mainly controlled by NS-striking secondary faults, which are perpendicular to the regional 3 and thus are not transcurrent faults but
rather extensional fractures. Also Aydin et al. (1990) observed that at
strike-slip fault zones, magma preferentially rises at the surface along
the associated secondary extensional structures rather than along the
main strike-slip fault segments (e.g. Fig. 32D). Scarrow et al. (1997)
found in Antarctica sets of shoshonitic dykes that were emplaced
along strike-slip faults in late Cretaceous times. Rossetti et al. (2000) illustrate how the effusive and intrusive rocks belonging to the McMurdo
Volcanic Group (Antarctica) were fed by dykes along a crustal-scale,
non-coaxial transtensional shear zone where the strike-slip component
increased over time. By analogue modelling, Corti et al. (2001) showed
that magma is emplaced at depth along faults parallel to the main shear
zone but uprises to the surface along cracks that are orthogonal to the
orientation of the extension. Finally, Eriksson et al. (2014) studying
the lftafjrur volcano (Iceland) by AMS techniques found that dykes
mostly propagated horizontally from a shallow magma chamber within
a transtensive shear zone. Although in other cases in the same Iceland,
AMS results show different directions of magma ow in dykes (Kissel
et al., 2010; Eriksson et al., 2011), the example of lftafjrur shows

As introduced before when discussing the Stromboli example, this


volcano shows other two zones of dyking (Fig. 26B). Between 85 and
20 ka BP, a NS-trending magma feeding zone developed in the SSW
side of the island, interpreted as the effect of a possibly southward extension of the magma chamber (Corazzato et al., 2008). After 13 ka
BP, sheets injected in the NW side of the volcano and along its summit
part, depicting a horse-shoe shape in plan view, open northwestward
(i.e. towards the sea) (Tibaldi, 2001) (Fig. 26B). These sheets follow
the escarpments created by a series of nested sector collapses
(circum-lateral collapse sheets) that affected the northwest volcano
side (Fig. 33), and the most recent failure produced the present shape
of the Sciara del Fuoco depression. The sheets' strike and dip tend to rotate along the margins of the sector collapse zone and their geometry is
rendered graphically in Fig. 34A, showing dyke attitude compared with
the local geometry of the collapse scarp as the angle between the two
dip directions. This indicates that all these dykes, with only two exceptions, intruded along planes striking parallel to the collapse scarp. Moreover, most dykes dip into the same direction as the collapse scarps, i.e.
towards the interior of the collapse depression. Along the uppermost
part of the collapse zone, the local orientation and location of the collapse scarp coincides with the NE-trending volcano-tectonic rift zone
(red zone in Fig. 33A), and here the dykes strike NESW. It is also
worth noting that in a section perpendicular to the sector collapse
scarp, dyke frequency is maximum at the scarp and decreases outward,
with dykes nally disappearing at 500 m of distance (Fig. 33B). All
sheets have been sampled (Fig. 33A); the geochemical characteristics
of the sheets cropping out along the southern part of the Sciara del
Fuoco depression are represented in Fig. 34B, and those located along
the northern and eastern part of the Sciara del Fuoco in Fig. 34C
(Corazzato et al., 2008). A total of 33 sheets (triangles) have the same
characteristics of the lava deposits of the Neostromboli volcano growth
phase and thus can be regarded as the shallow feeding system of that
volcanic phase. These sheets were emplaced following the rst huge
sector collapse towards NW that occurred about 13 ka BP. This sector
collapse clearly reects a major change in the shape of the volcano;
huge morphological changes inuence the orientation of the stress
due to the reorganisation of the gravity force following a new distribution of rock masses (Voight and Elsworth, 1997; Tibaldi, 2001). Rock
mass removal due to lateral failure produces: (1) debuttressing of the
cone ank; (2) new local directions of extension; (3) the formation of
a fracture pattern around the depression amphitheatre; and (4) the formation of lithological boundaries. These aspects have deserved an
increasing attention in recent times, in terms of new eld data, numerical modelling and analogue experiments. Aspects 1) and 2) are related
to the inuence of topography on dyke paths, whereas items 3) and
4) are more related to rock mechanics.
Regarding issues 1) and 2), results of nite difference numerical
modelling of the state of stress within the Stromboli volcano before
the rst huge sector collapse towards NW and after the failure, show
important changes in terms of both orientation of stress tensor and
stress magnitude (Fig. 35) (Tibaldi et al., 2008b). In particular, a closer
view of the area of the sector collapse shows the marked reorientation

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

115

Fig. 33. A. Aerial oblique view of the Sciara del Fuoco depression at Stromboli (Italy) generated by a series of lateral collapses, with all the intrusive sheets (yellow, with relative number of
sample), emplaced along the scarps. The main NESW volcano tectonic rift is represented by the red zone. B. Photo taken along the coast of Stromboli showing a section view of the lateral
collapse escarpment and intrusive sheets; the diagram quanties the frequency of dykes moving from the collapse depression outward. (For interpretation of the references to colour in
this gure legend, the reader is referred to the web version of this article.)

of the stress tensors close to the collapse scarps, with the minimum
compressive stress reaching a trend normal to the collapse surface.
This local conguration favours the intrusion of sheets parallel to the
collapse scarps and dipping steeply towards the depression, consistent
also with eld data summarised in Fig. 34A. Other examples of the inuence of local cone slope on dyke geometry have been observed at
Mt. Etna, Italy, during historical dyke-fed eruptions (McGuire and
Pullen, 1989; McGuire et al., 1990, 1991; Tibaldi and Groppelli, 2002).
During the 19781979 and 1983 magmatic events, feeder dykes propagated from the summit zone of Mt Etna towards the western wall of the
Valle del Bove. Geophysical data and structural data provided insights
into the changes in azimuth of feeder dykes at shallow depths, prior to
their intersection with the surface, as they deviated from an originally
southeasterly propagation direction, into a southerly one, parallel to

the western steep scarp of the Valle del Bove, at a distance of few tens
of meteres up to 500 m from the rock cliff (McGuire and Pullen, 1989;
McGuire et al., 1990, 1991). Also during the JulyAugust 2001 magmatic
event, eld data indicated a deviation of the strike of a shallow dyke that
followed the local topography (Tibaldi and Groppelli, 2002). A conguration similar to Stromboli has also been observed for the dykes intruded along lateral collapse scarps at Tenerife, Canary Islands (Gottsmann
et al., 2008; Delcamp et al., 2012), TahitiNui (French Polynesia)
(Hildenbrand et al., 2004), and Piton de la Fournaise (Reunion Island)
(Peltier et al., 2009). At Fogo Island (Cape Verde archipelago) a recent
NS swarm of dykes, emplaced below an older collapse structure,
show an en-chelon geometry associated with a seaward displacement
of the eastern ank of the volcano (Day et al., 1999). In the presence of a
high scarp, such as the one between the top of Vesuvius and Mt Somma,

116

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

Fig. 34. A. Geometry of intrusive sheets surveyed along the lateral collapse escarpments of
Stromboli volcano (Italy) around the Sciara del Fuoco depression, expressed as number of
dykes (Y axis) versus the angle between the dip direction of dyke and the dip direction of
the collapse scarp (X axis). B. Geochemical characteristics of the intrusive sheets cropping
out along the southern part of the Sciara del Fuoco depression, and C. along the northern
and eastern part of the Sciara del Fuoco depression. A total of 33 sheets (triangles) belong
to the Neostromboli volcano growth phase (label NEOSTR), 2 sheets belong to the previous Vancori growth phase and crop out far away from the Sciara del Fuoco depression
(see Vancori label in Fig. 32A for the location of these two sheets) and the other 3 sheets
were emplaced during post-Neostromboli growth phases.
A. Modied after Tibaldi (2003). C. Modied after Corazzato et al. (2008).

the difference in topography can create a zone of buttressing that hinders lateral dyke propagation across the scarp under ordinary excess
magmatic pressures (Acocella et al., 2006).
Also several analogue models indicate the differences in sheet conguration based on the volcano morphology, with the results shown
in Fig. 36. In the case of an intact cone, dykes propagate parallel to the
Hmax and perpendicular to the Hmin, which are radial and concentric,
respectively (Fig. 36, upper part) (McGuire and Pullen, 1989). If a lateral
collapse depression is cut into the cone, later intrusions follow
the scarps of the collapse depression if the initial intrusive source is located near the scarps, interpreted as the effect of unbuttressing and the
consequent reorganisation of the local stress eld (Fig. 36, middle)
(Acocella and Tibaldi, 2005). Similar results have been obtained for a
volcano with an unstable ank that has not collapsed yet (Walter and
Troll, 2003). If the initial intrusion is located below the collapse depression, the resulting dyke will propagate along the collapse axis, i.e. along
the mean points of the collapse in map view (Acocella and Tibaldi,
2005).
Regarding the other effects of the rock mass removal due to a lateral
failure, which are 3) the formation of a fracture pattern around the depression amphitheatre and 4) the formation of lithological boundaries,
these favour the capture of propagating sheets due to higher permeability and stoping effect, respectively. For example, a eld analysis conducted along the margins of the sector collapse that affected Ollague
volcano (ChileBolivia border) showed the very high density of fractures formed along the collapse scarps (Figs. 37AB) with respect to
the same lava deposits observed far away from the collapse zone
(Fig. 37C). These fractures may result from brittle deformation during
the sliding of the collapse blocks, as well as from debuttressing effect
due to rock mass removal. Preferential dyke intrusions along fractures
parallel to collapse scarps have been observed also at Piton de la
Fournaise (Reunion Island) by Letourneur et al. (2008). Once the
amphitheatre depression created by a sector collapse is inlled by successive volcano growth, a major lithological boundary may form at the
interface between the pre- and post-collapse deposits. In the case of a
strong difference in rock stiffness at this interface, an ascending dyke
can be diverted and become parallel to the lithological/mechanical
boundary.
In the case of elongated volcanoes, topography can also play a fundamental role. In very large volcanoes like at the Hawaiian islands, it has
been proposed that gravity forces can control the orientation of rectilinear dyke swarms that tend to be parallel to the volcano major axis (Fiske
and Jackson, 1972), once space for dyking is provided by slip of the volcano ank along deep faults (Dieterich, 1988). At some elongated volcanoes, anyway, the dyke pattern is represented by a central rectilinear
volcanic rift zone passing outwards into fan-shaped dyke systems at
the two opposite volcano anks (Fig. 36, bottom part), dened as diverging volcano-tectonic rifts (Tibaldi et al., 2014). In their pioneering
analogue experiments, Fiske and Jackson (1972) suggested that this geometry might be linked to the elongated shape of a volcano. Betterscaled models and eld data recently suggested that the formation of
these diverging rifts is not specically linked to substrate lithology and
mechanical behaviour. Volcanoes with diverging rifts have typical elongation (major/minor axis) b 0.88 and V N 10 km3 (mostly N300 km3),
and the central rift zone is normal to the regional Hmin (Tibaldi et al.,
2014). If the regional Hmin is oblique to the volcano elongation axis,
dyke geometry in the edice axial zone is controlled by elongation
and thus by local gravity 3, but dyke strike becomes perpendicular to
Hmin when dykes intrude the more external areas of the volcano. If a
dyke is injected under volcano anks with slope inclination b 50, at
the edice terminations magma paths diverge outwards and crosscut
slopes at high angle (Fig. 36). These authors point out the diverging volcanic rift system is an underestimated structural pattern that can be
more common than previously recognised. For example at Mt Etna the
diverging rift system is superimposed on other dyke patterns, like
the West rift, lying above hidden faults (Mattia et al., 2007) and the

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

117

Fig. 35. Results of nite difference numerical modelling by FLAC of the state of stress within the Stromboli volcano before the rst huge sector collapse towards NW and after the failure,
computed along a NW-striking section. On the left, the sections show the complete pre- and post-collapse stress congurations (upper and lower sections respectively), in terms of both
vertical stress contours and principal stress tensors (negative values indicate compression). A strong reorganisation in the stress distribution occurs, in terms of both magnitude and orientation. The right sections at a closer view show the marked reorientation of the stress tensors close to the collapse scarps, with the minimum compressive stress normal to the collapse
surface. This local conguration favours the intrusion of dykes parallel to the collapse scarps and dipping at high angles towards the depression, consistent also with eld evidence.
Modied after Tibaldi et al. (2008b).

dykes surrounding the Valle del Bove depression, whose geometry is


interpreted to have been guided by the depression debuttressing
(McGuire and Pullen, 1989). More clear eld examples of diverging
rifts can be seen at El Hierro (Canary Islands) (Gee et al., 2001),

Mt Cameroon (Favalli et al., 2012), Miyakejima (Nakada et al., 2005)


and Fuji (Takada, 1997) in Japan, Kliuchevskoi in Russia (Ozerov et al.,
1997), Nyamuragira in Congo (Hayashi et al., 1992), and Newberry in
the USA (MacLeod et al., 1995).

Fig. 36. Sketch of the main congurations of intrusive sheets at volcanoes under the inuence of topography and indication of the stress eld guiding the geometry of magma paths. Upper
part of the gure: circular intact cone; middle part: circular cone with a sector collapse depression; lower part: elongated volcano.

118

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

Fig. 37. Photos taken at the Ollague volcano (ChileBolivia border). A. Field view of the depression left by the Holocene sector collapse; B. photo taken close to the collapse scarp showing
the pervasive fracturing of the lava ows; C. photo of the same lava ows taken far away from the collapse depression.

The experiments of Tibaldi et al. (2014) also suggest that in the case
of steep slopes, the topography of a volcano exerts a dominant control
on magma path orientation, determining sheet rotation until they
become parallel to the local edice slope. The steep slope actually represents a debuttressing that produces a local 3 oriented perpendicularly to
the topography. Regarding the threshold value of slope steepness, these
experiments show that when sheets approach a slope dipping at 35,
they propagate perpendicularly to the slope. This has been observed
also with slopes as steep as 45 by Walter and Troll (2003). Hence, the
transition to slope-parallel sheets may occur with a slope angle N 45
50, but more research is needed to better constrain this value.
Walter et al. (2006) investigated the geometry of shallow magma
paths related to the deformation of large volcanic edices built by the

coalescence of adjacent cones, by reproducing the spreading of an edice composed of overlapping volcanoes. Their results suggest that
spreading edices of similar age that partially overlap tend to develop
a rift zone approximately perpendicular to the boundary of both volcanoes, and this causes the two edices to grow together and develop an
elongated topographic ridge. However, in the case of partially overlapping volcanoes of different ages that spread at different rates, the
resulting rift will be parallel to their boundary, causing the two edices
to structurally separate from each other.
Regarding general topographic effects, Gaffney and Damjanac
(2006) suggested that magma uprising along a fracture that runs from
a highland to an adjacent lowland, can be diverted away from the highland towards the lowland. This is in line with the results of Kervyn et al.

Fig. 38. Models of deformation of the surface topography above a shallow propagating dyke tip: A. uplift zones on each side of the dyke tip and no uplift above the dyke tip with propagation of normal faults and ssures at the surface. B. Uplift of the zone above the dyke tip with propagation of reverse faults and ssures at the surface (Gudmundsson et al., 2008). C. Photo
and D. sketch of the dramatic effects caused by the very shallow intrusion of a dyke in northern Iceland during the Kraa res of 19751984.
A. Modied after Mastin and Pollard (1988).

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

(2009) who investigated, by analogue modelling, the control of volcano


load on magma ascent and vent location. Their results show that if a
central conduit is blocked, the volcano loading stress favours eruption
of rising magma away from the volcano summit. Folch et al. (2000)
studied the numerical solutions of ground displacement caused by an
overpressurised magma chamber, nding that correct calculation and
interpretation of displacements can be reached only taking into consideration also topography. A recent paper by Maccaferri et al. (2014) developed numerical modelling of dyke propagation near the shoulders
of rift valleys. They found that the location of the shallow magma path
is governed by the competition between gravitational unloading pressure and tectonic stretching. When gravitational unloading dominates,
ascending dykes are diverted towards the rift shoulders, whereas
when tectonic stretching dominates, shallow dykes propagate within
rift. This means that rifts with mild topographic expression may concentrate dyke paths within the graben valleys.
7. Plumbing systems and surface deformation
Although it has been widely recognised that magma intrusion may
create regional uplift (McKenzie, 1984; Saunders et al., 2007; Ernst,
2014, and references therein), only recently the relations between shallow magma emplacement and local surface deformation eld have become clearer. Observations during episodes of dyke emplacement at
subaerial volcanic rift zones have shown the relation between dyking
and slip increment along existing faults and ssures and the formation
of new ssures (Abdallah et al., 1979; Bjornsson et al., 1979; Pollard
et al., 1983; Wright et al., 2006). More in detail, direct eld observations
(Larsen et al., 1979) and continuously recording measurements
(Hauksson, 1983), conrmed by more recent studies (Belachew et al.,
2011, and references therein), showed that fault and ssure motions
at the surface are correlated in time with the arrival of the earthquake
swarm front, several hours after the initiation of dyke propagation.
This suggested that faulting and ssuring were triggered by dyke intrusion (Rubin and Pollard, 1988). Above the propagating dyke tip, stress
concentration induces compression in the surrounding host rock and
consequent faulting up to, or beyond, the dyke plane at depth; this induces a typical deformation pattern at the surface given by: (1) uplift
of a region several kilometres across with a maximum value at the
edges of a central graben located above the dyke tip; (2) an abrupt transition from graben ank to graben oor; and (3) subsidence of a
concave-upward graben oor by one to several times the amount of
ank uplift (Rubin and Pollard, 1988). At the surface, near the dyke,
this process is expressed by two possible models: (1) uplift zones on
each side of the dyke tip and no uplift above the dyke tip (Fig. 38A)
(Pollard et al., 1983; Mastin and Pollard, 1988; Rubin and Pollard,
1988) and (2) uplift of the zone above the dyke tip (Fig. 38B)
(Gudmundsson et al., 2008; Abdelmalak et al., 2012). The effects can
be dramatic, causing heavy deformation as shown in a eld example
(Figs. 38CD) from northern Iceland, regarding one of the dyking and
rifting events that took place during the Kraa res of 19751984. In
model 1) the stress is not transferred directly above the dyke tip but it
is concentrated at its two sides, creating two areas with vertical ssures
and normal faults dipping towards the dyke. Two maxima of surface
ground breaking may occur in the case of a vertical dyke, one on each
side of the dyke plane, and are separated by a distance roughly equal
to twice the depth of the dyke top (Mastin and Pollard, 1988). An inclined sheet would produce asymmetrical uplift of the two side zones.
Subsidence above the dyke tip is not explained by the dyke alone, but
it requires the development of a graben bounded by normal faults that
extent down to the dyke tip level or even deeper. Once the propagating
crack in front of a dyke tip cuts the surface, the relative downward displacement over the dyke changes to upward displacement (Pollard
et al., 1983). For a layered crust with weak contacts, the stress magnitude and conguration above a propagating dyke may be more complicate and the straightforward inversion of geodetic surface data may lead

119

to imprecise results on the geometry and depth of the dyke


(Gudmundsson, 2003). Model 2) is valid only in the case of preexisting graben structures above a propagating dyke, and is more effective for layered host rocks. The effects of the intrusion are to open preexisting normal faults and, later on, to close the lower sectors of the
faults and encourage reverse slip along them. If the magma front has
propagated up into the graben, dyke emplacement uplifts the crustal
block between the boundary faults, forming a horst (Gudmundsson
et al., 2008).
Sill emplacement also requires important deformation of the host
rock that depends on the stiffness of the rock succession and the thickness, velocity and structure of the intrusion. Frequently, sills are saucershaped and consist of an axis-symmetrical at inner horizontal intrusion connected outward and upward to transgressive inclined sheets,
ending in a at outer sill (Planke et al., 2005; Polteau et al., 2008;
Galland et al., 2009; Galerne et al., 2011; Gudmundsson and Ltveit,
2014). Some authors proposed that saucer-shaped sills form along the
level of magma neutral buoyancy and the feeder dyke is located at one
side of the saucer structure (Fig. 39A) (Bradley, 1965; Francis, 1982;

Fig. 39. Existing models of saucer-shaped sill and laccolith emplacement mechanisms:
A. sill emplacement controlled at the level of neutral buoyancy (LNB): Sills are fed laterally
from one part of the outer sill; B. sill emplacement along horizontal discontinuity: Sills are
fed radially from the central inner sill. C. Gilbert's (1877) model for laccolith intrusion with
rigid strata and faulted periphery. D. Laccolith with continuous exure of overburden strata. E. Model of symmetric laccolith-induced deformation about intrusive plane. F. Mixed
exure-fault deformation at a laccolith periphery, inclined sheet climbing may initiate at
forced fold hinges or at a fault.
A. Modied after Bradley (1965) and Francis, 1982. B. Modied after Malthe-Srenssen
et al. (2004). D. Modied after Koch et al. (1981). E. Modied after Koch et al. (1981).
F. Modied after Thomson (2007).

120

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

Chevallier and Woodford, 1999; Goulty, 2005), whereas others claim


that the feeder dyke is located below the central part of the saucer structure (Fig. 39B) (Pollard and Johnson, 1973; Hansen et al., 2004; Galerne
et al., 2011 and references therein). In some large sill complexes it has
been observed that individual saucer-shaped sills are interconnected
to other overlying or underlying sills. These connections may represent
a model of plumbing system where magma does not migrate along vertical feeder dykes but moves along horizontal layers until it breaches the
rock succession along inclined sheets, and then rotates again to reach
horizontal attitude (Muirhead et al., 2012): This pattern resembles the
ramp and at structure of compressional tectonics. These magma
feeding structures have been named interconnected sill-feeding-sill
networks (Cartwright and Hansen, 2006). The rst models of host
rock deformation were proposed for the gradual growth of a single sill
into a laccolith. In case of rigid strata below the intrusion plane, two
main models suggest asymmetric deformation of the host rock where
the intrusion is compensated by uplift of the overburden; in one case
peripheral continuous faults accommodate the intrusion (Fig. 39C)
(Gilbert, 1877), and in the other case the sill/laccolith emplacement is
accommodated by peripheral bending of strata (Fig. 39D) (Koch et al.,
1981). Another model shows a laccolith-induced deformation that is
symmetric about the intrusive plane (modied after Koch et al., 1981).
More recent studies consider a possible mixed mode of deformation
by exural folding and faulting (Fig. 39E); the viscous dissipation of
magma along the length of sills may enable them to propagate faster
and induce non-elastic deformations in the surrounding rocks. This creates the conditions for faulting of the rock succession at the peripheral
zones of the sill and, eventually, inclined sheet climbing may initiate
at forced fold hinges or at a fault (Thomson, 2007). Since, in many instances, eld data indicate that laccoliths lack internal discontinuities,
suggesting that magma was injected as a single pulse or a series of
quickly coalescing pulses (Roni et al., 2014 and references therein), a
large amount of recent studies focused on the growth of a single sill
both by numerical computation (e.g.: Zenzri and Keer, 2001; Zhao
et al., 2008; Michaut, 2011 and references therein) and analogue experiments (e.g.: Dixon and Simpson, 1987; Roman-Berdiel et al., 1995). An
interesting case is represented by the laccoliths of Elba Island (Italy) that
also grew without internal discontinuities, suggesting continuous feeding of the magma (Westerman et al., 2015). These laccoliths are one
above the others but emplaced as separate sheets without coalescing.
They failed to form a composite pluton and instead gave rise to a
Christmas tree geometry, with each magma batch that created its
own room by lifting the rock overburden (Westerman et al., 2015).
Other recent eld studies indicate that, instead of growing in thickness or ramping upward, sills can give rise to stacked piles of horizontal
to sub-horizontal sheeted intrusions (Boudier et al., 1996; Menand,
2008; Tibaldi and Pasquar, 2008; Annen, 2009). The rst sill frequently
intrudes at a discontinuity given by the contact between two rock layers
with different stiffness, followed by a second sill that is emplaced at the
contact with the previous one. The multiple intrusion process may end,
as for example was the case at Maiden Creek (Horsman et al., 2005),
where a small intrusion was formed. However, if enough magma is
fed from below, the process may go on through the emplacement of
several stacked sills generating a sheeted laccolith (Menand, 2008;
Tibaldi and Pasquar, 2008; Tibaldi et al., 2008c). In the case of a single
sill, some InSAR-based modelling indicates that the emplacement is accompanied at the surface by uplift of circular to elliptical areas. For example, modelling of the 1994 unrest episode at Eyjafjallajokull
volcano (Iceland) suggests that a sill intrusion induced deformation
at the surface in the line of sight direction of the Radar satellite
(Pedersen and Sigmundsson, 2004). The interferometric fringes highlight a circular deformation pattern (Fig. 40A) that is consistent with
the modelling of an about 0.36-m-thick sill source, intruded at a depth
of 4.56.0 km (Fig. 40B); however, the authors acknowledge that further investigations are necessary to better constrain the data. At Yellowstone caldera (USA), since 1996 an expanding sill with a length of about

20 km intruded at a depth of 1418 km and produced about 6 cm of uplift of an elliptical area in plan view with an axis of more than 20 km
(Fig. 40C) (Wicks et al., 2006). The modelling of interferometry and
GPS data suggests the presence of a sill gently dipping towards NNW
that explains the asymmetric shape in plan view (Fig. 40D). Anyway,
also in this case prudence should be used since, based on eld examples
elsewhere, the large dimension of the sill in the horizontal plane does
not t with the very small uplift.
8. Plumbing systems at calderas
The plumbing systems at calderas are here reviewed separately due
to the complex structure of calderas and their marked morphological,
structural and dynamic differences in comparison to intact volcanic
cones. Several studies have focused on calderas all over the world, and
a good number of review papers have been recently published dealing
with: the structure of the caldera (Cole et al., 2005; Acocella, 2007),
the tectonic settings (Hughes and Mahood, 2011), the petrographic
characteristics of the caldera magma reservoir (Cashman and
Giordano, 2014), the characteristics of the ring faults (Geyer and
Mart, 2014), and caldera subsidence in the specic extensional tectonic
setting (Carlino et al., 2014). Here I focus on the geometry of intrusions
at calderas.
Calderas can be essentially associated with basaltic, peralkaline, andesiticdacitic and rhyolitic magmas. Basaltic calderas are linked to
shield volcanoes such as those at hotspots (e.g.: Hawaii and Galpagos
islands) (Simkin and Howard, 1970; Decker, 1987), or in other settings
such as convergent margins (e.g. Masaya Volcano in Nicaragua,
Williams and Stoiber, 1983). Peralkaline calderas are typically associated with zones of rifting (Cole et al., 2005) like at Pantelleria (Mahood,
and Hildreth, 1986) and Mayor Island, New Zealand (Cole, 1990). Andesiticdacitic calderas are mostly related to subduction zones like in
the case of Krakatau, Indonesia (Self and Rampino, 1981), and Santorini,
Greece (Druitt et al., 1999). Rhyolitic calderas are characterised by huge
volumes of erupted pyroclastic products and usually large collapse depressions. Examples comprise the Valles caldera, New Mexico, USA
(Smith and Bailey, 1968), the Cerro Galan, Argentina (Francis et al.,
1978), and the Campi Flegrei, Italy (Barberi et al., 1991).
A caldera forms mainly after a sufcient amount of magma has been
withdrawn during an eruption, and the consequent underpressure in
the magma chamber causes its roof to collapse (Lipman, 1997). Another
possibility is that overpressure within the magma chamber induces rst
doming and fracturing of the overburden, followed by magma migration and caldera collapse (Gudmundsson et al., 1997). Although the
rst model has been applied to several cases, it has also been criticised
in view of the fact that the underpressure in the magma chamber should
hinder the continuation of the eruption until the critical value to trigger
the collapse has been reached; another factor is the inconsistency, at
some calderas, between the volume of erupted material and the collapse volume, and the contradiction between the dip direction of the
theoretical caldera faults in comparison to eld data (Gudmundsson
and Nilsen, 2006).
From the point of view of the possible structure of a caldera, this may
result from different evolutionary processes that involve single or multiple collapses, each with possible different geometries, and also possible phases of resurgence. A categorisation was attempted by Lipman
(1997) who suggested ve end-member structural styles: Plate or piston, piecemeal, trapdoor, downsag and funnel (Fig. 41). Plate/piston collapse involves the subsidence of a coherent block of rock along a ring
fault (Fig. 41A), like at Creede caldera (Steven and Lipman, 1976). The
piecemeal case refers to a caldera where the collapsing block is
subdivided into a series of discrete secondary blocks (Fig. 41B) such as
the Scafell caldera, U.K. (Branney and Kokelaar, 1994). The trapdoor
style corresponds to an asymmetric collapse with block subsidence on
one side where a ring fault develops, and tilting on the hinged other
side (Fig. 41 C). Examples of the above are the Valles caldera, U.S.

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

121

Fig. 40. A. Interferometric fringes of the deformation eld from 06/09/1992 to 15/08/1995 that accompanied the 1994 unrest episode at Eyjafjallajokull volcano (Iceland) in the line of sight
direction of the Radar satellite, and B. articial fringes obtained by modelling a sill source, about 0.36-m-thick intruded at a depth of 4.56.0 km. C. Interferometric fringes of the deformation eld at Yellowstone caldera (USA) from the summer of 1996 to the summer of 2000 characterised by an expanding sill, and D. result of modelling of the source as a sill gently dipping
towards NNW.
B. Modied after Pedersen and Sigmundsson (2004). D. Modied after Wicks et al. (2006).

(Heiken et al., 1990), and Kumano caldera, Japan (Miura, 1999).


Downsagging takes place where ring faults do not form or do not
reach the surface (Fig. 41D), and the overburden deforms by bending
without fracturing. In this case there are no distinct caldera walls and

the caldera oor dips gently towards the collapse centre. Bolsena
(Italy) is an example of a downsag caldera (Walker, 1984), as well as
the Gross Brukkaros system in Namibia (Stachel et al., 1994). Funnel calderas occur through the complete destruction of the caldera oor that is

Fig. 41. Different styles of caldera collapse. (A) Plate or piston, (B) piecemeal, (C) trapdoor, (D) downsag, (E) funnel with complete destruction, and (F) funnel with concentric ring faults.
See text for details.
Redrawn after Lipman (1997).

122

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

Fig. 42. Relations between the size of magma chamber and of caldera. (A) In the case of caldera forming by overpressure, an expanding chamber with a sill-like shape (lower high/width
ratio) may induce steeply-dipping ring faults at the chamber edges that will guide the caldera collapse. The caldera and magma chamber size should correspond. (B) A laccolith-like shape
(higher H/W) can favour the nucleation of shear stress in correspondence of specic points of maximum curvature, inducing the development of ring faults. (C) An overpressurised magma
chamber with a laccolith-like shape may also produce arching of the overburden with development of local extension at the extrados and caldera occurrence. In cases (B) and (C) the
caldera width is much lower than the magma chamber size.
Redrawn after Aizawa et al. (2006).

Fig. 43. (A) Crater Lake caldera and (B) related model of magma propagation from the growing chamber through a series of steeply-dipping dykes diverging upward from the sides of the
magma chamber. (C) Toba caldera (northern Sumatra) and (D) related model of stacked sills linked by vertical dykes, superimposed on the distribution of the horizontally polarised shear
wave speed (shown above 20 km in depth), dashed line = the low-velocity area below the caldera that might have been affected by the super-eruption of 74 ka ago.
A. and B. After Karlstrom et al. (2015). C. and D. After Jaxybulatov et al. (2014).

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

broken into series of blocks that chaotically subside deeper towards the
collapse centre (Fig. 41E), or by blocks that are displaced downward
along concentric ring faults that become deeper towards the zone of
deepest collapse (Fig. 41F). An example of the latter type is the Guayabo
Caldera, Costa Rica (Halinan, 1993).
Although several publications have been dedicated to the study of
the structure and dynamics of calderas, a lot of work is still required to
clarify the geometry of their plumbing system. The necessity of further
investigations on the structure of sub-caldera magmatic systems is
even more important if we take into consideration the possible connection between location, dimension and shape of a magma chamber and
the related caldera; it has been shown that in the case of caldera forming
by underpressure of the magma chamber, the width of the caldera depression is similar to the diameter of the underlying chamber (Aizawa
et al., 2006 and references therein). In the case of caldera forming by
overpressure, an expanding chamber with a sill-like shape, i.e. with a
lower high/width (H/W) ratio, may induce the development of
steeply-dipping ring faults above the chamber edge that will guide the
caldera collapse (Fig. 42A). As a consequence, also in this case the caldera size and magma chamber size should correspond. In the case, instead,
of an overpressurised magma chamber with a laccolith-like shape
(higher H/W), the shape of the roof of the chamber may favour the nucleation of shear stress in correspondence of specic points of maximum curvature, inducing the development of ring faults (Fig. 42B)
(Mandl, 1988). A laccolithic shape of the magma chamber may also result in the arching of the overburden with the development of local extension at the extrados; the combination of this extension with the
upward-directed magma push may lead to the development of a caldera
(Fig. 42C) (Aizawa et al., 2006). In both cases, the caldera width is much
smaller than the magma chamber size.
Field data are scarce in comparison to other plumbing systems and
are mostly collected at eroded ancient calderas, like for example Scafell
(Branney and Kokelaar, 1994; Kokelaar et al., 2007) and Glencoe in the

123

U.K. (Moore and Kokelaar, 1998; Troll et al., 2002), and at younger calderas such as Rallier-du-Baty in Kerguelen Archipelago (Bonin et al.,
2004). At modern calderas, data are mostly of the geophysical and subordinately geological type, like at the Campi Flegrei (Piochi et al., 2014),
Miyakejima (Geshi et al., 2002; Geshi, 2009), Piton de la Fournaise
(Michon et al., 2007), Rabaul (Mori and McKee, 1987), Sierra Negra
(Jnsson, 2009), and Nyamulagira (Wauthier et al., 2013). Based on several eld and geophysical data, the classical model of caldera plumbing
system comprises a magma chamber and a system of transient conduits
that develop during caldera forming. This Standard Model as termed
by Gualda and Ghiorso (2013), is represented by a single, long-lived,
usually horizontally-elongated magma chamber, characterised by dominating melt. This model has been recently applied, for example, to Mt
Mazama (Oregon Cascades, U.S.), a long-lived (400 ka) volcanic centre
whose activity culminated with the 50 km3 climactic eruption of
7.7 ka BP that produced the Crater Lake caldera (Fig. 43A) (Bacon and
Lanphere, 2006; Wright et al., 2012). Based on eld data and numerical
work by Karlstrom et al. (2015), a centralised oblate spheroidal magma
chamber was responsible for the climactic eruption. The chamber
growth, fed by a deep magma inux, culminated in the caldera formation. Based on the distribution in space and time of the eruptive vents
preceding the Crater Lake caldera failure, Karlstrom et al. (2015) propose
that magma propagated from the growing chamber through a series of
steeply-dipping dykes diverging upward from the sides of the magma
chamber (Fig. 43B). This model, more related to silicic magmas, suggests
the focusing of dykes all around the perimetry of a growing magma
chamber, providing a plausible mechanism for clustering of eruptions
around the Mazama centre. The eruptions at the central main volcano
are fed by a vertical dyke that drains the chamber and releases accumulated overpressure; once this central event has taken place, eruptions
could return to proximal areas as the effective chamber volume and related deviatoric stresses are reduced. This model differs from the geometrical distribution of inclined sheets that propagate above a at,

Fig. 44. A. Model proposed for the plumbing system of resurgent calderas where several inclined sheets propagated upward from the magma chamber, and most became arrested at layer
contacts with contrasting mechanical properties. A number of inclined sheets intersected the ring faults that have been previously developed during the caldera collapse, and became
deected up along the fault to form multiple ring dykes. (B) Model of ring dykes directly linked with the underlying magma chamber.
A. After Browning and Gudmundsson (2015). B. After Saunders (2005).

124

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

mac magma chamber, as introduced in the previous sections (compare


with Figs. 10 and 17).
In recent years, there has been a growing evidence that sub-caldera
plumbing systems may be made of a series of superimposed sills interconnected by inclined sheets and vertical dykes. In particular, the existence of a complex magma storage region is postulated, where the
magma reservoir is composed of multiple melt lenses within a crystal
mush, or by multiple sills stacked into a completely crystalline rigid
framework (Cashman and Giordano, 2014). For example, in the deeply
eroded calderas of the southern Rocky Mountains, apart from the main
plutons, relatively minor intrusions near the calderas are characterised
by megadykes, up to some hundreds of metres wide, and tabular silllaccolith bodies (Lipman, 2007). In general, the presence of several
stacked tabular interconnected magma bodies can reconcile several observations of crystals, carried by the transporting melt, which have been
stored at a range of pressures and temperatures (Cashman and
Giordano, 2014). These crystals have been erupted when the tabular
bodies at different depths have been tapped. A modern example
comes from the felsic Toba caldera (northern Sumatra), which has
long been regarded as a supervolcano (Fig. 43C). Using ambient-noise
seismic tomography below the caldera, Jaxybulatov et al. (2014) observe an anisotropy that may reect a ne-scale layering linked to the
presence of many partially molten sills in the crust below a depth of
7 km (Fig. 43D).
Another model has been proposed for the plumbing system of resurgent calderas (Browning and Gudmundsson, 2015). It has been derived
from eld data, integrated with numerical modelling, at the eroded
Hafnarfjall volcano (western Iceland), an extinct stratovolcano with a
caldera of predominantly basaltic and subordinately andesitic composition. Here several inclined sheets propagated upward from the magma
chamber, and most became arrested at layer contacts with contrasting
mechanical properties (Fig. 44A). A number of inclined sheets
intersected the ring faults that have previously developed during caldera collapse, and became deected up along the fault to form multiple
ring dykes. This model differs from previous models of ring dykes,
which were directly linked with the underlying magma chamber (e.g.
Saunders, 2005) (Fig. 44B). Saunders (2005) also proposes that arching
of the caldera oor induces extension in the extrados, whereas in the intrados compression is caused by both arching and the force exerted laterally by the intruding ring dykes, preventing intrusions in the caldera
block. Several eroded calderas show that substantial amounts of
magma are intruded along ring faults that channel the magma ow upward (Clough et al., 1909; Bailey and Maufe, 1960; McCall and Bristow,
1965; Smith and Bailey, 1968; Almond, 1977; Sparks, 1988; Miura,
1999; Johnson et al., 2002; Sewell et al., 2012). In some case, ring
dykes have been channelling magma for a long time, so as to favour
the mingling of felsic and basic magmas, as observed at the Ossipee
ring complex (New Hampshire, U.S.) (Kennedy and Stix, 2007). Notwithstanding the frequent identication of this intrusion type, recent
studies have attempted to suggest alternative geometries at some
ring dyke complexes, although these cases remain controversial (e.g.
O'Driscoll et al., 2006; Stevenson et al., 2008).
In a study of several intrusive bodies and calderas in the western
Peninsular Ranges Batholith (Mexico and U.S.), Johnson et al. (2002) individuated the widespread presence of ring dykes. They suggested
that development of subvolcanic ring complexes in the Peninsular
Ranges Batholith involved a three-stage sequence: (1) a buoyant or
overpressured magma chamber induces fracturing of the overlying
rocks, with the formation of inward-dipping conical fractures and
cone sheets (Fig. 45A); (2) the successive loss of magma from the chamber causes collapse of the roof by near-vertical ring faults and ring dyke
intrusions (Fig. 45B); and (3) resurgence takes place due to chamber ination and/or intrusion of a nested pluton (Fig. 45C). A similar sequence
was found also at the deeply eroded, felsic complex of the Vallehermoso
Caldera (La Gomera, Canary Islands) by Rodrguez-Losada and
Martinez-Frias (2004). Johnson et al. (2002) also found some ring

Fig. 45. Sketch in section view of possible intrusive histories found in the western Peninsular Ranges Batholith (Mexico and U.S.) by Johnson et al. (2002). In the rst stage (A), a
set of conical fractures (black lines) above a rising magma chamber develops, with the
possible emplacement of cone sheets (red lines). In the successive stage (B), the collapse
of a caldera along ring-faults is followed by intrusion of ring dykes and volcanism. In
(C) a nested intrusion develops with possible further cone sheets. (For interpretation of
the references to colour in this gure legend, the reader is referred to the web version of
this article.)

complexes emplaced at depths of up to 18 km, suggesting that caldera


subsidence can extend to mid-crustal levels or that different processes
can produce ring dykes.
As we have seen, several models have been introduced, reecting
the possible variability of the structure of plumbing systems at calderas.
This variability has been linked to the diverse size and geometry of
magma chambers, and also to the possible structural style of the caldera
block. Their interaction gives rise to different congurations of the
plumbing system in the overburden that may be composed of inclined
sheets, stacked sills and dykes. A further complication may be due to
the interaction with regional tectonics. As an example, the Los Azufres
caldera complex, in the central part of the Mexican Volcanic Belt, is a
sub-circular depression (27 26 km) that resulted from several nested
collapses of latest MiocenePliocene age (Ferrari et al., 1991, 1993). The
caldera depression is lled by a uvio-lacustrine sequence and several
volcanic domes of PliocenePleistocene age, as well as cinder cones
and associated lava ows of Pleistocene age (Fig. 46A). Domes have a
dacitic to rhyolitic composition whereas cinder cones and lavas are basaltic. Some of the younger deposits are offset by normal faults mainly

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

125

Fig. 46. A. Distribution of silicic volcanic domes, basaltic cinder cones and normal faults at the Los Azufres caldera complex (central Mexican Volcanic Belt), suggesting the control of regional tectonics across the caldera depression. (B) The different cases of distribution of volcanoes with respect to a caldera give clues to the possible patterns of shallow magma plumbing
systems. See main text for details.
A. Redrawn after Ferrari et al. (1991). B. Taken from Walker (1984) and Geyer and Mart (2008).

striking EW to WSWENE, whereas the oldest deposits are also affected by a series of slip planes of different orientations that should represent segments of caldera ring faults. Some of the EW to WSWENE
faults show a very subordinate left-lateral strike-slip component, a geometry and kinematics consistent with the regional tectonics (Ferrari
et al., 1991). The distribution of vents and other data indicates that the
Los Azufres caldera complex was affected by post-caldera volcanism

with centres located both along the caldera rim and along regional faults
crossing the caldera depression (Fig. 46A). This interaction favoured the
presence of silicic centres in the middle of the caldera depression and
along the ring faults, whereas the new basaltic magmas within the caldera were guided by regional faults. As a consequence, the plumbing
system in this example should be represented by silicic dykes emplaced
along ring faults and regional faults, with the possibility that these dykes

126

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

are vertically connected with the felsic magma chamber, or might represent captured inclined sheets.
Regional tectonics can also have an inuence on the development of
ring dykes. As shown by numerical simulations of Walter (2008), an intruding ring-dyke may follow pre-existing faults and thus abruptly stop
or change its direction. This is the case of the Erongo complex (Namibia),
where a ring dyke formed only to the northwest of a SWNE-striking regional fault passing through the caldera (Wigand et al., 2004).
The Los Azufres example is also consistent with other cases suggesting that calderas are long-lasting volcanic complexes whose activity in
most cases does not end after caldera collapse: in fact, resurgence can
take place and/or volcanism can resume, guided by the development
of a plumbing system that can be different from the one that led to
the caldera-forming event. Geyer and Mart (2008), also integrating
data from Walker (1984), reviewed the distribution of post-caldera
vents at more than 160 Quaternary calderas in different geodynamic
settings, providing clues to the possible patterns of magma uprising.
They proposed nine categories of distribution of post-caldera volcanic
activity (surface distribution of vents in Fig. 46B). Assuming that the
shallow geometry of the plumbing system was mirrored by the different
distributions of volcanic centres, I suggest different magma paths in the
various cases (magma feeding systems in Fig. 46B): a single large or
small volcano can be located in different positions in the caldera, corresponding to single main conduits (cases 14). Vents along the ring faults
may be linked to ring dykes (case 5), as took place for example at the
20 30 km wide Tondano caldera (Indonesia) in 1952 and 1971
where ring-dyke intrusions and eruptive activity occurred at two or
more ring vents (Lecuyer et al., 1997). A circular or elliptical distribution
of vents outside the caldera suggests the existence of inclined cone
sheets (case 6). A dened straight line of vents across the caldera suggests a rectilinear dyke zone (case 7) or the control of regional faults
(case 8), as at the above-mentioned Los Azufres caldera, and at the
Rabaul caldera (Papua New Guinea) where eruptive activity simultaneously occurred at opposite sides in 1878 and 1937 (Mori and
McKee, 1987; Nairn et al., 1995). Finally, distributed vents inside the caldera may suggest conduits and dykes of different geometries linked to a
highly-fractured caldera block (case 9), as for example at the resurgent
Ischia Island caldera (Italy) (Tibaldi and Vezzoli, 1998).
9. Conclusions
The data presented in this work indicate that several parameters can
inuence the structure of a magma plumbing system. The classical
models of plumbing systems characterised by upward propagation of
magma by buoyancy forces have been questioned by more recent ndings that indicate magma overpressure as the main engine. The stress
eld induced by local magma pressure sources, in turn, interacts with
the rheological boundaries represented by rock layers with large differences in stiffness, and with physical discontinuities such as bedding,
joints and faults. This interaction may occur at any depth, as suggested
for example by magma sources in the substratum that are in an offset
position from the volcano. This is reected also in the available most
complete images of plumbing systems, as for example at Hawaii and
Mt Etna, which show complex arrays of vertical dykes, inclined sheets
and sills. Local isotropic stresses or time-frequent rotations of stress tensors might lead to intricate assemblages of intrusions of various geometries, as for example below Mt Redoubt volcano.
Most eld and geophysical recent research indicates not only that
magma moves along planar sheets and that circular/elliptical conduits
(in section view) are extremely rare, but there is increasing evidence
that also laccoliths and magma chambers are composed by stacked
planar sill intrusions. Similarly, vertical sheeted intrusions may characterise larger plutonic bodies. Sheets propagate by self-induced
hydrofracturing or by using pre-existent weakness planes, if these are
suitably oriented with respect to the stress eld. Most eld evidences
indicate that hydrofractures are primarily extension fractures whose

propagation can be modelled as mode I cracks, although eld observations also indicate that, in a much lower number of cases, magma overpressure may induce shear faults along which magma intrusion can take
place. These ndings open questions about the kinematics, geometry
and propagation modes of intrusive fractures; we have to admit that uncertainty still exists about the factors that control the different modes of
growth of small magma chambers vs. large igneous bodies. Further
studies, which may couple eld data and modelling, are required, especially integrated with a more in depth analysis of the inuence of the
rates of magma transport and accumulation in relation to diverse tectonic settings and fracture propagation at different depths. Moreover,
diapiric ascent is still recognised especially for granitic bodies, and
thus it is debatable how the deep distributed source region of meltlled pores is linked to the shallow sheeted plumbing system.
The presence of a shallow magma chamber is a prime candidate to
control the geometry of the sheets that typically display a circumferential pattern in plan view above the chamber, giving rise to centrallyinclined sheet swarms. Their arrangement is dictated by the presence of
an oblique stress tensor (Fig. 47, upper apex of the triangle). The local
oblique orientation of the 1 is a function of the shape and overpressure
of the chamber. The consequent 1 trajectories condition the arrangement of the intrusions that can range from radially-inclined sheets
(in vertical section view) diverging from a spherical chamber, to
centrally-inclined sheets with similar dip angle in the case of lobate
magma chambers. Above the central part of the chamber there is the
greatest chance to have subvertical to vertical conduits that drain
magma towards the axis of the volcano. This chance is increased by
the presence of a horizontal 3 below the volcano axis (Fig. 47, right
apex of the triangle) that can be induced by regional extensional tectonics or local deformation, for example due to volcano spreading (Borgia,
1994; van Wyk de Vries and Francis, 1997). With increasing distance
from the magma chamber, the effects of magma overpressure decrease
and the intrusive sheets can change their attitude from inclined sheets
to vertical dykes laterally with respect to the chamber. These dykes
can be parallel to each other, hence forming a swarm perpendicular to
the regional 3, or can assume a radial pattern if the inuence of regional
tectonics is poor. Above the magma chamber, inclined sheets and dykes
can bend into a sill-like attitude following a re-orientation of the stress
tensor (Fig. 47, left apex of the triangle). This can be caused by several
situations that involve the presence of stress barriers and major mechanical contrasts between different lithotypes (Menand, 2011).
Centrally-inclined sheet swarms are more widespread at mac
magma chambers, but have been also found at felsic bodies. Magma migration along sills and inclined sheets may occur at different scales,
going from tens of metres to tens of kilometres, with the potential of
producing eruptions in unexpected locations. Analysis of the literature
shows a great amount of works that use numerical or analogue modelling approaches to unravel these processes, but there is a comparative
lack of eld data. For example, only seventeen eld-based studies of
centrally-inclined sheet swarms are sufciently detailed to obtain quantitative data. Anchoring the results of modelling to quantitative ground
truth is fundamental to establish the thresholds of zones of inuence of
tectonic stresses versus other stress sources, and this might denitively
benet from investigations at eroded volcanic centres and exhumed
plumbing systems. Besides, large eld data sets should be integrated
in a more interdisciplinary way with geophysical and petro-chemical
data.
At shallower crustal levels, there may be a strong interference between tectonic stresses, the stresses related to volcano topography,
and the stresses exerted by magma overpressure. The tectonic stress
eld may affect both plumbing systems within volcanoes and below
them; if extensional stresses dominate, a rectilinear volcano-tectonic
rift zone can guide magma upwelling below the volcano as well as
across the cone with the injection of a swarm of parallel dykes (lower
left part of the graph of Fig. 48). In the case of a transcurrent regime,
dykes can follow the main strike-slip fault zone and/or secondary

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

127

Fig. 47. Graph that summarises some of the main parameters that inuence the upper part of a magma plumbing system. The apexes of the triangle are located in correspondence of stress
tensors with different orientations: the upper apex has oblique 1, 2 and 3; under this stress eld centrally-inclined sheets are favoured. The local oblique orientation of the 1 is a function of the shape and overpressure of the chamber. The consequent 1 trajectories condition the arrangement of the intrusions that can range from centrally-inclined sheets diverging from
a spherical chamber to centrally-inclined sheets with similar dip angle in the case of lobate magma chambers. With increasing distance form the magma chamber, or under an increasing
effect of 3, dykes dominate as in the right apex of the triangle where 2 and 3 are horizontal. Above the magma chamber, inclined sheets and dykes can bend to reach a sill attitude following a re-orientation of the stress tensor that locally has horizontal 1 and 2, as in the left apex of the triangle.

fractures. In the case of contractional tectonics, sheets may assume


more complex and variable geometries. Within a volcanic edice, if
the stresses induced by magma overpressure along the main central
conduit zone are sufciently high, a radial dyke pattern might develop.
The inuence of the stresses related to topography increases approaching the volcano slopes with more probable effects at distances
b0.5 km (increasing inuence towards the right side of the graph of

Fig. 48. Scheme of the most frequent sheet geometries as resulting from the interference
between different tectonic stress elds (stress tensors at Y axis) and the distribution of
the stresses related to topography (increasing inuence towards the right side of the
graph). The debuttressing effect of topography is larger at b0.5 km from the volcano
slope and for slopes steeper than 4050; in these cases sheets might reach an attitude
parallel to the slope.

Fig. 48), as can be inferred from the Mt Etna and Stromboli data. For
slopes steeper than 4050, sheets may assume an attitude parallel to
the slope.
Based on the available literature, it appears that much more studies
are required to elucidate the characteristics of magma feeding systems
in contractional tectonic settings, especially with an interdisciplinary
approach. Similarly, the effects of slopes on the trajectories of magma
paths need further studies based on eld data integrated with numerical and analogue modelling. Since eroded calderas are not very frequent
on Earth, the scarcity of eld data on the structure of their plumbing systems is even more noticeable.
In conclusion, the data illustrated in the present work indicate that
several parameters can inuence the structure and location of a
magma plumbing system, and that it is extremely difcult, if not impossible, to assign a given model to a volcano without a detailed complete
knowledge of its geological, structural, petro-geochemical and geomorphological characteristics. The ultimate magma path is governed by the
inner volcano structure, and by the stress eld in the host rock existing
prior to sheet emplacement and by changes in the stress eld induced
by the emplacement itself. As the assessment of the possible magma
paths leading to a new eruptive centre is one of the main goals of volcanic hazard studies, it is very important to better understand how far a
stress eld can be perturbed by the emplacement of previous sheets,
and why and when sheet propagation might be deviated or halted.
These questions should be better addressed by an integrated approach
that may combine natural examples with experimental works, and by
increasing the data sets on the distribution of mechanic properties inside volcanoes and of the active stress eld in the substrate and within
a volcanic edice. This knowledge can be achieved especially by desirable further geological and geophysical exploration and in-situ stress
measurements and observations that, coupled with numerical modelling and instrument monitoring of a volcano, can lead to a better understanding of these processes.

128

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

Acknowledgements
I thank the Editors Lionel Wilson and Joan Mart and the Managing
Editor Timothy Horscroft for requesting me to write this review, after
an invited talk I gave at the 2014 EGU Meeting in Vienna. My research
contribution to this review benetted from extensive eld work with
several colleagues, to mention Giorgio Pasquar, Alberto Renzulli,
Derek Rust, Federico Pasquar Mariotto, Claudia Corazzato, Fabio Bonali,
and many others. The research was carried out in the framework
of MIUR, FIRB, INGV and NATO projects, and under the aegis of the
International Lithosphere Program, Task Force II. F. Jim Cole and Agust
Gudmundsson are acknowledged for their useful comments on an earlier version of this paper. Pasquar Mariotto is also acknowledged for his
review of the English grammar.

References
Abdallah, A., Courtillot, V., Kasser, M., LeDain, A.Y., Lepine, J.C., Robineau, B., Ruegg, J.C.,
Tapponnier, P., Tarantola, A., 1979. Relevance of Afar seismicity and volcanism to
the mechanics of accreting plate boundaries. Nature 282, 1723.
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. 359360, 93105.
Acocella, V., 2007. Understanding caldera structure and development: an overview of analogue models compared to natural calderas. Earth Sci. Rev. 85 (3),
125160.
Acocella, V., 2014. Structural control on magmatism along divergent and convergent plate
boundaries: overview, model, problems. Earth Sci. Rev. 136, 226288.
Acocella, V., Neri, M., 2009. Dike propagation in volcanic edices: overview and possible
developments. Understanding Deformation and Stress in Active Volcanoes.
Tectonophysics, Special issue 471, pp. 6777.
Acocella, V., Tibaldi, A., 2005. Dike propagation driven by volcano collapse: a general
model tested at Stromboli, Italy. Geophys. Res. Lett. 32 (8). http://dx.doi.org/10.
1029/2004GL022248.
Acocella, V., Porreca, M., Neri, M., Massimi, E., Mattei, M., 2006. Propagation of dikes at
Vesuvio (Italy) and the effect of Mt. Somma. Geophys. Res. Lett. 33 (8).
Aizawa, K., Acocella, V., Yoshida, T., 2006. How the development of magma chambers affects collapse calderas: insights from an overview. Geol. Soc. Lond., Spec. Publ. 269
(1), 6581.
Aizawa, K., Koyama, T., Hase, H., Uyeshima, M., Kanda, W., Utsugi, M., Ryokei, Y., Yusuke,
Y., Takeshi, H., Kenichi, Y., Shintaro, K., Watanabe, A., Koji, M., Yasuo, O., 2014. Threedimensional resistivity structure and magma plumbing system of the Kirishima
Volcanoes as inferred from broadband magnetotelluric data. J. Geophys. Res. Solid
Earth 119, 198215.
Almond, D.C., 1977. The Sabaloka igneous complex, Sudan. Philos. Trans. R. Soc. Lond. 287,
595633.
Aloisi, M., Bonaccorso, A., Gambino, S., 2006. Imaging composite dyke propagation
(Etna, 2002 case). J. Geophys. Res. Solid Earth 111. http://dx.doi.org/10.1029/
2005JB003908.
Amelung, F., Day, S., 2002. InSAR observations of the 1995 Fogo, Cape Verde, eruption:
implications for the effects of collapse events upon island volcanoes. Geophys. Res.
Lett. 29 (12), 1606. http://dx.doi.org/10.1029/2001GL013760.
Anderson, E.M., 1936. Dynamics of formation of cone-sheets, ring-dikes, and cauldron
subsidence. Proc. Roy. Soc. Edinb. 56, 128157.
Anderson, E.M., 1938. The dynamics of sheet intrusion. Proc. Roy. Soc. Edinb. 58, 242251.
Anderson, E.M., 1951. The Dynamics of Faulting and Dyke Formation With Applications to
Britain. Oliver and Boyd, White Plains, New York.
Annen, C., 2009. From plutons to magma chambers: thermal constraints on the accumulation of eruptible silicic magma in the upper crust. Earth Planet. Sci. Lett. 284 (3),
409416.
Annen, C., Sparks, R.S.J., 2002. Effects of repetitive emplacement of basaltic intrusions on
thermal evolution and melt generation in the crust. Earth Planet. Sci. Lett. 203,
937955. http://dx.doi.org/10.1016/S0012-821X(02)00929-9.
Annen, C., Zellmer, G.F. (Eds.), 2008. Dynamics of Crustal Magma Transfer, Storage and
Differentiation. Geological Society, London, Special Publications 304, pp. 113.
Annen, C., Blundy, J.D., Sparks, R.S.J., 2006a. The genesis of intermediate and silicic
magmas in deep crustal hot zones. J. Petrol. 47 (3), 505539.
Annen, C., Blundy, J.D., Sparks, R.S.J., 2006b. The sources of granitic melt in deep hot zones.
Trans. R. Soc. Edinb. 97 (4), 297309.
Argnani, A., Serpelloni, E., Bonazzi, C., 2007. Pattern of deformation around the central
Aeolian Islands: evidence from multichannel seismics and GPS data. Terra Nova 19,
317323. http://dx.doi.org/10.1111/j.1365-3121.2007.00753.x.
Atherton, M.P., 1993. Granite magmatism. J. Geol. Soc. 150, 10091023.
Atkinson, B.K., Meredith, P.G., 1987. The theory of subcritical crack growth with application
to minerals and rocks. In: Atkinson, B.K. (Ed.), Fracture Mechanics of Rock. Academic,
New York, pp. 111166.
Aydin, A., Schultz, R.A., Campagna, D., 1990. Fault-normal dilatation in pull-apart basins:
implications for relationship between strike-slip fault and volcanic activity. In:
Boccaletti, M., Nur, A. (Eds.), Active and Recent Strike-slip Tectonics. Annales
Tectonicae Special Issue, pp. 4552.

Bacon, C.R., Lanphere, M.A., 2006. Eruptive history and geochronology of Mount Mazama
and the Crater Lake region, Oregon. Geol. Soc. Am. Bull. 118 (11), 1331.
Bagnardi, M., Amelung, F., 2012. Space-geodetic evidence for multiple magma reservoirs
and subvolcanic lateral intrusions at Fernandina volcano, Galpagos Islands.
J. Geophys. Res. 117, B10406. http://dx.doi.org/10.1029/2012JB009465.
Bahat, D., 1980. Hertzian fracture, a principal mechanism in the emplacement of the
British Tertiary intrusive centres. Geol. Mag. 117, 463470.
Bailey, E.B., Maufe, H.B., 1960. The geology of Ben Nevis and Glen Coe and the surrounding
country. Mem Geol Surv Scotlandp. 307.
Bailey, E.B., Clough, C.T., Wright, W.B., Richey, J.E., Wilson, G.V., 1924. The Tertiary and
post-tertiary geology of Mull, Loch Aline, and Oban. Geological Survey of Scotland
Memoir, Sheet. 44 ((Scotland): 445 pp.).
Baker, S., Amelung, F., 2012. Top-down ination and deation at the summit of Kilauea
volcano, Hawaii observed with InSAR. J. Geophys. Res. 117, B10406. http://dx.doi.
org/10.1029/2012JB009123.
Bald, N., Noe-Nygaard, A., Pedersen, K., 1971. The Kroksfjordur central volcano in NorthWest Iceland. Acta Naturalia Islandica II (10) (29 pages).
Barberi, F., Varet, J., 1970. The Erta Ale volcanic range (Danakil depression, northern Afar,
Ethiopia). Bull. Volcanol. 34, 848917.
Barberi, F., Cassano, E., La Torre, P., Sbrana, A., 1991. Structural evolution of Campi Flegrei
caldera in light of volcanological and geophysical data. J. Volcanol. Geotherm. Res. 48,
3349.
Barker, S.E., Malone, S.D., 1991. Magmatic system geometry at Mount St. Helens modeled
from the stress eld associated with post-eruptive earthquakes. J. Geophys. Res. 96
(B7), 11,88311,894. http://dx.doi.org/10.1029/91JB00430.
Bates, R., Jackson, J.A., 1987. Glossary of Geology. 3rd edition. McGraw-Hill Book Company
(788 pp.).
Bear, G., Beyth, M., Reches, Z., 1994. Dikes emplaced into fractured basement, Timna
Igneous Complex, Israel. J. Geophys. Res. 99, 2403924050.
Becerril, L., Galindo, I., Gudmundsson, A., Morales, J.M., 2013. Depth of origin of magma in
eruptions. Sci. Rep. 3.
Belachew, M., Ebinger, C., Cot, D., Keir, D., Rowland, J.V., Hammond, J.O.S., Ayele, A., 2011.
Comparison of dike intrusions in an incipient seaoor-spreading segment in
Afar, Ethiopia: seismicity perspectives. J. Geophys. Res. Solid Earth (19782012)
116 (B6).
Bell, B.R., Claydon, R.V., Rogers, G., 1994. The petrology and geochemistry of conesheets from the Cuillins Igneous Complex, Isle of Skye: evidence for combined
assimilation and fractional crystallization during lithospheric extension. J. Petrol.
35, 10551094.
Bellier, O., Sebrier, M., 1994. Relationship between tectonism and volcanism along the
Great Sumatran Fault zone deduced by SPOT image analyses. Tectonophysics 233,
215231.
Bellier, O., Bellon, H., Sebrier, M., Sutanto, M.R.C., 1999. KAr age of the Ranau tuffs;
implications for the Ranau Caldera emplacement and slip-partitioning in Sumatra
(Indonesia). Tectonophysics 312, 347359.
Benz, H.M., Chouet, B.A., Dawson, P.B., Lahr, J.C., Page, R.A., J.A.H, 1996. Three-dimensional
P and S wave velocity structure of Redoubt Volcano, Alaska. J. Geophys. Res. 101 (B4),
81118128.
Best, M.G., 1982. Igneous and Metamorphic Petrology. W.H. Freeman and Company, New
York (630 pp.).
Billings, M.P., 1972. Structural Geology. Prentice-Hall Inc., Englewood Cliffs, New Jersey,
USA (606 pp.).
Bistacchi, A., Tibaldi, A., Pasquar, F.A., Rust, D., 2012. The association of cone-sheets and
radial dykes: data from the Isle of Skye (UK), numerical modelling, and implications
for shallow magma chambers. Earth Planet. Sci. Lett. 339340, 4656.
Bjornsson, A., Johnsen, G., Sigurdsson, S., Thorbergsson, G., Tryggvason, E., 1979. Rifting of
the plate boundary in north Iceland 19751978. J. Geophys. Res. 84, 30293038.
Bonafede, M., Olivieri, M., 1995. Displacement and gravityanomaly produced by a shallow vertical dyke in a cohesionless medium. Geophys. J. Int. 123, 639652.
Bonafede, M., Rivalta, E., 1999. On tensile cracks close to and across the interface between
two welded elastic half-space. Geophys. J. Int. 138, 410434.
Bonali, F., Corazzato, C., Tibaldi, A., 2011. Identifying rift zones on volcanoes: an example
from La Runion Island, Indian Ocean. Bull. Volcanol. 73, 347366. http://dx.doi.org/
10.1007/s00445-010-0416-1.
Bonin, B., Ethien, R., Gerbe, M.C., Cottin, J.Y., Fraud, G., Gagnevin, D., Moine, B., 2004. The
Neogene to Recent Rallier-du-Baty nested ring complex, Kerguelen Archipelago
(TAAF, Indian Ocean): stratigraphy revisited, implications for cauldron subsidence
mechanisms. Geol. Soc. Lond., Spec. Publ. 234 (1), 125149.
Borgia, A., 1994. Dynamic basis of volcanic spreading. J. Geophys. Res. Solid Earth
(19782012) 99 (B9), 1779117804.
Boudier, F., Nicolas, A., Ildefonse, B., 1996. Magma chambers in the Oman ophiolite: fed
from the top and the bottom. Earth Planet. Sci. Lett. 144 (1), 239250.
Bradley, J., 1965. Intrusion of major dolerite sills. Trans. R. Soc. N. Z. 3, 2755.
Branney, M.J., Kokelaar, P., 1994. Volcanotectonic faulting, soft-state deformation,
and rheomorphism of tuffs during development of a piecemeal caldera, English
Lake District. Geol. Soc. Am. Bull. 106, 507530.
Bridgwater, D., Sutton, J., Watterson, J., 1974. Crustal downfolding associated with igneous
activity. Tectonophysics 21, 5777.
Browning, J., Gudmundsson, A., 2015. Caldera faults capture and deect inclined sheets:
an alternative mechanism of ring dike formation. Bull. Volcanol. 77 (1), 113.
Busby, C.J., Bassett, K.N., 2007. Volcanic facies architecture of an intra-arc strike-slip basin,
Santa Rita Mountains, Southern Arizona. Bull. Volcanol. 70, 85103.
Cabello, G.C., Garza, R.M.L.D., Argote, R.C., Flores, E., Ramrez, A.O., Rivera, H., Bhnel, J.
Leed, 2006. Geology and paleomagnetism of El Potrero pluton, Baja California:
understanding criteria for timing of deformation and evidence of pluton tilt during
batholith growth. Tectonophysics 424, 117.

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135


Calanchi, N., Romagnoli, C., Rossi, P.L., 1995. Morphostructural and some petrochemical
data from the submerged area around Alicudi and Filicudi volcanic islands (Aeolian
Arc, southern Tyrrhenian Sea). Mar. Geol. 123, 215238.
Canon-Tapia, E., Merle, O., 2006. Dyke nucleation and early growth from pressurized
magma chambers: insights from analogue models. J. Volcanol. Geotherm. Res. 158
(3), 207220.
Carlino, S., Tramelli, A., Somma, R., 2014. Caldera subsidence in extensional tectonics. Bull.
Volcanol. 76, 870. http://dx.doi.org/10.l007/s00445-014-0870-2.
Carracedo, J.C., 1994. The Canary-Islandsan example of structural control on the growth
of large oceanic-island volcanoes. J. Volcanol. Geotherm. Res. 60, 225241.
Cartwright, J., Hansen, D.M., 2006. Magma transport through the crust via interconnected
sill complexes. Geology 34 (11), 929932.
Cas, R.A.F., Wright, J.V., 1987. Volcanic Successions. Allen & Unwin, London (528 pp.).
Cashman, K.V., Giordano, G., 2014. Calderas and magma reservoirs. J. Volcanol. Geotherm.
Res. 288, 2845.
Castro, J.M., Dingwell, D.B., 2009. Rapid ascent of rhyolitic magma at Chaiten volcano,
Chile. Nature 461, 780784. http://dx.doi.org/10.1038/nature08458.
Cervelli, P., Miklius, A., 2003. The shallow magmatic system of Kilauea Volcano. U. S. Geol.
Surv. Prof. Pap. 1676, 149163.
Chadwick Jr., W.W., Dieterich, J.H., 1995. Mechanical modelling of circumferential and
radial dike intrusion on Galapagos volcanoes. J. Volcanol. Geotherm. Res. 66, 3752.
Chadwick, W.W., Jnsson, S., Geist, D.J., Poland, M., Johnson, D.J., Batt, S., Harpp, K.S., Ruz,
A., 2011. The May 2005 eruption of Fernandina volcano, Galpagos: the rst circumferential dike intrusion observed by GPS and InSAR. Bull. Volcanol. 73, 679697.
http://dx.doi.org/10.1007/s00445-010-0433-0.
Charlez, P.A., 1997. Rock Mechanics, 2: Petroleum Applications. Editions Technip, Paris.
Chaussard, E., Amelung, F., 2014. Regional controls on magma ascent and storage in volcanic arcs. Geochem. Geophys. Geosyst. 15, 14071418. http://dx.doi.org/10.1002/
2013GC005216.
Chen, Z., Jin, Z.H., Johnson, S.E., 2007. A perturbation solution for dyke propagation in an
elastic medium with graded density. Geophys. J. Int. 169 (1), 348356.
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), 4354.
Chiarabba, C., Pino, N.A., Ventura, G., Vilardo, G., 2004a. Structural features of the shallow
plumbing system of Vulcano Island Italy. Bull. Volcanol. 66, 477484.
Chiarabba, C., De Gori1, P., Patan, D., 2004b. The Mt. Etna plumbing system: the contribution of seismic tomography. Mt. Etna: Volcano Laboratory, Geophysical Monograph Series 143. AGU, pp. 191204.
Chouet, B., Saccorotti, G., Martini, M., Dawson, P., De Luca, G., Milana, G., Scarpa, R., 1997.
Source and path effects in the wave elds of tremor and explosions at Stromboli
Volcano, Italy. J. Geophys. Res. Solid Earth (19782012) 102 (B7), 1512915150.
Chouet, B., Dawson, P., Martini, M., 2008. Shallow-conduit dynamics at Stromboli
Volcano, Italy, imaged from waveform inversions. Geol. Soc. Lond., Spec. Publ. 307
(1), 5784.
Clague, D.A., 1987. Hawaiian xenolith populations, magma supply rates, and development
of magma chambers. Bull. Volcanol. 49, 577587.
Clague, D.A., Dixon, J.E., 2000. Extrinsic controls on the evolution of Hawaiian ocean island
volcanoes. Geochem. Geophys. Geosyst. 1 (4), 1010. http://dx.doi.org/10.1029/
1999GC000023.
Clemens, J.D., Mawer, C.K., 1992. Granitic magma transport by fracture propagation.
Tectonophysics 204, 339360.
Clemens, J.D., Petford, N., Mawer, C.K., 1997. Ascent mechanisms of granitic magmas:
causes and consequences. In: Holness, M.B. (Ed.), Deformation Enhanced Fluid Transport in the Earth's Crust and Mantle. Chapman & Hall, London, pp. 145172.
Clough, C.T.H., Maufe, H.B., Bailey, E.B., 1909. The cauldron subsidence of Glencoe and the
associated igneous phenomena. Q. J. Geol. Soc. Lond. 65, 611678.
Cole, J.W., 1990. Structural control and origin of volcanism in the Taupo volcanic zone,
New Zealand. Bull. Volcanol. 52, 445459.
Cole, J.W., Milner, D.M., Spinks, K.D., 2005. Calderas and caldera structures: a review.
Earth Sci. Rev. 69 (1), 126.
Corazzato, C., Menna, M., Tibaldi, A., Renzulli, A., Francalanci, L., Petrone, C.M., Vezzoli, L.,
Acocella, V., 2006. An integrated structural and petrochemical approach to unravel
dike injection conditions and volcano ank instability at Stromboli (Italy). In:
Thomson, K. (Ed.), Physical Geology of Subvolcanic Systems: Laccoliths, Sills, and
Dykes. Visual Geosciences, pp. 1114 http://dx.doi.org/10.1007/s10069-006-0002-z.
Corazzato, C., Francalanci, L., Menna, M., Petrone, C., Renzulli, A., Tibaldi, A., Vezzoli, L.,
2008. What does it guide sheet intrusion in volcanoes? Petrological and structural
characters of the Stromboli sheet complex, Italy. J. Volcanol. Geotherm. Res. 173,
2654.
Corry, C.E., 1988. Laccolithsmechanics of emplacement and growth. Geol. Soc. Am. Spec.
Pap. 220 (110 pp.).
Corti, G., Bonini, M., Innocenti, F., Manetti, P., Mulugeta, G., 2001. Centrifuge models simulating magma emplacement during oblique rifting. J. Geodyn. 31, 557576.
Cruden, A.R., 1998. On the emplacement of tabular granites. J. Geol. Soc. 155, 853862.
Curtie, K.L., Ferguson, J., 1970. The mechanism of intrusion of lamprophyre dykes indicated by offsetting of dykes. Tectonophysics 9 (525535), 1970.
Dahm, T., 2000. Numerical simulations of the propagation path and the arrest of uidlled fractures in the Earth. Geophys. J. Int. 141, 623638.
Daniels, K.A., Kavanagh, J.L., Menand, T., Stephen, J.S.R., 2012. The shapes of dikes:
evidence for the inuence of cooling and inelastic deformation. Geol. Soc. Am. Bull.
124 (78), 11021112.
Davis, G.H., Reynolds, S.J., Kluth, C., 1996. Structural Geology of Rocks and Regions. Second
edition. Wiley, New York (776 pp.).
Dawson, P., Whilldin, D., Chouet, B., 2004. Application of near real-time radial semblance to
locate the shallow magmatic conduit at Kilauea Volcano, Hawaii. Geophys. Res. Lett. 31.

129

Day, S.J., Heleno da Silva, S.I.N., Fonseca, J.F.B.D., 1999. A past giant lateral collapse and
present-day ank instability of Fogo, Cape Verde Islands. J. Volcanol. Geotherm. Res.
94, 191218.
Decker, R.W., 1987. Dynamics of Hawaiian volcanoes: an overview. Volcanism in Hawaii.
US Geol. Surv. Prof. Pap. 1350, 9971018.
del Potro, R., Dez, M., Blundy, J., Camacho, A.G., Gottsmann, J., 2013. Diapiric ascent of
silicic magma beneath the Bolivian Altiplano. Geophys. Res. Lett. 40 (10), 20442048.
Delaney, P.T., Gartner, A.E., 1997. Physical processes of shallow mac dike emplacement
near the San Rafael Swell, Utah. Geol. Soc. Am. Bull. 109, 11171192.
Delaney, P.T., Pollard, D.D., Ziony, J.I., Mckee, E.H., 1986. Field relations between dikes and
joints: emplacement processes and paleostress analyses. J. Geophys. Res. 91, 49204983.
Delcamp, A., Troll, V.R., van Wyk de Vries, B., Carracedo, J.C., Petronis, M.S., Prez-Torrado,
F.J., Deegan, F.M., 2012. Dykes and structures of the NE rift of Tenerife, Canary Islands:
a record of stabilisation and destabilisation of ocean island rift zones. Bull. Volcanol.
74, 963980. http://dx.doi.org/10.1007/s00445-012-0577-1.
Di Roberto, A., Rosi, M., Bertagnini, A., Marani, M.P., Gamberi, F., Del Principe, A., 2008.
Deep water gravity core from the Marsili Basin (Tyrrhenian Sea) records
PleistocenicHolocenic explosive events and instability of the Aeolian Archipelago,
(Italy). J. Volcanol. Geotherm. Res. 177, 133144. http://dx.doi.org/10.1016/j.
jvolgeores.2008.01.009.
Dieterich, J.H., 1988. Growth and persistence of Hawaiian volcanic rift zones. J. Geophys.
Res. Solid Earth 93, 42584270.
Dirks, P.H.G.M., Jelsma, H.A., 1998. Horizontal accretion and stabilization of the Archean
Zimbabwe Craton. Geology 26 (1), 1114.
Dixon, J.M., Simpson, D.G., 1987. Centrifuge modelling of laccolith intrusion. J. Struct. Geol.
9 (1), 87103.
D'Lemos, R.S., Brown, M., Strachan, R.A., 1992. Granite magma generation, ascent and emplacement within a transpressional orogen. J. Geol. Soc. Lond. 149, 487490.
Donoghue, S.L., Gamble, J.A., Palmer, A.S., Stewart, R.B., 1995. Magma mingling in an andesite pyroclastic ow of the Pourahu member, Ruapehu volcano, New Zealand.
J. Volcanol. Geotherm. Res. 68, 177191.
Druitt, T.H., Edwards, L., Mellors, R.M., Pyle, D.M., Sparks, R.S.J., Lanphere, M., Davies, M.,
Barriero, B., 1999. Santorini volcano. Geol. Soc. Lond. Mem. 19.
Dufek, J., Bergantz, G.W., 2005. Lower crustal magma genesis and preservation: a stochastic
framework for the evaluation of basaltcrust interaction. J. Petrol. 46 (11), 21672195.
Durrance, E.M., 1967. Photoelastic stress studies and their application to a mechanical
analysis of the tertiary ring-complex of Ardnamurchan, Argyllshire. Proc. Geol.
Assoc. 78, 289318.
Dvorak, J.J., Dzurisin, D., 1993. Variations in magma-supply rate at Kilauea volcano,
Hawaii. J. Geophys. Res. 98, 2225522268.
Ebinger, C.J., Casey, M., 2001. Continental breakup in magmatic provinces: an Ethiopian
example. Geology 29, 527530.
Emeleus, C.H., 2009. Ardnamurchan central complex, bedrock and supercial deposits. In:
1:25,000 Geology Series. British Geological Survey. Scale 1:25 000, 1 sheet.
Eriksson, P.I., Riishuus, M.S., Sigmundsson, F., Elming, S.A., 2011. Magma ow directions
inferred from eld evidence and magnetic fabric studies of the Streitishvarf composite dike in east Iceland. J. Volcanol. Geotherm. Res. 206, 3045.
Eriksson, P.I., Riishuus, M.S., Elming, S.., 2014. Magma ow and palaeo-stress deduced
from magnetic fabric analysis of the lftafjrur dyke swarm: implications for shallow crustal magma transport in Icelandic volcanic systems. Geol. Soc. Lond., Spec.
Publ. 396, SP3966.
Ernst, R.E., 2014. Large Igneous Provinces. Cambridge University Press.
Farris, D.W., Haeussler, P., Friedman, R., Paterson, S.R., Saltus, R.W., Ayuso, R., 2006.
Emplacement of the Kodiak batholith and slab-window migration. Geol. Soc. Am.
Bull. 118, 13601376.
Favalli, M., Tarquini, S., Papale, P., Fornaciai, A., Boschi, E., 2012. Lava ow hazard and risk
at Mt. Cameroon volcano. Bull. Volcanol. 74, 423439. http://dx.doi.org/10.1007/
s00445-011-0540-6.
Ferrari, L., Garduno, V.H., Pasquar, G., Tibaldi, A., 1991. Geological evolution of Los
Azufres Caldera as a response to the regional tectonics. J. Volcanol. Geotherm. Res.
47, 129148.
Ferrari, L., Garduno, V.H., Pasquar, G., Tibaldi, A., 1993. The Los Azufres Caldera, Mexico:
the result of multiple nested collapses. J. Volcanol. Geotherm. Res. 56, 345349.
Ferr, E.C., Gleizes, G., Djouadi, M.T., Bouchez, J.L., Ugodulunwa, F.X.O., 1997. Drainage and
emplacement of magmas along an inclined transcurrent shear zone: petrophysical
evidence from a granitecharnockite pluton (Rahama, Nigeria). In: Bouchez, J.L.,
Hutton, D.H.W., Stephens, W.E. (Eds.), Granite: From Segregation of Melt to Emplacement Fabrics vol 8. Kluwer, Dordrecht, pp. 253273.
Ferr, E.C., Gleizes, G., Caby, R., 2002. Tectonics and post-collisional granite emplacement
in an obliquely convergent orogen: the Trans-Saharan belt, Eastern Nigeria. Precambrian
Res. 114, 199219.
Ferr, E.C., Galland, O., Montanari, D., Kalakay, T.J., 2012. Granite magma migration and
emplacement along thrusts. Int. J. Earth Sci. 101 (7), 16731688.
Fialko, Y.A., Rubin, A.M., 1998. Thermodynamics of lateral dike propagation, Implications
for crustal accretion at slow spreading mid-ocean ridges. J. Geophys. Res. 103 (B2),
25012514.
Fiske, R.S., Jackson, E.D., 1972. Orientation and growth of Hawaiian volcanic rifts: the effect of regional structure and gravitational stresses. Proc. R. Soc. Lond. 329, 299326.
Fiske, R.S., Kinoshita, W.T., 1969. Ination of Kilauea Volcano prior to its 19671968 eruption. Science 165, 341349.
Folch, A., Mart, J., 1998. The generation of overpressure in felsic magma chambers by replenishment. Earth Planet. Sci. Lett. 163, 301314. http://dx.doi.org/10.1016/S0012821X(98)00196-4.
Folch, A., Fernndez, J., Rundle, J.B., Mart, J., 2000. Ground deformation in a viscoelastic
medium composed of a layer overlying a half-space: a comparison between point
and extended sources. Geophys. J. Int. 140 (1), 3750.

130

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

Francis, E.H., 1982. Magma and sediment I. Emplacement mechanism of late Carboniferous tholeiite sills in northern Britain. J. Geol. Soc. Lond. 139 (1), 120.
Francis, P.W., Hammill, M., Kretzschmar, G., Thorpe, R.S., 1978. The Cerro Galan Caldera,
north-west Argentina and its tectonic setting. Nature 274, 749751.
Franke, W., 1992. Phanerozoic structures and events in Central Europe. In: Bundell, D.,
Freeman, R., Mueller, S. (Eds.), A Continent Revealed. Cambridge University Press,
The European Geotraverse, pp. 164180.
Frey, F.A., Wise, W.S., Garcia, M.O., West, H., Kwon, S.T., Kennedy, A., 1990. Evolution of
Mauna Kea volcano, Hawaii: petrologic and geochemical constraints on post-shield
volcanism. J. Geophys. Res. Solid Earth 95 (B2), 12711300.
Fridleifsson, I.B., 1977. Distribution of large basaltic intrusions in the Icelandic crust and
the nature of the layer 2layer 3 boundary. Geol. Socam Bull. 11, 16891693.
Fuchs, K., Bonjer, K.P., Gajewski, D., Lschen, E., Prodehl, C., Sandmeier, K.J., Wilhelm, H.,
1987. Crustal evolution of the Rhine-graben area. 1. Exploring the lower crust in
the Rhine graben rift by unied geophysical experiments. Tectonophysics 141
(13), 261275.
Fujita, E., Ukawa, M., Yamamoto, E., 2004. Subsurface cyclic magma sill expansions in the
2000 Miyakejima volcano eruption: possibility of two-phase ow oscillation.
J. Geophys. Res. http://dx.doi.org/10.1029/2003JB002556.
Gaffney, E.S., Damjanac, B., 2006. Localization of volcanic activity: topographic effects on
dike propagation, eruption and conduit formation. Geophys. Res. Lett. 33, L14313.
http://dx.doi.org/10.1029/2006GL026852.
Gaffney, E.S., Damjanac, B., Valentine, G.A., 2007. Localization of volcanic activity: 2. Effects
of pre-existing structure. Earth Planet. Sci. Lett. 263, 323338.
Galerne, C.Y., Galland, O., Neumann, E.-R., Planke, S., 2011. 3D relationships between sills
and their feeders: evidence from the Golden Valley Sill Complex (Karoo Basin) and
experimental modelling. J. Volcanol. Geotherm. Res. 202, 189199.
Galland, O., Cobbold, P.R., de Bremond d'Ars, J., Hallot, E., 2007a. Rise and emplacement of
magma during horizontal shortening of the brittle crust: insights from experimental
modelling. J. Geophys. Res. http://dx.doi.org/10.1029/2006JB004604.
Galland, O., Hallot, E., Cobbold, P.R., Ruffet, G., de Bremod d'Ars, J., 2007b. Volcanism in a
compressional Andean setting: a structural and geochronological study of Tromen
volcano (Neuquen province, Argentina). Tectonics http://dx.doi.org/10.1029/
2006TC002011.
Galland, O., Planke, S., Neumann, E.R., Malthe-Srenssen, A., 2009. Experimental modelling of shallow magma emplacement: application to saucer-shaped intrusions.
Earth Planet. Sci. Lett. 277 (34), 373383.
Galland, O., Burchardt, S., Hallot, E., Mourgues, R., Bulois, C., 2014. Dynamics of dikes
versus cone sheets in volcanic systems. J. Geophys. Res. Solid Earth 119 (8),
61786192.
Garagash, D., Detournay, E., 2000. The tip region of a uid-driven fracture in an elastic medium. ASME J. Appl. Mech. 67, 183192.
Garca, A., Fernndez-Ros, A., Berrocoso, M., Marrero, J.M., Prates, G., De la Cruz-Reyna, S.,
Ortiz, R., 2014. Magma displacements under insular volcanic elds, applications to
eruption forecasting: El Hierro, Canary Islands, 20112013. Geophys. J. Int. http://
dx.doi.org/10.1093/gji/ggt505.
Gautneb, H., Gudmundsson, A., Oskarsson, N., 1989. Structure, petrochemistry and evolution of a sheet swarm in an Icelandic central volcano. Geol. Mag. 126, 659673.
Gee, M.J.R., Masson, D.G., Watts, A.B., Mitchell, N.C., 2001. Offshore continuation of volcanic rift zones, El Hierro, Canary Islands. J. Volcanol. Geotherm. Res. 105, 107119.
Geshi, N., 2005. Structural development of dike swarms controlled by the change of
magma supply rate: the cone sheets and parallel dike swarms of the Miocene
Otoge igneous complex, Central Japan. J. Volcanol. Geotherm. Res. 141 (3), 267281.
Geshi, N., 2008. Vertical and lateral propagation of radial dikes inferred from the owdirection analysis of the radial dike swarm in Komochi Volcano, Central Japan.
J. Volcanol. Geotherm. Res. 173 (1), 122134.
Geshi, N., 2009. Asymmetric growth of collapsed caldera by oblique subsidence during
the 2000 eruption of Miyakejima, Japan. Earth Planet. Sci. Lett. 280, 148158.
http://dx.doi.org/10.1016/j.epsl.2009.01.027.
Geshi, N., Shimano, T., Chiba, T., Nakada, S., 2002. Caldera collapse during the eruption of
Miyakejima Volcano, Japan. Bull. Volcanol. 64, 5568.
Geyer, A., Mart, J., 2008. The new worldwide collapse caldera database (CCDB): a tool for
studying and understanding caldera processes. J. Volcanol. Geotherm. Res. 175 (3),
334354.
Geyer, A., Mart, J., 2014. A short review of our current understanding of the development
of ring faults during collapse caldera formation. Front. Earth Sci. Volcanol. 2, 22.
Gilbert, G.K., 1877. Report on the Geology of the Henry Mountains. U.S. Government
Printing Ofce, Washington D.C. (160 pp.).
Glazner, A.F., 1991. Plutonism, oblique subduction, and continental growth: an example
from the Mesozoic of California. Geology 19, 784786.
Goes, S., Giardini, D., Jenny, S., Hollenstein, C., Kahle, H.-G., Geiger, A., 2004. A recent tectonic reorganization in the south-central Mediterranean. Earth Planet. Sci. Lett. 226,
335345. http://dx.doi.org/10.1016/j.epsl.2004.07.038.
Gonzlez, P.J., Samsonov, S.V., Pepe, S., Tiampo, K.F., Tizzani, P., Casu, F., Fernndez, J.,
Camacho, A.G., Sansosti, E., 2013. Magma storage and migration associated with the
20112012 El Hierro eruption: implications for crustal magmatic systems at oceanic
island volcanoes. J. Geophys. Res. Solid Earth 118 (8), 43614377.
Gottsmann, J., Camacho, A.G., Mart, J., Wooller, L., Fernndez, J., Garcia, A., Rymer, H.,
2008. Shallow structure beneath the Central Volcanic Complex of Tenerife from
new gravity data: implications for its evolution and recent reactivation. Phys. Earth
Planet. Inter. 168 (3), 212230.
Goulty, N.R., 2005. Emplacement mechanism of the Great Whin and Midland Valley dolerite sills. J. Geol. Soc. 162, 10471056.
Grandin, R., Socquet, A., Doubre, C., Jacques, E., King, G.C.P., 2012. Elastic thickness control
of lateral dyke intrusion at mid-ocean ridges. Earth Planet. Sci. Lett. 319320, 8395.
http://dx.doi.org/10.1016/j.epsl.2011.12.011.

Greene, A.R., Garcia, M.O., Pietruszka, A.J., Weis, D., Marske, J.P., Vollinger, M.J., Eiler, J.,
2013. Temporal geochemical variations in lavas from Klauea's Pu'u '' eruption
(19832010): cyclic variations from melting of source heterogeneities. Geochem.
Geophys. Geosyst. 14 (11), 48494873.
Gretener, P.E., 1969. On the mechanics of the intrusion of sills. Can. J. Earth Sci. 6,
14151419.
Grosls, E.B., 2007. Magma reservoir failure on the terrestrial planets: assessing the importance of gravitational loading in simple elastic models. J. Volcanol. Geotherm.
Res. 166 (2), 4775.
Gualda, G.A.R., Ghiorso, M.S., 2013. The Bishop Tuff giant magma body: an alternative to
the Standard Model. Contrib. Mineral. Petrol. 166, 755775.
Gudmundsson, A., 1983. Form and dimensions of dykes in eastern Iceland. Tectonophysics
95, 295307.
Gudmundsson, A., 1986. Formation of crustal magma chambers in Iceland. Geology 14,
164166.
Gudmundsson, A., 1987. Formation and mechanics of magma reservoirs in Iceland.
Geophys. J. R. Astron. Soc. Lond. 91, 2741.
Gudmundsson, A., 1988. Effect of tensile stress concentration around magma chambers
on intrusion and extrusion frequencies. J. Volcanol. Geotherm. Res. 35, 179194.
Gudmundsson, A., 1990. Emplacement of dykes, sills and crustal magma chambers at divergent plate boundaries. Tectonophysics 176, 257275.
Gudmundsson, A., 1995. Infrastructure and mechanics of volcanic systems in Iceland.
J. Volcanol. Geotherm. Res. 64 (1), 122.
Gudmundsson, A., 1998. Magma chambers modeled as cavities explain the formation of
rift zone central volcanoes and their eruption and intrusion statistics. J. Geophys.
Res. 103 (B4), 74017412.
Gudmundsson, A., 2002. Emplacement and arrest of sheets and dykes in central volcanoes. J. Volcanol. Geotherm. Res. 116, 279298.
Gudmundsson, A., 2003. Surface stresses associated with arrested dykes in rift zones. Bull.
Volcanol. 65, 606619. http://dx.doi.org/10.1007/s00445-003-0289-7.
Gudmundsson, A., 2006. How local stresses control magma-chamber ruptures, dyke injections, and eruptions in composite volcanoes. Earth Sci. Rev. 79, 131.
Gudmundsson, A., 2011. Deection of dykes into sills at discontinuities and magmachamber formation. Tectonophysics 500 (1), 5064.
Gudmundsson, A., 2012. Magma chambers: formation, local stresses, excess pressures,
and compartments. J. Volcanol. Geotherm. Res. 237238, 1941.
Gudmundsson, A., Brenner, S.L., 2001. How hydrofractures become arrested. Terra Nova
13, 456462.
Gudmundsson, A., Brenner, S.L., 2005. On the conditions of sheet injections and eruptions
in stratovolcanoes. Bull. Volcanol. 67, 768782.
Gudmundsson, A., Ltveit, I.F., 2014. Sills as fractured hydrocarbon reservoirs: examples
and models. Geol. Soc. Lond., Spec. Publ. 374 (1), 251271.
Gudmundsson, A., Nilsen, K., 2006. Ring faults in composite volcanoes: structures,
models, and stress elds associated with their formation. J. Geol. Soc. Lond.
269, 83108.
Gudmundsson, A., Marti, J., Turon, E., 1997. Stress elds generating ring faults in volcanoes. Geophys. Res. Lett. 24, 15591562.
Gudmundsson, A., Berg, S.S., Lyslo, K.B., Skurtveit, E., 2001. Fracture networks and uid
transport in active fault zones. J. Struct. Geol. 23, 343353.
Gudmundsson, A., Friese, N., Galindo, I., Philipp, S.L., 2008. Dike-induced reverse faulting
in a graben. Geology 36, 123126. http://dx.doi.org/10.1130/G24185A.1.
Gudmundsson, A., Pasquar Mariotto, F.A., Tibaldi, A., 2014. Dykes, sills, laccoliths, and inclined sheets in Iceland. In: Breitkreuz, C., Rocchi, S. (Eds.), Laccoliths, Sills and Dykes,
Physical Geology of Shallow Level Magmatic Systems. Advances in Volcanology series
(in press).
Halinan, S., 1993. Non-chaotic collapse at funnel calderas: gravity study of the ring fractures at Guayabo Caldera, Costa Rica. Geology 21, 367370.
Hall, A., 1987. Igneous Petrology. John Wiley & Sons Inc., New York (573 pp.).
Halliday, A.N., Aftalion, M., Parsons, I., Dickin, A.P., Johnson, M.R.W., 1987. Syn-orogenic
alkaline magmatism and its relationship to the Moine Thrust Zone and the thermal
state of the Lithosphere in NW Scotland. J. Geol. Soc. 144 (4), 611617.
Hamilton, W.B., 1995. Subduction systems and magmatism. In: Smellie, J.R. (Ed.), Volcanism Associated With Extension to Consuming Plate Margins. Geol. Soc. London
Spec. Publ. vol. 81, pp. 328.
Hansen, D.M., 2006. The morphology of intrusion-related vent structures and their implications for constraining the timing of intrusive events along the NE Atlantic margin.
J. Geol. Soc. 163, 789800.
Hansen, D.M., Cartwright, J.A., Thomas, D., 2004. 3D seismic analysis of the geometry
of igneous sills and sill junctions relationships. In: Davies, R.J., Cartwright, J.A.,
Stewart, S.A., Lappin, M., Underhill, J.R. (Eds.), 3D Seismic Technology: Application
to the Exploration of Sedimentary Basins. Memoirs. Geological Society, London,
pp. 199208.
Hardee, H.C., 1982. Incipient magma chamber formation as a result of repetitive intrusions. Bull. Volcanol. 45 (1), 149.
Harker, A., 1904. The Tertiary igneous rocks of Skye. Mem. Geol. Surv. 1481.
Harrison, T.N., Brown, P.E., Dempster, T.J., Hurron, D.H.W., Becker, S.M., 1990. Granite
magmatism and extensional tectonics in southern Greenland. Geol. J. 25, 287293.
Hatayama, Y., et al., 1980. Chigaku Jiten (Geological Dictionary). The Association for Geological Collaboration, Heibonsha K.K., Tokyo (1612 pp., (in Japanese)).
Hauksson, E., 1983. Episodic rifting and volcanism at Kraa in north Iceland: growth of
large ground ssures along the plate boundary. J. Geophys. Res. 88, 626636.
Hayashi, S., Kasahara, M., Tanaka, K., Hamaguchi, H., Zana, N., 1992. Major chemistry of
recent eruptive products from Nyamuragire volcano, Africa (19761989).
Tectonophysics 209, 273276.
Heiken, G., Goff, F., Gardner, J.N., Baldridge, W.S., 1990. The Valles/Toledo Caldera Complex, Jemez Volcanic Field, New Mexico. Annu. Rev. Earth Planet. Sci. 18, 2753.

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135


Hildenbrand, A., Gillot, P., Le Roy, I., 2004. Volcano-tectonic and geochemical evolution of
an oceanic intra-plate volcano: TahitiNui (French Polynesia). Earth Planet. Sci. Lett.
217, 349365.
Hildner, E., Kluegel, A., Hansteen, T.H., 2012. Barometry of lavas from the 1951 eruption of
Fogo, Cape Verde Islands: implications for historic and prehistoric magma plumbing
systems. J. Volcanol. Geotherm. Res. 217218, 7390. http://dx.doi.org/10.1016/j.
volgeores.2011.12.014.
Hildreth, W., 1981. Gradients in silicic magma chambers: implications for lithospheric
magmatism. J. Geophys. Res. 86, 1015310192.
Hildreth, W., 1987. New perspectives on the eruption of 1912 in the Valley of Ten Thousand
Smokes, Katmai National Park, Alaska. Bull. Volcanol. 49, 680693.
Hills, E.S., 1972. Elements of Structural Geology. second edition. John Wiley & Sons, New
York.
Holohan, E.P., van Wyk de Vries, B., Troll, V.R., 2007. Analogue models of caldera collapse
in strike-slip tectonic regimes. Bull. Volcanol. http://dx.doi.org/10.1007/s00445-0070166-x.
Horsman, E., Tikoff, B., Morgan, S., 2005. Emplacement-related fabric and multiple sheets
in the Maiden Creek sill, Henry Mountains, Utah, USA. J. Struct. Geol. 27 (8),
14261444.
Hughes, G.R., Mahood, G.A., 2011. Silicic calderas in arc settings: characteristics, distribution, and tectonic controls. Geol. Soc. Am. Bull. 123 (78), 15771595.
Huppert, H., Sparks, R.S., 1981. The uid dynamics of a basaltic magma chamber
replenished by inux of hot, dense ultrabasic magma. Contrib. Mineral. Petrol. 75
(3), 279289. http://dx.doi.org/10.1007/bf01166768.
Hutton, D.H.W., 1982. A tectonic model for the emplacement of the Main Donegal granite,
NW Ireland. J. Geol. Soc. Lond. 139, 615631.
Hutton, D.H.W., 2009. Insights into magmatism in volcanic margins: bridge structures and
a new mechanism of basic sill emplacementTheron Mountains, Antarctica. Pet.
Geosci. 15 (3), 269278.
Hutton, D.H.W., Dempster, T.J., Brown, P.E., Becker, S.D., 1990. A new mechanism of granite emplacement intrusion in active extensional shear zones. Nature 343, 452455.
Ishizuka, O., Geshi, N., Itoh, J., Kawanabe, Y., Tuzino, T., 2008. The magmatic plumbing of
the submarine Hachijo NW volcanic chain, Hachijojima, Japan: long-distance magma
transport? J. Geophys. Res. 113, B08S08. http://dx.doi.org/10.1029/2007JB005325.
Ito, G., Martel, S.J., 2002. Focusing of magma in the upper mantle through dike interaction.
J. Geophys. Res. 107 (B10), 2223. http://dx.doi.org/10.1029/2001JB000251.
Jaxybulatov, K., Shapiro, N.M., Koulakov, I., Mordret, A., Lands, M., Sens-Schnfelder, C.,
2014. A large magmatic sill complex beneath the Toba caldera. Science 346, 617619.
Jellinek, A.M., DePaolo, D.J., 2003. A model for the origin of large silicic magma chambers:
precursors of caldera-forming eruptions. Bull. Volcanol. 65, 363381. http://dx.doi.
org/10.1007/s00445-003-0277-y.
Jin, Z.H., Johnson, S.E., 2008. Magmadriven multiple dike propagation and fracture toughness of crustal rocks. J. Geophys. Res. Solid Earth (19782012) 113 (B3). http://dx.doi.
org/10.1029/2006JB004761.
Johnson, R.B., 1961. Patterns and origin of radial dike swarms associated with West
Spanish peak and Dike Mountain, South-Central Colorado. Geol. Soc. Am. Bull. 72,
579590.
Johnson, S.E., Paterson, S.R., Tate, M.C., 1999. Structure and emplacement history of
multiple-center, cone-sheet-bearing ring complex: the Zarza Intrusive Complex,
Baja California, Mexico. Geol. Soc. Am. Bull. 111 (4), 607619.
Johnson, S.E., Schmidt, K.L., Tate, M.C., 2002. Ring complexes in the Peninsular Range
Batholith, Mexico and the USA: magma plumbing systems in the middle and upper
crust. Lithos 61, 187208.
Johnson, D.J., Eggers, A.A., Bagnardi, M., Battaglia, M., Poland, M.P., Miklius, A., 2010.
Shallow magma accumulation at Klauea Volcano, Hawaii, revealed by microgravity
surveys. Geology 38 (12), 11391142.
Jnsson, S., 2009. Stress interaction between magma accumulation and trapdoor faulting
on Sierra Negra volcano, Galpagos. Tectonophysics 471, 3644. http://dx.doi.org/10.
1016/j.tecto.2008.08.005.
Karlstrom, L., Dufek, J., Manga, M., 2009. Organization of volcanic plumbing through magmatic lensing by magma chambers and volcanic loads. J. Geophys. Res. 114, B10204.
http://dx.doi.org/10.1029/2009JB006339.
Karlstrom, L., Dufek, J., Manga, M., 2010. Magma chamber stability in arc and continental
crust. J. Volcanol. Geotherm. Res. 190 (3), 249270.
Karlstrom, L., Wright, H.M., Bacon, C.R., 2015. The effect of pressurized magma chamber
growth on melt migration and pre-caldera vent locations through time at Mount
Mazama, Crater Lake, Oregon. Earth Planet. Sci. Lett. 412, 209219.
Kauahikaua, J.P., 1993. Geophysical characteristics of the hydrothermal systems of Kilauea
Volcano, Hawaii. Geothermics 22, 271299.
Kauahikaua, J.P., Hildenbrand, T., Webring, M., 2000. Deep magmatic structures of Hawaiian
volcanoes, imaged by three-dimensional gravity models. Geology 28, 883886.
Kavanagh, J.L., Menand, T., Sparks, R.S.J., 2006. An experimental investigation of sill formation and propagation in layered elastic media. Earth Planet. Sci. Lett. 245, 799813.
Kennedy, B., Stix, J., 2007. Magmatic processes associated with caldera collapse at Ossipee
ring dyke, New Hampshire. Geol. Soc. Am. Bull. 119 (12), 317.
Kervyn, M., Ernst, G.G.J., van Wyk, B., de Vries, L., Mathieu, P. Jacobs, 2009. Volcano load
control on dyke propagation and vent distribution: insights from analogue modeling.
J. Geophys. Res. 114, B03401. http://dx.doi.org/10.1029/2008JB005653.
Kissel, C., Laj, C., Sigurdsson, H., Guillou, H., 2010. Emplacement of magma in Eastern
Iceland dikes: insights from magnetic fabric and rock magnetic analyses. J Volcanol
Geoth Res 191, 7992.
Klausen, M.B., 2004. Geometry and mode of emplacement of the Thverartindur cone
sheet swarm, SE Iceland. J. Volcanol. Geotherm. Res. 138, 185204.
Klausen, M.B., 2006. Geometry and mode of emplacement of dike swarms around the
Birnudalstindur igneous centre, SE Iceland. J. Volcanol. Geotherm. Res. 151 (4),
340356.

131

Klein, F.W., Koyanagi, R.Y., Nakata, J.S., Tanigawa, W.R., 1987. The seismicity of Kilauea's
magma system. U. S. Geol. Surv. Prof. Pap. 350, 10191186.
Klgel, A., Hansteen, T.H., Galipp, K., 2005. Magma storage and underplating beneath
Cumbre Vieja volcano, La Palma (Canary Islands). Earth Planet. Sci. Lett. 236 (1),
211226.
Koch, F.G., Johnson, A.M., Pollard, D.D., 1981. Monoclinal bending of strata over laccolithic
intrusions. Tectonophysics 74 (3), T21T31.
Koehler, R.D., Farrell, R.-E., Burns, P.A.C., Combellick, R.A., 2012. Quaternary faults and
folds in Alaska: a digital database. Alaska Division of Geological & Geophysical Surveys, Miscellaneous Publication 141 (31 pp.).
Kokelaar, P., Raine, P., Branney, M., 2007. Incursion of a large-volume, spatter-bearing
pyroclastic density current into a caldera lake: Pavey Ark ignimbrite, Scafell caldera,
England. Bull. Volcanol. 70, 2354. http://dx.doi.org/10.1007/s00445-007-0118-5.
Komorowski, J.C., 2002. The January 2002 ank eruption of Nyiragongo volcano (DRC):
chronology, evidence for a tectonic rift trigger and impact of lava ows on the city
of Goma. Acta Vulcanol. 14, 2557.
Khn, D., Dahm, T., 2008. Numerical modelling of dyke interaction and its inuence on
oceanic crust formation. Tectonophysics 447 (1), 5365.
La Delfa, S., Patane, G., Clocchiatti, R., Joron, J.L., Tanguy, J.C., 2001. Activity of Mount Etna
preceding the February 1999 ssure eruption: inferred mechanism from seismological and geochemical data. J. Volcanol. Geotherm. Res. 105, 121139.
Lagarde, J.L., Brun, J.P., Gapais, D., 1990. Formation of epizonal granitic plutons by in situ
assemblage of laterally expanding magma. C. R. Acad. Sci. Paris 310 (II), 11091114.
Lagmay, A.M.F., van Wyk de Vries, B., Kerle, N., Pyle, D.M., 2000. Volcano instability induced by strike-slip faulting. Bull. Volcanol. 62, 331346.
Lahr, J.C., Chouet, B.A., Stephens, C.D., Power, J.A., Page, R.A., 1994. Earthquake classication, location and error analysis in a volcanic environment implications for the
magmatic system of the 19891990 eruptions at Redoubt volcano, Alaska.
J. Volcanol. Geotherm. Res. 62, 137151.
Larsen, G., Gronvold, K., Thorarinsson, S., 1979. Volcanic eruption through a geothermal
borehole at Namafjall, Iceland. Nature 278, 707710.
Le Bas, M.J., 1971. Cone-sheets as a mechanism of uplift. Geol. Mag. 108 (05), 373376.
Leake, B.E., 1990. Granite magmas: their sources, initiation and consequences of emplacement. J. Geol. Soc. Lond. 147, 579589.
Lecuyer, F., Bellier, O., Gourgaud, A., Vincent, P.M., 1997. Tectonique active du Nord-Est de
Sulawesi (Indonesie) et controle structural de la caldeira de Tondano. C. R. Acad. Sci.
Paris 325 (8), 607613.
Lefort, P., 1981. Manaslu leucoganite a collision signature of the Himalaya a model for
its genesis and emplacement. J. Geophys. Res. 86, 545568.
Legrand, D., Calahorrano, A., Guillier, B., Rivera, L., Ruiz, M., Villagomez, D., Yepes, H., 2002.
Stress tensor analysis of the 19981999 tectonic swarm of northern Quito related to
the volcanic swarm of Guagua Pichincha volcano, Ecuador. Tectonophysics 344,
1536.
Legros, F., Kelfoun, K., Mart, J., 2000. The inuence of conduit geometry on the dynamics
of caldera-forming eruptions. Earth Planet. Sci. Lett. 179 (1), 5361.
Letourneur, L., Peltier, A., Staudacher, T., Gudmundsson, A., 2008. The effects of rock heterogeneities on dyke paths and asymmetric ground deformation: the example of
Piton de la Fournaise (Runion Island). J. Volcanol. Geotherm. Res. 173 (3), 289302.
Lin, G., Amelung, F., Lavallee, Y., Okubo, P.G., 2014. Seismic evidence for a crustal magma
reservoir beneath the upper east rift zone of Kilauea volcano, Hawaii. Geology http://
dx.doi.org/10.1130/G35001.1.
Lipman, P.W., 1997. Subsidence of ash-ow calderas: relation to caldera size and magmachamber geometry. Bull. Volcanol. 59, 198218.
Lipman, P.W., 2007. Incremental assembly and prolonged consolidation of Cordilleran
magma chambers: Evidence from the Southern Rocky Mountain volcanic eld.
Geosphere 3 (1), 4270.
Lister, J.R., 1990. Buoyancy-driven uid fracture: the effects of material toughness and of
low-viscosity precursors. J. Fluid Mech. 210, 263280.
Lister, J.R., Kerr, R.C., 1991. Fluidmechanical models of crack propagation and their
application to magma transport in dykes. J. Geophys. Res. Solid Earth 96 (B6),
1004910077.
Lloyd, A.S., Ruprecht, P., Hauri, E.H., Rose, W., Gonnermann, H.M., Plank, T., 2014.
NanoSIMS results from olivine-hosted melt embayments: magma ascent rate during
explosive basaltic eruptions. J. Volcanol. Geotherm. Res. 283, 118.
Lowenstern, J.B., Wallmann, P.C., Pollard, D.D., 1991. The west Mageik lake sill complex as
an analogue for magma transport during the 1912 eruption at the valley of Ten
Thousand Smokes, Alaska. Geophys. Res. Lett. 18 (8), 15691572.
Maale, S., 1987. The generation and shape of feeder dykes from mantle sources. Contrib.
Mineral. Petrol. 96 (1), 4755.
Maale, S., Scheie, ., 1982. The permeability controlled accumulation of primary magma.
Contrib. Mineral. Petrol. 81 (4), 350357.
Maccaferri, F., Rivalta, E., Keir, D., Acocella, V., 2014. Off-rift volcanism in rift zones determined by crustal unloading. Nat. Geosci. 7, 297300. http://dx.doi.org/10.1038/
NGEO2110.
MacLeod, N.S., Sherrod, D.R., Chitwood, L.A., Jensen, R.A., 1995. Geologic map of Newberry
Volcano, Deschutes, Klamath, and Lake Counties, Oregon: U.S. Geological Survey
Miscellaneous Investigations Series Map I-2455, 2 sheets, scale 1:62,500, pamphlet,
23 p.
Magee, C., Stevenson, C., O'Driscoll, B., Schoeld, N., McDermott, K., 2012. An alternative
emplacement model for the classic Ardnamurchan cone sheet swarm, NW Scotland,
involving lateral magma supply via regional dykes. J. Struct. Geol. 43, 7391.
Mahood, G.A., Hildreth, W., 1986. Geology of the peralkaline volcano at Pantelleria, Strait
of Sicily. Bull. Volcanol. 48, 143172.
Malthe-Srenssen, A., Planke, S., Svensen, H., Jamtveit, B., 2004. Formation of saucer
shaped sills. In: Breitkreuz, C., Petford, N. (Eds.), Physical Geology of High-level Magmatic Systems: Geological Society. Special Publication, London, pp. 215227.

132

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

Manconi, A., Longpr, M.-A., Walter, T.R., Troll, V.R., Hansteen, T.H., 2009. The effects of
ank collapses on volcano plumbing systems. Geology 37 (12), 10991102. http://
dx.doi.org/10.1130/G30104A.1.
Mandl, G., 1988. Mechanics of Tectonic Faulting: Models and Basic Concepts. Elsevier,
Amsterdam (407 pp.).
Marcotte, S.B., Klepeis, K.A., Clarke, G.L., Gehrels, G., Hollis, J.A., 2005. Intra-arc
transpression in the lower crust and its relationship to magmatism in a Mesozoic
magmatic arc. Tectonophysics 407, 135163.
Marra, F., 2001. Strike-slip faulting and block rotation: a possible triggering mechanism
for lava ows in the Alban Hills? J. Struct. Geol. 23, 127141.
Marsh, B., 1989. On the convective style and vigour in sheet-like magmac hambersd.
Petrol 30, 479530.
Marsh, B.D., 1982. On the mechanics of igneous diapirism, stoping, and zone-melting. Am.
J. Sci. 282, 808855.
Marsh, B.D., 2000. Magma chambers. Encyclopedia of volcanoes 1, 191206.
Marsh, B.D., 2004. A magmatic mush column Rosetta Stone: the McMurdo dry valleys of
Antarctica. EOS Trans. Am. Geophys. Union 85, 497508.
Mart, J., Geyer, A., 2009. Central vs. ank eruptions at TeidePico Viejo twin stratovolcanoes (Tenerife, Canary Islands). J. Volcanol. Geotherm. Res. 181, 4760.
Mart, J., Mitjavila, J., Roca, E., Aparicio, A., 1992. Cenozoic magmatism of the Valencia
trough (western Mediterranean): relationship between structural evolution and volcanism. Tectonophysics 203 (1), 145165.
Massonnet, D., Sigmundsson, F., 2000. Remote sensing of volcano deformation by radar
interferometry from various satellites. Geophys. Monogr. Ser. 116, 207221.
Mastin, L.G., Pollard, D.D., 1988. Surface deformation and shallow dike intrusion processes
at Inyo Craters, Long Valley, California. J. Geophys. Res. 1322113235.
Mathieu, L., van Wyk de Vries, B., 2009. Edice and substrata deformation induced by intrusive complexes and gravitational loading in the Mull volcano (Scotland). Bull.
Volcanol. 71 (10), 11331148. http://dx.doi.org/10.1007/s00445-009-0295-5.
Mattia, M., Patane, D., Aloisi, M., Amore, M., 2007. Faulting on the western ank of Mt Etna
and magma intrusions in the shallow crust. Terra Nova 19, 8994.
Mazzuoli, R., Tortorici, L., Ventura, G., 1995. Oblique rifting in Salina, Lipari and Vulcano
islands (Aeolian islands, southern Italy). Terra Nova 7, 444452.
McCall, G.J.H., Bristow, C.M., 1965. An introductory account of Suswa volcano, Kenya. Bull.
Volcanol. 28, 333367.
McGuire, W.J., Pullen, A.D., 1989. Location and orientation of eruptive ssures and
feederdykes at Mount Etna; inuence of gravitational and regional tectonic stress regimes. J. Volcanol. Geotherm. Res. 38.3, 325344.
McGuire, W.J., Pullen, A.D., Saunders, S.J., 1990. Recent dyke-induced large-scale block
movement at Mt. Etna and potential slope failure. Nature 343, 357359.
McGuire, W., Murray, B., Pullen, A.D., Saunders, S.J., 1991. Ground deformation monitoring
at Mt. Etna; evidence for dyke emplacement and slope instability. J. Geol. Soc. London
148, 577583.
McKenzie, D., 1984. A possible mechanism for epeirogenic uplift. Nature 307 (5952),
616618.
McKenzie, D., McKenzie, J.M., Saunders, R.S., 1992. Dike emplacement on Venus and on
Earth. J. Geophys. Res. 97, 15,97715,990.
McLeod, P., Tait, S., 1999. The growth of dykes from magma chambers. J. Volcanol.
Geotherm. Res. 92 (3), 231245.
McNulty, B.A., Tong, W., Tobisch, O.T., 1996. Assembly of a dyke-fed magma chamber: the
Jackass Lakes pluton, central Sierra Nevada, California. Geol. Soc. Am. Bull. 108 (8),
926940.
Memeti, V., Paterson, S.R., Matzel, J., Mundil, R., Okaya, D., 2010. Magmatic lobes as
snapshots of magma chamber growth and evolution in large, composite batholiths:
an example from the Tuolumne Intrusion, Sierra Nevada, CA. GSA Bull. http://dx.doi.
org/10.1130/B30004.1.
Menand, T., 2008. The mechanics and dynamics of sills in layered elastic rocks and their
implications for the growth of laccoliths and other igneous complexes. Earth Planet.
Sci. Lett. 267 (1), 9399. http://dx.doi.org/10.1016/j.epsl.2007.11.043.
Menand, T., 2011. Physical controls and depth of emplacement of igneous bodies: a review. Tectonophysics 500, 1119.
Menand, T., Tait, S.R., 2002. The propagation of a buoyant liquid-lled ssure from a
source under constant pressure: an experimental approach. J. Geophys. Res. 107
(B11), 2306.
Menand, T., Daniels, K.A., Benghiat, P., 2010. Dyke propagation and sill formation in
a compressive tectonic environment. J. Geophys. Res. Solid Earth (19782012)
115 (B8).
Meriaux, C., Jaupart, C., 1998. Crack propagation through an elastic plate. J. Geophys. Res.
103 (B8), 1829518314.
Meriaux, C., Lister, J.R., 2002. Calculation of dike trajectories from volcanic centers.
J. Geophys. Res. 107 (B4), 2077. http://dx.doi.org/10.1029/2001JB000436.
Michaut, C., 2011. Dynamics of magmatic intrusions in the upper crust: theory and applications to laccoliths on Earth and the Moon. J. Geophys. Res. Solid Earth (19782012)
116 (B5).
Michon, L., Saint-Ange, F., Bachelery, P., Villeneuve, N., Staudacher, T., 2007. Role of the
structural inheritance of the oceanic lithosphere in the magmato-tectonic evolution
of Piton de la Fournaise volcano (La Runion Island). J. Geophys. Res. 112, B04205.
http://dx.doi.org/10.1029/2006JB004598.
Miller, C.F., Miller, J.S., 2002. Contrasting stratied plutons exposed in tilt blocks, Eldorado
Mountains, Colorado River rift, Nevada, USA. Lithos 61, 209224.
Miller, R.B., Paterson, S.R., 1999. In defense of magmatic diapirs. J. Struct. Geol. 21,
11611173.
Miller, R.B., Paterson, S.R., 2001. Construction of mid-crustal sheeted plutons: examples
from the north Cascades, Washington. Geol. Soc. Am. Bull. 113 (11), 14231442.
Mitjavila, J., Marti, J., Soriano, C., 1997. Magmatic evolution and tectonic setting of the Iberian Pyrite Belt volcanism. J. Petrol. 38 (6), 727755.

Miura, D., 1999. Arcuate pyroclastic conduits, ring faults, and coherent oor at Kumano
caldera, southwest Honshu, Japan. J. Volcanol. Geotherm. Res. 92 (3), 271294.
Miura, S., Ueki, S., Sato, T., Tachibana, K., Hamaguchi, H., 2000. Crustal deformation associated with the 1998 seismo-volcanic crisis of Iwate Volcano, Northeastern Japan, as
observed by a dense GPS network. Earth Planets Space 52, 10031008.
Mkweli, S., Kamber, B., Berger, M., 1995. Westward continuation of the cratonLimpopo
Belt tectonic break in Zimbabwe and new age constraints on the timing of the thrusting. J. Geol. Soc. 152 (1), 7783.
Moore, I., Kokelaar, P., 1998. Tectonically controlled piecemeal caldera collapse: a case
study of Glencoe volcano, Scotland. Geol. Soc. Am. Bull. 110, 14481466.
Moore, J.G., Normark, W.R., Holcomb, R.T., 1994. Giant Hawaiian landslides. Annu. Rev.
Earth Planet. Sci. 22, 119144.
Mori, J., McKee, C., 1987. Outward-dipping ring-fault structure at Rabaul Caldera as shown
by earthquake locations. Science 235, 193195. http://dx.doi.org/10.1126/science.
235.4785.193.
Muirhead, J.D., Airoldi, G., Rowland, J.V., White, J.D., 2012. Interconnected sills and inclined sheet intrusions control shallow magma transport in the Ferrar large igneous
province, Antarctica. Geol. Soc. Am. Bull. 124 (12), 162180.
Muller, O.H., 1986. Changing stresses during emplacement of the radial dike swarm at
Spanish Peaks, Colorado. Geology 14 (2), 157159.
Muller, O.H., Pollard, D.D., 1977. The stress state near Spanish Peaks, Colorado, determined from a dike pattern. Pure Appl. Geophys. 115, 6986.
Murase, T., Mcbirney, A.R., 1973. Properties of some common igneous rocks and their
melts at high temperatures. Geol. Soc. Am. Bull. 84, 35633592.
Nairn, I.A., McKee, C.O., Talai, B., Wood, C.P., 1995. Geology and eruptive history of the
Rabaul caldera area, Papua New Guinea. J. Volcanol. Geotherm. Res. 69 (34),
255284.
Nairn, I.A., Kobayashi, T., Nakagawa, M., 1998. The ~10 ka multiple vent pyroclastic eruption sequence at Tongariro Volcanic Centre, Taupo Volcanic Zone, New Zealand. Part
1. Eruptive processes during regional extension. J. Volcanol. Geotherm. Res. 86,
1944.
Nakada, S., Nagai, M., Kaneko, T., Nozawa, A., Suzuki-Kamata, K., 2005. Chronology and
products of the 2000 eruption of Miyakejima Volcano, Japan. Bull. Volcanol. 67,
205218.
Nakamura, K., 1977. Volcanoes as possible indicators of tectonic stress orientation:
principle and proposal. J. Volcanol. Geotherm. Res. 2, 116.
Nakamura, K., Uyeda, S., 1980. Stress gradient in arc-back arc regions and plate subduction. J. Geophys. Res. 85, 64196428.
Nakamura, K., Jacob, K.H., Davies, J.N., 1977. Volcanoes as possible indicators of tectonic
stress orientation Aleutinian and Alaska. Pure Appl. Geophys. 115, 87112.
Nakamura, K., Plakfer, G., Jacob, K.H., Davies, J.N., 1980. A tectonic stress trajectory map of
Alaska using information from volcanoes and faults. Bull. Earthquake Res. Inst. 55,
89100.
Neri, G., Barberi, G., Orecchio, B., Mostaccio, A., 2003. Seismic strain and seismogenic stress
regimes in the crust of the southern Tyrrhenian region. Earth Planet. Sci. Lett. 213,
97112. http://dx.doi.org/10.1016/S0012-821X(03)00293-0.
Newman, A.V., Dixon, T.H., Gourmelen, N., 2006. A four-dimensional viscoelastic deformation model for Long Valley caldera, California, between 1995 and 2000. J. Volcanol.
Geotherm. Res. 150, 244269. http://dx.doi.org/10.1016/j.jvolgeores.2005.07.017.
Nicholson, R., Pollard, D.D., 1985. Dilation and linkage of echelon cracks. J. Struct. Geol. 7,
583590.
Nicolas, A., Boudier, F., Ildefonse, B., 1994. Dyke patterns in diapirs beneath oceanic ridges:
the Oman Ophiolite. In: Ryan, M. (Ed.), Magmatic Systems. Academic Press, New
York, pp. 7795.
Nunn, J., 1996. Buoyancy-driven propagation of isolated uid-lled fractures: implications for uid transport in the Gulf of Mexico. J. Geophys. Res. 101, 29632970.
Od, H., 1957. Mechanical analysis of the dike pattern of the Spanish Peaks area, Colorado.
Geol. Soc. Am. Bull. 68, 567576.
O'Driscoll, B., Troll, V.R., Reavy, R.J., Turner, P., 2006. The Great Eucrite intrusion of
Ardnamurchan, Scotland: reevaluating the ring-dike concept. Geology 34 (3),
189192.
Olivier, P., Ameglio, L., Richen, H., Vadeboin, F., 1999. Emplacement of the Aya Variscan
granitic pluton (Basque Pyrenees) in a dextral transcurrent regime inferred from a
combined magneto-structural and gravimetric study. J. Geol. Soc. Lond. 156,
9911002.
Ozerov, A.Yu., Ariskin, A.A., Kyle, P., Bogoyavlenskaya, G.E., Karpenko, S.F., 1997. Petrological
geochemical model for genetic relationships between basaltic and andesitic magmatism
of Klyuchevskoi and Bezymyannyi volcanoes, Kamchatka. Petrology 5, 550569.
Parsons, T., Sleep, N.H., Thompson, G.A., 1992. Host rock rheology controls on the
emplacement of tabular intrusions: implications for underplating of extended crust.
Tectonics 11, 13481356.
Pasquar, F.A., Tibaldi, A., 2003. Do transcurrent faults guide volcano growth? The case of
NW Bicol Volcanic Arc, Luzon, Philippines. Terra Nova 15 (3), 204212.
Pasquar, F.A., Tibaldi, A., 2007. Structure of a sheet-laccolith system revealing the interplay between tectonic and magma stresses at Stardalur Volcano, Iceland. J. Volcanol.
Geotherm. Res. 161, 131150.
Pasquar, G., Francalanci, L., Garduno, V.H., Tibaldi, A., 1993. Structure and geological
evolution of the Stromboli volcano, Aeolian islands, Italy. Acta Volcanol. Pisa 3,
7989.
Paterson, S.R., Miller, R.B., 1998. Mid-crustal magmatic sheets in the Cascades Mountains,
Washington: implications for magma ascent. J. Struct. Geol. 20, 13451363.
Paterson, S.R., Vernon, R.H., 1995. Bursting the bubble of ballooning plutons: a return to
nested diapers emplaced by multiple processes. GSA Bull. 107, 13561380.
Paterson, S.R., Tobisch, O.T., Morand, V.J., 1990. The inuence of large ductile shear zones
on the emplacement and deformation of the Wyangala Batholith, SE Australia.
J. Struct. Geol. 12 (5/6), 639650.

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135


Paterson, S.R., Memeti, V., Putirka, K., Paige, M., 2010. Volcano-plutonic plumbing systems: a comparison of the Jurassic Guadalupe Igneous Complex to the Cretaceous
Tuolumne batholith, central Sierra Nevada, California. GIC Field Trip Guide (41 pages).
Pedersen, R., Sigmundsson, F., 2004. InSAR based sill model links spatially offset areas of
deformation and seismicity for the 1994 unrest episode at Eyjafjallajokull volcano,
Iceland. Geophys. Res. Lett. 31, L14610. http://dx.doi.org/10.1029/2004GL020368.
Peltier, A., Ferrazzini, V., Staudacher, T., Bachelery, P., 2005. Imaging the dynamics of dyke
propagation prior to the 20002003 ank eruptions at Piton de La Fournaise, Reunion
Island. Geophys. Res. Lett. 32. http://dx.doi.org/10.1029/2005GL023720.
Peltier, A., Bachlery, P., Staudacher, T., 2009. Magma transport and storage at Piton de La
Fournaise (La Reunion) between 1972 and 2007: a review of geophysical and geochemical data. J. Volcanol. Geotherm. Res. 184, 93108. http://dx.doi.org/10.1016/j.
jvolgeores.2008.12.008.
Petford, N., Atherton, M.P., 1992. Granitoid emplacement and deformation along a major
crustal lineament: the Cordillera Blanca, Peru. Tectonophysics 205, 171185.
Petford, N., Gallagher, K., 2001. Partial melting of mac (amphibolitic) lower crust by periodic inux of basaltic magma. Earth Planet. Sci. Lett. 5983, 117.
Petford, N., Cruden, A.R., Mccaffrey, K.J.W., Vigneresse, J.L., 2000. Granite magma formation, transport and emplacement in the Earth's crust. Nature 408, 669673.
Phillips, W.J., 1974. The dynamic emplacement of cone sheets. Tectonophysics 24,
6984.
Pinel, V., Jaupart, C., 2004. Magma storage and horizontal dyke injection beneath a volcanic edice. Earth Planet. Sci. Lett. 221, 245262.
Piochi, M., Kilburn, C.R.J., Di Vito, M.A., Mormone, A., Tramelli, A., Troise, C., De Natale, G.,
2014. The volcanic and geothermally active Campi Flegrei caldera: an integrated multidisciplinary image of its buried structure. Int. J. Earth Sci. 103 (2), 401421.
Pitcher, W.S., 1978. The anatomy of a batholith. J. Geol. Soc. Lond. 135, 157182.
Pitcher, W.S., 1979. The nature, ascent and emplacement of granitic magmas. J. Geol. Soc.
136, 627662.
Pitcher, W.S., Berger, A.R., 1972. The Geology of Donegal: A Study of Granite Emplacement
and Unroong. Wiley, New York (435 pp.).
Planke, S., Rasmussen, T., Rey, S.S., Myklebust, R., 2005. Seismic characteristics and distribution of volcanic intrusions and hydrothermal vent complexes in the Vring and
Mre basins. In: Dor, A.G., Vining, B.A. (Eds.), Petroleum Geology: North-West
Europe and Global Perspectives. Proceedings of the 6th Petroleum Geology Conference. Geological Society, London, pp. 833844.
Poland, M.P., Fink, J.H., Tauxe, L., 2004. Patterns of magma ow in segmented silicic dikes
at Summer Coon volcano, Colorado: AMS and thin section analysis. Earth Planet. Sci.
Lett. 219, 155169.
Poland, M.P., Moats, W.P., Fink, J.H., 2008. A model for radial dike emplacement in composite cones based on observations from Summer Coon volcano, Colorado, USA.
Bull. Volcanol. 70, 861875. http://dx.doi.org/10.1007/s00445-007-0175-9.
Pollard, D.D., 1987. Elementary fracture mechanics applied to the structural interpretation
of dykes. Mac Dyke Swarms 34, 524.
Pollard, D.D., Johnson, A.M., 1973. Mechanics of growth of some laccolithic intrusions in
the Henry Mountains, Utah. II. Bending and failure of overburden layers and sill formation. Tectonophysics 18, 311354.
Pollard, D.D., Muller, O.H., Dockstader, D.R., 1975. The form and growth of ngered sheet
intrusions. Geol. Soc. Am. Bull. 86 (3), 351363.
Pollard, D.D., Delaney, P.T., Dufeld, W.A., Endo, E.T., Okamura, A.T., 1983. Surface deformation in volcanic rift zones. Tectonophysics 94, 541584.
Polteau, S., Mazzini, A., Galland, O., Planke, S., Malthe-Sorenssen, A., 2008. Saucer-shaped
intrusions: occurrences, emplacement and implications. Earth Planet. Sci. Lett. 266
(12), 195204.
Porreca, M., Acocella, V., Massimi, E., Mattei, M., Funiciello, R., De Benedetti, A.A., 2006.
Geometric and kinematic features of the dike complex at Mt. Somma, Vesuvio
(Italy). Earth Planet. Sci. Lett. 245, 389407. http://dx.doi.org/10.1016/j.epsl.2006.
02.027.
Preston, R.J., 2001. Composite minor intrusions as windows into subvolcanic magma reservoir processes: mineralogical and geochemical evidence for complex magmatic
plumbing systems in the British Tertiary Igneous Province. J. Geol. Soc. 158, 4758.
Quareni, F., Ventura, G., Mulargia, F., 2001. Numerical modelling of the transition from ssure to central-type activity on volcanoes: a case study from Salina Island, Italy. Phys.
Earth Planet. Inter. 124, 213221.
Quick, J.E.S., Sinigoi, S., Mayer, A., 1994. Emplacement dynamics of a large mac intrusion
in the lower crust, IvreaVerbano zone, Northern Italy. J. Geophys. Res. 99,
2155921573.
Ramelow, J., Riller, U., Romer, R.L., Oncken, O., 2006. Kinematic link between episodic
trapdoor collapse of the Negra Muerta Caldera and motion on the OlacapatoEl
Toro Fault Zone, southern central Andes. Int. J. Earth Sci. (Geol. Rundsch.) 95, 529541.
Ranalli, G., 1995. Rheology of the Earth. 2nd ed Chapman Hall, London, p. 413.
Richey, J.E., Thomas, H.H., 1930. The Geology of Ardnamurchan, North-west Mull and
Coll. Memoir of the Geological Survey of Great Britain, Sheet 51 and 52, Scotland
(393 pp.).
Rickwood, P.C., 1990. The anatomy of a dyke and the determination of propagation and
magma ow directions. In: Parker, A.J., Rickwood, P.C., Tucker, D.H. (Eds.), Mac
Dykes and Emplacement Mechanisms. Balkema, Rotterdam, pp. 81100.
Riller, U., Petrinovic, I., Ramelow, J., Strecker, M., Oncken, O., 2001. Late Cenozoic tectonism, collapse caldera and plateau formation in the central Andes. Earth Planet. Sci.
Lett. 188, 299311.
Rittmann, A., 1962. Volcanoes and Their Activity. Interscience Publishers.
Robson, G.R., Barr, K.G., 1964. The effect of stress on faulting and minor intrusions in the
vicinity of a magma body. Bull. Volcanol. 27, 315330.
Rodrguez-Losada, J.A., Martinez-Frias, J., 2004. The felsic complex of the Vallehermoso
Caldera: interior of an ancient volcanic system (La Gomera, Canary Islands).
J. Volcanol. Geotherm. Res. 137 (4), 261284.

133

Roman, D.C., Gardine, M.D., 2013. Seismological evidence for long-term and rapidly accelerating magma pressurization preceding the 2013 eruption of Redoubt Volcano,
Alaska. Earth Planet. Sci. Lett. 371372, 226234.
Roman-Berdiel, T., Gapais, D., Brun, J.P., 1995. Analogue models of laccolith formation.
J. Struct. Geol. 17 (9), 13371346.
Roni, E., Westerman, D.S., Dini, A., Stevenson, C., Rocchi, S., 2014. Feeding and growth of a
dykelaccolith system (Elba Island, Italy) from AMS and mineral fabric data. J. Geol.
Soc. 171 (3), 413424.
Roper, S.M., Lister, J.R., 2005. Buoyancy-driven crack propagation from an overpressure
source. J. Fluid Mech. 536, 7998.
Rosenberg, C.L., 2004. Shear zones and magma ascent: a model based on a review of the
Tertiary magmatism in the Alps. Tectonics http://dx.doi.org/10.1029/2003TC001526.
Rossetti, F., Storti, F., Salvini, F., 2000. Cenozoic noncoaxial transtension along the western
shoulder of the Ross Sea, Antarctica, and the emplacement of McMurdo dyke arrays.
Terra Nova 12, 6066.
Rubin, A.M., 1993. Dykes vs. diapirs in viscoelastic rock. Earth Planet. Sci. Lett. 119,
641659.
Rubin, A.M., 1995a. Propagation of magma-lled crack. Annu. Rev. Earth Planet. Sci. 23,
287336.
Rubin, A.M., 1995b. Getting granite dikes out of the source region. J. Geophys. Res. 100
(B4), 59115929. http://dx.doi.org/10.1029/94JB02942.
Rubin, A.M., Pollard, D.D., 1987. Origins of blade-like dikes in volcanic rift zones. In:
Decker, R.W., Wight, T.L., Stuffer, P.H. (Eds.), Volcanism in Hawaii. US Geological
Survey Professional Papers 1350, pp. 14491470.
Rubin, A.M., Pollard, D.D., 1988. Dike-induced faulting in rift zones of Iceland and Afar.
Geology 16 (5), 413417.
Ruprecht, P., Plank, T., 2013. Feeding andesitic eruptions with a high-speed connection
from the mantle. Nature 500, 6872. http://dx.doi.org/10.1038/nature12342.
Russo, G., Giberti, G., 2000. Mechanical stability of Mt. Vesuvius volcano effects of
asymmetries on the stress eld. Surv. Geophys. 21, 407421.
Russo, G., Giberti, G., Sartoris, G., 1996. The inuence of regional stresses on the mechanical stability of volcanoes: Stromboli (Italy). In: McGuire, W.J., Jones, A.P., Neuberg, J.
(Eds.), Geological Society Special Publication. Volcano Instability on the Earth and
Other Planets 110, pp. 6575.
Russo, G., Giberti, G., Sartoris, G., 1997. Numerical modeling of surface deformation and
mechanical stability of Vesuvius volcano, Italy. J. Geophys. Res. 102, 2478524800.
Ryan, M.P., 1987. Neutral buoyancy and the mechanical evolution of magmatic systems.
In: Mysen, B.O. (Ed.), Magmatic Processes: Physicochemical Principles. Spec. Publ.
Geochem. Soc. 1, pp. 259287.
Ryan, M.P., 1988. The mechanics and three-dimensional internal structure of active magmatic systems: Kilauea Volcano, Hawaii. J. Geophys. Res. Solid Earth 93 (B5),
42134248.
Saint Blanquat, M., Tikoff, B., Teyssier, C., Vigneresse, J.L., 1998. Transpressional kinematics
and magmatic arcs. In: Holdsworth, R.E., Strachan, R.A., Dewey, J.F. (Eds.), Continental
Transpressional and Transtensional Tectonics. Special Publications vol. 135. Geological Society of London, pp. 327340.
Sanchez, J.J., Wyss, M., Mcnutt, S.R., 2004. Temporalspatial variations of stress at Redoubt
volcano, Alaska, inferred from inversion of fault plane solutions. J. Volcanol.
Geotherm. Res. 130, 130.
Sartoris, G., Pozzi, J.P., Phillippe, C., Mouel, J.L.L., 1990. Mechanical stability of shallow
magma chambers. J. Geophys. Res. 95 (B4), 51415151. http://dx.doi.org/10.1029/
JB095iB04p05141.
Saunders, S.J., 2005. The possible contribution of circumferential fault intrusion to caldera
resurgence. Bull. Volcanol. 67 (1), 5771.
Saunders, A.D., Jones, S.M., Morgan, L.A., Pierce, K., Widdowson, M., Xu, Y.G., 2007. Regional uplift associated with continental large igneous provinces: the roles of mantle
plumes and the lithosphere. Chem. Geol. 241 (3), 282318.
Scarrow, J.H., Vaughan, A.P.M., Leat, P.T., 1997. Ridge-trench collision-induced switching
of arc tectonics and magma sources: clues from Antarctic Peninsula mac dykes.
Terra Nova 9 (56), 255259.
Schirnick, C., van den Bogaard, P., Schmincke, H.-U., 1999. Cone sheet formation and intrusive growth of an oceanic island the Miocene Tejeda complex on Gran Canaria
(Canary Islands). Geology 27 (3), 207210.
Schmidt, C.J., Smedes, H.W., O'neill, J.M., 1990. Syncompressional emplacement of the
Boulder and Tobacco Root Batholiths (Montana, USA) by pull-apart along old fault
zones. Geol. J. 25, 305318.
Schoeld, N.J., Brown, D.J., Magee, C., Stevenson, C.T., 2012. Sill morphology and comparison
of brittle and non-brittle emplacement mechanisms. J. Geol. Soc. 169 (2), 127141.
Secor, D., Pollard, D., 1975. On the stability of open hydraulic fractures in the Earth's crust.
Geophys. Res. Lett. 2, 510513.
Self, S., Rampino, M.R., 1981. The 1883 eruption of Krakatau. Nature 294, 699704.
Sewell, R.J., Tang, D.L., Campbell, S.D.G., 2012. Volcanicplutonic connections in a tilted
nested caldera complex in Hong Kong. Geochem. Geophys. Geosyst. 13 (1). http://
dx.doi.org/10.1029/2011GC003865.
Shaw, H.R., 1965. Comments on viscosity, crystal settling and convection in granitic systems. Am. J. Sci. 263, 120152.
Shaw, H.R., 1980. Fracture mechanisms of magma transport from the mantle to the surface. In: Hargreaves, R.B. (Ed.), Physics of Magmatic Processes. Princeton University
Press, Princeton, pp. 201264.
Shimura, T., 1992. Intrusion of granitic magma and uplift tectonics of the Hidaka metamorphic belt, Hokkaido. J. Geol. Soc. Jpn. 98 (1), 120.
Sigmundsson, F., 2006. Magma does the splits. Nature 442, 251252.
Simkin, T., Howard, K.A., 1970. Caldera collapse in the Galapagos Islands, 1968. Science
169, 429437.
Sinton, J.M., Detrick, R.S., 1992. Midocean ridge magma chambers. J. Geophys. Res. Solid
Earth 97.B1 (19782012), 197216.

134

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135

Smith, R.P., 1987. Dyke emplacement at Spanish Peaks, Colorado. In: Halls, H.C., Fahrig,
W.F. (Eds.), Mac Dyke Swarms. Geological Association of Canada Special Paper 34,
pp. 4754.
Smith, R.L., Bailey, R.A., 1968. Resurgent cauldrons. Geol. Soc. Am. Mem. 116, 613662.
Sparks, R.S.J., 1988. Petrology and geochemistry of the Loch Ba ring-dyke, Mull (NW
Scotland): an example of the extreme differentiation of tholeiitic magmas. Contrib.
Mineral. Petrol. 100, 446461.
Sparks, R.S.J., Huppert, H.E., Turner, J.S., Sakuyama, M., O'Hara, J., 1984. The uid dynamics
of evolving magma chambers [and discussion]. Philos. Trans. R. Soc. London, Ser. A
310, 511534.
Spence, D.A., Sharp, P., 1985. Self-similar solutions for elastohydrodynamic cavity ow.
Proc. R. Soc. Lond. A400, 289313.
Spence, D.A., Sharp, P., Turcotte, D.L., 1987. Buoyancy-driven crack propagation: a mechanism for magma migration. J. Fluid Mech. 174, 135153.
Stachel, T., Brey, G., Stanistreet, I.G., 1994. Gross Brukkaros the unusual intracaldera
sediments and their magmatic components. Commun. Geol. Surv. Namibia 9, 2341.
Steven, T.A., Lipman, P.W., 1976. Calderas of the San Juan volcanic eld, southwestern
Colorado. U. S. Geol. Surv. Prof. Pap. 95 (35 pp.).
Stevens, B., 1911. The laws of intrusions. Bull. Am. Inst. Min. Eng. 41, 123.
Stevenson, C.T., O'Driscoll, B., Holohan, E.P., Couchman, R., Reavy, R.J., Andrews, G.D., 2008.
The structure, fabrics and AMS of the Slieve Gullion ring-complex, Northern Ireland:
testing the ring-dyke emplacement model. Geol. Soc. Lond., Spec. Publ. 302 (1),
159184.
Tait, S., Jaupart, C., Vergniolle, S., 1989. Pressure, gas content and eruption periodicity of a
shallow, crystallizing magma chamber. Earth Planet. Sci. Lett. 92, 107123. http://dx.
doi.org/10.1016/0012-821X(89)90025-3.
Takada, A., 1990. Experimental study on propagation of liquid-lled crack in gelatin:
shape and velocity in hydrostatic stress conditions. J. Geophys. Res. 95, 84718481.
Takada, A., 1997. Cyclic ank-vent and central-vent eruption patterns. Bull. Volcanol. 58,
539556.
Takada, A., 1999. Variations in magma supply and magma partitioning: the role of tectonic settings. J. Volcanol. Geotherm. Res. 93 (1), 93110.
Thomson, K., 2007. Determining magma ow in sills, dykes and laccoliths and their implications for sill emplacement mechanisms. Bull. Volcanol. 70 (2), 183201.
Tibaldi, A., 1995. Morphology of pyroclastic cones and tectonics. J. Geophys. Res. 100
(B12), 24,52124,535.
Tibaldi, A., 1996. Mutual inuence of diking and collapses at Stromboli volcano, Aeolian
Arc, Italy. Geol. Soc. Spec. Pub. 110, 5563.
Tibaldi, A., 2001. Multiple sector collapses at Stromboli volcano, Italy: how they work.
Bull. Volcanol. 63 (2/3), 112125.
Tibaldi, A., 2003. Inuence of volcanic cone morphology on dykes, Stromboli, Italy.
J. Volcanol. Geotherm. Res. 126, 7995.
Tibaldi, A., 2004. Major changes in volcano behaviour after a sector collapse: insights from
Stromboli, Italy. Terra Nova 16, 28.
Tibaldi, A., 2008. Contractional tectonics and magma paths in volcanoes. J. Volcanol.
Geotherm. Res. 176, 291301.
Tibaldi, A., Groppelli, G., 2002. Volcano-tectonic activity along structures of the unstable
NE ank of Mt. Etna (Italy) and their possible origin. J. Volcanol. Geotherm. Res.
115 (3), 277302.
Tibaldi, A., Pasquar, F.A., 2008. A new mode of inner volcano growth: the ower intrusive structure. Earth Planet. Sci. Lett. 271, 202208.
Tibaldi, A., Romero, Leon J., 2000. Morphometry of Late PleistoceneHolocene faulting in
the southern Andes of Colombia and volcano-tectonic relationships. Tectonics 19 (2),
358377.
Tibaldi, A., Vezzoli, L., 1998. The space problem of caldera resurgence: an example from
Ischia Island, Italy. Geogr. Rundsch. 87, 5366.
Tibaldi, A., Lagmay, A.M.F., Ponomareva, V.V., 2005. Effects of basement structural and
stratigraphic heritages on volcano behaviour and implications for human activities
(the UNESCO/IUGS/IGCP project 455). Episodes. J. Int. Geosci. 28 (3), 113.
Tibaldi, A., Corazzato, C., Kozhurin, A., Lagmay, A.F.M., Pasquar, F.A., Ponomareva, V., Rust,
D., Tormey, D., Vezzoli, L., 2008a. Inuence of substrate tectonic heritage on the evolution of volcanoes: predicting sites of ank eruptions, lateral collapses, and erosion.
Glob. Planet. Chang. 61, 151174.
Tibaldi, A., Vezzoli, L., Pasquar, F.A., Rust, D., 2008b. Strike-slip fault tectonics and the emplacement of sheetlaccolith systems: the Thverfell case study (SW Iceland). J. Struct.
Geol. 30, 274290.
Tibaldi, A., Corazzato, C., Apuani, T., Pasquar, F.A., Vezzoli, L., 2008c. Geologicalstructural
framework of Stromboli Volcano, past collapses, and the possible inuence on the
events of the 200203 crisis. In: Calvari, S., Inguaggiato, S., Puglisi, G., Ripepe, M.,
Rosi, M. (Eds.), The Stromboli Volcano: An Integrated Study of the 20022003 Eruption. Geophysical Monograph Series 182. AGU. ISBN: 978-0-87590-447-4, pp. 517.
Tibaldi, A., Corazzato, C., Gamberi, F., Marani, M., 2009. Subaerial-submarine evidence of
structures feeding magma to Stromboli Volcano, Italy, and relations with edice
ank failure and creep. Tectonophysics 469, 112136.
Tibaldi, A., Pasquar, F.A., Tormey, D., 2010. Volcanism in reverse and strike-slip fault settings. In: Cloetingh, S., Negendank, J. (Eds.), New Frontiers in Integrated Solid
Earth Sciences. Springer-Verlag, pp. 315348 http://dx.doi.org/10.1007/978-90-4812737-5.
Tibaldi, A., Pasquar, A.F., Rust, D., 2011. New insights into the cone sheet structure of the
Cuillin Complex, Isle of Skye, Scotland. J. Geol. Soc. 168, 689704.
Tibaldi, A., Bonali, F.L., Pasquar, F.A., Rust, D., Cavallo, A., D'Urso, A., 2013. Structure of regional dykes and local cone sheets in the MidhyrnaLysuskard area, Snaefellsnes
Peninsula (NW Iceland). Bull. Volcanol. 75, 764. http://dx.doi.org/10.1007/s00445013-0764-8.
Tibaldi, A., Bonali, F.L., Corazzato, C., 2014. The diverging volcanic rift system. Tectonophysics
611, 94113.

Tobisch, O.T., Paterson, S.R., 1990. The Yarra granite: an intradeformational pluton associated with ductile thrusting, Lachlan Fold Belt, southeastern Australia. Geol. Soc. Am.
Bull. 102, 693703.
Toyoshima, T., Komatsu, M., Shimura, T., 1994. Tectonic evolution of lower crustal rocks in
an exposed magmatic arc section in the Hidaka metamorphic belt, Hokkaido, northern Japan. Island Arc 3, 182198.
Troll, V.R., Walter, T.R., Schmincke, H.-U., 2002. Cyclic caldera collapse: piston or piecemeal subsidence? Field and experimental evidence. Geology 30, 135138.
Upton, B.G.J., Wright, J.B., 1961. Intrusions of gabbro and granophire in the Snaefelsness,
western Iceland. Geol. Mag. 98 (6), 488492.
Valentine, A.G., Krogh, K.E.C., 2006. Emplacement of shallow dikes and sills beneath a
small basaltic volcanic center the role of pre-existing structure (Paiute Ridge,
southern Nevada, USA). Earth Planet. Sci. Lett. 246, 217230.
van Wyk de Vries, B., Francis, P.W., 1997. Catastrophic collapse at stratovolcanoes induced
by gradual volcano spreading. Nature 387 (6631), 387390.
van Wyk de Vries, B., Matela, R., 1998. Styles of volcano-induced deformation: numerical
models of substratum exure, spreading and extrusion. J. Volcanol. Geotherm. Res.
81, 118.
Vigneresse, J.L., Bouchez, J.L., 1997. Successive granitic magma batches during pluton emplacement: the case study of Cabeza de Araya, Spain. J. Petrol. 38, 17671776.
Voight, B., Elsworth, D., 1997. Failure of volcano slopes. Geotechnique 47, 131.
Walker, G.P.L., 1958. Geology of the Reydarfjordur Area, Eastern Iceland. Q. J. Geol. Soc.
114 (1/4), 367391.
Walker, G.P.L., 1960. Zeolite zones and dike distribution in relation to the structure of the
basalts of eastern Iceland. J. Geol. 68, 515528.
Walker, G.P.L., 1974. The structure of Eastern Iceland. In: Kristjnsson, L. (Ed.),
Geodynamics of Iceland and the North Atlantic Area. Reidel, Dordrecht, pp. 177188.
Walker, G.P.L., 1975. Intrusive sheet swarms and the identity of Crustal Layer 3 in Iceland.
J. Geol. Soc. Lond. 131, 143161.
Walker, G.P.L., 1984. Downsag calderas, ring faults, caldera sizes, and incremental caldera
growth. J. Geophys. Res. 89B, 84078416.
Walker, G.P.L., 1999. Volcanic rift zones and their intrusion swarms. J. Volcanol.
Geotherm. Res. 94, 2134.
Wallmann, P.C., Pollard, D.D., Hildreth, W., Eichelberger, J.C., 1990. New structural limits
on magma chamber locations at the Valley of Ten Thousand Smokes, Katmai National
Park, Alaska. Geology 18, 12401243.
Walter, T.R., 2008. Facilitating dike intrusions into ring-faults. Dev. Volcanol. 10, 351374.
Walter, T.R., Schmincke, H.-U., 2002. Rifting, recurrent landsliding and Miocene structural
reorganization on NW-Tenerife (Canary Islands). Int. J. Earth Sci. 91, 615628.
Walter, T.R., Troll, V.R., 2003. Experiments on rift zone evolution in unstable volcanic edices. J. Volcanol. Geotherm. Res. 127, 107120.
Walter, T.R., Klgel, A., Mnn, S., 2006. Gravitational spreading and formation of new rift
zones on overlapping volcanoes. Terra Nova 18, 2633.
Warpinski, H.R., 1985. Measurement of width and pressure in a propagating hydraulic
fracture. J. Soc. Petrol. Eng. 4654.
Watanabe, T., Koyaguchi, T., Seno, T., 1999. Tectonic stress controls on ascent and emplacement of magmas. J. Volcanol. Geotherm. Res. 91, 6578.
Watson, J.V., 1984. The ending of the Caledonian orogeny in Scotland. J. Geol. Soc. Lond.
141, 193214.
Wauthier, C., Cayol, V., Poland, M., Kervyn, F., d'Oreye, N., Hooper, A., Samsonov, S.,
Tiampo, K., Smets, B., 2013. Nyamulagira's magma plumbing system inferred from
15 years of InSAR. Geol. Soc. Lond., Spec. Publ. 380 (1), 3965.
Weertman, J., 1971. Theory of water-lled crevasses in glaciers applied to vertical magma
transport beneath oceanic ridges. J. Geophys. Res. 76, 11711183.
Weertman, J., 1980. The stopping of a rising, liquid-lled crack in the Earth's crust by a
freely slipping horizontal joint. J. Geophys. Res. 85 (B2), 967976.
Weinberg, R.F., Sial, A.N., Mariano, G., 2004. Close spatial relationship between plutons
and shear zones. Geology 32, 377380.
Wenzel, F., Brun, J.P., 1991. A deep reection seismic line across the Northern Rhine
Graben. Earth Planet. Sci. Lett. 104 (24), 140150.
Westerman, D.S., Rocchi, S., Dini, A., Farina, F., Roni, E., 2015. Rise and fall of a multi-sheet
intrusive complex, Elba Island, Italy. Advs in Volcanology. Springer http://dx.doi.org/
10.1007/11157_2014_5.
White, A.J.R., Chappell, B.W., 1977. Ultrametamorphism and granitoid genesis.
Tectonophysics 43, 722.
White, R.S., Smith, L.K., Roberts, A.W., Christie, P.A.F., Kusznir, N.J., Roberts, A.M., Healy, D.,
Spitzer, R., Chappell, A., Eccles, J.D., Fletcher, R., Hurst, N., Lunnon, Z., Parkin, C.J.,
Tymms, V.J., 2008. Lower-crustal intrusion on the North Atlantic continental margin.
Nature 452 (7186), 460464. http://dx.doi.org/10.1038/nature06687.
Whitehead, J.A., Luther, D.S., 1975. Dynamics of laboratory diapir and plume models.
J. Geophys. Res. 80, 705717.
Wickham, S.M., 1987. Segregation and emplacement of granitic magmas. J. Geol. Soc.
Lond. 144, 281297.
Wicks, C.W., Thatcher, W., Dzurisin, D., Svarc, J., 2006. Uplift, thermal unrest and magma
intrusion at Yellowstone caldera. Nature 440. http://dx.doi.org/10.1038/nature04507.
Wiebe, R.A., Collins, W.J., 1998. Depositional features and stratigraphic sections in granitic
plutons: implications for the emplacement and crystallization of granitic magma.
J. Struct. Geol. 20, 12731289.
Wigand, M., Schmitt, A.K., Trumbull, R.B., Villa, I.M., Emmermann, R., 2004. Short-lived
magmatic activity in an anorogenic subvolcanic complex: 40Ar/39Ar and ion microprobe UPb zircon dating of the Erongo, Damaraland, Namibia. J. Volcanol. Geotherm.
Res. 130 (34), 285305.
Williams, S.N., Stoiber, R.E., 1983. Masaya-type caldera redened as the mac analogue
of Krakatau-type caldera. EOS Trans. Am. Geophys. Union 64 (45), 877.
Wright, T.L., 1971. Chemistry of Kilauea and Mauna Loa lava in space and time. U.S. Geol.
Surv. Prof. Pap. 735, 140.

A. Tibaldi / Journal of Volcanology and Geothermal Research 298 (2015) 85135


Wright, T.J., Ebinger, C., Biggs, J., Ayele, A., Yirgu, G., Keir, D., Stork, A., 2006. Magmamaintained rift segmentation at continental rupture in the 2005 Afar dyking episode.
Nature 442, 291294. http://dx.doi.org/10.1038/nature04978.
Wright, H.M.N., Bacon, C.R., Vazquez, J.A., Sisson, T.W., 2012. Sixty thousand years of magmatic volatile history before the caldera-forming eruption of Mount Mazama, Crater
Lake, Oregon. Contrib. Mineral. Petrol. 164, 10271052.
Wylie, J.J., Helfrich, K.R., Dade, B., Lister, J.R., Philips, K., 1999. Flow localization in ssure
eruptions. Bull. Volcanol. 60, 432440. http://dx.doi.org/10.1007/s004450050243.
Yamaoka, K., Kawamura, M., Kimata, F., Fujii, N., Kudo, T., 2005. Dike intrusion associated
with the 2000 eruption of Miyakejima Volcano, Japan. Bull. Volcanol. 67, 231242.
Yew, C.H., 1997. Mechanics of Hydraulic Fracturing. Gulf Publishing, Houston TX.
Zak, J., Paterson, S.R., 2006. Roof and walls of the Red Mountain Creek pluton, eastern
Sierra Nevada, California (USA): implications for process zones during pluton emplacement. J. Struct. Geol. 28, 575587.
Zellmer, G.F., 2008. Some rst-order observations on magma transfer from mantle wedge
to upper crust at volcanic arcs. Geol. Soc. Lond., Spec. Publ. 304 (1), 1531.

135

Zellmer, G.F., 2009. Petrogenesis of Sr-rich adakitic rocks at volcanic arcs: insights from
global variations of eruptive style with plate convergence rates and surface heat
ux. J. Geol. Soc. 166 (4), 725734.
Zellmer, G.F., Annen, C., 2008. An introduction to magma dynamics. In: Annen, C., Zellmer,
G.F. (Eds.), Dynamics of Crustal Magma Transfer, Storage and Differentiation. Special
Publications 304. Geological Society, London, pp. 113.
Zenzri, H., Keer, L.M., 2001. Mechanical analyses of the emplacement of laccoliths and
lopoliths. J. Geophys. Res. Solid Earth (19782012) 106 (B7), 1378113792.
Zhao, C., Hobbs, B.E., Ord, A., Peng, S., 2008. Particle simulation of spontaneous crack generation associated with the laccolithic type of magma intrusion processes. Int.
J. Numer. Methods Eng. 75 (10), 11721193.
Ziv, A., Rubin, A.M., Agnon, A., 2000. Stability of dyke intrusion along preexisting fractures.
J. Geophys. Res. 105 (B3), 59405961.

Anda mungkin juga menyukai