ARTICLE IN PRESS - Earth-prints and Neri...Volcanoes Topography Tectonic setting ... show characteristic radial and/or circumferential patterns ... location and orientation of …

Tectonophysics xxx (2008) xxx–xxx

TECTO-124367; No of Pages 11

Contents lists available at ScienceDirect

Tectonophysics

j ourna l homepage: www.e lsev ie r.com/ locate / tecto

ARTICLE IN PRESS

Dike propagation in volcanic edifices: Overview and possible developments

V. Acocella a,⁎, M. Neri b,1

a Dip. Scienze Geologiche Roma Tre. Largo S.L. Murialdo 1, 00146, Roma, Italyb Istituto Nazionale di Geofisica e Vulcanologia, Piazza Roma 2, 95123, Catania, Italy

⁎ Corresponding author. Tel.: +39 06 57338043; fax: +E-mail addresses: [email protected] (V. Acocella)

1 Tel.: +39 095 7165858; fax: +39 095 435801.

0040-1951/$ – see front matter © 2008 Published by Edoi:10.1016/j.tecto.2008.10.002

Please cite this article as: Acocella, V., Neri,(2008), doi:10.1016/j.tecto.2008.10.002

a b s t r a c t

Article history:

Keywords:DikesVolcanoesTopographyTectonic settingEruptions

; therefore, better knowledge of dike propagation is necessary to improve ourunderstanding of how magma is transferred and extruded at volcanoes. This study presents an overview ofdike patterns and the factors controlling dike propagation within volcanic edifices. Largely based onpublished data, three main types of dikes (regional, circumferential and radial) are illustrated and discussed.Dike pattern data from 25 volcanic edifices in different settings are compared to derive semi-quantitativerelationships between the topography (relief, shape, height, and presence of sector collapses) of the volcano,tectonic setting (presence of a regional stress field), and mean composition (SiO2 content). The overviewdemonstrates how dike propagation in a volcano is not a random process; rather, it depends from thefollowing factors (listed in order of importance): the presence of relief, the shape of the edifice and regionaltectonic control. We find that taller volcanoes develop longer radial dikes, whose (mainly lateral)propagation is independent of the composition of magma or the aspect ratio of the edifice. Future research,starting from these preliminary evaluations, should be devoted to identifying dike propagation paths andlikely locations of vent formation at specific volcanoes, to better aid hazards assessment.

© 2008 Published by Elsevier B.V.

1. Introduction

Understanding magma ascent and extrusion at volcanoes is a crucialstep tominimizinghazards associatedwithvolcanic unrest. Eruptions areoften fed by dikes, as observed at numerous active volcanoes worldwide,for example, Afar (Sigmundsson, 2006), Cerro Negro (Nicaragua; LaFemina et al., 2004), Miyakejima (Japan; Ueda et al., 2005), Iwate (Japan;Sato and Hamaguchi, 2006), Kilauea (Hawaii; Desmarais and Segall,2007), Montserrat (Lesser Antilles; Mattioli et al., 1998), Piton de laFournaise (Reunion Island; Cayol and Cornet, 1998), Nyiragongo (Congo;Komorowski et al., 2002), Etna (Italy; Bousquet and Lanzafame, 2001),and Stromboli (Italy; Acocella et al., 2006a). In many of these episodes,dikes ruptured the surface close to urban areas, feeding eruptive ventsand sometimes even causing landslides and tsunamis (Komorowski et al.,2002; Billi et al., 2003; Behncke et al., 2005; Calvari et al., 2005). Theseand other examples illustrate that to improve our understanding ofmagma transport and eruption, and associated consequences, it isfundamental to advance knowledge of dike propagation.

The mechanisms of dike propagation in the crust have been thesubject of many theoretical studies in the past several decades (e.g.,Anderson, 1936; Ode, 1957; Pollard, 1973; Pollard and Muller, 1976;Delaney et al., 1986). The orientation of a dike is controlled by the

39 06 57338201., [email protected] (M. Neri).

lsevier B.V.

M., Dike propagation in volc

orientation of the principal stresses, with the dike orthogonal to theleast compressive stress in the crust (e.g. Nakamura, 1977; Rubin andPollard, 1988). This relation is best demonstrated in absence ofprominent relief, as in flat rift zones along divergent plate boundaries(Iceland, Afar). In such locations dike propagation may be heavilyinfluenced by stiffness contrasts within the host rock (Gudmundsson,2006, and references therein).

The presence of a volcanic edifice, with some relief, complicatesthis simple dependence on the regional tectonic setting, introducingsignificant deviations from expected patterns. Loading by the edificefocuses the stresses above the center of a magma chamber, promotingthe development of a central vent system (Pinel and Jaupart, 2003). Inaddition, dikes and/or fissure eruptions at many volcanic edificesshow characteristic radial and/or circumferential patterns (e.g.Chadwick and Howard, 1991; Takada, 1997), suggesting control by alocal stress field imposed by a pressurizedmagma reservoir and/or theload of the edifice. In particular, the latter effect becomes predominantwith increasing volcano height (McGuire and Pullen, 1989). Thelocation and orientation of the dikes may be also controlled by theshape of the edifice (Fiske and Jackson,1972), or the presence of scarpsalong the volcano slopes, commonly produced by sector collapses (e.g.McGuire and Pullen, 1989; Tibaldi, 2003; Walter et al., 2005a).Therefore, while dike propagation in areas without prominent reliefis usually controlled by regional tectonism, the propagation of dikes involcanic edifices seems to depend upon the shape and topography ofthe edifice, as well as the stress conditions within shallow magmareservoirs.

anic edifices: Overview and possible developments, Tectonophysics

2 V. Acocella, M. Neri / Tectonophysics xxx (2008) xxx–xxx

ARTICLE IN PRESS

This study aims at producing an overview of the factors controllingthe propagation of dikes within volcanic edifices. Because eruptivefissures and rift zones on volcanic edifices are dike-fed, we includedata from such activity in our analysis. Largely based on publisheddata, the types of dikes and eruptive fissures in volcanic edifices arediscussed to derive general, semi-quantitative insights into dikebehavior as a function of the edifice shape, topography, structuralsetting, and volcano composition.

2. Overview of dike patterns and propagation

2.1. Type of dikes within a volcanic edifice and related conditions offormation

The most common patterns of dikes (or eruptive fissures) involcanic regions can be largely categorized as belonging to one ofthree classes: regional, circumferential and radial. While regionaldikes result from the influence of a far-field (i.e. regional) stress,circumferential and radial dikes result from a near-field (or local)stress imposed by the presence of a pressurizedmagma reservoir and/or the load of the volcano. The main features of each of these con-figurations are summarized below.

2.1.1. Regional dikesRegional dikes within volcanic edifices are aligned with the

regional tectonic stress field, that is, perpendicular to the leastcompressive stress. Several volcanoes are characterized by a dikecomplex consistent with such a far-field stress. Most of these arelocated along the axis of extensional rift zones, in oceanic (Askja andKrafla, in Iceland; Gudmundsson and Nilsen, 2006, and referencestherein) and continental settings (Erta Ale, Ale Bagu, Dabbahu andGabho, in Afar, Nyiragongo and Nyamuragira, in Congo, Tongariro, inNew Zealand, early stage of Etna, in Italy; Barberi and Varet, 1970;Nairn et al., 1998; Komorowski et al., 2002; Corsaro et al., 2002;Wright et al., 2006). Common features are the development of focusedand clear rift zones, usually departing from the volcano summit and

Fig. 1. a) Focused regional dike-fed fissures along the rift zones of Erta Ale volcano, Afar (afte(after Ida, 1995).

Please cite this article as: Acocella, V., Neri, M., Dike propagation in volc(2008), doi:10.1016/j.tecto.2008.10.002

parallel to the regional extensional structures (Fig. 1a). The regionaldikes within volcanic edifices in rift zones are usually subvertical tosteeply dipping, several meters thick (Gudmundsson, 1998); at times,as Hawaii, dikes may constitute more of 50% of the rift zone (Walker,1987, and references therein). The development of regionally-controlled rift zones is also common to arc stratovolcanoes incompressional settings, such as Reventador, Ecuador (Tibaldi, 2005);Izu-Oshima (Ida, 1995; Fig. 1b); Iwate and Chokai, Japan (Nakano andTsuchiya, 1992; Ida, 1995; Sato and Hamaguchi, 2006); and Batur,Indonesia (Newhall and Dzurisin, 1988, and references therein). Themain difference between regional dike patterns in volcanic edifices inextensional and compressional settings lies in the fact that the formermay often propagate beyond the base of the volcano, reachingconsiderable distances along the rift. This is the case for regional dikesformed in volcanoes along the divergent plate boundaries in Icelandand the Afro-Arabian Rift (Gudmundsson, 1995; Ebinger and Casey,2001; Sigmundsson, 2006).

Independent from their tectonic setting, regional dikes withinvolcanic edifices may undergo vertical (Sato and Hamaguchi, 2006) orlateral (Wright et al., 2006; Paquet et al., 2007, and references therein)propagation. Lateral propagation usually occurs when the magmareaches the level of neutral buoyancy within the host rock (Rubin andPollard,1987;Morita et al., 2006),which, formostmagmas, occurswhenthe host rock density is between 2300 and 2700 kg/m3 (Pinel andJaupart, 2000). The load of a volcanic edifice enhances lateral pro-pagationof regional dikes beneath thevolcano (Pinel and Jaupart, 2004),which may partly explain the occurrence of regional, laterallypropagating dikes in central volcanoes and within related magmaticsegments along the axis of oceanic (Gudmundsson, 1995) andcontinental (Ebinger and Casey, 2001;Wright et al., 2006; Sigmundsson,2006) rifts.

2.1.2. Circumferential dikesCircumferential dikes form arcuate patterns concentric to the

volcanic edifice, especially at summit craters and calderas, where thedikes may be also associated with pre-existing faults or fractures.

r Acocella, 2006). b) Dispersed regional dike-fed fissures and vents at Izu-Oshima, Japan

anic edifices: Overview and possible developments, Tectonophysics

3V. Acocella, M. Neri / Tectonophysics xxx (2008) xxx–xxx

ARTICLE IN PRESS

Circumferential dikes have been mostly reported along the sides ofseveral calderas, includingMedicine Lake (California; Donnelly-Nolan,1990) Kilauea (Hawaii; Neal and Lockwood, 2003), Suswa (Kenya;Skilling, 1993, and references therein) and Galàpagos Islands (Ecua-dor; Chadwick and Howard, 1991). In particular, the best-known ex-amples of circumferential dikes at active volcanoes are found aroundthe calderas of shield volcanoes in the western Galápagos archipelago,especially Cerro Azul, Darwin, Fernandina, and Wolf (Fig. 2; Chadwickand Howard,1991). Their developmentmay be related to overpressurewithin a shallow diapir-shapedmagma reservoir, which also promotesradial dike formation on the lower flanks of the volcanoes (Chadwickand Dieterich, 1995). An additional factor for the development ofcircumferential dikes around summit calderas may be the topographicexpression of the caldera, which may re-orient the least compressivestress radially (Munro and Rowland, 1996); a similar reorientation ofthe stress trajectories has been observed at volcanoes characterized bysignificant scarps, like those related to sector collapses (Fiske andJackson, 1972; McGuire and Pullen, 1989; Acocella and Tibaldi, 2005).

At extinct caldera complexes there is widespread evidence for thepresence of circumferential dikes lying at a deeper level within thevolcanic edifice (Yoshida, 1984; Troll et al., 2000; Kennedy and Styx,2007), even though the nature of some of these is debated (Driscollet al., 2006, and references therein). These circumferential dikesusually form above the periphery of a shallow magma chamber,resulting from an increase (cone sheets) or decrease (ring dikes) in themagmatic pressure (Gudmundsson, 2006, and references therein).

2.1.3. Radial dikesThis group of dikes has a characteristic radial distribution with

regard to the axis of the volcanic edifice. The radial pattern may beisotropic, with a similar frequency of dikes in every direction, or,more often, anisotropic, with clustering along preferred orientations.Examples of the first group include Fernandina, Galápagos (Chadwickand Dieterich, 1995); Summer Coon volcano, Colorado (Poland et al.,2004, 2008); Kliuchevskoi, Kamchatka (Takada, 1997); as well asextra-terrestrial volcanoes (Krassilnikov and Head, 2003). Examplesfrom the second group are more numerous, including Vesuvio, Etna,and Stromboli, Italy (Fig. 3; Acocella and Neri, 2003; Tibaldi, 2003;Acocella et al., 2006b); Spanish Peaks, Colorado (Ode, 1957); Hekla,Iceland (Gudmundsson et al., 1992, and references therein); Fuji andSakurajima, Japan (Takada,1997, and references therein; Takada et al.,2007); and Erta Ale, Ethiopia (Acocella, 2006). The rift zone passingthrough the summit of some of these volcanoes (Hekla, Erta Ale),

Fig. 2. Circumferential and radial eruptive fissures (solid lines) at Fernandina volcano,Galápagos.

Please cite this article as: Acocella, V., Neri, M., Dike propagation in volc(2008), doi:10.1016/j.tecto.2008.10.002

even though associated to a regional stress field, resembles an end-member type of highly-clustered radial configuration of dikes (seealso Section 3.1). The anisotropic distribution of radial dikes mayresult, in general, from the influence of a regional stress field whichorients most dikes perpendicular to the least compressive stress(Nakamura, 1977). Anisotropic distributions of radial dikes are alsofound at volcanoes in intraplate settings, in absence of a dominantregional stress field. In such locations, the development of rift zonesmay be controlled by the stability of the flank of the volcano and/orthe presence of nearby volcanoes (e.g. Fiske and Jackson, 1972;Walter et al., 2006). Examples include Kilauea and Mauna Loa,Hawai'i (Decker, 1987), Fogo, Capo Verde (Day et al., 1999), and Pitonthe La Fournaise (Carter et al., 2007, and references therein). In theCanarian archipelago (Marinoni and Gudmundsson, 2000; Acostaet al., 2003, and references therein;Walter, 2003), radial dikes clusterin rift zones that form triple-arms spaced 120° apart and focused onthe summit conduit. This configuration is interpreted to result fromthe shallow emplacement of magma below the volcano summit and/or the growth of nearby spreading volcanoes, promoting connectingrift zones (Walter, 2003, and references therein).

Two mechanisms favour the development of radial dikes in avolcano. The first is the distribution of the gravitational stresses due tothe load of the edifice. This controls the trajectories of the maximumcompressive stress, which becomes subparallel to the slope of thevolcano (Dieterich, 1988), while the minimum compressive stress istangential (Fig. 4a; Acocella and Tibaldi, 2005, and references therein).The larger and taller the volcano, therefore, the stronger is themaximum local stress (McGuire and Pullen, 1989). Locally, the pro-pagation path of radial dikes may be controlled by the presence ofsignificant scarps or other topographic irregularities on the volcanoflanks. In this case, the trajectory of the maximum compressive stressat the scarp margin varies, becoming subparallel to the scarp; theminimum compressive stress becomes perpendicular to the directionof the scarp (Fig. 4b). As a result, dikes tend to propagate parallel to themajor scarps on volcanoes, as at Stromboli and Etna, Italy (McGuireand Pullen, 1989; Ferrari et al., 1991; Tibaldi, 2003; Neri et al., 2004;Acocella and Tibaldi, 2005; Rust et al., 2005; Walter et al., 2005b; Neriand Acocella, 2006; Neri et al., 2007). Similarly, elongated volcanicedifices will be characterized by a maximum compressive stressoriented parallel to the major axis of the edifice, these conditions willresult in the development of dikes oriented parallel to the elongationof the edifice (Fig. 4c; Fiske and Jackson, 1972). Notable examples ofrift zones parallel to themajor elongation of the edifice includeMaunaLoa and Kilauea volcanoes, Hawai'i (Decker, 1987), even though theshape of both volcanoes may be controlled by lateral instability. Thesecond mechanism controlling the development of radial dikes isradially-oriented maximum compressive stress due to pressurizationin a subsurface magma reservoir (Knopf, 1936; Ode, 1957). In this case,radial dikes form to accommodate the enlargement of the circumfer-ence of the edifice with relief, due to volcano inflation (Acocella et al.,2001, and references therein).

Radial dikes may propagate vertically or laterally from thevolcano's central conduit along the slope of the volcano. The me-chanism of lateral propagation of dikes from the summit of a volcanicedifice remains poorly understood. Observational evidencesuggests that propagation direction may depend in part on theclosure/opening of the central conduit (Fig. 5; Acocella et al., 2006b).When the central conduit is closed or solidified, magma is emplaced inthe edifice by means of vertically propagating dikes; in the upper partof the edifice, along the frozen conduit, these may follow gravitationalstresses, becoming radial. Conversely, the lateral propagation of dikesis widespread, but not exclusive (Acocella and Neri, 2003; Lanzafameet al., 2003), in volcanoes characterized by an open summit conduit.Here, magma in the upper part of the conduit degasses and,becoming denser, intrudes laterally, propagating downslope. Notableexamples of volcanoes with such a behaviour include Tenerife

anic edifices: Overview and possible developments, Tectonophysics

Fig.

3.Ex

amples

ofradial

erup

tive

fissures.a

)Etna

,period19

00–20

05;insetb)

show

sfissureorientation;

insetc)

show

sthethreemainrift

zone

s.d)

Vesuv

io,p

eriod16

31–19

44;insetesh

owsfissureorientation.

4 V. Acocella, M. Neri / Tectonophysics xxx (2008) xxx–xxx

ARTICLE IN PRESS

Please cite this article as: Acocella, V., Neri, M., Dike propagation in volcanic edifices: Overview and possible developments, Tectonophysics(2008), doi:10.1016/j.tecto.2008.10.002

Fig. 5. Dikes mostly propagate laterally (a) when the summit conduit is open, and themagma has already degassed, and vertically (b) when the conduit is closed.

Fig. 4. Most common shapes for volcanic edifices with relief (2D, map view) and related local stress conditions (σHMAX=maximum horizontal stress; σHmin=minimum horizontalstress). (a) cone, (b) cone with sector collapse, (c) ridge. The maximum horizontal stress is radial in a cone, slightly diverging in a cone with sector collapse and elongated in a ridge.

5V. Acocella, M. Neri / Tectonophysics xxx (2008) xxx–xxx

ARTICLE IN PRESS

Please cite this article as: Acocella, V., Neri, M., Dike propagation in volc(2008), doi:10.1016/j.tecto.2008.10.002

(Soriano et al., 2008), Etna (Acocella and Neri, 2003), Stromboli(Acocella et al., 2006a) and Vesuvio (between 1631 and 1944; Acocellaet al., 2006b). Estimates from Etna and Vesuvio suggest that the meanalong-strike top surface of a laterally propagating dike has a moderatedownslope dip, on the order of 10°–15° (Acocella et al., 2006c, andreferences therein). This mechanism of emplacement of radial dikes isusually limited to the upper part of the edifice, below its slope.Therefore, the specific conditions controlling it (i.e. opening/closure ofthe conduit) need not to be the same as those of the radial dikespropagating at the base of the volcanic edifice, as for example observedat Summer Coon Volcano (Poland et al., 2008).

2.2. What controls dike propagation in different volcanic edifices

The features discussed above establish general guidelines for dikepropagation in volcanic edifices. The relations between these features(e.g., topography, regional vs. local stress field) and other factors (e.g.,shape of the volcano, composition of magma) in the context of dikepropagation has not yet been investigated in detail. Here, we aim tominimizing this gap, with a semi-quantitative analysis of the relationsbetween various factors related to dike emplacement. We considerseveral features, listed in Table 1, related to dike emplacement at 25active volcanic centers. Those features include: 1) height (H1) to thebase of the edifice, including any submerged portion. 2) Aspect ratio(A) of the edifice (where A=height/width). 3) Eccentricity (E) of theedifice (where E=minimum elongation/maximum elongation).4) Mean SiO2 content of the erupted magmas, which is expected toapproximate, to a first order, average magma viscosity. It is possiblethat dikes have far from average compositions, but this possibility hasnot been taken into account in this study. 5) Maximum length (L)reached by the dikes or, more commonly, the eruptive fissures in avolcano; in both cases, it is possible that the real length of the dikemay be larger than that of its visible part, so this length has tobe considered as minimum value. 6) Difference in height (Hd)between the highest and lowest part of the longest fissure oroutcropping dike; this value refers to the subaerial part of the dikeor fissure and thus may significantly differ from H1. 7) Frequency of

anic edifices: Overview and possible developments, Tectonophysics

Table 1Mean features which may be related to the dike patterns of selected volcanoes

Volcano H1 A E SiO2 L Hd RT RR References(km) (wt.%) (km) (km)

Hekla (Iceland) 1.5 0.08 0.37 48±2 9 0.8 1 1 10, 27Krafla (Iceland) 0.6 0.042 0.5 48±3 12 0.2 1 1 27Erta Ale (Ethiopia) 0.8 0.042 0.70 49±1 4.5 0.3 1 1 1, 12, 24Nyamuragira (Congo) 1 0.071 0.85 45±2 12 0.9 1 0.56 5Nyiragongo (Congo) 2 0.11 0.70 40±8 20 1.8 1 0.625 21, 32Etna (Italy) 3.3 0.12 0.83 50±3 16 2.6 1 0.37 17, 26Vesuvio (Italy) 1.2 0.082 0.85 57±5 4 1 1 0.57 25Stromboli (Italy) 2.9 0.16 0.76 52±1 3 0.9 0.77 0.39 33Kliuchevskoi (Russia) 4.7 0.26 0.92 (54) 15 3.7 1 0.18 9, 10Terevaka (Chile) 3.5 0.031 0.73 51±3 8 0.4 0.96 0 28Lonquimay (Chile) 2.8 0.076 0.53 58±5 11 1 1 0.77 34Villarrica (Chile) 2.8 0.078 0.75 53±5 11 1.4 1 0.75 22, 34Fuji (Japan) 3.7 0.3 0.95 58±8 13 2.7 1 0.45 10, 23, 31Sakurajima (Japan) 1.2 0.1 0.73 62±5 4 1 1 0.40 10Izu-Oshima (Japan) 1.5 0.047 0.6 4 0.7 1 0.83 6, 10Miyake (Japan) 1.3 0.087 0.82 53±3 4.8 0.8 1 0.42 14, 15Chokai (Japan) 2.1 0.129 0.69 58±5 10 1.6 1 0.69 20, 34Fernandina (Galapagos) 4.4 0.041 0.79 48±2 14 1.3 0.6 0 4, 7Wolf (Galapagos) 4.6 0.089 0.78 48±2 16 1.5 0.66 0 4, 7Darwin (Galapagos) 4.3 0.056 0.94 48±2 15 1.1 0.59 0 4, 16Fogo (Capo Verde) 6.7 0.108 0.92 42±2 16 2.7 0.9 0 13, 18Piton (Reunion) 6.5 0.109 0.74 48±3 12 2.1 1 0 8, 10, 30Kilauea (Hawaii) 5.7 0.015 0.37 48±2 17⁎ 1.1 1 0 2, 3, 10, 19Mauna Loa (Hawaii) 8.5 0.067 0.59 49±2 22⁎ 1.8 1 0 2, 3, 10Tenerife (main; Spain) 7.1 0.075 0.69 50±7 28 2.7 1 0 11, 29

H1=total height to the base of the edifice (including any portion below sea level); A=aspect ratio of edifice (where A=height/width); E=eccentricity of edifice (where E=minimumelongation/maximum elongation); L=maximum length reached by the dikes or eruptive fissures; Hd=difference in height between the highest and lowest subaerial parts of thelongest fissure or outcropping dike; RT=frequency of the radial vs. circumferential dikes or fissures (where RT=number of radial dikes /number of radial+circumferential dikes);RR=frequency of regional dikes vs. radial+circumferential dikes (where RR=number of regional dikes /number of regional+radial+circumferential dikes). (data from: (1) Barberiet al., 1980; (2) Holcomb, 1987; (3) Lockwood and Lipman, 1987; (4) Chadwick and Howard, 1991; (5) Hayashi et al., 1992; (6) Ida, 1995; (7) Munro and Rowland 1996; (8) Albarèdeet al., 1997; (9) Ozerov et al., 1997; (10) Takada, 1997; (11) Ablay et al., 1998; (12) Barrat et al., 1998; (13) Day et al., 1999, and references therein; (14) Amma-Miyasaka and Nakagawa,2002; (15) Geshi et al., 2002; (16) Naumann et al., 2002; (17) Acocella and Neri, 2003; (18) Doucelance et al., 2003; (19) Thornber et al., 2003; (20) Kondo et al., 2004; (21) Platz et al.,2004; (22) Witter et al., 2004; (23) Yamamoto et al., 2005; (24) Acocella, 2006; (25) Acocella et al., 2006b; (26) Allard et al., 2006, and references therein; (27) Gudmundsson andNilsen, 2006; (28) Vezzoli and Acocella, 2006; (29) Carracedo et al., 2007; (30) Carter et al., 2007, and references therein; (31) Takada et al., 2007; (32) Tedesco et al., 2007, andreferences therein; (33) Corazzato et al., 2008 and (34) Nakamura, 1977). Asterisk refers to values of length of dikes or fissures in portions of the volcano least affected by flank slip.

6 V. Acocella, M. Neri / Tectonophysics xxx (2008) xxx–xxx

ARTICLE IN PRESS

the radial vs. circumferential dikes or fissures in a volcano, designatedRT (where RT=numberof radial dikes/numberof radial+circumferentialdikes). 8) Frequency of regional dikes vs. radial+circumferential dikesin a volcano, designated RR (where RR=number of regional dikes/number of regional+radial+circumferential dikes). The 25 activevolcanoes have been selected based on availability of data (frompreviously published studies), variability of edifice type (e.g., calderas,shield volcanoes, composite volcanoes), and tectonic setting (e.g.,regional extension, regional compression, hot spots). The generalrelations between these features are summarized in Fig. 6.

The distribution of the RR values for the selected volcanoes shows acluster at RR=0, where radial or circumferential systems are dominantand regional dikes are absent (Fig. 6a). These volcanoes are ocean islandshields related to hot spot activity and away fromplate boundarieswithstrong regional stress fields. Another at RR∼0.5 represents stratovol-canoes with similar proportions of local- and regional-controlled dikesor fissures. A third cluster of data, approaching RR=1, corresponds toshield volcanoes along the axis of oceanic rifts, characterized bya strongregional control on dike emplacement. The histogram shown in Fig. 6atherefore indicates that the regional setting influences the dike patternof a volcano with different degrees of intensity.

The distribution of RT values for the selected volcanoes suggeststhat volcanic edifices are largely dominated, at surface, by dikes orfissures with a radial attitude (RT=1; Fig. 6b). A minor number ofvolcanoes have a significant component of circumferential dikes orfissures, but RT values are never lower than 0.5. Fig. 6b thus confirmsthat circumferential dikes on the surface of active volcanoes areunusual, in contrast to observations of dike patterns below the surface(see Section 2.1.2). The discrepancy between surface and depth dikeorientations suggests that a radial most compressive stress, induced

Please cite this article as: Acocella, V., Neri, M., Dike propagation in volc(2008), doi:10.1016/j.tecto.2008.10.002

by the relief of the volcanic edifice, prevails at shallow depths,inhibiting the upward propagation of circumferential dikes and/orrotating these towards the surface.

There is a poor correlation between the degree of regional controlon dike or fissure emplacement (RR) and the eccentricity of the base ofthe volcanic edifice (Fig. 6c). In fact, strongly elongated volcanicedifices are not always controlled by regional tectonics. However,neglecting Kilauea and Mauna Loa (arrows in Fig. 6c), whose rift zoneshapes are significantly influenced by the slip of the SE flank of Hawaiiand the topography of existing volcanoes (Fiske and Jackson, 1972;Walter et al., 2006, and references therein), a more significantcorrelation is obtained (dashed line in Fig. 6c). This relationship isstrengthened by the fact that, except for hot spot volcanoes, theedifices are usually elongated normal to the regional minimum com-pressive stress. Therefore, for volcanoes that are isolated and withoutmajor flank slip, the data suggest that there may be a generalcorrelation between the elongation of a volcano and the pattern ofregionally-controlled dikes.

There is also an inverse correlation between the regional tectoniccontrol on dike emplacement (RR) and the maximum height of avolcano above its base (H1; Fig. 6d). Volcanoes located in hot spotsettings have not been considered in this diagram, as lacking of anyregional control. The inverse correlation suggests that taller volcanoesare characterized by a distribution of radial dikes or fissures whichtends to be more isotropic, implying that the control of a regionalstress field fades with the size of the volcano. In general, volcanoestaller than 3 km do not show significant evidence of regional tectoniccontrol on dike emplacement. Similar conclusions have beensuggested for Etna, where it has been proposed that the regionaltectonic control on dike propagation in the uppermost part of the

anic edifices: Overview and possible developments, Tectonophysics

Fig. 6. Relations between factors controlling dike emplacement at selected volcanoes (see Table 1). a) Distribution of RR (number of regional dikes /number of regional+radial andcircumferential dikes); b) Distribution of RT (number of radial dikes /number or radial+circumferential dikes); c) Eccentricity E of volcanoes vs. RR; d) RR of volcanoes vs. their heightH1; e) H1 of volcanoes vs. the maximum length of their dikes or fissures L; f) differential height of the fissure Hd vs. L; g) Mean SiO2 value of the magmatic products of a volcano vs L;h) Aspect ratio A (thickness/width) of a volcano vs. L; i) A vs. Hd.

7V. Acocella, M. Neri / Tectonophysics xxx (2008) xxx–xxx

ARTICLE IN PRESS

edifice is replaced by a local, topographic stress field (McGuire andPullen, 1989).

The development of radial dikes was analysed as a function of H1,Hd, A and SiO2. There is a direct proportion between the maximumlength of an eruptive fissure or dike (L) and the total height of thevolcano (Fig. 6e). The association of longer fissures or dikes with tallervolcanoes suggests topographic control on dike propagation. Similar

Please cite this article as: Acocella, V., Neri, M., Dike propagation in volc(2008), doi:10.1016/j.tecto.2008.10.002

results are suggested by Fig. 6f, which indicates a direct proportion-ality between L and the differential vertical height of the fissure ordike. The behaviour may be associated with dikes propagatinglaterally downslope, from the summit of an open conduit. In fact, itis expected that the shallower the dike separates from the centralconduit, the higher its potential energy and propagation force will be;therefore the dike will propagate laterally a greater distance. There is

anic edifices: Overview and possible developments, Tectonophysics

8 V. Acocella, M. Neri / Tectonophysics xxx (2008) xxx–xxx

ARTICLE IN PRESS

clear proof of such lateral propagation of dikes only at ∼30% of theconsidered volcanoes (Table 1), so further evidence is needed togeneralize such behaviour.

The correlation between the maximum length of an eruptivefissure or dike (L) and the SiO2 content of the related magma (Fig. 6g)is poor, indicating that the composition and/or viscosity of the magmadoes not significantly limit dike propagation. This is in agreementwithrecent data from Stromboli, showing that the petrochemical featuresof the magma, including viscosity, have a very low influence on thegeometry of dike propagation (Corazzato et al., 2008).

There is also no correlation between the maximum length of aneruptive fissure or dike (L) and the aspect ratio of the volcanic edifice(A; Fig. 6h). Apparently, the dip of the slope of the volcano, related tothe aspect ratio of the edifice, does not influence dike propagation.

Finally, there is a direct proportionality between the aspect ratio ofthe volcanic edifice (A) and the differential vertical height of theeruptive fissure (Hd; Fig. 6i). The result implies that steeper volcanoesare also associated with eruptive fissures with larger difference inheight; therefore, even though the length of the dike is not controlledby the steepness of the volcano (Fig. 6h), the drop in altitude of thedike may depend from the dip of the slope of the volcano.

3. Discussion

3.1. General features of dike propagation

The overview above indicates that the propagation path followed bydikeswithin avolcanic edifice is influencedby several factors.More thanone of these factors may act simultaneously, producing complex dikepatterns, and the role of each factor on dike propagation is summarized

Fig. 7. Main features, seen as end-members, controlling dike patterns at volcanoes. Dike pattorder of importance. Right side shows schematic map view of typical dike patterns (with re

Please cite this article as: Acocella, V., Neri, M., Dike propagation in volc(2008), doi:10.1016/j.tecto.2008.10.002

in Fig. 7. Below,we consider each factor independently, as end-member;however, corresponding examples are not always found in nature, giventhe complex interactions between the factors or the specificity of someconditions. Examples given (italic in Fig. 7) are supposed to provide thebest available approximation, even though the dike patterns at thosevolcanoes may result from more than a single mechanism.

The most important first-order control on dike propagationwithinan edifice is topographic relief. Topography always introduces a local,gravitational stress field which is superimposed over other local (e.g.,induced by a shallow magma reservoir) or regional stress field. Thegeneral consequence of significant topography is the development of aradial (more or less isotropic) dike pattern. The lack of topographicrelief may lead to circumferential and/or regional dike patterns.

Edifices without significant relief may develop subparallel dikeswarms consistent with the regional tectonic trend, as exemplified atKrafla, Iceland. Inabsenceof a regional stressfield, thedikepatternmaybecontrolled by the presence of topographic irregularities, such as calderas;circumferential dike systemsmay form at the periphery of the caldera. Toour knowledge, there is not any particular example of active volcanowithexposed circumferential dikes related to the lack of relief and regionalstress field. Another possible control on dike propagation is that inducedby a shallow magma reservoir; circumferential dikes are still mostlyexpected to form. Underpressure conditions within the reservoir willgenerate ring dikes, whereas overpressure conditions will generate conesheets (Anderson, 1936; Phillips, 1974). Again, we are not aware of anyclear dike pattern resulting from a magma chamber and the absence oftopographic relief and a regional stress field in an active volcano. There-fore, these conditions, even though feasible, remain largely theoretical.

Edifices with topographic relief are characterized by a radial mostcompressive stress andwill bedominatedby radial dikes. Anydeparture

erns at given volcanoes result from the features listed in the upper part of the figure, inlief=grey; without relief=white).

anic edifices: Overview and possible developments, Tectonophysics

9V. Acocella, M. Neri / Tectonophysics xxx (2008) xxx–xxx

ARTICLE IN PRESS

from an isotropic radial pattern and clustering of the dikes alongpreferred orientations depends upon the shape of the edifice and, to alesser extent, the regional stress field. If a sector collapse scar or othermajor scarp is present, radial dikes will partly focus around and parallelthe relief, as observed at the Sciara del Fuoco flank collapse at Stromboliand the Valle del Bove at Etna (McGuire and Pullen, 1989; Acocella andTibaldi, 2005).

In the case of subcircular or elongated edifices, the dike patternwilldependupon the presence or absenceof a regional stressfield.Withouta regional stress field, dikes follow the major axis of the edifice, as atKilauea and Mauna Loa, or develop characteristic triple configurationsat 120°, as at Tenerife. With a regional stress field, elongated volcanoesdevelop dikes focused along preferred regional orientations (as at ErtaAle, Afar, or Hekla, Iceland) or simply dike swarms that parallel themost compressive stress, as at Izu-Oshima (Fig. 1). The first situationcan be interpreted as an end-member type represented by a highlyasymmetric radial configuration of dikes, focused along a direction.The second situation, where the dikes are more dispersed, regards theemplacement of non-radial dike swarms. The development of oneconfiguration (focused) or the other (dispersed) may depend from theintensity of the regional stress. At oceanic and continental divergentplate boundaries (exemplified by Erta Ale and Hekla), extensionalstress focuses strain on the axis of the rift, where volcanoes are usuallylocated. Extension is therefore restricted to a narrowaxial zone passingthrough the volcano summit. Conversely, at arc or back-arc settings (forexample, at Izu-Oshima; Takada, 1997), weaker tectonic extensionresults in more dispersed extensional structures, forming wider riftzones (Macdonald, 1998, and references therein). As a consequence,dike swarms may be found throughout the volcanic edifice and notrestricted to the axial zone of the volcano.

Similar to elongated edifices, the dike pattern at subcircular edificesalso depends upon the presence or absence of a regional stress field.Without any regional stress field, the dikes follow a radial path, as atFogo (Capo Verde). With a regional stress field, the dike pattern mayalso depend upon the height of the edifice. Dike propagation at shortvolcanoesmaybe still partly influenced bya far-field stress, developingradial patterns with a clustering along directions parallel to theregional most compressive stress (for example, Nyamuragira, Nyir-agongo, and Miyakejima). Dike propagation at taller volcanoes is lessinfluenced by a regional stress field, displaying amore (though seldomfully) isotropic radial pattern (examples include Etna, Fuji andKliuchevskoi). The boundary that defines the relation between theheight of a volcano and any anisotropy of the pattern of its radial dikesis not sharp, as several volcanoes (e.g., Vesuvio and Sakurajima),display intermediate values of H1 and RR.

The final dike pattern on a volcano may also depend upon thesuperimposition of different evolutionary stages of the edifice. In fact,the development of the main rift zones may be also due to lateralcollapses or stress variations (Walter et al., 2005a; Tibaldi et al., 2006;Corazzato et al., 2008, and references therein).

The dike patterns described above may be more or lesspronounced depending upon the length reached by the dikes. Asintroduced above, the most important parameter controlling theextent of dike propagation is probably the height of the edifice. Otherfactors, such as the composition of magma or the aspect ratio of theedifice, have moderate or negligible influence (Fig. 6). The comparisonwith known active cases (Hawaii, Tenerife, Etna, Vesuvio, Stromboli)suggests that these conditions may significantly apply to laterallypropagating dikes. This common process of lateral dike propagationmay be related to the opening of the central conduit, even though amore advanced study is needed (Acocella et al., 2006b).

3.2. Implications for hazard and future research

The factors above suggest that the emplacement of dikes in agiven volcano may be predicted based on knowledge of the most

Please cite this article as: Acocella, V., Neri, M., Dike propagation in volc(2008), doi:10.1016/j.tecto.2008.10.002

important controlling parameters on dike propagation. Studies thataccount for dike emplacement and eruption are essential in anyproject concerning hazard mitigation at active volcanoes. In fact,many volcanic eruptions are fed by dikes, and dike-fed vents maydevelop at significant distances from the summit of a volcano.Therefore, understanding the controls on dike propagation is crucialto predict where a vent may develop and what hazards may result.

On these premises, we suggest that future research on dikeemplacement should focus on the following points. 1) A betterdefinition of the mechanical conditions for the lateral developmentof dikes originating from the central conduit of a volcanic edifice.Such dikes are a common feature inmany types of volcano, regardlessof tectonic setting. 2) The evaluation of the potential propagationpaths of dikes at individual volcanoes. To the latter aim, the nature ofthe boundary conditions of individual volcanic systems need to bedefined, including the opening or closure of the central conduit (e.g.,Acocella et al., 2006b), the stress distribution within the volcanicedifice (e.g., Dieterich, 1988), also as a function of its topography(Acocella et al., 2006a), the magmatic pressure in subsurface reser-voirs (e.g. Gudmundsson, 2002, 2003), the presence of central oreccentric reservoirs (e.g. Neri et al., 2005, and references therein), therecent eruptive history (e.g. Behncke and Neri, 2003; Behncke et al.,2005; Allard et al., 2006), the occurrence of pre-eruptive earthquakes(Nostro et al., 1998; Walter et al., 2005b) and the presence of layerswith different stiffness or mechanical properties (Gudmundsson,2006, and references therein).

For example, investigations at Vesuvio, Stromboli and Etna suggestthat most of the recent dikes propagate laterally from the centralconduit,with a path largely controlled by the topography, as significantscarps (Neri et al., 2005, and references therein; Acocella et al., 2006a,b,c; Neri et al., 2008). These studies may constitute a starting point to tryto evaluate with more precision the future propagation paths of thedikes as a function of different boundary conditions in these volcanoes.More in general, the propagation paths of dikes, as well as their likelylengths, may be estimated at a given volcano, allowing for the creationof hazard maps with the probability of dike-fed activity in differentareas of the volcano. Such an approachmay predict dike propagation ata volcano, considering not only the distribution of the most recentvents or fissures, but especially understanding why and under whichconditions these dikes and fissures developed.

4. Conclusions

We describe the formation of three types of dikes (regional, circum-ferential and radial) in relation to several factors, including volcanotopography (positive or negative relief, shape, height, sector collapses),tectonic setting (presence of a regional stress field) and mean com-position (SiO2 content). Data from 25 volcanic edifices in differentsettings indicate that dike propagation depends, in order of importance,upon the presence of any relief, the shape of the edifice, and thepresence of a regional stress field. Taller volcanoes develop longer dikeswith radial orientations, with negligible effects of the composition ofmagma or the aspect ratio of the volcano. Our overview demonstratesthat the style of dike emplacementmaybepredicted at givenvolcanoes.Future research and hazards assessments should evaluate the possiblepropagation paths of dikes at specific volcanoes, as well as the role ofthe boundaryconditions of the system, including the openingor closureof conduit, the regional stress trajectorieswithin an edifice, themagma-tic pressure, the recent eruptive history, the occurrence of pre-eruptiveearthquakes, and the presence of layers with different stiffness.

Acknowledgements

T. Yoshida and V. Zanon kindly provided useful data and informa-tion. Constructive reviews from M. Poland and A. Tibaldi significantlyimproved thepaper. Partly fundedwithDPC-INGV funds (LAVAProject).

anic edifices: Overview and possible developments, Tectonophysics

10 V. Acocella, M. Neri / Tectonophysics xxx (2008) xxx–xxx

ARTICLE IN PRESS

References

Ablay, G.J., Carroll, M.R., Palmer, M.R., Martí, J., Sparks, R.S.J., 1998. Basanite–phonolitelineages of the Teide-Pico Viejo volcanic complex, Tenerife, Canary Islands. J. Petrol.39, 905–936.

Acocella, V., 2006. Regional and local tectonics at Erta Ale caldera, Afar (Ethiopia).J. Struct. Geol. 28, 1808–1820.

Acocella, V., Neri, M., 2003. What makes flank eruptions? The 2001 Etna eruption and itspossible triggeringmechanisms. Bull. Volcanol. 65, 517–529. doi:10.1007/s00445-003-0280-3.

Acocella, V., Tibaldi, A., 2005. Dike propagation driven by volcano collapse: a generalmodel tested at Stromboli, Italy. Geophys. Res. Lett. 32, L08308. doi:10.1029/2004GL022248.

Acocella, V., Cifelli, F., Funiciello, R., 2001. The control of overburden thickness onresurgent domes: insights from analogue models. J. Volcanol. Geotherm. Res. 111,137–153.

Acocella, V., Neri, M., Scarlato, P., 2006a. Understanding shallow magma emplacementat volcanoes: orthogonal feeder dikes during the 2002–2003 Stromboli (Italy)eruption. Geophys. Res. Lett. 33, L17310. doi:10.1029/2006GL026862.

Acocella, V., Porreca, M., Neri, M., Mattei, M., Funiciello, R., 2006b. Fissure eruptions atMount Vesuvius (Italy): insights on the shallow propagation of dikes at volcanoes.Geology 34, 673–676. doi:10.1130/G22552.1.

Acocella, V., Porreca, M., Neri, M., Massimi, E., Mattei, M., 2006c. Propagation of dikes atVesuvio (Italy) and the effect of Mt. Somma. Geophys. Res. Lett. 33, L08301.doi:10.1029/2005GL025590.

Acosta, J., Uchupi, E., Smith, D., Munoz, A., Herranz, P., Palomo, C., Llanes, P., Ballesteros,M., Zee Working Group, 2003. Comparison of volcanic rifts on La Palma and ElHierro, Canary Islands and the Island of Hawaii. Marine Geophys. Res. 24, 59–90.

Albarède, F., Luais, B., Fitton, G., Semet, M., Kaminski, E., Upton, B.G.J., Bachelery, P.,Cheminee, J.-L., 1997. The geochemical regimes of Piton de la Fournaise volcano(Reunion) during the last 530000 years. J. Petrol. 38, 171–201.

Allard, P., Behncke, B., D'Amico, S., Neri, M., Gambino, S., 2006. Mount Etna 1993–2005.Anatomy of an evolving eruptive cycle. Earth Sci. Rev. 78, 85–114. doi:10.1016/j.earscirev.2006.04.002.

Amma-Miyasaka, M., Nakagawa, M., 2002. Origin of anorthite and olivine megacrysts inisland-arc tholeiites: petrological study of 1940 and 1962 ejecta from Miyake-Jimavolcano, Izu-mariana arc. J. Volcanol. Geotherm. Res. 117, 263–283.

Anderson, E.M., 1936. The dynamics of the formation of cone-sheets, ring dykes andcauldron subsidence. Proc. Roy. Soc. Edin. 56, 128–157.

Barberi, F., Varet, J., 1970. The Erta Ale Volcanic Range (Danakil depression, northernAfar, Ethiopia). Bull. Volcanol. 34, 848–917.

Barberi, F., Civetta, L., Varet, J., 1980. Sr isotopic composition of Afar volcanics and itsimplications for mantle evolution. Earth Planet. Sci. Lett. 50, 247–259.

Barrat, J.A., Fourcade, S., Jahn, B.M., Chemirke, J.L., Capdevila, R., 1998. Isotope (Sr, Nd, Pb,O) and trace-element geochemistry of volcanics from the Erta' Ale range (Ethiopia).J. Volcanol. Geotherm. Res. 80, 85–100.

Behncke, B., Neri, M., 2003. Cycles and trends in the recent eruptive behaviour of MountEtna (Italy). Can. J. Earth Sci. 40, 1405–1411. doi:10.1139/E03-052.

Behncke, B., Neri,M.,Nagay, A., 2005. Lavaflowhazard atMount Etna (Italy): newdata froma GIS-based study. In Kinematics and dynamics of lava flows, edited by Manga, M.,Ventura, G., Spec. Pap. Geol. Soc.Am., 396-13,187–205. doi:10.1130/0-8137-2396-5.189.

Billi, A., Acocella, V., Funiciello, R., Giordano, G., Lanzafame, G., Neri, M., 2003.Mechanism for ground surface fracturing and incipient slope failure associatedwith the 2001 eruption of Mt. Etna, Italy: analysis of ephemeral field data.J. Volcanol. Geotherm. Res. 122, 281–294.

Bousquet, J.C., Lanzafame, G., 2001. Nouvelle interprétation des fractures des éruptionslatérales de l'Etna: conséquences pour son cadre tectonique. Bull. Soc. Geol. Fr. 172,455–467.

Calvari, S., Spampinato, L., Lodato, L., Harris, A.J.L., Patrick, M.R., Dehn, J., Burton, M.R.,Andronico, D., 2005. Chronology and complex volcanic processes during the 2002–2003 flank eruption at Stromboli volcano (Italy) reconstructed from directobservations and surveys with a handheld thermal camera. J. Geophys. Res. 110.doi:10.1029/2004JB003129.

Carracedo, J.C., Rodriguez Badiola, E., Guillou, H., Paterne, M., Scaillet, S., Perez Torrado,F.J., Paris, R., Fra-Paleo, U., Hansen, A., 2007. Eruptive and structural history of Teidevolcano and rift zones of Tenerife, Canary Islands. GSA Bull. 119, 1027–1051.

Carter, A., van Wyk de Vries, B., Kelfoun, K., Bachelery, P., Briole, P., 2007. Pits, rifts andslumps: the summit structure of Piton de la Fournaise. Bull. Volcanol. doi:10.1007/s00445-006-0103-4.

Cayol, V., Cornet, F.H., 1998. Three-dimensional modelling of the 1983–1984 eruption atpiton de la Fournaise Volcano, Reunion Island. J. Geophys. Res. 103, 18025–18037.

Chadwick, W.W., Howard, K.A., 1991. The pattern of circumferential and radial eruptivefissures on the volcanoes of Fernandina and Isabela islands, Galapagos. Bull.Volcanol. 53, 259–275.

Chadwick, J.R., Dieterich, J.H., 1995. Mechanical modeling of circumferential and radialdike intrusion on Galapagos volcanoes. J. Volcanol. Geotherm. Res. 66, 37–52.

Corazzato, C., Francalanci, L., Menna, M., Petrone, M.C., Renzulli, A., Tibaldi, A., Vezzoli,L., 2008. What controls sheet intrusion in volcanoes? Structure and Petrology of theStromboli sheet complex, Italy. J. Volcanol. Geotherm. Res. 173, 26–54.

Corsaro, R.A., Neri, M., Pompilio, M., 2002. Paleo-environmental and volcano-tectonicevolution of the south-eastern flank of Mt. Etna during the last 225 ka inferred fromvolcanic succession of the «Timpe», Acireale, Sicily. J. Volcanol. Geotherm. Res. 113,289–306.

Day, S.J., Heleno da Silva, S.I.N., Fonseca, J.F.B.D., 1999. A past giant lateral collapse andpresent-day flank instability of Fogo, Cape Verde Islands. J. Volcanol. Geotherm. Res.94, 191–218.

Please cite this article as: Acocella, V., Neri, M., Dike propagation in volc(2008), doi:10.1016/j.tecto.2008.10.002

Decker, R.W., 1987. Dynamics of Hawaiian volcanoes: an overview. U.S. Geological Surv.Prof. Pap. 1350, 997–1018.

Delaney, P.T., Pollard, D.D., Ziony, J.I., McKee, E.H., 1986. Field relations between dikesand joints — emplacement processes and paleostress analysis. J. Geophys. Res. 91,4920–4938.

Desmarais, E.K., Segall, P., 2007. Transient deformation following the 30 January 1997dike intrusion at Kilauea volcano, Hawaii. Bull. Volcanol. 69, 253–363.

Dieterich, J.H., 1988. Growth and persistence of Hawaiian volcanic rift zones. J. Geophys.Res. 93, 4258–4270.

Donnelly-Nolan, J.M., 1990. Post 11000 year volcanism at Medicine Lake volcano,cascade Range, Northern California. J. Geophys. Res. 95, 19693–19704.

Doucelance, R., Escrig, S., Moreira, M., Gariépy, C., Kurz, M.D., 2003. Pb–Sr–He isotopeand trace element geochemistry of the Cape Verde Archipelago. Geochim.Cosmochim. Acta 67, 3717–3733.

Driscoll, B.O., Troll, V.R., Reavy, R.J., Turner, P., 2006. The Great Eucrite intrusion ofArdnamurchan, Scotland, reevaluating the ring-dike concept. Geology 34, 189–192.

Ebinger, C.J., Casey, M., 2001. Continental breakup in magmatic provinces: an Ethiopianexample. Geology 29, 527–530.

Ferrari, L., Garduno, V.H., Neri, M., 1991. I dicchi della Valle del Bove, Etna: un metodoper stimare le dilatazioni di un apparato vulcanico. Mem. Soc. Geol. It. 47, 495–508.

Fiske, R.S., Jackson, E.D., 1972. Orientation and growth of Hawaiian volcanic rifts: theeffect of regional structure and gravitational stresses. Proc. R. Soc. Lond. 329,299–326.

Geshi, N., Shimano, T., Chiba, T., Nakada, S., 2002. Caldera collapse during the 2000eruption of Miyakejima volcano, Japan. Bull. Volcanol. 64, 55–68.

Gudmundsson, A., 1995. Infrastructure and mechanics of volcanic systems in Iceland.J. Volcanol. Geotherm. Res. 64, 1–22.

Gudmundsson, A., 1998. Magma chambers modeled as cavities explain the formation ofrift zone central volcanoes and their eruption and intrusion statistics. J. Geophys.Res. 103, 7401–7412.

Gudmundsson, A., 2002. Emplacement and arrest of sheets and dykes in centralvolcanoes. J. Volcanol. Geotherm. Res. 116, 279–298.

Gudmundsson, A., 2003. Surface stresses associated with arrested dykes in rift zones.Bull. Volcanol. 65, 606–619.

Gudmundsson, A., 2006. How local stresses control magma-chamber ruptures, dykeinjections, and eruptions in composite volcanoes. Earth Sci. Rev. 79, 1–31.

Gudmundsson, A., Nilsen, K., 2006. Ring faults in composite volcanoes: structures,models, and stress fields associated with their formation. J. Geol. Soc. Lond. 269,83–108.

Gudmundsson, A., Oskarsson, N., Gronvold, K., Saemundsson, K., Sigurdsson, O.,Stefansson, R., Gislason, S.R., Einarsson, P., Brandsdottir, B., Larsen, G., Johannesson,H., Thordarson, T., 1992. The 1991 eruption of Hekla, Iceland. Bull. Volcanol. 54,238–246.

Kennedy, B., Styx, J., 2007. Magmatic processes associatedwith caldera collapse at Ossipeering dyke, New Hampshire. Bull. Geol. Soc. Am. 119, 3–17. doi:10.1130/B25980.1.

Komorowski, J.C., et al., 2002. The January 2002 flank eruption of Nyiragongo volcano(DRC): Chronology, evidence for a tectonic rift trigger and impact of lava flows onthe city of Goma. Acta Vulcanol. 14, 25–57.

Kondo, H., Tanaka, K., Mizuochi, Y., Ninomiya, A., 2004. Long-term changes in thedistribution and chemistry of middle Miocene to Quaternary volcanism in theChokai-Kurikoma area across the northeast Japan Arc. Island Arc 13, 18–46.

Knopf, A., 1936. Igneous geology of the Spanish Peaks region, Colorado. Geol. Soc. Am.Bull. 68, 1727–1784.

Krassilnikov, A.S., Head, J.W., 2003. Novae on Venus: geology, classification, evolution.J. Geophys. Res. 108, 5108. doi:10.1029/2002JE001983.

Hayashi, S., Kasahara, M., Tanaka, K., Hamaguchi, H., Zana, N., 1992. Major chemistry ofrecent eruptive products from Nyamuragire volcano, Africa (1976–1989). Tectono-physics 209, 273–276.

Holcomb, R.T., 1987. Eruptive history and long-term behavior of Kilauea volcano. U.S.Geol. Survey Prof. Pap. 1350, 261–350.

Ida, Y., 1995. Magma chamber and eruptive processes at Izu-Oshima volcano, Japan:buoyancy control of magma migration. J. Volcanol. Geotherm. Res. 66, 53–67.

La Femina, P.C., Condor, C.B., Hill, B.E., Strauch, W., Saballos, J.A., 2004. Magma-tectonicinteractions in Nicaragua: the 1999 seismic swarm and eruption of Cerro Negrovolcano. J. Volcanol. Geotherm. Res. 137, 187–199.

Lanzafame, G., Neri, M., Acocella, V., Billi, A., Funiciello, R., Giordano, G., 2003. Structuralfeatures of the July–August 2001 Mount Etna eruption: evidence for a complexmagmatic system. J. Geol. Soc. Lond. 160, 531–544.

Lockwood, J.P., Lipman, P.W., 1987. Holocene eruptive history of Mauna Loa volcano. U.S.Geol. Surv. Prof. Pap. 1350, 509–535.

Macdonald, K.C., 1998. Linkages between faulting, volcanism, hydrothermal activity andsegmentation on fast spreading centres. In: Buck, W.R., Delaney, P.T., Karson, J.A.,Lagabrielle, Y. (Eds.), Faulting and Magmatism at Mid-Ocean Ridges. Am. Geophys.Un., Washington, pp. 27–58.

Marinoni, L.B., Gudmundsson, A., 2000. Dykes, faults and paleostresses in the Teno andAnaga massifs of Tenerife (Canary Islands). J. Volcanol. Geotherm. Res. 103, 83–103.

Mattioli, G.S., Dixon, T.H., Farina, F., Howell, E.S., Jansma, P.E., Smith, A.L., 1998. GPSmeasurement of surface deformation around Soufriere Hills volcano, Montserrat,from October 1995 to July 1996. Geophys. Res. Lett. 25, 3417–3420.

McGuire, W.J., Pullen, A.D., 1989. Location and orientation of eruptive fissures andfeeder-dykes at Mount Etna: influence of gravitational and regional stress regimes.J. Volcanol. Geotherm. Res. 38, 325–344.

Morita, Y., Nakao, S., Hayashi, Y., 2006. A quantitative approach to the dike intrusionprocess inferred from a joint analysis of geodetic and seismological data for the1998 earthquake swarm of the east coast of Izu Peninsula, central Japan. J. Geophys.Res. 111, B06208. doi:10.1029/2005JB003860.

anic edifices: Overview and possible developments, Tectonophysics

11V. Acocella, M. Neri / Tectonophysics xxx (2008) xxx–xxx

ARTICLE IN PRESS

Munro, D.C., Rowland, S.K., 1996. Caldera morphology in the western Galapagos andimplications for volcano eruptive behaviour and mechanism of caldera formation.J. Volcanol. Geotherm. Res. 72, 85–100.

Nakamura, K., 1977. Volcanoes as possible indicators of tectonic stress orientation:principle and proposal. J. Volcanol. Geotherm. Res. 2, 1–16.

Nakano, S., Tsuchiya, N., 1992. Geology of the Chokaisan and Fukura district. withgeological sheet map at 1:50,000. Geol. Surv. Japan 138 pp. (in Japanese withEnglish abstract 7p).

Nairn, I.A., Kobayashi, T., Nakagawa, M., 1998. The ∼10 ka multiple vent pyroclasticeruption sequence at Tongariro Volcanic Centre, Taupo Volcanic Zone, NewZealand. Part 1. Eruptive processes during regional extension. J. Volcanol.Geotherm. Res. 86, 19–44.

Naumann, T., Geist, D., Kurz, M., 2002. Petrology and geochemistry of Volcán Cerro Azul:petrologic diversity among the western Galápagos volcanoes. J. Petrol. 43, 859–883.

Neal, C.A., Lockwood, J.P., 2003. Geologic map of the summit region of Kilauea Volcano,Hawaii. USGS map MI-2759.

Neri, M., Acocella, V., 2006. The 2004–05 Etna eruption: implications for flankdeformation and structural behaviour of the volcano. J. Volcanol. Geotherm. Res.158, 195–206. doi:10.1016/j.jvolgeores.2006.04.022.

Neri, M., Acocella, V., Behncke, B., 2004. The role of the Pernicana fault system in thespreading of Mt. Etna (Italy) during the 2002–2003 eruption. Bull. Volcanol. 66,417–430. doi:10.1007/s00445-003-0322-x.

Neri, M., Acocella, V., Behncke, B., Maiolino, V., Ursino, A., Velardita, R., 2005.Contrasting triggering mechanisms of the 2001 and 2002–2003 eruptions ofMount Etna (Italy). J. Volcanol. Geotherm. Res. 144, 235–255.

Neri, M., Guglielmino, F., Rust, D., 2007. Flank instability on Mount Etna: Radon, radarinterferometry, and geodetic data from the southwestern boundary of the unstablesector. J. Geophys. Res. 112, B04410. doi:10.1029/2006JB0047.

Neri, M., Lanzafame, G., Acocella, V., 2008. Dike emplacement and related hazard involcanoes with sector collapse: the Stromboli 2007 eruption. J. Geol. Soc. Lond.in press.

Newhall, C.G., Dzurisin, D., 1988. Historical unrest at large calderas of the world. U.S.Geological Survey. 1109 pp.

Nostro, C., Stein, R.S., Cocco, M., Belardinelli, M.E., Marzocchi, W., 1998. Two-waycoupling between Vesuvius eruptions and southern Apennine earthquakes, Italy, byelastic stress transfer. J. Geophys. Res. 103, 24487–24504.

Ode, H., 1957. Mechanical analysis of the dike pattern of the Spanish Peaks area,Colorado. Geol. Soc. Am. Bull. 68, 567–575.

Ozerov, A.Yu., Ariskin, A.A., Kyle, Ph., Bogoyavlenskaya, G.E., Karpenko, S.F., 1997.Petrological–geochemical model for genetic relationships between basaltic andandesitic magmatism of Klyuchevskoi and Bezymyannyi volcanoes, Kamchatka.Petrology 5, 550–569.

Paquet, F., Dauteuil, O., Hallot, E., Moreau, F., 2007. Tectonics and magma dynamicscoupling in a dyke swarm of Iceland. J. Struct. Geol. 29, 1477–1493.

Phillips, W.J., 1974. The dynamic emplacement of cone sheets. Tectonophysics 24,69–84.

Pinel, V., Jaupart, C., 2000. the effect of edifice load on magma scent beneath a volcano.Phil. Trans. R. Soc. Lond. 358, 1515–1532.

Pinel, V., Jaupart, C., 2003. Magma chamber behavior beneath a volcanic edifice.J. Geophys. Res. 108, 2072. doi:10.1029/2002JB001751.

Pinel, V., Jaupart, C., 2004. Magma storage and horizontal dike injection beneath avolcanic edifice. Earth Planet. Sci. Lett. 221, 245–262.

Platz, T., Foley, S.F., Andrè, L., 2004. Low-pressure fractionation of the Nyitagongovolcanic rocks, Virunga Province, D.R. Congo. J. Volcanol. Geotherm. Res. 136,269–295.

Pollard, D.D., 1973. Derivation and evaluation of a mechanical model for sheetintrusions. Tectonophysics 19, 233–269.

Pollard, D.D., Muller, O.H., 1976. Effect of gradients in regional stress and magmapressure on form of sheet intrusions in cross-section. J. Geophys. Res. 81, 975–984.

Poland, M.P., Fink, J.H., Tauxe, L., 2004. Patterns of magma flow in segmented silicicdikes at Summer Coon volcano, Colorado: AMS and thin section analysis. EarthPlanet. Sci. Lett. 219, 155–169.

Poland, M.P., Moats, W.P., Fink, J.H., 2008. A model for radial dike emplacement incomposite cones based on observations from Summer Coon volcano, Colorado, USA.Bull. Volcanol 70, 861–875. doi:10.1007/s00445-007-0175-9.

Rubin, A.M., Pollard, D.D.,1987. Origin of blade-like dikes in volcanic rift zones. U.S. Geol.Surv. Prof. Pap. 1350, 1449–1470.

Please cite this article as: Acocella, V., Neri, M., Dike propagation in volc(2008), doi:10.1016/j.tecto.2008.10.002

Rubin, A.M., Pollard, D.D., 1988. Dike-induced faulting in rift zones of Iceland and Afar.Geology 16, 413–417.

Rust, D., Behncke, B., Neri, M., Ciocanel, A., 2005. Nested zones of instability in theMount Etna volcanic edifice, Sicily. J. Volcanol. Geotherm. Res. 155, 137–153.doi:10.1016/j.jvolgeores.2004.11.021.

Sato, M., Hamaguchi, H., 2006. Weak long-lived round deformation related to Iwatevolcanism revealed by Bayesian decomposition of strain, tilt and positioning data.J. Volcanol. Geotherm. Res. 155, 244–262.

Sigmundsson, F., 2006. Magma does the splits. Nature 442, 251–252.Skilling, I.P., 1993. Incremental caldera collapse of Suswa volcano, Gregory Rift Valley,

Kenya. J. Geol. Soc. London 150, 885–896.Soriano, C., Beamud, E., Garcés,M., 2008.Magmaflow indikes from rift zonesof thebasaltic

shield of Tenerife, Canary Islands: implications for the emplacement of buoyantmagma. J. Volcanol. Geotherm. Res. 173, 55–68. doi:10.1016/j.jvolgeores.2008.01.007.

Takada, A., 1997. Cyclic flank-vent and central-vent eruption patterns. Bull. Volcanol. 58,539–556.

Takada, A., Ishizuka, Y., Nakano, S., Yamamoto, T., Kobayashi, M., Suzuki, Y., 2007.Characteristic and evolution inferred from eruptive fissures of Fuji volcano, Japan.In: Aramaki, S. (Ed.), Fuji Volcano. Volcanol. Soc., Japan, Tokyo, pp. 183–202.

Tedesco, D., Vaselli, O., Papale, P., Carn, S.A., Voltaggio, M., Sawyer, G.M., Durieux, J.,Kasereka, M., Tassi, F., 2007. January 2002 volcano-tectonic eruption of Nyiragongovolcano, Democratic Republic of Congo. J. Geophys. Res. 112, B09202. doi:10.1029/2006JB004762.

Tibaldi, A., 2003. Influence of cone morphology on dikes, Stromboli, Italy. J. Volcanol.Geotherm. Res. 126, 79–95.

Tibaldi, A., 2005. Volcanism in compressional tectonic settings. Is it possible? Geophys.Res. Lett. 32. doi:10.1029/2004GL021798.

Tibaldi, A., Bistacchi, A., Pasquare, F.A., Vezzoli, L., 2006. Extensional tectonics andvolcano lateral collapses: insights from Ollague volcano (Chile–Bolivia) andanalogue modelling. Terra Nova 18, 282–289.

Thornber, C.R., Heliker, C., Sherrod, D.R., Kauahikaua, J.P., Miklius, A., Okubo, P.G.,Trusdell, F.A., Budahn, J.R., Ridley, W.I., Meeker, G.P., 2003. Kilauea East Rift Zonemagmatism: an episode 54 perspective. J. Petrol. 44, 1525–1559.

Troll, V.R., Emeleus, C.H., Donaldson, C.H., 2000. Caldera formation in the Rum CentralIgneous Complex, Scotland. Bull. Volcanol. 62, 301–317.

Ueda, H., Fujita, E., Ukawa, M., Yamamoto, E., Irwan, M., Kimata, F., 2005. Magmaintrusion and discharge process at the initial stage of the 2000 activity ofMiyakejima, Central Japan, inferred from tilt and GPS data. Geophys. J. Int. 161,891–906.

Vezzoli, L., Acocella, V., 2006. Geological and structural evolution of Easter Island(Chile). Proceedings of EGU meeting, Vienna (Austria), April 2006, p. 509. Abstractvolume pag.

Walker, G.P.L., 1987. The dike complex of Koolau volcano, Oahu: internal structure of aHawaiian rift zone. U.S. Geol. Surv. Prof. Pap. 1350, 961–993.

Walter, T.R., 2003. Buttressing and fractional spreading of Tenerife, an experimentalapproach on the formation of rift zones. Geophys. Res. Lett. 30, 1296. doi:10.1029/2002GL016610.

Walter, T.R., Troll, V.R., Cailleau, B., Belousov, A., Schmincke, H.U., Amelung, F., Bogaard,P., 2005a. Rift zone reorganization through flank instability in ocean islandvolcanoes: an example from Tenerife, Canary Islands. Bull. Volcanol. 67, 281–291.

Walter, T.R., Acocella, V., Neri, M., Amelung, F., 2005b. Feedback processes betweenmagmatism and E-flank movement at Mt. Etna (Italy) during the 2002–2003eruption. J. Geophys. Res. 110, B10205. doi:10.1029/2005JB003688.

Walter, T.R., Klugel, A., Munn, S., 2006. Gravitational spreading and formation of new riftzones on overlapping volcanoes. Terra Nova 18, 26–33.

Witter, J.B., Kress, V.C., Delmelle, P., Stix, J., 2004. Volatile degassing, petrology andmagma dynamics of the Villarrica Lava Lake, Southern Chile. J. Volcanol. Geotherm.Res. 134, 303–337.

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, 291–294. doi:10.1038/nature04978.

Yamamoto, T., Takada, A., Ishizuka, Y., Miyaji, N., Tajima, Y., 2005. Basaltic pyroclasticflows of Fuji volcano, Japan: characteristics of their deposits and their origin. Bull.Volcanol. 67, 622–633.

Yoshida, T., 1984. Tertiary Ishizuchi Cauldron, Southwestern Japan Arc: formation byring fracture subsidence. J. Geophys. Res. 89, 8502–8510.

anic edifices: Overview and possible developments, Tectonophysics