Portal de Periódicos da CAPES

New mafic magma refilling a quiescent volcano: Evidence from He-Ne-Ar isotopes during the 2011-2012 unrest at Santorini, Greece

2015; Wiley; Volume: 16; Issue: 3 Linguagem: Inglês

10.1002/2014gc005653

1525-2027

Andrea Luca Rizzo, F. Barberi, Maria Luisa Carapezza, A. Di Piazza, L. Francalanci, Francesco Sortino, W. D’Alessandro,

Geochemistry and Geologic Mapping

Geochemistry, Geophysics, GeosystemsVolume 16, Issue 3 p. 798-814 Research ArticleFree Access New mafic magma refilling a quiescent volcano: Evidence from He-Ne-Ar isotopes during the 2011–2012 unrest at Santorini, Greece A. L. Rizzo, Corresponding Author A. L. Rizzo Istituto Nazionale di Geofisica e Vulcanologia, Palermo, ItalyCorrespondence to: A. L. Rizzo, andrea.rizzo@ingv.itSearch for more papers by this authorF. Barberi, F. Barberi Dipartimento di Scienze, Università di Roma Tre, Rome, ItalySearch for more papers by this authorM. L. Carapezza, M. L. Carapezza Istituto Nazionale di Geofisica e Vulcanologia, Rome, ItalySearch for more papers by this authorA. Di Piazza, A. Di Piazza Istituto Nazionale di Geofisica e Vulcanologia, Rome, ItalySearch for more papers by this authorL. Francalanci, L. Francalanci Dipartimento di Scienze della Terra, Università degli Studi di Firenze, Florence, ItalySearch for more papers by this authorF. Sortino, F. Sortino Istituto Nazionale di Geofisica e Vulcanologia, Palermo, ItalySearch for more papers by this authorW. D'Alessandro, W. D'Alessandro Istituto Nazionale di Geofisica e Vulcanologia, Palermo, ItalySearch for more papers by this author A. L. Rizzo, Corresponding Author A. L. Rizzo Istituto Nazionale di Geofisica e Vulcanologia, Palermo, ItalyCorrespondence to: A. L. Rizzo, andrea.rizzo@ingv.itSearch for more papers by this authorF. Barberi, F. Barberi Dipartimento di Scienze, Università di Roma Tre, Rome, ItalySearch for more papers by this authorM. L. Carapezza, M. L. Carapezza Istituto Nazionale di Geofisica e Vulcanologia, Rome, ItalySearch for more papers by this authorA. Di Piazza, A. Di Piazza Istituto Nazionale di Geofisica e Vulcanologia, Rome, ItalySearch for more papers by this authorL. Francalanci, L. Francalanci Dipartimento di Scienze della Terra, Università degli Studi di Firenze, Florence, ItalySearch for more papers by this authorF. Sortino, F. Sortino Istituto Nazionale di Geofisica e Vulcanologia, Palermo, ItalySearch for more papers by this authorW. D'Alessandro, W. D'Alessandro Istituto Nazionale di Geofisica e Vulcanologia, Palermo, ItalySearch for more papers by this author First published: 26 February 2015 https://doi.org/10.1002/2014GC005653Citations: 51AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinked InRedditWechat Abstract In 2011–2012, Santorini was characterized by seismic-geodetic-geochemical unrest, which was unprecedented since the most-recent eruption occurred in 1950 and led to fear an eruption was imminent. This unrest offered a chance for investigating the processes leading to volcanic reactivation and the compositional characteristics of involved magma. We have thus analyzed the He-Ne-Ar-isotope composition of fluid inclusions in olivines and clinopyroxenes from cumulate mafic enclaves hosted in cogenetic dacitic lavas of the 1570–1573 and 1925–1928 eruptions of Nea Kameni. These unique data on Aegean volcanism were compared with those of gases collected in quiescent periods and during the unrest. The 3He/4He ratios (3.1–4.0 Ra) are significantly lower than the typical arc-volcano values (R/Ra ∼ 7–8), suggesting the occurrence of magma contamination in Santorini plumbing system, which would further modify the 3He/4He ratio of parental magmas generated in the local metasomatized mantle. The 3He/4He values of enclaves (3.1–3.6 Ra) are comparable to those measured in gases during quiescent periods, confirming that enclaves reflect the He-isotope signature of magma residing at shallow depths and feeding passive degassing. A significant increase in soil CO2 flux from Nea Kameni and anomalous compositional variations in the fumaroles were identified during the unrest, accordingly with previous studies. Simultaneously, 3He/4He ratios up to 4.0 Ra were also measured, demonstrating that the unrest was due to the intrusion into the shallow plumbing system of a more-primitive 3He-rich magma, which is even volatile richer and less contaminated than mafic magma erupted as enclaves. This new intrusion did not however trigger an eruption. Key Points: We studied the He-Ne-Ar isotopes in mafic enclaves and in gases from Santorini The 3He/4He ratios of gases and enclaves is in the range 3.0–4.0 Ra The 2011–2012 unrest at Santorini was due to the intrusion of a new mafic magma 1 Introduction New intrusions of mafic magma into the shallow plumbing system of a quiescent volcano are generally believed to reactivate its dynamic processes, possibly also triggering an eruption [Sparks et al., 1977; Pallister et al., 1992; Folch and Martì, 1998; Burgisser and Bergantz, 2011]. The relationship between such intrusions and the probability of an eruption is still debated [e.g., Gudmundsson, 2012; Huber et al., 2011]. Indeed, the time lag between a new magma input and an eruption can vary, or a system can simply adjust to the change without actually erupting [e.g., Cooper and Kent, 2014]. Thus, understanding magma recharge dynamics in quiescent volcanoes is fundamental to dealing with problems of volcanic hazards, and it is critically important in the study of plumbing systems associated with explosive eruptions. Investigating the 2011–2012 unrest at Santorini, Greece, provided an excellent opportunity to better understand how volcanic reactivation is linked with magmatic processes, and the time scale of this relationship. It has been recently estimated that the large silicic system of Santorini can change from a quiescent state to a Plinian eruption in less than 100 years by several recharges of the magma reservoir [Druitt et al., 2012], whereas historic dacitic lava effusions can be triggered by mafic inputs on 1 month time scales [Martin et al., 2008]. Santorini is the site of the "Minoan" violent Plinian silicic eruption that occurred around 1650 B.C., which covered all of the emerged nearby land with a thick tephra, and caused a caldera collapse with an associated tsunami [Heiken and McCoy, 1984; Druitt et al., 1999, 2012; Druitt, 2014]. Post-Minoan volcanic activity inside the caldera was first reported by Strabo in 197 B.C. and starting from 46 A.D. up to 1950, two volcanic islets—Palea Kameni (PK) and Nea Kameni (NK)—progressively formed in the center of the caldera (Figure 1) by effusive dacitic activity associated with mild Vulcanian explosions [Fouqué, 1879; Pyle and Elliott, 2006]. The explosive eruption of PK in 726 A.D. caused considerable damage at Santorini [Vougioukalakis and Fytikas, 2005]. Another active submarine volcano (Kolumbo) is located outside the caldera 7 km to the northeast of Santorini. Its most-recent explosive eruption occurred in 1649–1650 and killed 70 people and thousands of animals on Santorini due to ash fallout, volcanic gas clouds, and a tsunami [Fouqué, 1879; Vougioukalakis and Fytikas, 2005]. All of the historic eruptions of Santorini occurred from vents located on two northeast-southwest volcano-tectonic lines, the Kameni line running from the Kameni islets to the main island of Thira, and the northern Kolumbo line (Figure 1b). In recent years (up to 2011), these two lines have exhibited markedly contrasting behaviors with alternating seismicity from one line to the other [Dimitriadis et al., 2010; Stiros et al., 2010]. Figure 1Open in figure viewerPowerPoint (a) Geodynamic sketch of the subduction front (Ionian trench) of the Hellenic arc. (b) Inset of Santorini island and location of the Kameni and Kolumbo active volcano-tectonic lines on Santorini. The star indicates the center of the 2011–2012 geodetic unrest [Newman et al., 2012; Parks et al., 2012; Papoutsis et al., 2013]. (c) Geological map of the Kameni islands [after Pyle and Elliott, 2006]; the red circles indicate craters. Locations of the fumarolic and bubbling gas emission (blue triangle) and of the sites where enclaves in dacitic lava have been sampled (orange dots) are also indicated. In January 2011, anomalous seismicity occurred with local magnitude (ML) values up to 3.2 inside the caldera on the Kameni line, and at the same time inflation was observed with a radial pattern from the center of the caldera (Figure 1b) [Newman et al., 2012; Parks et al., 2012; Papoutsis et al., 2013]. Marked variations were also recorded at NK in the chemistry of fumaroles [Tassi et al., 2013] and in the rate of diffuse degassing [Carapezza et al., 2012; Parks et al., 2013], indicative of an increasing input of magmatic gases. These seismic, geodetic, and geochemical indications of unrest, which were unprecedented at Santorini since at least the 1950 eruption, were possibly related to the inflation of a magmatic source and could have been predictive of a volcanic reactivation. Nevertheless, the dynamics and composition of magma causing the inflation and new gas outputs were not known. More than 100,000 visitors are usually present on Santorini Island during the tourist season, which makes it crucial to perform accurate evaluations of the eruption hazard, which mainly depends on the type of magmas in the volcanic system and on their dynamic processes. In particular, it is important to identify if the 2011–2012 unrest was caused by the ascent of an evolved and degassed magma (as proposed by Parks et al. [2012]) or by a new intrusion of mafic and volatile-rich magma generating overpressure in the shallow system [e.g., Folch and Martì, 1998]. In order to address these questions, we investigated the He-Ne-Ar isotope composition of fluid inclusions retained in olivine and clinopyroxene crystals of some mafic enclaves included in the dacitic lavas of the 1570–1573 and 1925–1928 eruptions of NK (Figure 1c). These are the first measurements of 3He/4He ratio performed in rocks at Santorini, and they yield unique data on Aegean volcanism. We then compared the He-Ne-Ar isotope composition of the current emissions of fumarolic gases with that of the magmas feeding the historic NK eruptions, represented by the He-Ne-Ar isotopes analyzed in the fluid inclusions of their mineral phases. Noble gases are among the most sensitive geochemical tracers for establishing the origin of fluids. In particular, the 3He/4He ratio is widely accepted as a powerful tool for characterizing the features of a mantle/magma source [e.g., Nuccio et al., 2008; Correale et al., 2012, 2014; Martelli et al., 2014] and, when monitored over time, is highly sensitive to magma dynamics at depth. These characteristics make the 3He/4He ratio helpful in volcanic surveillance and in the evaluation of volcanic hazards [e.g., Caracausi et al., 2003; Capasso et al., 2005; Rizzo et al., 2006, 2009; Paonita et al., 2012]. 2 The 2011–2012 Unrest at Santorini Santorini volcano entered a quiescent state after the 1950 eruption. The only activity was the continuous emission of mostly hydrothermal gases from fumaroles and hot springs located in the Kameni islands and from low-temperature vents sited in the north and south basins of the caldera [Sigurdsson et al., 2006]. No geochemical anomalies were recorded in the period 1994–2010, which was characterized by the mean soil CO2 flux at NK remaining within a fixed range (i.e., 18–42 g m−2 d−1) [Chiodini et al., 1998; Parks et al., 2013]. Conversely, a small-scale and totally aseismic inflation was recorded during 1994–2000 in the northwest of Santorini, and attributed to slow magma recharge of the shallow plumbing system [Stiros et al., 2010; Papageorgiou et al., 2011; Saltogianni et al., 2012]. This was followed by low-magnitude earthquakes in 2003 beneath the submarine Kolumbo volcano [Dimitriadis et al., 2009], suggesting that most of the magmatic activity occurred in that area and involved the Kolumbo line (Figure 1b). Recent investigations have discovered the presence of high-temperature vents (up to 220°C) located at the bottom of the Kolumbo crater that continuously discharge virtually pure CO2 [e.g., Sigurdsson et al., 2006; Carey et al., 2013]. The period of quiescence at Santorini lasted until early 2011, when the geophysical monitoring system recorded a new phase of unrest characterized by seismic activity and ground deformation (mostly detected by GPS and InSAR) [Newman et al., 2012; Parks et al., 2012; Papoutsis et al., 2013]. A significant seismic swarm started inside the caldera on 9 January 2011, with most of the earthquakes occurring at depths between 1 and 6 km, and with ML values 5) in January 2012 that occurred south of Christiana [Feuillet, 2013], which is one of the volcanoes of the Santorini volcanic field aligned along a ∼N40°E direction [Vougioukalakis and Fytikas, 2005]. The consensus was that the composition of intruded magma was more likely to be an evolved dacite [Parks et al., 2012] than a mafic magma. The 2011–2012 unrest was also characterized by some geochemical anomalies regarding either the composition of fumarolic gases or the soil gas flux [Parks et al., 2013; Tassi et al., 2013]. An increase in magmatic gases such as H2 was observed at NK fumaroles from May 2011 to February 2012 [Tassi et al., 2013]. In the summit area of NK, Parks et al. [2013] recorded an increase in the diffusive emission rate of CO2 between September 2010 and January 2012. A convective heat pulse from depth due to a magma intrusion was hypothesized for explaining the compositional evolution of the fumaroles, although changes in local permeability in response to seismic activity and ground deformations could also have explained the observed variations and the increased rate of CO2 degassing [Parks et al., 2013; Tassi et al., 2013]. All of the recorded anomalies ceased in February 2012, indicating that a new period of quiescence had been reestablished. 3 Volcanological and Petrological Background Santorini belongs to the Southern Aegean volcanic arc developing north of the convergent plate boundary between the African lithosphere and the Eurasian Plate (the Ionian trench; Figure 1a). The subaerial part of Santorini volcano comprises five islands that are the remnants of a series of volcanic collapses (Thira, Thirasia, and Aspronisi) or were formed by historic to recent activity inside the caldera (PK and NK) (Figure 1b). Subaerial volcanic activity at Thira began about 650 ka ago and has continued without significant interruption to the present day [Druitt et al., 1999]. Several caldera-forming acidic explosive eruptions, such as the most-recent Minoan parossistic eruption, have been alternated with mafic/intermediate extrusive and effusive eruptions [Druitt et al., 1999]. Santorini volcanic rocks have a calc-alkaline to high-K calc-alkaline affinity, and range in composition from basalts to rhyolites (Figure 2) [Druitt et al., 1999]. They display the typical trace element patterns of igneous rocks from a subduction-related tectonic setting (Figure 3). Sr, Nd, Pb, and O isotopes of Santorini rocks have a characteristic volcanic arc signature, with a variability mainly ascribed to subduction-related mantle-source metasomatism plus crustal assimilation processes. The genesis of Santorini parental magmas is indeed ascribed to high degrees of partial melting of a depleted mantle wedge, metasomatized by subducted sediment melts and minor aqueous fluids [Francalanci et al., 1998, 2007; Clift and Blusztajn, 1999; Zellmer et al., 2000; Bailey et al., 2009; Vaggelli et al., 2009]. Assimilation of crustal rocks during magma ascent seems to play an important role in the evolution of Santorini magmas, as reflected in several proposed models of assimilation and fractional crystallization involving Santorini basalts as mafic end-members and crustal basement rocks as contaminants [Druitt et al., 1999, and references therein; Bailey et al., 2009; Fabbro et al., 2013]. Processes of mixing and mingling between mafic (basaltic to andesitic) and more-evolved magmas (dacitic to rhyolitic) have been also identified based on the large compositional variability in the pyroclastic units and on the presence of mafic magmatic enclaves in more-evolved eruptive products. These processes imply that shallow reservoirs hosting evolved magmas are refilled by mafic magmas forming compositionally zoned magma chambers. Polybaric evolution of Santorini magmas has been also suggested [e.g., Druitt et al., 1999, 2012; Druitt, 2014]. Figure 2Open in figure viewerPowerPoint K2O versus SiO2 classification diagram for the studied enclaves and host lavas. Pink and orange fields indicate the compositional ranges of Santorini and Kameni rocks, respectively (data from Francalanci et al. [1998], Druitt et al. [1999], Zellmer et al. [2000], and Bailey et al. [2009]). Figure 3Open in figure viewerPowerPoint Patterns of trace elements normalized to primordial mantle [Sun and McDonough, 1989] measured in bulk mafic enclaves and their host dacitic rocks. Gray area indicates the variation range of trace elements in bulk rocks from Santorini [data after Zellmer et al., 2000; Bailey et al., 2009]. 4 Petrochemical Characteristics and Evolution of the Post-Minoan Kameni Magmas Lava flows and domes from PK (46–47 and 726 A.D.) and NK islets (six eruptive events from 1570 to 1950) (Figure 1) have dacitic compositions (SiO2 = 65–70 wt %; Figure 2) and low phenocryst contents (<20 vol %) of plagioclase, pyroxenes, Ti-magnetite, and ilmenite [Washington, 1926; Huijsmans, 1985; Barton and Huijsmans, 1986; Francalanci et al., 1998]. The dacitic lavas contain a variety of mafic enclaves (SiO2 = 49–62%; Figure 2), ranging in size from 30 to 0.5 mm were handpicked from the fraction and then cleaned with diluted acid (6% HNO3) and washed with deionized water. A sieved aliquot of mineral weighing 1.0–2.2 g was put in a stainless-steel bowl and placed in a crusher capable of loading up to six samples simultaneously. The system was heated under pumping conditions for 48–72 h at 130°C in order to reach an ultrahigh vacuum (10−9 mbar). Samples were then crushed at room temperature (22°C), and the gas released by the opening of fluid inclusions was purified in a stainless-steel preparation line in order to remove all species in the gas mixture other than noble gases. He, Ne, and Ar were cryogenically separated and admitted into mass spectrometers. The 3He/4He ratio and 20Ne content were analyzed by a Helix SFT mass spectrometer, while the Ar-isotope composition was analyzed by an Argus mass spectrometer (both devices from GVI) following previously developed procedures [e.g., Nuccio et al., 2008; Correale et al., 2012, 2014; Martelli et al., 2014]. The analytical uncertainty in the determination of elemental He, Ne, and Ar contents was generally 5%. Typical blanks for He and Ne are below 10−15 mol, while for Ar are below 10−14 mol. Considering that Ar is strongly affected by the presence of air, the error bar for each recalculated 40Ar* was calculated on the basis of the analytical error estimated for each 40Ar/36Ar ratio. 6 Results and Discussion 6.1 Composition of the Studied NK Enclaves and Host Lavas The chemical compositions of the studied cumulate enclaves and their host lavas from 1570–1573 to 1925–1928 eruptive events are reported in Table 1. Enclaves are basalts with SiO2 contents of 49–52 wt % (Figure 2), but they do not represent a melt composition due to their high content of cumulus crystals. Similarly to all of the Kameni lavas, their host rocks have dacitic compositions with SiO2 contents of 65–67 wt % (Figure 2). Mineral phases are mainly constituted by olivine, pyroxene, and plagioclase. Olivine crystals from the enclaves of both the 1570–1573 and 1925–1928 eruptive events are not zoned, and are among the most-forsteritic olivines (Fo76–80) of the Kameni enclaves [Conticelli et al., 1998; Martin et al., 2006]. Similarly, the Mg# value (i.e., molecular Mg/(Mg + Fe + Mn)) of the analyzed clinopyroxenes is the highest among all of the clinopyroxenes of the enclaves, ranging between 0.74 and 0.82 in the cores and 0.71–0.82 in the rims. The selected enclaves also have plagioclase with high anorthite content (An88–92). These mineral compositions agree with their crystallization from mafic magmas, as proposed for the origin of the Kameni cumulate enclaves [Francalanci et al., 1998; Petrone et al., 2013a, 2013b]. Table 1. Whole-Rock Compositions of the Studied NK Enclaves and Their Host Lavasaa b.d.l.: below detection limit; "-": not determined. Major elements are reported in wt %, while trace elements are in ppm. Major and trace elements were determined by X-ray fluorescence and Neutron Activation Analyses (asterisk) at the Department of Earth Science of the University of Florence. Sr isotope data are mainly not available on these particular samples, but these reported values (authors' unpublished data) are estimated to be very close to the Sr isotope ratios of these samples. Date 1925–1928 1570–1573 Sample KAM24 KAM23 KAM106B KAM 105 KAM89 KAM 88 Lithotype Cumulate enclave Host lava Cumulate enclave Host lava Cumulate enclave Host lava SiO2 51.92 65.79 51.14 65.44 49.17 66.68 TiO2 0.70 0.78 0.73 0.81 0.62 0.69 Al2O3 19.11 15.70 18.44 15.59 20.52 15.24 Fe2O3 1.84 1.22 1.85 1.97 2.79 1.06 FeO 6.14 3.88 6.74 3.44 4.68 3.56 MnO 0.16 0.15 0.18 0.16 0.15 0.14 MgO 7.06 1.38 7.98 1.37 6.63 1.40 CaO 9.48 3.77 9.47 4.00 12.43 3.77 Na2O 2.72 4.95 2.69 5.08 2.29 5.09 K2O 0.53 1.78 0.48 1.73 0.34 1.79 P2O5 0.05 0.12 0.06 0.13 0.05 0.12 LOI 0.29 0.47 0.25 0.29 0.33 0.46 SUM 100.00 99.99 100.01 100.01 100.00 100.01 Sc* 32.2 13.8 - 15.7 - 14.5 V 243 56 232 52 211 40 Cr 52 b.d.l. 59 13 60 6 Co 35 10 39 9 30 10 Ni 35 3 42 5 29 5 Cu 66 20 43 23 58 19 Zn 68 82 70 81 58 74 Rb 14 68 10 68 7 68 Sr 203 165 201 169 232 160 Y 20 44 25 45 21 44 Zr 82 214 79 217 60 215 Nb 3 9 4 9 4 9 Ba 130 361 122 349 93 362 La 10 22 10 24 4 23 La* 9.1 22.9 - 27 - 23 Ce 14 49 17 48 17 47 Ce* 19 41 - 52 - 48 Nd 10 23 10 23 10 24 Nd* 9.0 18 - 23 - 23 Sm* 2.7 5.1 - 6.9 - 6.6 Eu* 0.88 1.3 - 1.51 - 1.35 Tb* 0.39 0.72 - 1.1 - 1.1 Yb* 2.1 4.1 - 4.9 - 5.0 Lu* 0.28 0.74 - 0.88 - 0.63 Hf* 1.8 4.6 - 5.4 - 5.6 Ta* 0.2 0.6 - 0.74 - 0.68 Pb 5 11 3 10 6 10 Th 3 12 b.d.l. 12 1 13 Th* 2.8 11.1 - 12.3 - 12.1 U 1 3 b.d.l. 3 b.d.l. 3 87Sr/86Sr 0.70484–0.70487bb Isotopic range of enclaves from the previous (1866–1870) and successive (1939–1940) eruptive events. 0.704775cc Isotope value analysed on this lava sample from Francalanci et al. [1998]. 0.70484–0.70487bb Isotopic range of enclaves from the previous (1866–1870) and successive (1939–1940) eruptive events. - 0.704814dd Value of another cumulitic enclave from the same 1570–1573 eruptive event. 0.704712ee Value of another lava sample from the same 1570–1573 eruptive event. 2s 0.000009cc Isotope value analysed on this lava sample from Francalanci et al. [1998]. - 0.000007dd Value of another cumulitic enclave from the same 1570–1573 eruptive event. 0.000012ee Value of another lava sample from the same 1570–1573 eruptive event.