Article history:Received 26 March 2010Accepted 4 August 2010Available online 13 August 2010
Keywords:Plumbing systemPetrologyGeochemistryTectonicsAlkali transfer
The enrichment in potassium shown by the basic lavas erupted at Mount Etna volcano after 1971 (K2Omax~2.2 wt.%) has been considered by previous researchers to be too high to be related to simple crystalfractionation and instead linked this high K2O content to either crustal assimilation or changes in the magmasource. Unfortunately all existing models for the post-1971 K2O enrichment fail to explain the phenomenonsatisfactorily leaving the question still open.We present a critical re-examination of published data for major elements (633 whole rock analyses), traceelements (376 whole rock analyses) and isotopic ratios (136 87Sr/86Sr analyses), for historical and pre-historical lavas. Potassium enrichment is not limited to the products of the last 35 years. A comparableincrease in potassium is noticed in lavas erupted during the pre-historic phase of the recent Mongibello (K2Omax~2.5 wt.%) and in lavas related to the early phase of the ancient Mongibello (K2O max~3 wt.%).Moreover, data from melt inclusions in olivines from the 2001 and 2002 eruptions, reveal that potassiumcontents remain constant for melts with entrapment pressure between 490 to 100 MPa and increasesignificantly in melts entrapped at pressures below~100 MPa.We propose that supercritical fluids coming from deeper magmas and carrying alkali Cl-complexes migratethrough basic to intermediate magmas residing in the shallow feeding system. As chlorine exsolves andleaves the system alkalis are released contributing to the observed potassium enrichment of the shallowmagma. Fluctuations of the volatiles influx throughout time are likely related to the magma supply rate.Considering that the amount of magma entering a plumbing system is determined by the rate of regionalextension, the flux of alkali Cl-complexes entering the melt might be related to an extensional regime actingin the Etnean area.
Mount Etna volcano (3340 m a.s.l.) located in eastern Sicily (Italy)is one of the most active volcanoes in the world with a continuousdegassing through the summit craters and frequent eruptionsoccurring mostly from its flanks and from the two sub-terminal (NEand SE) craters. The scientific literature regarding the petrology ofEtnean volcanics has shown that after 1971 eruptions mostlyproduced lavas with trachybasaltic to trachyandesitic compositions,characterized by systematically higher K contents (K2O~2.2 wt.%)than lavas of otherwise comparable compositions erupted pre-1971(K2O~1.5 wt.%). Moreover the K increase is accompanied by aprogressive Rb, Cs, and 87Sr/86Sr increase (Joron and Treuil, 1984;Clocchiatti et al., 1988; Condomines et al., 1995).
The origin of this geochemical feature cannot be explained bysimple crystal fractionation, and its emergence at one definite period
of the volcano evolution has puzzled volcanologists for years, and hasresulted in several models that attempt to explain this enrichment.Some of the earliest workers suggested that the enrichment in K, Rband Cs displayed by the post 1971 Etnean products could result fromthe assimilation of sediments underlying the volcanic edifice or fromselective contamination of magma during its migration to the surface,indicating meteoric water as an alkali exchanger between basementrock and magma (Joron and Treuil, 1984; Clocchiatti et al., 1988;Condomines et al., 1995; Tonarini et al., 1995). Other authorsinterpreted the post-1971 K enrichment as being inherited from thepartial melting of a chemically and isotopically heterogeneous mantlesource (Armienti et al., 1989; Barbieri et al., 1993; Corsaro andCristofolini, 1996). Tanguy et al. (1997), related the K increase to amixing ofmantle-derived-magmaswith K and Si-rich partial melts (asdefined by Schiano and Clocchiatti, 1994), and suggested that ageneral increase of K2O/Na2O is displayed by the entire alkalinesequence and that it becomes more noticeable after 1971. Morerecently, the K increase has been interpreted as the signature of a fluidcomponent derived from the subducting Ionian slab in the magmasource (Schiano et al., 2001; Tonarini et al., 2001).
2001–2005Viccaro and Cristofolini (2008) 12 12 12Viccaro et al. (2006) 7 7Spilliaert et al. (2006a) 60Métrich et al. (2004) 4Ferlito et al. (2008a) 17Corsaro et al. (2007) 14 14Clocchiatti et al. (2004) 14
1971–1999Viccaro and Cristofolini (2008) 4 4 4Trigila et al. (1990) 7Treuil and Joron (1994) 28Tanguy et al. (1997) 1 1 1Tanguy and Kieffer (1977) 5Scott (1983) 17 17La Delfa et al. (2001) 14 8Joron and Treuil (1984) 9 9Distefano (2007) 10Corsaro and Pompilio (2004c) 17Condomines et al. (1995) 6 6Barbieri et al. (1993) 17Armienti et al. (1995) 41 28Armienti et al. (1994b) 46 27Armienti et al. (1990) 7 2Armienti et al. (1989) 11 11 11
HistoricViccaro and Cristofolini (2008) 8 8 8Tanguy et al. (2007) 28 28Tanguy et al. (1997) 5 5 4Joron and Treuil (1984) 4 4Corsaro et al. (1996) 9 9 6Corsaro and Cristofolini (1993) 70Condomines et al. (1995) 18 16Barbieri et al. (1993) 28
R.M. UndatedTanguy et al. (2007) 13 13Tanguy et al. (1997) 1 1 1Corsaro and Cristofolini (1993) 77
R.M. Mt. MalettoTanguy et al. (1997) 1 1 1Armienti et al. (1988) 2 2 1
R.M. Mt. SpagnoloTanguy et al. (1997) 1 1 1Armienti et al. (1988) 1 1 1
A.M. (Ellittico)Ferlito (1994) 45 45D'Orazio et al. (1997) 68 68 16Cristofolini et al. (2002) 47
Figures in columns represent the number of analyses for each reference. A.M. is ancientMongibello; R.M. is recent Mongibello.
643C. Ferlito, G. Lanzafame / Lithos 119 (2010) 642–650
In this work we assess the problem of the post 1971 K enrichmentby starting from a basic question: does it characterize the last 35 yearsonly, or can it be recognized at other times throughout the evolutionof Mount Etna volcano? To answer this question, we have investigatedthe stratigraphic record of the Mongibello center and collectedpublished petrochemical data for ancient Mongibello (Ellittico) andrecent Mongibello (present day eruptive volcano) up to the 2005eruption. The critical appraisal of this large data set reveals that Kenrichment is not exclusive to recent (post-1971) eruptions and thatcomparable features can be traced throughout the stratigraphicsuccession. Recent studies on the 2001 volcanics have proposed thatthe K enrichment of the magma is related to the activity of magmaticfluids percolating through the magma bodies residing at shallow levelin the plumbing system (Ferlito et al., 2008). In this work we proposethat the differentiation induced by volatiles can be considered as aneffective process that contributes anomalous increase of K content inlavas and that it might have acted not only after 1971 but also duringearlier phases of the volcano evolution.
2. A critical appraisal of petrochemical data from recent andancient Mongibello lavas
Etnean activity can be traced back to~500 ka ago with productsranging fromtholeiitic to differentiatedalkaline basalts (cf. Tanguy et al.,1997, 2007). However, a complete database ofmajor and trace elementsfor many historic and prehistoric lavas is lacking. Most of the availablepetrochemical data are from the products erupted in the last 50 years,whereas relatively few data cover the past activity.
We have broadened our investigation and have gathered petro-logical data on lavas of ancient Mongibello (Ellittico; 60 to 15 ka -Branca et al., 2008 and references therein) and recent Mongibello(15 ka to 2005 eruption). We have taken into account data fromliterature as well as from doctoral theses (Table 1), and assembled adatabasewith 633whole rock analyses for major elements, 376 wholerock analyses for trace elements, 136 isotopic analyses of 87Sr/86Sr and64 major element analyses of melt inclusions in olivines. TheMongibello is the biggest volcano of the entire Etnean succession(Romano, 1982; Cristofolini et al., 1991; Branca et al., 2008). Itsvolcanics in the TAS diagram (total alkali vs. silica - Le Maitre, 1989)span the fields from basalt to trachyte with a great abundance oftrachybasalts and basaltic trachyandesites (Fig. 1).
Differentiation of Etnean magmas occurs at depth throughfractionation of mafic phases such as olivine, clinopyroxene and Ti-magnetite.Moreover themagmamoving in the open conduit system isefficiently differentiated by the polybaric crystallization of plagioclasecausing an efficient differentiation (Cristofolini, 1973; Corsaro andCristofolini, 1993; Joron and Treuil, 1984; Armienti et al., 1988; Tanguyet al., 1997, 2007). The examination of the entire major element dataset suggests that Mg content tends to increase with time (time span ofca 60 ka) (Fig. 2), pointing toward “primitive” compositions repre-sented by the Mg rich sub-aphyric alkali basalts of Mt. Spagnolo andMt. Maletto (Armienti et al., 1988; Tanguy et al., 1997, 2007). A shifttoward Mg-rich magmas has also been noticed in recently eruptedlavas and related by Clocchiatti et al. (2004), Métrich et al. (2004) andViccaro et al. (2006) to the ingress into the plumbing system of newmantle-derivedmagma. The tendency to eruptmore basic andMg-richmagmas with time is not restricted to the post-1971 activity. In fact,products of the Ellittico, Recent Mongibello and historic, contain lessMg than the most recent ones, but can be traced back to recentMongibello and Ellittico. Potassiumdisplays a positive correlationwithSiO2 since, due to its lowmineral/melt partition coefficient (e.g. olivine0.0068; clinopyroxene 0.038; cf. Arth, 1976), crystal fractionation ofmafic phases will produce a K enrichment in the residual liquid. Theeffect of fractionation on the K amount for Etnean magmas can bequantified using simulations with the thermodynamic algorithm ofMELTS code (release v5.0; Ghiorso and Sack, 1995; Asimow and
Ghiorso, 1998). By cooling of a liquid with basaltic composition(SiO2~46.60 wt.%; K2O~1.25 wt.%) from 1100 °C, at constant pressureof 100 MPa and 2.5 wt.% of H2O (common conditions for shallow-residing Etnean magmas), we obtain a liquid with hawaiitic compo-sition (SiO2~50.20 wt.%; K2O~1.80 wt.%) crystallizing~4 wt.% ofolivine, ~ 17 wt.% clinopyroxene, ~ 6 wt.% Ca-rich plagioclase,and~3 wt.% spinel. The computed composition is in good agreementwith the proportions of observed phenocrysts in the magma(Cristofolini et al., 1991; Armienti et al., 1990, 1994a, b; D'Orazio etal., 1997; Viccaro et al., 2006; Corsaro et al., 2007). The K enrichmentduring differentiation can be regarded as ΔK2O/ΔSiO2 where Δ is the
Fig. 1. Total alkali vs. SiO2 (TAS) classification diagram (Le Maitre, 1989). For all the 598analyses of the ancient Mongibello (Ellittico) and recent Mongibello, symbols are asfollows: crosses, eccentric cones of Mount Spagnolo and Mount Maletto; diamonds,Ellittico; triangles, pre-historical and historical recent Mongibello; boxes, 1971–1999;circles, 2001–2005. All lavas belong to the Na alkaline series, only the 2001–2005 lavascan be considered as K trachybasalts according to K2O=Na2O - 1.5 (Zanettin, 1984).The majority of samples fall in the fields of the trachybasalts and basaltictrachyandesites, only the Ellittico lavas are more evolved falling in the trachyandesitesand trachytes fields. The lavas of the eccentric cones of Mt. Spagnolo and Mt. Malettorepresent the most basic compositions, falling in the trachybasaltic and basaltic field.
644 C. Ferlito, G. Lanzafame / Lithos 119 (2010) 642–650
difference between the maximum and minimum values in wt.%respectively for K2O and SiO2, in the computed differentiation ΔK2O/ΔSiO2b0.2. Significantly, the K increment for most of the rocks ofMongibello follows a characteristic low K trend (K2O max 1.9 wt.%;ΔK2O/ΔSiO2~0.1–0.2 dashed line in Fig. 3a and b), driven by crystalfractionation of mafic phases. However, a good 30% of the recentMongibello and Ellittico rocks display a high K trend (K2O max 2.9 wt.%; ΔK2O/ΔSiO2~0.4–0.6 solid line in Fig. 3a and b), too high to beexplained by simple fractional crystallization. Even for fractionation
Fig. 2. CaO/Al2O3 vs. MgO diagram. For all the analyses in the data set, symbols as inFig. 1. A tendency may be seen for different volcanic phases throughout the evolution ofMongibello in the last 60 ka, of a Mg content increase and high CaO/Al2O3, which can berelated to a progressive ingress in the plumbing system of primitive magma batches.This hypothesis is verified at least for the most recent products which are associatedwith very basic and volatile rich melt inclusions (3.4 wt.% H2O and 4000 ppm CO2)(Métrich et al., 2004; Ferlito et al., 2008). The most primitive samples belong to theeccentric pre-historic eruptions of Mt. Spagnolo and Mt. Maletto.
greater than the amount suggested by the petrographic evidence (seereferences cited above) we cannot obtain compositions comparable tothose shownby the high K trend (Fig. 3a and b), since as crystallizationproceeds K enrichment will be accompanied by an increase in SiO2,Al2O3 and other major elements. The post-1971 rocks present high Kcontent at relatively low SiO2 content suggesting that it might not berelated to themagma differentiation during the upwardmigration butcould be a characteristic inherited from the partial melting of themantle.
Finally, earlier authors associated the post-1971 K enrichmentwith an increase of elements with analogous geochemical behaviourand having low crystal/melt partition coefficients, such as Th, Cs, Baand Rb (Joron and Treuil, 1984; Clocchiatti et al., 1988; Condomines etal., 1995). In Fig. 4 it can be seen that these elements display a positivecorrelation with SiO2, which can be explained by their accumulationin the residual melt during crystal fractionation. In particular, O'Hara(1977) argued that the continuous refilling of high level magmachambers with fresh basaltic magma could lead to an appreciableincrease of the most incompatible elements. Their enrichmentobserved in Etnean lavas might be partly driven by the refillinginvoked by O'Hara (1977) (cf. also Spera and Bohrson, 2004 andreferences therein). However, the incompatible element increase iscorrelated with MgO decrease (except for the post-1971 Rb, Fig. 4),and it is not as great as the increase in K.
3. A critical review of the existing K enrichment hypotheses
A difficulty with the hypothesis that K enrichment (of post 1971lavas) could be directly caused by the crustal assimilation (Joron andTreuil, 1984; Clocchiatti et al., 1988; Condomines et al., 1995; Tonariniet al., 1995), stems from the low K contents of the sedimentsunderlying the volcano. These sediments are dominated by thecarbonate rocks of the Hyblean plateau and by the carbonate andquartz-rich arenites of the Appennninic-Maghrebian chain (Lentini,1982 and references therein). To increase K content up to~2.2 wt.%, asin the post-1971 lavas, or~3 wt.% as for the Ellittico lavas (Fig. 3a andb), requires the assimilation of a significant quantity of sedimentaryrock. But K2O contents, up to 2.9 wt.%, are found only in mica-bearingflyschoid sediments, such as the Capo d'Orlando and the Piedimonteformations (Clocchiatti et al., 1988), which crop out only north of thevolcanic pile and are not found among the xenoliths sometimesembedded in Etnean lavas. Selective contamination through meteoricwater would also be unlikely to act on large scale, since it would affectonly the magma occupying the uppermost portion of the plumbingsystem (above 1 km depth). At such shallow levels all Etneanmagmasare normally volatile-saturated and therefore unable to incorporateexternal fluids (Métrich et al., 2004; Corsaro and Pompilio, 2004a;Allard et al., 2006). Additionally, these depths are far from the level atwhich the magma might reside for significant periods of time (Murruet al., 1999). In any case, the contamination model was formulatedexclusively for post-1971 lavas and does not explain the K fluctuationsfound in the older lavas (Fig. 3a and b).
The alternative idea that the K increase in the post-1971 lavas is acharacteristic inherited by the partial melting of a chemically andisotopically heterogeneous mantle source (Armienti et al., 1989;Barbieri et al., 1993; Corsaro and Cristofolini, 1996) has beensupported by 87Sr/86Sr and other isotopic ratios (Condomines et al.,1995; Schiano et al., 2001). Unfortunately, whole rock isotopic dataare not available for the entire Mongibello succession (Armienti et al.,1988, 1990; Condomines et al., 1995; Tanguy et al., 1997). As shown inFig. 5a, K2O increases considerably in the Ellittico lavas with arelatively small variation in 87Sr/86Sr (0.703226 - 0.703471); Fig. 5bconfirms that the Sr isotopic ratio increases in the post-1971 lavas (upto 0.703670), although the pre-historic aphiric basalts of Mt. Malettoand Mt Spagnolo display the entire Sr isotopic variation (0.703270 -0.703740). Such variability of 87Sr/86Sr of Ellittico and recent
Fig. 3. K2O vs. SiO2 diagram. For (a) the recent Mongibello and (b) Ellittico, symbols as in Fig. 1. Dashed arrows indicate “normal” K growth ΔK2O/ΔSiO2b0.2 due to crystalfractionation, solid arrows indicate anomalous K growth ΔK2O/ΔSiO2~0.4–0.6. See Section 2 for explanation.
645C. Ferlito, G. Lanzafame / Lithos 119 (2010) 642–650
Mongibello can be explained by a heterogeneous mantle sourceregion (cf. Viccaro and Cristofolini, 2008 and references therein), butoverall a correlation of isotopic ratios with the anomalous Kenrichment is not evident.
Finally, K enrichment was also interpreted as the signature of afluid component in the magma source derived from the subductingIonian slab (Schiano et al., 2001; Tonarini et al., 2001). The partialmelting of a peridotite metasomatized from a subducting slabcomponent (Schiano et al., 2001 and references therein) or the
Fig. 4. Incompatible trace element vs. MgO. Data sets are different for di
mixing between primary liquid and K-Si-rich melts could produce Krich primitive magmas (Tanguy et al., 1997). However, these modelsstill fail to explain that (a) in the Mongibello data set there are no Krich primitive lavas; (b) K enrichment does not have any definitetemporal trends; (c) other incompatible trace elements such as Cs, Rb,Th and Ba, do not mirror the K enrichment trend; and (d) K-enrichedlavas form recognizable clusters, corresponding to peculiar phases inthe evolution of the volcano (Fig. 3a and b). Within each cluster theenrichment in K is rather discontinuous: in the Ellittico sequence high
fferent elements. Symbols as in Fig. 1. See Section 2 for explanation.
Fig. 5. (a). K2O vs. 87Sr/86Sr. Symbols as in Fig. 1. (b) 87Sr/86Sr variation range for different phases of Mongibello evolution. See Section 3 for explanation.
646 C. Ferlito, G. Lanzafame / Lithos 119 (2010) 642–650
K lavas are found interbedded with low K lavas (Cristofolini et al.,2002); the pre-historic and historic lavas of recent Mongibello,supposed as typically low K, display about 30% of high K lavasthroughout the entire succession (Fig. 3a); even the lavas eruptedafter 1971 are not always K enriched, in fact during the eruption of2002 low K lavas were contemporaneously erupted along with high Klavas (Ferlito et al., 2009). All the above considerations make itdifficult to explain the K enrichment as a geochemical featureexclusively derived from the source.
4. Discussion
4.1. The degassing plumbing system
A useful proxy for the rates of magma supply and accumulation in theEtnean plumbing system is the SO2 flux, systematically measured bymeans of COSPEC since 1987 (Caltabiano et al., 2004 and referencestherein). Allard (1997) and Allard et al. (2006) used the total amount ofSO2 in the gas plume to estimate the accumulated volume of degassedmagma within the plumbing system. The results are surprising: about 4timesmoremagmahaddegassed than extruded, implying that in the longterm volatiles undergoing degassing come predominantly from un-erupted magma. Based on the radionuclide activity of 210Pb, 210Bi and210Po, Le Cloarec and Pennisi (2001) presented a sophisticated evaluationof the volume of the shallow degassing magma reservoir. They proposedthat for the period 1983–1995 only 15–20% of degassing magma waserupted. This un-erupted magma solidified at depth forming large high-velocity bodies of solidified dykes and crystal cumulates, extendingvertically across the sedimentary basement (Hirn et al., 1991; Laigle et al.,2000). Amajor low-velocity zonewas found through seismic refraction atabout 20–25 km of depth b.s.l. (Sharp et al., 1980), and interpreted as themain magma reservoir (1600 km3 of magma; assuming 14% of moltenrock). Murru et al. (1999) identified two low-velocity zones at 12±3 kmand at 5±2 km of depth b.s.l., and related them to shallow magmaponding. Recently, the simultaneous eruption of compositionally distinctmagmas during the 2001 and 2002–2003 events has revealed that risingmagma can involve different storage volumes and conduits that developindependently from each other (Clocchiatti et al., 2004; Métrich et al.,2004; Viccaro et al., 2006; Spilliaert et al., 2006a, b; Ferlito et al., 2008,2009).
4.2. The volatile induced differentiation
Geophysical and petrological data indicate that the plumbingsystem is continuously filled withmagma producing amassive releaseof volatiles which are transported as supercritical fluids and risethrough the overlyingmagma until, as pressure decreases, a gas phaseforms. During their ascent and/or storage in the sub-volcanicplumbing system magma batches undergo chemical differentiationmostly driven by crystal fractionation, which can be efficientlyaccompanied by other processes where volatiles become important(cf. Philpotts, 1990). The mass transfer of volatile components inigneous environments can involve mobile elements and produce a“volatile-induced differentiation” (Boudreau et al., 1986; Hedenquistand Lowenstern, 1994 and references therein; Keppler and Wyllie,1990, 1991; Rae et al., 1996; Webster, 1992, 1997; Wilkinson et al.,1996). Some authors have suggested the possible role of volatiles inthe evolution of peralkaline acid products (e.g., comendites; Bohrsonand Reid, 1997; Davies andMacDonald, 1987; MacDonald et al., 1987;Taylor et al., 1981).
Crucial in the transfer of metallic elements are the halogens; inparticular, Cl− is able to extract metals (i.e. K, Ti, Fe) as chloridecomplexes from the melt (Symonds et al., 1994 and reference therein;Kodosky andKeith, 1993; Shinohara et al., 1989 and references therein).Ferlito et al. (2008) interpreted the anomalous increase in Ti, Fe, P,and particularly of K and Cl in the glassy shards of 2001 tephra, asdue to a “volatile-induced differentiation” of a magma batch at depthof 6 km b.s.l.. They relate the process to three essential conditions:(1) the presence of a deep magma able to provide free migratingvolatiles; (2) a crust sufficiently permeable to permit the migration;(3) the presence of shallow-level magma batch(es) into which themetallic cations can be released by the exsolving volatiles.
4.3. Magma degassing and K enrichment of Etnean lavas
Data frommelt inclusions (Spilliaert et al., 2006b) indicate thatH2Oin magma is very abundant at depth: 3.50 wt.% at 490 MPa (~18 kmaccording to the density proposed by Corsaro and Pompilio, 2004b),whereas erupted lavas have lower H2O contents (~0.5–1 wt.%,references cited above) implying that magma loses its volatiles duringascent. For example for 1.7 km3 of magma solidified at depth in theperiod 1987–2000 (conservative value of Fig. 6a), the degassing of just
Fig. 6. (a) Cumulated erupted and degassed magma for the period 1987–2000. Erupted magma from Caltabiano et al. (2004); degassed magma, found using the equation of Allard(1997) Qd(m3)=Qs/[S∙ρ∙(1-X)]; Qs=SO2 flux (from Caltabiano et al., 2004), S=initial sulfur content in magma (we used 0.28 wt.% fromMétrich et al., 2004; Spilliaert et al., 2006a,b), ρ=magma density (we used 2.31 g/cm3 obtained for dykes of the Valle del Bove from Ferlito and Nicotra, 2007), X=crystal fractions (we used 0 and 30%). (b) Content of gas(CO2, H2O, S) in melt inclusions in olivines of 2001 and 2002 vs. pressure of entrapment.
Fig. 7. Cl, K2O/SiO2 and Na2O/SiO2 vs. pressure of entrapment of melt inclusions. SeeSection 4.2 for explanation.
647C. Ferlito, G. Lanzafame / Lithos 119 (2010) 642–650
1 wt.% H2O would produce 66×106 tons of H2O, which will migratetogether with halogens (in particular Cl) through the sedimentaryrocks of the basement, the volcanic pile and the shallow batches ofmagma.
According to Frank et al. (2003) Cl-rich fluids will interact with themagma extracting alkalis:
where HClfluid is a component of the fluid phase and (Na, K)OHmelt is acomponent in the basaltic melt, whereas (Na, K)Clmelt,fluid are Clbearing complexes free to migrate into the plumbing system andpartition between the magma and the supercritical fluid. Stelling et al.(2008) have experimentally proven that maximum Cl content for atrachybasaltic melt of Mt. Etna (trachybasalt of 2001 eruption at200 MPa) is 4 wt.% and that molar partition coefficient of Cl betweenfluids and silicate melt is strongly dependent on the bulk Cl,decreasing from 14 to 0.04 as Cl leaves the system as gaseous phase.This implies that at decreasing pressure, Cl will progressively leave thesystem and the remaining Cl will have an increasing affinity for thesilicate melt. Particularly interesting are the compositions of meltinclusions entrapped in olivines at various pressures (490 - 10 MPa)(Métrich et al., 2004; Spilliaert et al., 2006a) (Fig. 7). For entrapmentpressures ranging from 490 to 100 MPa, the Cl content in melt doesnot vary significantly (0.14–0.17 wt.%), whereas for pressures below100 MPa (≤4 km of depth) Cl in melts increases to 0.21 wt.% (Stellinget al., 2008). At 100 MPa Cl begins to exsolve, forms a gaseous phaseand can be removed from the magma (Spilliaert et al., 2006b).
To evaluate the behaviour of alkali and metals in melt inclusionswe must recognize that droplets of melts entrapped in olivinesbecome more silica-rich as the pressure decreases (Métrich et al.,2004; Spilliaert et al., 2006a). In order to avoid effects of differenti-ation we used K2O/SiO2 and Na2O/SiO2 ratios. In Fig. 7, alkalis follow apattern similar to Cl, remaining constant from 490 to 100 MPa andincreasing significantly for pressures below 100 MPa. The increase ofK and Na is higher than the increase in SiO2 and cannot be explainedby crystal fractionation.
We argue that alkalis can be extracted from the magma residing inthe lower portion of the plumbing system, form Cl-complexes ((Na, K)Clmelt,fluid cf. Frank et al., 2003) and migrate convectively upward assupercritical fluids (Fig. 8). As pressure decreases, the Cl solubilitythreshold (100 MPa according to Spilliaert et al., 2006a, b) is reachedand Cl will exsolve in the form of HCl, forming a gaseous phase and
leave the magma. Due to their low volatility in the vapor phase(Pokrovski et al., 2005) alkalis cannot leave with Cl in the gaseousphase and will remain in the melt, thus increasing their concentrationin magmas residing above the exsolution threshold for Cl.
Both K and Nawill be transported from the lower to the upper partof the plumbing system. In magmas with basic to intermediatecompositions Na will be easily accommodated within the lattice of thecrystallizing plagioclase (Armienti et al., 1990; Cashman, 1990 andreferences therein), which will become progressively more albitic(e.g. bytownite-labradorite for hawaiites; bytownite-andesine formugearites; andesine for benmoreites and trachytes - D'Orazio et al.,1997). Plagioclase crystals will grow in the convectively movingmagma of the shallow plumbing system (b50 MPa) and will produceconcentrically zoned phenocrysts (Cashman, 1993; Armienti et al.,1994a), which will eventually fractionate. Under these conditions theNa transferred to the shallow plumbing system will not increase its
Fig. 8. Scheme of the present-day plumbing system beneath Mount Etna volcano: deepmagma reservoir is the source of volatiles (S 0.35 wt.%; H2O 3.6 wt.%; CO2 4000 ppm;Aiuppa et al., 2007). See Section 4.3 for explanation.
648 C. Ferlito, G. Lanzafame / Lithos 119 (2010) 642–650
content in the residing magmas. Conversely, for magmas with basic tointermediate compositions, K is much more incompatible than Na,entering in the plagioclase only in small amount (partition coefficientplagioclase/melt 0.170, Arth, 1976). Potassium will therefore not beremoved by the fractionation occurring in the upper portion of theplumbing system, and will accumulate in the residual melt and so canexplain the observed K enrichment.
This model in which the K enrichment is closely associated withvolatile influx could explain the irregular occurrence of K and itsenrichment in high emission rate volcanic phases (Ellittico, pre-historic Mongibello and post-1971). Furthermore, recent Mongibelloand Ellittico, K-enriched lavas have SiO2 contents of 51 to 53 wt.% (cf.Fig. 3a and b), indicating that the K enrichment is accompanied by amoderate degree of fractional crystallization, and that the volatile-induced K enrichment becomes significant onlywhen shallow batchesof magma can reside for enough time to be infiltrated by largeamounts of fluids. In the post-1971 lavas, K enrichment is observed inmore basic compositions (SiO2~48 wt.%), implying that in present-day Etna the primitive magma entering the plumbing systemmust bemore basic than before 1971. This is in agreement with thecompositions of the melt inclusions found in the 2001 lavas, inwhich those entrapped at depth have SiO2b44 wt.% (Métrich et al.,2004; Spilliaert et al., 2006a, b).
Catalano et al. (2004) have pointed out the importance of tectoniccontrol in magma output rate for the last 200 ka, demonstrating thatthe transition from strike-slip to extensional regime that occurred125 ka ago has resulted in a significant increase in the amount ofmagma being intruded and erupted on Etna. We propose that theprogressive ingress of basic magma in the plumbing system of Ellitticoand recent Mongibello and after 1971 as well as the fluctuatingrecurrence of K-rich lavas may be controlled by extensional tectonicsin the Etnean region. However, during the early Ellittico period the Kenrichment was much more prominent, and can be associated withthe elevated growth rate during the Elittico volcanic phase, whichproduced a huge volcanic edifice (Branca et al., 2008). This leads to theconclusion that the degree of extension in the Etnean region is notcontinuous in time but fluctuates and this is consistent with theirregular growth of the edifice. The occurrences of K-enriched lavas
throughout the Etnean succession may indicate periods of highmagma influx within the plumbing system related to increments ofthe extensional regime of the entire Etnean area.
5. Concluding remarks
The critical appraisal of petrological data of Etnean lavas eruptedover the last 60 ka has shown that K enrichment is not exclusive to thepost-1971 lavas but is also common in the earlier lavas. All modelsaimed at explaining this K enrichment (sediments assimilation;mantle heterogeneity; subduction related components) do not takeinto account its intermittent character and its associationwith basic tointermediate whole-rock compositions. Volatile-induced differentia-tion can provide a good explanation of these occurrences and can beregarded as an important, though not exclusive, mechanism toexplain K increment. In particular, the contribution of a composition-ally heterogeneousmantle cannot be excluded, since inputs of gas richmagma, necessary to cause volatile-induced differentiation, can beassociated with an increase of mantle melting rate. As the source ofvolatiles can be related to the ingress of basic and more primitivemagmas in the feeding system, we might infer that the potassiumenriched lavas are associated with phases of intense extensionaltectonics during the evolution of the Etnean volcano.
Acknowledgements
We are thankful to Dr. Tommaso Caltabiano for providing us dataon gas emissions and for the fruitful discussions. We are also indebtedto G. Pokrovski, W. Bohrson and A. Kerr for their thoughtful and con-structive reviews. A final thank you goes to Liz Simpson and Paul H.Reitan, for their help with English text.
Appendix A. Supplementary data
Supplementary data to this article can be found online atdoi:10.1016/j.lithos.2010.08.006.
References
Aiuppa, A., Moretti, R., Federico, C., Giudice, G., Guerrieri, S., Liuzzo, M., Papale, P.,Shinohara, H., Valenza, M., 2007. Forecasting Etna eruptions by real-timeobservation of volcanic gas composition. Geology 35 (12), 1115–1118.
Allard, P., 1997. Endogenous magma degassing and storage at Mount Etna. GeophysicalResearch Letters 24, 2219–2222.
Allard, P., Behncke, B., D'Amico, S., Neri, M., Gambino, S., 2006. Mount Etna 1993–2005:anatomy of an evolving eruptive cycle. Earth Science Reviews 78, 85–114.
Armienti, P., Innocenti, F., Petrini, R., Pompilio, M., Villari, L., 1988. Sub-aphyric alkalibasalt from Mt. Etna: inferences on the depth and composition of the sourcemagma. Rendiconti della Società Italiana di Mineralogia e Petrologia 43, 877–891.
Armienti, P., Innocenti, F., Petrini, R., Pompilio, M., Villari, L., 1989. Petrology and Sr-Ndisotope geochemistry of recent lavas fromMt. Etna: bearing on the volcano feedingsystem. Journal of Volcanology and Geothermal Research 39, 315–327.
Armienti, P., Calvari, S., Innocenti, F., Petrini, R., Pompilio, M., Villari, L., 1990.Petrography and chemical composition. In: Barberi, F., Bertagnini, A., Landi, P.(Eds.), Mt. Etna: the 1989 eruption. CNR GNV Special Issue, Giardini, Pisa, pp. 30–33.
Armienti, P., Pareschi, M.T., Innocenti, F., Pompilio, M., 1994a. Effects of magma storageand ascent on the kinetics of crystal growth. The case of the 1991–93 Mt. Etnaeruption. Contributions to Mineralogy and Petrology 115, 402–414.
Armienti, P., Clocchiatti, R., D'Orazio, M., Innocenti, F., Petrini, R., Pompilio, M., Tonarini,S., Villari, L., 1994b. The long-standing 1991-1993 Mount Etna eruption:petrography and geochemistry of lavas. Acta Vulcanologica 4, 15–28.
Armienti, P., Tonarini, S., D'Orazio, M., Innocenti, F., Pompilio, M., Petrini, R., 1995.Geochmical and isotopic monitoring of Mt. Etna 1989–1993 eruptive activity:bearing on the shallow feeding system. Journal of Volcanology and GeothermalResearch 64, 95–115.
Arth, J.G. 1976 Behavior of trace element during magmatic processes – a summary oftheoretical models and their application Journal Res. U.S. Geol. Surv. 4 pp. 41-47. InRollinson H., 1993. Using geochemical data. Pearson edu. Ltd., pp 352.
Barbieri, M., Cristofolini, R., Delitala, M.C., Fornaseri, M., Romano, R., Taddeucci, A.,Tolomeo, L., 1993. Geochemical and Sr-isotope data on historic lavas of Mount Etna.Journal of Volcanology and Geothermal Research 56, 57–69.
649C. Ferlito, G. Lanzafame / Lithos 119 (2010) 642–650
Bohrson, W.A., Reid, M.R., 1997. Genesis of silicic perakaline volcanic rocks in an oceanisland setting by crustal melting and open system processes: Socorro Island,Mexico. Journal of Petrology 38, 1137–1166.
Boudreau, A.E., Mathez, E.A., McCallum, I.S., 1986. Halogen geochemistry of theStillwater and Bushveld complexes: evidence for transport of the platinum groupelements by Cl-rich fluids. Journal of Petrology 27, 627–645.
Branca, S., Coltelli, M., De Beni, E., Wijbran, J., 2008. Geological evolution of Mount Etnavolcano (Italy) from earliest products until the first central volcanism (between500 and 100 ka ago) inferred from geochronological and stratigraphic data.International Journal of Earth Sciences 97, 135–152.
Caltabiano, T., Burton, M., Giammanco, S., Allard, P., Bruno, N., Murè, F., Romano, R.,2004. Volcanic gas emission from the summit craters and flanks of Mt. Etna, 1987–2000. In: Calvari, S., Bonaccorso, A., Coltelli, M., Del Negro, C., Falsaperla, S. (Eds.),Mt. Etna: Volcano Laboratory, Geophysical Monograph Series, 143. AGU, Washing-ton DC, pp. 111–128.
Cashman, K.V., 1990. Textural constrains on the kinetics of crystallization of igneousrocks. In: Nicholls, J., Russell, J. (Eds.), Modern Methods of Igneous Petrology:Understanding Magmatic Processes: Mineralogical Society of America, 24,pp. 259–314.
Cashman, K.V., 1993. Relationship between crystallization and cooling rate insight fromtextural studies of dikes. Contributions to Mineralogy and Petrology 113, 126–142.
Catalano, S., Torrisi, S., Ferlito, C., 2004. The relationship between Late Quaternarydeformation and volcanism of Mt. Etna (eastern Sicily): new evidence from thesedimentary substratum in the Catania region. Journal of Volcanology andGeothermal Research 132, 311–334.
Clocchiatti, R., Joron, J.L., Treuil, M., 1988. The role of selective alkali contamination inthe evolution of recent historical lavas of Mt. Etna. Journal of Volcanology andGeothermal Research 34, 241–249.
Clocchiatti, R., Condomines, M., Guénot, N., Tanguy, J.C., 2004. Magma changes at MountEtna: the 2001 and 2002–2003 eruptions. Earth and Planetary Science Letters 226,397–414.
Condomines, M., Tanguy, J.C., Michaud, V., 1995. Magma dynamics at Mt Etna:constraints from U-Th-Ra-Pb radioactive disequilibria and Sr isotopes in historicallavas. Earth and Planetary Science Letters 132, 25–41.
Corsaro, R.A., Cristofolini, R., 1993. Nuovi dati petrochimici ed isotopici sullasuccessione del Mongibello Recente (M.te Etna). Bollettino dell'Accademia Gioeniadi Scienze Naturali, 26 (341), pp. 185–225.
Corsaro, R.A., Cristofolini, R., 1996. Origin and differentiation of recent basaltic magmasfrom Mount Etna. Mineralogy and Petrology 57, 1–21.
Corsaro, R.A., Cristofolini, R., Patanè, L., 1996. The 1669 eruption at Mount Etna:chronology, petrology and geochemistry, with inferences on themagma source andascent mechanisms. Bulletin of Volcanology 58, 348–358.
Corsaro, R.A., Pompilio, M., 2004a. 2004a. Magma dynamics at Mount Etna. In:Bonaccorso, A., Calvari, S., Coltelli, M., Del Negro, C., Falsaperla, S. (Eds.), Mt. Etna:Volcano Laboratory, Geophysical Monograph Series, 143. AGU, Washington DC, pp.91–110.
Corsaro, R.A., Pompilio, M., 2004b. Buoyancy-controlled eruption of magmas at Mt Etna.Terra Nova 16, 16–22.
Corsaro, R.A., Pompilio, M., 2004c. Magma dynamics in the shallowplumbing systemofMt. Etna as recorded by compositional variations in volcanics of recent summitactivity (1995-1999). Journal of Volcanology and Geothermal Research 137,53–71.
Corsaro, R.A., Miraglia, L., Pompilio, M., 2007. Petrologic evidence of a complexplumbing system feeding the July–August 2001 eruption of Mt. Etna, Sicily, Italy.Bulletin of Volcanology 69, 401–421.
Cristofolini, R., 1973. Recent trends in the study of Etna. Philosophical Transactions ofthe Royal Society of London A 274, 17–34.
Cristofolini, R., Corsaro, R.A., Ferlito, C., 1991. Variazioni petrolchimiche nellasuccessione etnea: un riesame in base a nuovi dati da campioni di superficie e dasondaggi. Acta Vulcanologica 1, 25–37.
Cristofolini, R., De Rosa, R.C., Tripolo, M., 2002. Variazioni petrografiche e geochimichedella sequenza vulcanica affiornate nell'area di Serra Giannicola Grande (Distrettovulcanico etneo). Abs 82° Congresso SIMP, Università della Calabria, Italy, Cosenza.
Davies, G.R., MacDonald, R., 1987. Crustal influences in the petrogenesis of the Naivashabasalt-comendite complex: combined trace element and Sr-Nd-Pb isotopeconstraints. Journal of Petrology 28, 1009–1031.
Distefano, S., 2007. Caratteri petrografici e chimici dei prodotti emessi dalla BoccaNuova (Etna) nel 1999. Laurea Thesis (in Italian), University of Catania, Italy
D'Orazio, M., Tonarini, S., Innocenti, F., Pompilio, M., 1997. Northern Valle del Bovevolcanic succession (Mt. Etna, Sicily): petrography, geochemistry and Sr-Ndisotope data. Acta Vulcanologica 9, 73–86.
Ferlito, C., 1994. Geologia e petrologia delle successioni di vulcaniti affioranti nella partesud-occidentale della Valle del Bove (Etna). Ph.D. Thesis, University of Catania,Italy.
Ferlito, C., Nicotra, E., 2007. The dykes swarm of Mt. Calanna as an example of coherentintrusion complex within the alkaline sequence of Mt. Etna (Italy). 23rd GeneralAssembly IUGG, July 2-13, Perugia, Italy.
Ferlito, C., Viccaro, M., Cristofolini, R., 2008. Volatile induced differentiation in theplumbing system of Mt. Etna volcano (Italy): evidence from glass tephra of the2001 eruption. Bulletin of Volcanology 70 (4), 455–473.
Ferlito, C., Coltorti, M., Cristofolini, R., Giacomoni, P.P., 2009. The contemporaneousemission of low-K and high-K trachybasalts and the role of the NE rift during the2002 eruptive event, Mt. Etna, Italy. Bulletin of Volcanology 71 (5), 575–587.
Frank, M.R., Candela, P.A., Piccoli, P.M., 2003. Alkali exchange equilibria between asilicate melt and coexisting magmatic volatile phase: an experimental study at800 °C and 100 MPa. Geochimica et Cosmochimica Acta 67, 1415–1427.
Ghiorso, M.S., Sack, R.O., 1995. Chemical mass transfer in magmatic processes 4. Arevised and internally consistent thermodynamic model for the interpolation andextrapolation of liquid-solid equilibria in magmatic systems at elevated tempera-tures and pressures. Contributions to Mineralogy and Petrology 119, 197–212.
Hedenquist, J.W., Lowenstern, J.B., 1994. The role of magmas in the formation ofhydrothermal ore deposits. Nature 370, 519–527.
Hirn, A., Nercessian, A., Sapin, M., Ferrucci, F., Wittlinger, G., 1991. Seismicheterogeneity of Mt Etna: structure and activity. Geophysical Journal International105, 139–153.
Joron, J.L., Treuil, M., 1984. Etude géochimique et pétrogenése des laves e l'Etna, Sicilie,Italie. Bulletin of Volcanology 47 (4), 1125–1144.
Keppler, H., Wyllie, P.J., 1990. Role of fluids in transport of uranium and thorium inmagmatic processes. Nature 348, 531–533.
Keppler, H., Wyllie, P.J., 1991. Partitioning of Cu, Sn, Mo, W, U, and Th between melt andaqueous fluid in the system haplogranite-H2O-HCl and haplogranite-H2O-HF.Contributions to Mineralogy and Petrology 109, 139–150.
Kodosky, L.G., Keith, T.E.C., 1993. Factors controlling the geochemical evolution offumarolic encrustations, Valley of Ten Thousand Smokes, Alaska. Journal ofVolcanology and Geothermal Research 55, 185–200.
La Delfa, S., Patanè, G., Clocchiatti, R., Joron, J.L., Tanguy, J.C., 2001. Activity of MountEtna preceding the February 1999 fissure eruption: inferred mechanism fromseismological and geochemical data. Journal of Volcanology and GeothermalResearch 105, 121–139.
Laigle, M., Hirn, A., Sapin, M., Lépine, J.C., Diaz, J., Gallart, J., Nicolich, R., 2000. MountEtna dense array local earthquake P and S tomography and implications forvolcanic plumbing. Journal of Geophysical Research 5, 21633–21646.
Le Cloarec, M.F., Pennisi, M., 2001. Radionuclides and sulfur content in Mount Etnaplume in 1983–1995: new constraints on the magma feeding system. Journal ofVolcanology and Geothermal Research 108, 141–155.
Le Maitre, R.W., 1989. A Classification of Igneous Rocks and Glossary of Terms.Recommendations of the IUGS Subcommission of Systematic of Igneous rocks.Blackwell Scientific, London, p. 193.
Lentini, F., 1982. The geology of the Mt. Etna basement. Memorie della SocietàGeologica Italiana 23, 7–25.
MacDonald, R., Davies, G.R., Bliss, C.M., Leat, P.T., Bailey, D.K., Smith, R.L., 1987.Geochemistry of high-silica peralkaline rhyolites, Naivasha, Kenya Rift Valley.Journal of Petrology 28, 979–1008.
Métrich, N., Allard, P., Spilliaert, N., Andronico, D., Burton, M., 2004. 2001 flankeruption of the alkali- and volatile-rich primitive basalt responsible for MountEtna's evolution in the last three decades. Earth and Planetary Science Letters 228,1–17.
Murru, M., Montuori, C., Wyss, M., Privitera, E., 1999. The locations of magma chambersat Mt. Etna, Italy, mapped by b-values. Geophysical Research Letters 26,2553–2556.
O'Hara, M.J., 1977. Geochemical evolution during fractional crystallization of aperiodically refilled magma chamber. Nature 266, 503–507.
Philpotts, A.R., 1990. Principles of igneous and metamorphic petrology. Prentice-Hall,pp. 235–258.
Pokrovski, G.S., Roux, J., Harrichoury, J.-C., 2005. Fluid density control on vapour–liquidpartitioning of metals in hydrothermal systems. Geology 33, 657–660.
Rae, D.A., Coulson, I.M., Chambers, A.D., 1996. Metasomatism in the North Qoroq centre,South Greenland: apatite chemistry and rare earth element transport.MineralogicalMagazine 60, 207–220.
Romano, R., 1982. Succession of the volcanic activity in the etnean area. Memorie dellaSocietà Geologica Italiana 23, 27–48.
Schiano, P., Clocchiatti, R., 1994. Worldwide occurrence of silica-rich melts in sub-continental and sub-oceanic mantle minerals. Nature 368, 621–624.
Schiano, P., Clocchiatti, R., Ottolini, L., Busà, T., 2001. Transition of Mount Etna lavasfrom a mantle-plume to an island-arc magmatic source. Nature 412, 900–904.
Scott, S.C., 1983. Variations in lava composition during the March 1981 eruption ofMount Etna and the implication of a compositional comparison with earlier historiceruption. Bulletin of Volcanology 46 (4), 393–412.
Sharp, A.D.L., Davis, P.M., Gray, F., 1980. A low velocity zone beneath Mount Etna andmagma storage. Nature 287, 587–591.
Shinohara, H., Iiyama, J.T., Matsuo, S., 1989. Partition of chlorine compounds betweensilicate melt and hydrothermal solutions. Geochimica et Cosmochimica Acta 53,2617–2630.
Spera, F.J., Bohrson, W.A., 2004. Open-system magma chamber evolution: an Energy-constrained Geochemical Model Incorporating the Effects of Concurrent Eruption,Recharge, Variable Assimilation and Fractional Crystallization (EC-E0RA_FC).Journal of Petrology 45 (12), 2459–2480.
Spilliaert, N., Allard, P., Métrich, N., Sobolev, A.V., 2006a. Melt inclusion record of theconditions of ascent, degassing, and extrusion of volatile-rich alkali basalt duringthe powerful 2002 flank eruption of Mount Etna (Italy). Journal of GeophysicalResearch 111, B04203.
Spilliaert, N., Métrich, N., Allard, P., 2006b. S-Cl-F degassing pattern of water rich alkalibasalt: modelling and relationship with eruption styles on Mount Etna Volcano.Earth and Planetary Science Letters 248, 772–786.
Stelling, J., Botcharnikov, R.E., Beermann, O., Nowak, M., 2008. Solubility of H2O- andchlorine-bearing fluids in basaltic melt of Mount Etna at T=1050-1250°C andP=200 MPa. Chemical Geology 256 (3–4), 102–110.
Tanguy, J.C., Kieffer, G., 1977. The 1974 eruption of Mount Etna. Bulletin of Volcanology40 (4), 239–252.
650 C. Ferlito, G. Lanzafame / Lithos 119 (2010) 642–650
Tanguy, J.C., Condomines, M., Kieffer, G., 1997. Evolution of Mount Etna: constraints onthe present feeling system and eruptive mechanism. Journal of Volcanology andGeothermal Research 75, 221–250.
Tanguy, J.C., Condomines, M., Le Goff, M., Chillemi, V., La Delfa, S., Patanè, G., 2007. MountEtna Eruption of the last 2, 750 years: revised chronology and location througharcheomagnetic and 226Ra-230Th dating. Bulletin of Volcanology 70, 55–83.
Taylor, R.P., Strong, D.F., Fryer, B.J., 1981. Volatile control of contrasting trace elementdistributions in peralkaline granitic and volcanic rocks. Contributions to Mineral-ogy and Petrology 77, 267–271.
Tonarini, S., Armenti, P., D'Orazio, M., Innocenti, F., Pompilio, M., Petrini, R., 1995.Geochemical and isotopic monitoring of Mt. Etna 1989–1993 eruptive activity:bearing on the shallow feeding system. Journal of Volcanology and GeothermalResearch 64, 95–115.
Tonarini, S., Armenti, P., D'Orazio, M., Innocenti, F., 2001. Subduction-like fluids in thegenesis of Mt. Etna magmas: evidence from boron isotopes and fluid mobileelements. Earth and Planetary Science Letters 192, 471–483.
Treuil, M., Joron, J.L., 1994. Etude Géochimique des elements en traces dans les lavesémises au cours de l'éruption 1991-1993 de l'Etna Mise en evidence descontributions de la source, de la fusion partielle, de la differenciation et desmodalités de transfert des magmas. Acta Vulcanologica 4, 29–45.
Trigila, R., Spera, F.J., Aurisicchio, C., 1990. The 1983 Mount Etna eruption:thermochemical and dynalmical inferences. Contributions to Mineralogy andPetrology 104, 594–608.
Viccaro, M., Ferlito, C., Cortesogno, L., Cristofolini, R., Gaggero, L., 2006. Magma mixingduring the 2001 event at Mount Etna (Italy): effect on the eruptive dynamics.Journal of Volcanology and Geothermal Research 149, 139–159.
Viccaro, M., Cristofolini, R., 2008. Nature of mantle heterogeneity and its role in theshort-term geochemical and volcanological evolution of Mt. Etna (Italy). Lithos 105(3–4), 272–288.
Webster, J.D., 1992. Fluid-melt interactions involving Cl-rich granites; experimentalstudies from 2 to 8 kbar. Geochimica et Cosmochimica Acta 56, 659–678.
Webster, J.D., 1997. Exsolution of magmatic volatile phases from Cl-enrichedmineralizing granitic magmas and implications for oremetal transport. Geochimicaet Cosmochimica Acta 61, 1017–1029.
Wilkinson, J.J., Nolan, J., Rankin, A.H., 1996. Silicothermal fluid: a novel medium formass transport in the lithosphere. Geology 24, 1059–1062.
Zanettin, B., 1984. Proposed new chemical classification of volcanic rocks. Episodes 7(4), 19–20.
