New insights into magma dynamics during last ... - Wiley Online Library

Geochemistry Geophysics Geosystems

3

G

Research Letter Volume 7, Number 6 8 June 2006 Q06008, doi:10.1029/2005GC001175

AN ELECTRONIC JOURNAL OF THE EARTH SCIENCES Published by AGU and the Geochemical Society

ISSN: 1525-2027

New insights into magma dynamics during last two eruptions of Mount Etna as inferred by geochemical monitoring from 2002 to 2005 A. Rizzo, A. Caracausi, R. Favara, M. Martelli, A. Paonita, and M. Paternoster Istituto Nazionale di Geofisica e Vulcanologia—Sezione di Palermo, Via Ugo La Malfa 153, Palermo I-90146, Italy ([email protected])

P. M. Nuccio and A. Rosciglione Dipartimento CFTA, Universita` di Palermo, Via Archirafi 36, Palermo I-90123, Italy

[1] Two distinct eruptive events characterize the volcanic activity at Mount Etna during the 2002 to 2005 period. We identified signals of magma ascent preceding these eruptions by geochemical monitoring of both chemical composition and He-isotope ratio of gas emissions from five locations in the peripheral area of the volcano. The geochemical signals are interpreted using the models proposed by Caracausi et al. (2003a, 2003b) and allow identification of episodes of magma ascent and estimation of the pressures of degassing magma. As observed for the 2001 eruption (Caracausi et al., 2003b), magma ascent probably triggered the onset of the 2002–2003 eruption, and minor events of magma ascent were observed between May and December 2003. In contrast to the previous two eruptions, the 2004–2005 eruption was not preceded by significant geochemical signals of volcanic unrest, suggesting that this eruption was mainly triggered by the failure of the upper portion of the volcanic edifice under the magmatic hydrostatic pressure in the conduits. High 3He/4He ratio revealed new volatile-rich magma accumulation. The 2002–2003 eruption was preceded by a much shorter period of new magma accumulation from deep levels of the feeding system. Few minor signals of magma migration were detected at some of the sites during the months preceding the 2004–2005 eruption, suggesting that the degassed 3He-depleted magma resident in the volcanic conduits was not replaced by new volatile-rich magma. This is in agreement with the lack of explosive activity during the 2004–2005 eruption and with petrologic observations that the parent magma probably erupted in 2000 and 2001. New geochemical signals of magma ascent from the deep reservoir have been identified since June 2005, indicating that the volcanic activity of Mount Etna is evolving toward new pre-eruptive conditions. Components: 6783 words, 5 figures, 2 tables. Keywords: chemical composition; geochemical monitoring; He isotope; magma migration. Index Terms: 1009 Geochemistry: Geochemical modeling (3610, 8410); 8434 Volcanology: Magma migration and fragmentation. Received 3 November 2005; Revised 7 March 2006; Accepted 17 March 2006; Published 8 June 2006. Rizzo, A., A. Caracausi, R. Favara, M. Martelli, A. Paonita, M. Paternoster, P. M. Nuccio, and A. Rosciglione (2006), New insights into magma dynamics during last two eruptions of Mount Etna as inferred by geochemical monitoring from 2002 to 2005, Geochem. Geophys. Geosyst., 7, Q06008, doi:10.1029/2005GC001175.

1. Introduction [2] Mount Etna has been characterized in the last 5–6 years by frequent and sudden eruptions with Copyright 2006 by the American Geophysical Union

many new and unusual geochemical and petrologic features [Clocchiatti et al., 2004; Andronico et al., 2005; Corsaro and Miraglia, 2005] that have significantly modified the morphology of the upper 1 of 12


3

G

rizzo et al.: mount etna magma dynamics

part of the volcanic edifice. Regular geochemical monitoring of the Mount Etna volcano has been aimed at relating the variations observed in plume chemistry [Bruno et al., 2001; Aiuppa et al., 2004a, 2004b, 2005; Allard et al., 2004], soil CO2 flux [Giammanco et al., 1998a, 1998b; Alparone et al., 2004], and both the chemical and the He-isotope composition of gas emitted [Caracausi et al., 2003a, 2003b], with the degassing mechanism and eruption dynamics. Caracausi et al. [2003a] investigated the 3He/4He ratio in gas emissions measured twice per month from five locations in the peripheral area of the volcano. They identified synchronous variations in the compositions of gas sampled at sites located 60 km apart. They attributed these variations to magma degassing process, and suggested that the Etnean plumbing system is much more extensive than previously thought. They also detected pulses of ascending magma that probably provided the engine for the onset of the 2001 eruption. Caracausi et al. [2003b] investigated the chemical composition of the same peripheral gas emissions, and found that gas discharged during magma degassing interacts with local and shallow hydrothermal aquifers and causes the selective dissolution of CO2. They proposed a quantitative model for determining the pristine magmatic gas composition. Briefly, they considered that each species in the magmatic gases dissolves into an infinitesimal parcel of gas-free hot water up to saturation, with the process being incrementally repeated until the results are consistent with the chemical composition measured in the collected gases. The recalculated He/CO2 and He/Ne ratios displayed variations over time that were synchronous at all the monitored sites. On the basis of numerical simulations of the Etnean magma degassing, these variations have been attributed to volatile outgassing following the ascent and depressurization of magma [Caracausi et al., 2003b]. This allowed the authors to determine the depth of magma batches feeding the sampled emissions, with their results being in good agreement with the two main magma bodies at depths of about 10 and 3 km b.s.l., as identified by geophysical investigations [e.g., Murru et al., 1999; De Gori et al., 2005]. [3] In this paper, we describe data from geochemical monitoring of peripheral gas discharges in the period 2002–2005, and explain the observed variations on the basis of the above mentioned models of chemical composition as well as 3He/4He ratios of the gases, as already monitored by Caracausi et al. [2003a, 2003b] up to the end of 2001. We

10.1029/2005GC001175

identify the signals indicative of magma migrations toward the shallowest levels of the Etnean plumbing system during the investigated period, which show that the 2002–2003 eruption was preceded by magma injections into the shallow storage volume, and that minor signals were detectable before the onset of the 2004–2005 eruption. Finally, we show that a new magma migration toward a depth of about 4 km b.s.l. occurred between June and August 2005.

2. Eruptive Activity During 2002–2005 [4] Following the 2001 eruption, Etna was almost quiescent until March 2002, when the summit craters reactivated [Patane` et al., 2003; Andronico et al., 2005]. Strombolian activity (at the Bocca Nuova and Northeast Craters) together with shallow seismicity and deformation then preceded the 2002–2003 eruption [Andronico et al., 2005; Neri et al., 2005]. This eruption began on 26 October 2002 and was preceded by a seismic swarm lasting 2 hours, while a 1-km-long eruptive fracture opened on the southern flank, at 2850–2600 m a.s.l. [Acocella et al., 2003]. Early in the morning of 27 October 2002, a second eruptive fracture opened on the northeastern flank of the volcano at a similar altitude as the southern flank fracture, showing lava fountains coupled with violent Strombolian activity [Andronico et al., 2005]. The lava effusion followed explosive activity and produced a 1-km-long flow. The petrography and geochemistry of the erupted products indicate that the lava on the northeastern flank was partially degassed during its storage at shallow depths of the plumbing system. The 2002–2003 volcanic products show features similar in petrography as the 2001 eruptive products emitted from the vents located on the upper portion of the volcano (higher than 2700 m a.s.l.). In particular, they are mineralogically similar to trachybasalts erupted during the past decades by the summit craters. They are porphyritic, showing 30 – 40% phenocrysts [Clocchiatti et al., 2004]. Conversely, the magma erupted from the southern fissure is relatively undegassed, and can be linked to the deeper portion of the plumbing system. Its petrographic features were close to those of the magma feeding the 2001 eruptive fissure located along the lower southern flank (2600 – 2100 m a.s.l. [Andronico et al., 2005]). These lavas are more basic and show a much smaller amount of phenocrysts (5–10%), with the occasional occurrence of amphibole and orthopyroxene, and abundant sandstone inclusions 2 of 12


3

G


[Clocchiatti et al., 2004]. It is noteworthy that the exceptional explosive activity characterizing the 2002–2003 eruption (among the most intense in recent times) significantly modified the morphology of the upper parts of the volcano with the growing of two distinct cones [Patane` et al., 2003; Andronico et al., 2005]. Furthermore, having different magmas feeding the same eruptive activity is extraordinary, with the last such identified event (apart from the 2001 eruption) occurring 300 years ago [Andronico et al., 2005]. [5] The 2002–2003 eruption on 28 January 2003 was followed by a long period of quiescence and on 7 September 2004 a new effusive eruption started. This event was not preceded by seismic activity or ground deformations [Burton et al., 2005; Patane` et al., 2005]. The lava flow fed a fissure system that opened on the lower eastern flank of the Southeast Crater, and maintained almost steady conditions until the eruption ended on 8 March 2005 [Corsaro and Miraglia, 2005]. Combined geophysical, petrological and geochemical studies [e.g., Burton et al., 2005; Corsaro and Miraglia, 2005; Patane` et al., 2005] suggest that this eruption stored in the shallow part of the plumbing system since the volcanic activity in 2000 and 2001. Moreover, the absence of significant explosive activity during this eruption indicates that the erupted magma was extremely degassed [Corsaro and Miraglia, 2005].

3. Analytical Techniques [6] The determination of the helium isotopic composition was carried out on a static vacuum mass spectrometer (VG-5400TFT, VG Isotopes) modified for the simultaneous detection of 3He and 4He ion beam, in order to reduce the error of the 3 He/4He measurements in 3He-poor gases (e.g., Etnean volcanic gases) down to very low values (an average of ±0.05 Ra). The 3He/4He ratio was determined by measuring 3He on the Daly detector and 4He in the Faraday cup. The 3He/4He ratios have been corrected for the atmospheric contamination [Sano and Wakita, 1985]. The correction was generally negligible and the corrected values are reported as R/Ra values (where Ra is equal to 1.39 106). The neon content was determined by the comparison of peak intensity of the 20Ne of air and that of the gas sample. [7] Chemical abundances of He, O2, N2, CH4 and CO2 were measured with a Perkin Elmer Autosystem XL gas-chromatograph equipped with a 4 m

10.1029/2005GC001175

Carboxen 1000 column and double detector (Hot Wire Detector followed by a methanizer and Flame Ionisation Detector). Analytical errors were ±5% for He and ±3% for the other species.

4. Results and Discussion [8] We have investigated the volcanic activity of Mount Etna from the end of the 2001 eruption until October 2005. Gas samples were collected from four sites (Stadio, P39, Vallone Salato, and Fondachello) located around Mount Etna and from one (Naftia) that is 40 km south of the Mount Etna border at the limit with the Hyblean volcanics (Figure 1), with a sampling frequency ranging from twice a month during quiescence and up to twice a week during periods of volcanic unrest. The chemical composition as well as the helium isotope ratios are listed in Table 1.

4.1. He Isotopes [9] The long-term record of 3He/4He ratios (as R/Ra) measured in the sampled gases are shown in Figure 2. Most of the analyzed gases have 4He/20Ne ratios in excess of 300, well above that in the air (4He/20Ne = 0.318) and hence ruling out any significant air contamination. The measured 3He/4He ratios showed a large variability, well outside the overall error (±0.05 Ra). Most of the He-isotopes variations are coherent between the sites. Similar coherence was observed in the time period as observed between 1996 and 2001 [Caracausi et al., 2003a], even if the monitored emissions have differences in the average 3He/4He ratios. These authors used the He-isotope variations at constant CH4 concentrations as evidence for the lack of interaction with crustal fluids. Instead the spatial and temporal variations in 3 He/ 4 He may be explained by magma degassing, which would cause kinetic fractionation of helium isotopes during the exsolution of the gas phase. During bubble growth, He isotopes can undergo kinetic fractionation as the lighter 3He diffuses into the growing bubbles at a higher rate than 4He. The residual melts are consequently depleted of 3He, relative to 4He [Nuccio and Valenza, 1998; Caracausi et al., 2003a]. This process can be modeled by the following equation, which is similar to that proposed by Hoefs [1987] for Rayleigh distillation: R=R0 ¼ Fðk1Þ ;

ð1Þ

where R is the 3He/4He ratio in the melt, F is the residual fraction of helium after a degassing event, 3 of 12


3

G


10.1029/2005GC001175

Figure 1. Satellite image of Sicily and the Mount Etna ash plume on 28 October 2002 just after the onset of the 2002 –2003 eruption (modified from the MODIS Rapid Response System, NASA). Sketch map of the location of sampling sites. The characteristics of the sampling sites are detailed by Caracausi et al. [2003a].

R0 refers to initial conditions (namely 3He/4He in melt before the degassing event), and k is the kinetic fractionation factor (k = [4/3]1/2). Equation (1) predicts that degassing will decrease the 3 He/4He ratio of the melt. [10] The lower 3He/4He ratio of the emissions at the Stadio and Fondachello sites indicates these are fed by more-degassed portions of the plumbing system, especially compared to site P39. The different extent of evolution of the magma batches inside the same plumbing system may be explained by a system of fractures and dykes (and not simple magma chambers). This plumbing system is fed directly from the asthenospheric mantle. This is in agreement with geophysical observations [Hirn et al., 1997; Chiarabba et al., 2004; De Gori et al., 2005]. Thus gas discharges (separated by several kilometers) originate from magma bodies that are only partially in communication and hence experiencing differing extents of degassing. [11] In this respect, lava of the 2002–2003 eruption, for which the 3He/4He ratio in the fluid inclusions of olivine phenocrysts was 6.6 ± 0.16 Ra (average of N = 16 samples; Natalie MacLean and Finlay Stuart, personal communication), would be fed by magma batches that were more degassed than that contemporaneously feeding site P39 that show higher values.

[12] Kinetic fractionation could also be the key process underlying the temporal 3He/4He variations observed at the monitored sites. Decreasing 3 He/4He values are due to progressive magma degassing driven by magma ascent and decompression toward the surface, while high 3He/4He ratio in the record are either discharge of bubbles kinetically enriched in 3He during magma vesiculation or outgassing of new volatile-rich magma coming from deep levels of the plumbing system [Nuccio and Valenza, 1998; Caracausi et al., 2003a]. The contemporaneous occurrence of the isotopic variations at all the monitored sites would be indicative of an extensive involvement of the plumbing system during the degassing and/or refilling events. As inferred by Caracausi et al. [2003a], the Etnean feeding system appears to be much wider than previously reported [Sharp et al., 1980; Hirn et al., 1997; Murru et al., 1999], with an alignment in a northeast-southwest direction [Sharp et al., 1980] that would support the synchronous variations observed at the Naftia and Fondachello sites, which are separated by 70 km. Conversely, minor magma input from depth may only partially involve the plumbing system, causing synchronous variations at only two or three of the monitored sites. In this respect, the correlation of the R/Ra variations among the sites can be used to qualitatively estimate the magnitude of magma dynamics events in the plumbing system. 4 of 12


3

G


10.1029/2005GC001175

Table 1 (Representative Sample). Chemical and Helium Isotopic Composition of the Sampled Gasesa [The full Table 1 is available in the HTML version of this article at http://www.g-cubed.org] Date

He

Ne

01/09/02 01/22/02 02/05/02 02/14/02 02/27/02 03/12/02 03/26/02 04/09/02 04/23/02 05/07/02 05/21/02 06/04/02 06/17/02 07/01/02 07/16/02 07/31/02 08/13/02 08/28/02 09/10/02 09/24/02 10/10/02 10/22/02 10/27/02 10/31/02 11/03/02 11/08/02 11/11/02 11/14/02 11/18/02 11/21/02 11/25/02 12/02/02 12/11/02 12/17/02 12/24/02 12/27/02 01/08/03 01/16/03 01/23/03 01/30/03 02/05/03 02/22/03 03/12/03 04/02/03 04/16/03 05/06/03 05/30/03 06/13/03 06/24/03 07/03/03 07/17/03 07/28/03 08/07/03 08/19/03 09/03/03 09/30/03 10/16/03 10/28/03

75 93 96 88 84 72 77 72 67 74 75 83 59 90 69 78 91 67 73 85 78 86 105 76 82 68 93 63 73 77 71 79 70 74 72 74 75 59 72 75 66 61 60 66 66 55 56 60 57 57 58 60 56 56 56 54 50

0.065 0.049 0.050 0.084 0.218 0.175 0.018 0.063 0.035 0.078 0.039 0.071 0.108 0.084 0.093 0.171 0.277 0.139 0.188 0.139 0.093 0.081 0.090 0.117 0.089 0.114 0.220 0.118 0.056 0.038 0.035 0.102 0.107 0.082 0.054 0.077 0.058 0.131 0.040 0.030 0.078 0.050 0.021 0.047 0.061 0.084 0.082 0.068 0.044 0.054 0.105 0.047 0.085 0.267 0.045 1.045 2.045

N2

CH4

CO2

R/Ra

P39 Site 1.1 0.9 0.7 1.1 1.0 0.7 0.0 0.6 0.9 0.4 0.6 0.6 0.8 0.7 0.7 0.8 1.1 1.3 1.0 1.3 0.7 0.8 1.1 0.8 0.8 0.7 1.0 0.5 0.8 6.6 0.6 0.6 0.6 0.5 0.5 0.5 0.6 0.5 0.5 0.5 0.5 0.6 0.5 0.7 0.7 0.6 0.6 0.6 0.6 0.7 0.8 0.9 0.7 4.2 0.5 0.5 0.3

12.6 12.9 12.5 12.5 12.8 12.8 12.4 12.2 12.5 12.5 11.9 12.3 11.9 12.1 11.9 12.2 12.3 12.2 12.5 12.3 12.4 12.3 13.1 12.5 12.5 12.8 13.3 12.3 12.8 12.3 12.4 13.0 12.0 12.2 12.4 12.3 12.0 12.1 12.3 12.4 12.5 12.6 13.0 13.4 12.8 13.5 14.5 14.4 14.1 10.7 11.0 13.0 13.1 11.1 13.6 13.9 13.9

86.1 86.6 86.4 86.6 86.4 86.6 89.4 87.9 86.5 87.6 87.6 87.2 87.3 87.2 87.3 87.1 86.4 86.5 86.8 86.7 87.0 87.0 85.8 86.6 86.7 87.8 85.5 87.3 86.9 80.5 87.2 86.7 87.6 87.8 87.3 87.4 87.6 88.1 87.4 87.6 87.2 87.6 87.1 86.3 86.7 86.2 85.0 85.2 86.1 89.3 88.9 86.3 87.0 84.5 85.6 85.7 85.8

7.01 7.07 7.03 7.04 7.12 7.01 7.00 6.97 6.91 6.89 6.94 6.90 6.90 6.83 6.70 6.65 6.73 6.83 6.91 6.99 7.17 7.02 6.91 6.85 6.93 6.94 6.90 6.89 6.97 7.00 6.95 6.98 6.94 7.02 7.09 7.23 6.89 7.02 7.00 7.06 6.95 7.06 7.12 6.98 6.95 6.63 6.83 6.96 7.01 6.74 6.74 6.96 6.98 6.91 6.98 7.01 7.00

a Oxygen is not reported because it is generally negligible, except for a few neon-rich samples that were air contaminated. Helium and neon are expressed in ppmVol, while N2, CH4, and CO2 are in %vol. R/Ra values are reported already corrected for the small amount of air contamination.

5 of 12


3

G


10.1029/2005GC001175

Figure 2. Variations of the 3He/4He ratio (as R/Ra) from June 2001 to October 2005 at all five sampling sites. R/Ra values are corrected for the small amount of air contamination on the basis of the 4He/20Ne ratio [Sano and Wakita, 1985]. Average error bar (±0.05 Ra) is also shown. Vertical lines include the last three eruptive periods (2001, 2002 – 2003, and 2004 – 2005). Data of 2001 are from Caracausi et al. [2003a]. The sampling procedures are detailed by Caracausi et al. [2003b].

[13] The above considerations enabled us to identify magma migrations toward the surface. Following the major variations recorded before and during the July 2001 eruption [Caracausi et al., 2003a], new and significant variations of the 3He/4He values were observed during the last 4 months of 2001 (Figure 2), and these have been interpreted as episodes of magma ascent within the plumbing system. The 3He/4He variations also indicated a magma migration toward the surface and associated degassing from February to August 2002 (Figure 2), which finished just before the onset of the 2002–2003 eruption (27 October 2002). A further input of magma probably occurred close to the end of the eruptive period (January 2003). The asynchronous He-isotope record from the different sites would suggest that only a part of the plumbing system experienced a new magma input. Following the eruption, new episodes of magma recharge from May to September 2003 were evident in the gases collected from all the sites, with a behavior similar to that following the 2001 eruption; no significant variations were subsequently observed. The most-recent variation of the He-isotope ratio was recorded from June to August 2005, when it reached its highest value in 2 years, suggesting a new episode of magma migration toward the surface. [14] To elucidate the long-term behavior of the Etnean plumbing system, we considered 9 years

of 3He/4He ratio data recorded at the Naftia and Vallone Salato sites beginning from 1997, including data from Caracausi et al. [2003a]. Trend lines for the data using moving averages of 12 points (equating to about 6 months) are plotted in Figure 3; note that similar results were observed for the gas discharges from the other sites (Stadio, Fondachello, and P39). These trend lines clearly show that the main variations occurred almost simultaneously at both sampling sites. Decreasing 3 He/4He with time is attributed to progressive magma degassing, while the increasing 3He/4He with time a gradual replenishment of new volatilerich melts with a relatively high 3He/4He ratio. [15] The variations of the average 3He/4He values observed between 1997 and the first few months of 1999 (Figure 3) suggest a recharge of new volatilerich magma (followed by its progressive degassing) involving the entire plumbing system, which probably fed the explosive and effusive activity that took place at Mount Etna during that period [Alparone et al., 2003, and references therein]. The average 3He/4He ratios increased from the first half of 1999 till August 2001 (Figure 3), suggesting a new massive recharge of the plumbing system by nondegassed magma in accordance with the conclusions reached by Caracausi et al. [2003a] from long-term monitoring at site P39. The new magma exhibited a progressive degassing from the end of 2001 until September 2002, after which a new 6 of 12


3

G


10.1029/2005GC001175

Figure 3. Twelve-point moving average (equating to about 6 months) of R/Ra from 1996 to 2005. Only the data for the Naftia and Vallone Salato sites are shown, since similar behavior was observed at all the sampling sites. Yellow bar represents an intense phase when explosive and eruptive activity occurred at all the craters. Vertical lines are as in Figure 2.

replenishment of volatile-rich magma occurred that both preceded and accompanied the 2002–2003 eruption. Subsequent small and localized variations in a general degassing phase were recorded until the first half of 2005, crossing the 2004–2005 effusive eruption, with a new and ongoing magma recharge apparently starting in June 2005. [16] Figure 3 suggests decoupling of the deep magma supply to the plumbing system of the volcano and the superficial occurrence of an eruption. The 2001 and 2002–2003 eruptions occurred during a deep recharging phase of the plumbing system but ended while magma supply to the deep levels of the plumbing system was still ongoing. The 2004–2005 eruption appears to have lacked magma recharging at a shallow depth, with the eruption probably being due to failure of the upper parts of the volcanic edifice [Burton et al., 2005; Corsaro and Miraglia, 2005].

4.2. Chemical Composition [17] In addition to the He isotopic composition, we also monitored the chemical composition of the discharged gases (Table 1). As already observed by Caracausi et al. [2003a, 2003b] for the pre-2001 gasses, the gases are of magmatic origin, and are normally dominated by CO2, and have a variable CH4 content. The CH4 content progressively increased from 0.4% at Naftia site up to 90% at the Fondachello site. Oxygen was generally present at only trace level in all the collected samples. He

ranged from 30 ppmV at the Stadio site to more than 700 ppmV at the Fondachello site, while Ne varied from a few ppbV to 0.7 ppmV. The gas released from the magma interacts with hot, shallow hydrothermal aquifers during its rise toward the surface, causing significant changes in its chemical composition [D’Alessandro et al., 1997; Caracausi et al., 2003b]. Thus the variability of methane content, as well as that of He and Ne content, is due to the selective dissolution of CO2 in water that induces a relative enrichment of the less-soluble species in the residual gas phase. Caracausi et al. [2003b] modeled the interaction of the ascending Etnean magmatic gas through an aquifer, and their quantitatively evaluation allowed them to calculate the pristine magmatic gas composition prior to interaction. Briefly, they considered that each species in the magmatic gases dissolves into an infinitesimal parcel of gas-free hot water up to saturation, with the process being incrementally repeated until the results are consistent with the chemical composition measured in the collected gases. [18] As observed by Caracausi et al. [2003b], the corrected He/Ne and He/CO2 ratios display synchronous variations over time, which can be attributed to magma degassing caused by ascent and decompression of the magma body. The volatiles dissolved exhibit different solubilities in silicate melt, and thus the composition of the exsolved vapor varies as a function of the extent of degassing 7 of 12


3

G


10.1029/2005GC001175

Table 2. Main Variations of He/Ne and He/CO2 Ratios Over the Investigated Period and Related Initial and Final Degassing Pressures of the Magmaa Date of Variations

Restored DHe/CO2b

Measured DHe/Ne

Computed Pressure in MPa

P39 Site Start End Start End Start End Start End Start End

08/03/01 08/16/01 10/09/01 10/29/01 08/28/02 10/27/02 10/09/02 10/27/02 05/30/03 07/17/03

Startc Endc Startc Endc Start End

06/28/01 07/27/01 08/16/01 09/18/01 07/05/05 08/03/05

Startc Endc Startc Endc Startc Endc Startc Endc Start End Start End Start End Start End Start End Start End Start End

07/27/01 08/03/01 09/07/01 09/14/01 10/02/01 10/23/01 01/22/02 02/27/02 07/16/02 08/13/02 08/28/02 09/24/02 10/10/02 10/22/02 10/28/02 10/31/02 01/23/03 01/30/03 09/08/04 10/12/04 07/05/05 08/25/05

Startc Endc Startc Endc Startc Endc Start End Start End Start End Start End

07/27/01 08/13/01 09/07/01 10/02/01 10/23/01 11/07/01 07/01/02 10/27/02 08/19/03 09/03/03 12/01/03 12/30/03 09/22/04 10/12/04

0.471

0.699

0.349

0.635

0.319

0.600

0.214

0.577

0.281

0.498

474 150 471 198 453 208 491 276 401 212

Vallone Salato Site 0.389

0.500

0.387

0.626

0.235

0.529

444 176 443 177 447 249

Fondachello Site 0.277

0.472

0.194

0.400

0.354

0.449

0.244

0.402

0.387

0.438

0.543

0.646

0.336

0.451

0.355

0.404

0.371

0.690

0.149

0.314

0.315

0.438

381 206 388 249 332 161 354 210 314 147 407 119 342 171 305 154 503 200 364 261 342 179

Stadio Site 0.404

0.571

0.478

0.549

0.310

0.493

0.758

0.694

0.244

0.543

0.429

0.429

0.182

0.500

398 162 356 129 382 193 394 64 452 245 294 130 466 288

8 of 12


3

G


10.1029/2005GC001175

Figure 4. Synchronous increases in the He/Ne and He/CO2 ratios at the Fondachello site around the 2002 – 2003 eruption (shown by the vertical red lines). The variations recorded from July 2002 to January 2003 enabled magma migration in the plumbing system to be identified (see text). Both ratios are corrected for shallow dissolution processes [Caracausi et al., 2003b].

and of the magma pressure [Nuccio and Paonita, 2001]. Caracausi et al. [2003b] calculated that both He/Ne and He/CO2 ratios increase with increasing extent of degassing because CO2 and Ne are both less soluble than He. By simultaneously incorporating the temporal variation of both He/Ne and He/CO2 in the degassing model, we were able to estimate the initial and final degassing pressures of the magma body (Table 2). [19] During the investigated period, we detected several synchronous increases in the He/Ne and He/CO2 ratios (Figure 4), most of which occurred around the 2002–2003 eruption. On the basis of the above interpretation, we consider that the He/ Ne and He/CO2 covariation indicates events of magma migration in the Etnean plumbing system, and we have quantified the related degassing pressures using the method of Caracausi et al. [2003b]. The calculations suggest the occurrence of magma-ascent events from the deep reservoir toward the shallowest one (Table 2, Figure 5). In the periods preceding and following the onset of the 2002–2003 eruption, several inputs of magma from deep levels of the plumbing system were evident from the synchronous increases in chemical ratios. In agreement with the data on He

isotopes, several magma-ascent events toward a depth of 3 –5 km were evident in the months following the 2001 eruption until the end of that year (Figure 5). No subsequent magma input was observed until July 2002, except for a tail event during January and February 2002 that was recorded only at the Fondachello site. New magma inputs from about 10-km deep to about 2-km deep occurred from July to October 2002 (Figure 5). It is noteworthy that these migrations were recorded from a few months to a few days before the onset of the 2002–2003 eruption, suggesting a causeand-effect relation related to the high eruptive potential of the volcano. During this eruptive period, magma ascent was only detected at the Fondachello site at the end of the eruption. Similarly to the July 2001 eruption [Caracausi et al., 2003b], the 2002–2003 eruption also lasted only a few months, most likely because (after the onset) limited magma inputs occurred at shallow depth until January 2003 (Figure 4). Few episodes of magma ascent have been identified since the end of the eruption, and only during the period from May to December 2003. Particularly, and consistent with inferences from measurements of He isotopes, two magma migrations were recorded at the Stadio site, indicating a magma batch rising from 12 km b.s.l.

Notes to Table 2. a Pressures were computed following the method proposed by Caracausi et al. [2003b], which uses each covariation of He/Ne and He/CO2 ratio into a proper degassing model for the Etnean basalt. Isothermal, open system degassing was considered, starting from a CO2-dominated (CO2 85 – 95 mol%, H2O 15 – 5 mol%) gas phase at 700 MPa and 1200°C [Caracausi et al., 2003b]. Errors due to uncertainties in the initial conditions are generally within ±15%. b Displayed values of restored DHe/CO2 have been computed after the restoring of pristine ratios corrected for dissolution process in shallow aquifers. For further details, see the text and also Caracausi et al. [2003b]. c Data from Caracausi et al. [2003b].

9 of 12


3

G


10.1029/2005GC001175

Figure 5. Assessment of the degassing pressures of magma feeding the monitored gas emissions. Naftia data are excluded because this location is fed by magma intruding into the lateral part of the plumbing system. Solid and open symbols represent the computed initial and final pressures, respectively. Red arrows, in example, indicate two episodes of magma migration. Two gray bars indicate the location of magma reservoirs identified by recent geophysical investigations [e.g., Murru et al., 1999; De Gori et al., 2005]. Yellow bars represent the main important phases of 3He/4He ratio (as R/Ra) variations. Further details about the calculation of degassing pressures are given by Caracausi et al. [2003b]. Vertical lines are as in Figure 2.

toward about 2 km b.s.l. (see arrows in Figure 5). No subsequent events were recorded until the onset of the 2004–2005 eruption (on 7 September 2004), when magma depressurization probably occurred from 12 to 6 km b.s.l. Magma ascent during this period probably did not directly trigger the effusion activity, as it was not accompanied by new inputs of magma at shallow levels. This possibly induced a destabilization of the plumbing system at shallow levels. Finally, in agreement with our He-isotope data, a new magma-migration episode was recorded between July and August 2005.

4.3. Comparison With Volcanological and Geophysical Evidence [20] The combination of our geochemical data with volcanological observations of eruptive activity in recent years at Mount Etna allow us to detect forerunner signals of magma migration toward the shallowest levels of the plumbing system. In particular, both isotope (3He/4He ratio) and chemical (He/Ne and He/CO2 ratios) data suggested that a new recharge phase of the shallowest levels of the plumbing system (at 3–5 km b.s.l.) occurred after the end of the 2001 eruption. This inference is strongly supported by geophysical data. In fact, Patane` et al. [2003] found that the seismicity rate remained high after the eruption and that a renewal area dilatation by means of geodetic data was

observed, which they interpreted as a new magma accumulation at shallow depths (3 to 5 km). The reactivation of summit craters which started in March 2002 [Andronico et al., 2005] may further support this interpretation. Our geochemical signals obtained from July to October 2002 suggest that new batches of volatile-rich magma migrated toward the surface, forecasting the onset of the 2002–2003 eruption of a few months (Figures 2 and 5). Even though it was preceded by very few geophysical precursors [Monaco et al., 2005], ground deformations [Aloisi et al., 2003; Patane` et al., 2005] and seismic activity [Gambino et al., 2004; Monaco et al., 2005] indicated that a continuous magma refilling occurred from February 2002. Nevertheless, the most evident geophysical signals were tilt data and seismic tremors recorded for a few hours before the onset of the eruption [Aloisi et al., 2003; Gambino et al., 2004; Patane` et al., 2005]. From May 2003 until the end of 2003, new phases of magma migrations toward 2 km depth b.s.l. have been identified from geochemical data (Figures 2 and 5). Only minor variations of the geochemical parameters were subsequently observed, while Patane` et al. [2005] recorded a recharging phase from June 2003 through GPS and seismic data, which they considered a possible cause of the last eruption. Nevertheless, several authors [e.g., Burton et al., 2005; 10 of 12


3

G


Corsaro and Miraglia, 2005] agree that the 2004– 2005 eruption was notable for the lack of clear signals associated with magmatic processes such as seismic swarms, volcanic tremors, and sustained magmatic gas emissions. More likely, it was triggered by the failure of the upper portion of the volcanic edifice under magmatic hydrostatic pressure in the conduits. Such pressure at a shallow depth might be induced by the magma still resident at shallow levels since the last eruption (2002– 2003). The absence of explosive activity at the vents is evidence against the intrusion of new magma from depth. In fact, the petrography and glass composition of samples collected during the effusion suggest that the magma feeding the eruption was stored at the shallow levels of the plumbing system during the 2000 and 2001 activity [Corsaro and Miraglia, 2005]. This is strongly supported by our long-term monitoring of R/Ra average values, which has not revealed a new recharge of volatile-rich magma before the 2004– 2005 eruption (Figure 3), in contrast to what happened during the previous two eruptions. Our geochemical monitoring from June to August 2005 identified a new magma migration toward about 3 km b.s.l. This episode was followed by an increase of the seismic activity that was still ongoing as at 31 October 2005 (see http://www.ingv.it/ terremoti/terremoti.html for the latest status). We believe that Mount Etna could rapidly evolve into new pre-eruptive conditions.

5. Conclusions [21] Geochemical monitoring of Mount Etna during the period 2002–2005 combined with previously developed degassing models of the Etnean system [Caracausi et al., 2003a, 2003b] has allowed us to identify several episodes of magma migrations inside the plumbing system from the deep reservoir toward the shallowest one. The main important evidence is summarized as follows: [22] 1. After the July 2001 eruption, several inputs of magma from deep levels of the plumbing system were recorded until the end of 2001. Subsequent new events of magma recharge were detected for a few months before the onset of the 2002–2003 eruption, in agreement with geophysical investigations that found a high seismicity just after the 2001 eruption and a progressive inflation of the volcanic edifice beginning from February 2002. We hypothesize that this phase of magma recharging induced a high pressure at shallow depths in the system and triggered the eruption.

10.1029/2005GC001175

[23] 2. Following the 2002–2003 eruption, new episodes of magma migrations toward the surface were observed between May and December 2003. Subsequent minor geochemical signals were recorded until the onset of the 2004–2005 eruption, when a small magma-ascent event occurred. We infer that the 2004–2005 eruption was probably due to an edifice collapse along the already fractured sector of the volcano induced by the still-high hydrostatic pressure of magma in the conduits, as also confirmed by other authors [Burton et al., 2005; Corsaro and Miraglia, 2005]. [24] 3. Long-term monitoring of R/Ra average values has allowed us to elucidate the main phases of magma recharging in the period 1997–2005. In particular, we infer that the 2001 and 2002–2003 eruptions were preceded by recharge of new volatile-rich magma, while the 2004–2005 eruption was not supplied by undegassed magma, as confirmed by the absence of explosive activity. The petrologic features of the emitted lava strongly support our inferences, considering that the magma erupted during the last eruption was a degassed portion of the magma that erupted during the summit activity in 2000 and 2001. [25] Finally, we point out that a new magma migration inside the feeding system was observed from June to August 2005, followed by an increase of the seismicity that was still ongoing as at 31 October 2005. We therefore consider that close attention should be paid to the recharging phase of the Etnean plumbing system, which is evolving toward new pre-eruptive conditions.

Acknowledgments [26] We wish to thank F. Salerno and M. Tantillo for technical support during the analyses in the gas chromatographic and noble gas spectrometric laboratories. We are also grateful to F. Grassa for his support during fieldwork. We finally thank Vincent Salters, David R. Hilton, Finlay Stuart, and an anonymous reviewer for providing significant suggestions that improved the clearness of the manuscript. English text was revised two times by English Science Editing.

References Acocella, V., B. Behncke, M. Neri, and S. D’Amico (2003), Link between major flank slip and 2002 – 2003 eruption at Mt. Etna (Italy), Geophys. Res. Lett., 30(24), 2286, doi:10.1029/2003GL018642. Aiuppa, A., M. Burton, F. Mure`, and S. Inguaggiato (2004a), Intercomparison of volcanic gas monitoring methodologies performed on Vulcano Island, Italy, Geophys. Res. Lett., 31, L02610, doi:10.1029/2003GL018651. 11 of 12


3

G


Aiuppa, A., C. Federico, G. Giudice, S. Gurrieri, A. Paonita, and M. Valenza (2004b), Plume chemistry provides insights into the mechanisms of sulfur and halogen degassing at basaltic volcanoes, Earth Planet. Sci. Lett., 222, 469 – 483. Aiuppa, A., S. Inguaggiato, A. J. S. McGonigle, M. O’Dwyer, C. Oppenheimer, M. J. Padgett, D. Rouwet, and M. Valenza (2005), H2S fluxes from Mt. Etna, Stromboli, and Vulcano (Italy) and implications for the sulfur budget at volcanoes, Geochim. Cosmochim. Acta, 69(7), 1861 – 1871. Allard, P., M. Burton, and F. Mure´ (2004), Spectroscopic evidence for a lava fountain driven by previously accumulated magmatic gases, Nature, 433, 407 – 410. Aloisi, M., A. Bonaccorso, S. Gambino, M. Mattia, and G. Puglisi (2003), Etna 2002 eruption imaged from continuous tilt and GPS data, Geophys. Res. Lett., 30(23), 2214, doi:10.1029/2003GL018896. Alparone, S., D. Andronico, L. Lodato, and T. Sgroi (2003), Relationship between tremor and volcanic activity during the Southeast Crater eruption on Mount Etna in early 2000, J. Geophys. Res., 108(B5), 2241, doi:10.1029/2002JB001866. Alparone, S., D. Andronico, S. Giammanco, and L. Lodato (2004), A multidisciplinary approach to detect active pathways for magma migration and eruption at Mt. Etna (Sicily, Italy) before the 2001 and 2002 – 2003 eruptions, J. Volcanol. Geotherm. Res., 136, 121 – 140. Andronico, D., et al. (2005), A multi-disciplinary study of the 2002 – 03 Etna eruption: Insight into a complex plumbing system, Bull. Volcanol., 67, 314 – 330. Bruno, N., T. Caltabiano, S. Giammanco, and R. Romano (2001), Degassing of SO2 and CO2 at Mount Etna (Sicily) as an indicator of pre-eruptive ascent and shallow emplacement of magma, J. Volcanol. Geotherm. Res., 110, 137 – 153. Burton, M. R., et al. (2005), Etna 2004 – 2005: An archetype for geodynamically-controlled effusive eruptions, Geophys. Res. Lett., 32, L09303, doi:10.1029/2005GL022527. Caracausi, A., R. Favara, S. Giammanco, F. Italiano, A. Paonita, G. Pecoraino, A. Rizzo, and P. M. Nuccio (2003a), Mount Etna: Geochemical signals of magma ascent and unusually extensive plumbing system, Geophys. Res. Lett., 30(2), 1057, doi:10.1029/2002GL015463. Caracausi, A., F. Italiano, A. Paonita, A. Rizzo, and P. M. Nuccio (2003b), Evidence of deep magma degassing and ascent by geochemistry of peripheral gas emissions at Mount Etna (Italy): Assessment of the magmatic reservoir pressure, J. Geophys. Res., 108(B10), 2463, doi:10.1029/ 2002JB002095. Chiarabba, C., P. De Gori, and D. Patane´ (2004), The Mount Etna plumbing system: The contribution of seismic tomography, in Mount Etna: Volcano Laboratory, Geophys. Monogr. Ser., vol. 143, edited by S. Calvari et al., pp. 191 – 204, Washington, D. C. Clocchiatti, R., M. Condomines, N. Gue´not, and J. C. Tanguy (2004), Magma changes at Mount Etna: The 2001 and 2002 – 2003 eruptions, Earth Planet. Sci. Lett., 226, 397 – 414. Corsaro, R. A., and L. Miraglia (2005), Dynamics of 2004 – 2005 Mt. Etna effusive eruption as inferred from petrologic monitoring, Geophys. Res. Lett., 32, L13302, doi:10.1029/ 2005GL022347.

10.1029/2005GC001175

D’Alessandro, W., S. De Gregorio, G. Dongarra`, S. Guerrieri, F. Parello, and B. Parisi (1997), Chemical and isotopic characterization of the gases of Mount Etna (Italy), J. Volcanol. Geotherm. Res., 78, 65 – 76. De Gori, P., C. Chiarabba, and D. Patane` (2005), Qp structure of Mount Etna: Constraints for the physics of the plumbing system, J. Geophys. Res., 110, B05303, doi:10.1029/ 2003JB002875. Gambino, S., A. Mostaccio, D. Patane`, L. Scarfı`, and A. Ursino (2004), High-precision locations of the microseismicity preceding the 2002 – 2003 Mt. Etna eruption, Geophys. Res. Lett., 31, L18604, doi:10.1029/2004GL020499. Giammanco, S., S. Inguaggiato, and M. Valenza (1998a), Soil and fumarole gases of Mount Etna: Geochemistry and relations with volcanic activity, J. Volcanol. Geotherm. Res., 81, 297 – 310. Giammanco, S., S. Gurrieri, and M. Valenza (1998b), Anomalous soil CO2 degassing in relation to faults and eruptive fissures on Mount Etna (Sicily, Italy), Bull. Volcanol., 60, 252 – 259. Hirn, A., R. Nicolich, J. Gallart, M. Laigle, and L. Cernobori (1997), Roots of Etna volcano in faults of great earthquakes, Earth Planet. Sci. Lett., 148, 171 – 191. Hoefs, J. (1987), Stable Isotope Geochemistry, 241 pp., Springer, New York. Monaco, C., S. Catalano, O. Cocina, G. De Guidi, C. Ferlito, S. Gresta, C. Musumeci, and L. Tortorici (2005), Tectonic control on the eruptive dynamics at Mt. Etna Volcano (Sicily) during the 2001 and the 2002 – 2003 eruptions, J. Volcanol. Geotherm. Res., 144, 211 – 233. Murru, M., C. Montuori, M. Wyss, and E. Privitera (1999), The locations of magma chambers at Mt. Etna, Italy, mapped by b-values, Geophys. Res. Lett., 16, 2553 – 2556. Neri, M., V. Acocella, B. Behncke, V. Maiolino, A. Ursino, and R. Velardita (2005), Contrasting triggering of the 2001 and 2002 – 2003 eruptions of Mount Etna (Italy), J. Volcanol. Geotherm. Res., 144, 235 – 255. Nuccio, P. M., and A. Paonita (2001), Magmatic degassing of multicomponent vapors and assessment of magma depth: Application to Vulcano Island (Italy), Earth Planet. Sci. Lett., 193, 467 – 481. Nuccio, P. M., and V. Valenza (1998), Magma degassing and geochemical detection of its ascent, Water-Rock Interaction, edited by G. B. Arehart and J. R. Hulston, pp. 475 – 478, A. A. Balkema, Brookfield, Vt. Patane`, D., P. De Gori, C. Chiarabba, and A. Bonaccorso (2003), Magma ascent and the pressurization of the Mount Etna’s volcanic system, Science, 299, 2061 – 2063. Patane`, D., M. Mattia, and M. Aloisi (2005), Shallow intrusive processes during 2002 – 2004 and current volcanic activity on Mt. Etna, Geophys. Res. Lett., 32, L06302, doi:10.1029/ 2004GL021773. Sano, Y., and H. Wakita (1985), Geographical distribution of 3 He/4He in Japan: Implications for arc tectonics and incipient magmatism, J. Geophys. Res., 90, 8729 – 8741. Sharp, A. D. L., P. M. Davis, and F. Gray (1980), A low velocity zone beneath Mt. Etna and magma storage, Nature, 287, 587 – 591.

12 of 12