Volcanic plume fingerprint in the groundwater of a persistently degassing basaltic volcano: Mt. Etna

Chemical Geology 433 (2016) 68–80

Contents lists available at ScienceDirect

Chemical Geology journal homepage: www.elsevier.com/locate/chemgeo

Volcanic plume fingerprint in the groundwater of a persistently degassing basaltic volcano: Mt. Etna M. Liotta ⁎, W. D'Alessandro, S. Bellomo, L. Brusca Istituto Nazionale di Geofisica e Vulcanologia (INGV), Sezione di Palermo, via La Malfa 153, 90146, Palermo, Italy

a r t i c l e

i n f o

Article history: Received 15 September 2015 Received in revised form 8 March 2016 Accepted 31 March 2016 Available online 11 April 2016 Keywords: Mt. Etna Groundwater Volcanic plume Chemical weathering

a b s t r a c t The chemical composition of the groundwater at Mt. Etna was investigated in order to determine the extent to which the persistent plume affects the chemical composition of circulating waters. Samples from 31 springs and wells were collected during June and July 2014 and analyzed for their chemical compositions. The content of dissolved elements derives from the bulk deposition (wet and dry deposition) at the recharge areas as well as from the weathering of volcanic rocks during the infiltration and transport of groundwater. In its early phase, the chemical weathering of volcanic rocks and ashes is promoted by the acid rain that characterizes the area and subsequently by the huge amount of deep magmatic carbon dioxide (CO2) coming up through the volcanic edifice and dissolving in the water. The high content of chlorine is mainly derived from interactions between the plume and rainwater, while the total alkalinity can be completely ascribed to the dissociation of carbonic acid (H2CO3) after the hydration of CO2. The relative contributions of plume-derived elements/ weathering and CO2-driven weathering has been computed for each element. In addition, the comparison between the chemical compositions of the bulk deposition and of groundwater provides a new understanding about the mobility of volatile elements. The proposed approach has revealed that the persistent plume strongly affects the chemical composition of groundwater at Mt. Etna and probably also at other volcanoes characterized by huge open-conduit degassing activity. © 2016 Elsevier B.V. All rights reserved.

1. Introduction The chemical composition of groundwater circulating in volcanic systems is often the result of gas-rock-water interaction processes occurring within the volcanic edifice. In closed-conduit volcanoes, gases exsolving from magma cannot freely escape towards the atmosphere and intensively interact with the surrounding water (including condensed volcanic steam) and rocks producing highly saline solutions (Giggenbach, 1987, 1988). Fumaroles, geysers, and thermal waters often characterize the hydrothermal systems developing in such volcanic systems. All these manifestations can provide useful information on the thermodynamic conditions of the hydrothermal system and are studied worldwide in order to understand how the hydrothermal systems work as well as to recognize the occurrence of unrest events (e.g. Todesco, 2009; Sorey et al., 2003; Villemant et al., 2005; Lowenstern et al., 2006; Chiodini et al., 2010; Crider et al., 2011; Moretti et al., 2013). When the hydrothermal system is confined within a small volume, it can be highly reactive to the input of deep magmatic fluids (Villemant et al., 2014). Conversely, in open-conduit volcanoes, most of exsolved gases from magma freely escape towards the ⁎ Corresponding author. E-mail address: [email protected] (M. Liotta).

http://dx.doi.org/10.1016/j.chemgeo.2016.03.032 0009-2541/© 2016 Elsevier B.V. All rights reserved.

atmosphere and only minor fractions are emitted as diffuse degassing from the soils and/or are dissolved in circulating groundwater (e.g. D'Alessandro et al. 1997a; Grassa et al., 2008). At Mt. Etna, groundwaters mainly circulate in the permeable volcanites that overlie impermeable terrains composed by allochthonous series of flyschs and by post-orogenic clayey sediments (Lentini, 1982). For these waters the influence of hydrothermal water-rock interaction was ruled out by Aiuppa et al. (2000, and reference therein). However, a deeper aquifer may be hosted by carbonate units, which are covered by the impermeable terrains. Chiodini et al. (1996) observed that about 20 km south-west of Mt. Etna craters, at the contact between volcanic and sedimentary formations, mud volcanoes discharge large quantities of gases, mainly CO2, small quantities of saline cold brines and mud. These cold brines come up from a deep hydrothermal system and are hydrologically separated from the shallow groundwaters having also a lot of dissolved CO2 of deep origin but showing no evidence of hydrothermal reactions that feed many springs in the same area (Chiodini et al., 1996). Based on noble gases and CO2 concentrations, Caracausi et al. (2003) computed that magmatic gases occurring in the mud volcanoes, exsolve at depths in the range 3-10 km below sea level. Furthermore, the summit area of Mt. Etna hosts a fumarole field that has been recently studied in order to define and model its chemical and isotopic compositions and their relationship

M. Liotta et al. / Chemical Geology 433 (2016) 68–80

with magma degassing (Martelli et al., 2008; Liotta et al., 2010; Liotta et al., 2012; Paonita et al., 2012; Rizzo et al., 2013). The fumarolized area covers few hundreds of square meters around the main craters and there are no further identified fumaroles in the remaining part of the volcanic edifice. Concerning carbon emission, Mt. Etna is one of the greatest emitter of volcanic CO2 to the atmosphere (Allard et al., 1991; Gerlach, 1991; D'Alessandro et al. 1997b, Rizzo et al. 2014) mainly through the summit craters and only about 10% being emitted as soil degassing through the flanks and as CO2 dissolved in groundwater (D'Alessandro et al. 1997b, Brusca et al. 2001). A persistent volcanic plume results in a large contribution of volcanogenic elements to the bulk deposition (Aiuppa et al., 2006; Calabrese et al., 2011). These emissions significantly impact on soils and vegetation (Notcutt and Davies, 1989; Bellomo et al., 2007; Floor et al., 2011; D'Alessandro et al., 2012; Martin et al., 2012; Calabrese et al., 2015; Calabrese and D'Alessandro, 2015; James et al., 2016). The volcanic plume consists of solid particles, acidic droplets, and gaseous species, of which gaseous species represent the most relevant source of volcanogenic elements under quiescent degassing conditions (Martin et al., 2008). The volcanic aerosols at Mt. Etna have been studied by Bergametti et al. (1984); Allen et al. (2006); Martin et al. (2008), and Calabrese et al. (2011), with large differences in the reported results, which probably reflect the use of different sampling techniques and/or the variability of the volcanic aerosol emission. During rainfall the acidic gases and aerosol interact rapidly with droplets lowering the pH of the rain. During this process the dissolution and dissociation of the most acidic gases, such as hydrogen chloride (HCl), hydrogen iodide (HI), and hydrogen bromide (HBr) relative to sulfur dioxide (SO2), carbon dioxide (CO2), and hydrogen fluoride (HF) is favored (Aiuppa et al., 2006). Therefore, rainwater falling on the downwind (eastern) area of the volcano is abundant in chlorine (Cl), and the sulfur (S)/Cl ratio of the bulk deposition differs markedly from that of the volcanic plume (Aiuppa et al., 2006) depending on the different dissociation constants of the acidic species. As a general rule, the most acidic species are enriched in rainwater. Despite this, fluoride (F−) is also abundant in the bulk deposition around Mt. Etna as the result of the interaction between HF and small ash particles. In fact, leachates of volcanic ash indicate that F− is easily adsorbed by ash, which results in its enrichment in leaching solutions (Bagnato et al., 2011). The S, Cl, and F are the most-enriched elements on the surface of ash particles compared to its bulk composition, and they are in the form of sulfate and halide salts (Delmelle et al., 2005, 2007; Jones and Gislason, 2008). This implies that the dry deposition of large amounts of ash can be responsible for the input of these elements to the hydrological cycle as a consequence of natural leaching of the ash by rainwater. Despite the clear evidence of a meteoric recharge that is strongly influenced by the plume, its contribution to circulating groundwater has not been adequately recognized previously (e.g., D'Alessandro et al., 2011). In this paper we demonstrate that the persistent volcanic plume at Mt. Etna represents the main source of Cl and bromine (Br) dissolved in cold groundwaters hosted in the permeable volcanites. We describe the effects of the plume on the chemical weathering of rocks and its element contribution that, coupled with CO2-driven weathering (CDW), are responsible for the chemical composition of the circulating groundwater. 2. Hydrogeological setting and climate The petrological evolution of the volcanic edifice indicates that after an initial phase characterized by subalkaline products, the magma evolved toward an alkaline composition (Corsaro and Pompilio, 2004). Most of the present edifice belongs to the “Mongibello” unit (Branca et al., 2011 and reference therein) composed of differentiated alkaline lavas. These basaltic rocks form the volcanic edifice that develops on an impermeable sedimentary basement. Due to this peculiar structure, Mt. Etna is the largest aquifer in Sicily, and hence represents an

69

important hydrological source. The volcanic edifice covers an area of about 1,200 km2, and its complex geological evolution is reflected in the structure of the aquifer (Ferrara and Pappalardo, 2008). The whole volcanic edifice can been divided in the three main hydrological structures corresponding to the northern, eastern, and southwestern flanks of Mt. Etna and further subdivided into secondary hydrological subunits (Ferrara and Pappalardo, 2008). The morphology of Mt Etna determines a very characteristic climate relative to the surrounding areas. Favalli et al. (2004) observed that during the night, downslope surface winds force the plume to follow the steep morphology, whereas during the day, very frequent NNW synoptic winds impact on the cone and are contrasted by strong SE sea breezes and anabatic winds, with the consequent formation of convective ascending currents. The persistent plume further modifies the local climate. The impact of volcanic activity on meteorological conditions was highlighted by Durbin and Henderson-Sellers (1981) and Chester and Duncan (1985), who described how sea breezes, anabatic winds, and the volcanic plume favor the formation of cumulus clouds within and around “Valle del Bove” (on the eastern flank). These peculiar features are also reflected in the vegetation cover, with the eastern flank characterized by acidophilic species (Bazan et al., 2010). As a result of such climatic features, the rainfall is highest on the eastern flank of Mt. Etna at about 800 m, which represents the main recharge area for this sector of the volcano (D'Alessandro et al., 2004). The isohyet map of precipitation, which is publicly available at http://www. osservatorioacque.it/?cmd=page&id=dati_elab_cartepioggia&tpl= default, clearly indicates that the yearly precipitation on the eastern flank is two or three times higher than that on the western flank, and hence represents an area with a large amount of precipitable water (Zhu et al., 2008). Under such conditions, clouds could be characterized by high deuterium-excess values, with the Mediterranean Sea being a classic example of a reservoir producing water vapor characterized by a high deuterium excess (Gat and Carmi, 1970; Gat et al., 2003). This also results in infiltrating water maintaining such an isotopic fingerprint (Liotta et al., 2013). The isotopic compositions of oxygen and deuterium have allowed identification of the meteoric origin of circulating waters (D'Alessandro et al., 2004; Liotta et al., 2013). Another important source of meteoric water that infiltrates toward the aquifers is the snow cover. At Mt. Etna this represents a large amount of solid water that melts slowly, since the top of the volcano is covered by snow from autumn until spring. 3. Sampling and analytical methods Groundwater samples were collected from springs and wells along the eastern and southern flanks of Mt. Etna (Fig. 1). Almost all sampling points belong to three different hydrological subunits, two on the eastern flank and one on the southwestern one. All samples have been collected in two sampling campaigns in June and July 2014. The summer period was chosen because in that period wells are permanently pumped. All sampling sites have been previously sampled and analyzed for major ions and water isotopes for longer periods. Thirteen sampling sites at present belong to the monthly sampling program for volcanic surveillance of INGV-Pa and further 4 belonged to it in the past while the remaining were sampled quarterly for three years (INGV database). Almost all do not show seasonal variations and instead show long-term trends generally affecting the total solute content and not the relative ion ratios (D'Alessandro et al., 2011). All of the sampled groundwater circulates in basaltic rocks. The temperature, conductivity, pH, and redox potential (Eh) were measured in the field using portable instruments: a conductivity meter (Orion 150A +, Thermo Scientific) equipped with an integrated temperature sensor electrode, and a pH meter (Orion Star A121, Thermo Scientific) equipped with a Hamilton Mecotrode pH electrode and with a Hamilton Oxitrode Pt 120 electrode. The calibration routines for pH (involving the use of pH 4 and 7 buffer solutions) and conductivity were performed each day before sampling.

70

M. Liotta et al. / Chemical Geology 433 (2016) 68–80

Table 1 Chemical composition of groundwater. The concentrations are expressed in mmol/L except alkalinity which is expressed in meq/L. SAMPLE

ID

pH

MUSMECI FARO A. ROSSA SOLICCHIATA VALCORRENTE ROMITO S26 S65 CHERUBINO S. LEONARDELLO A .GRASSA PRIMOTI BAGLIO ERCOLINO MASARACCHIO ARDIZZONE ELLERA RAFFO 1 RAFFO 2 SERAFICA S. VITO S59 P5 DIFESA PEDARA ROCCA CAMPANA S. MARIA PIANO ELISI MURI ANTICHI RANIERI S. GIACOMO P31 ILICE

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

6.99 7.08 6.52 7.02 6.72 6.80 6.64 6.89 6.82 6.08 6.41 6.43 6.28 7.46 6.50 7.19 6.20 6.25 6.17 6.27 6.29 6.06 7.27 7.04 6.67 6.24 7.09 7.12 6.50 6.03 6.24

COND (μS/cm)

Eh (mV)

1210 1125 1524 1534 1301 1230 1854 1010 1069 1570 1790 1297 1050 703 1031 753 1140 1041 1136 981 910 951 401 582 696 812 422 459 565 738 331

77 0 170 42 169 −83 180 101 28 13 165 46 95 120 121 135 163 190 167 130 187 190 380 163 350 450 630 125 95 280 178

The total alkalinity was determined after sampling by titration with 0.1 M HCl. The samples were analyzed at Istituto Nazionale di Geofisica e Vulcanologia (Palermo Department) for major anions and cations using an ion chromatography system (1100 IC, Dionex) in suppressed mode and equipped with an anion column (AS14A) and a precolumn (AG14A) that works under continuous flow of carbonate–bicarbonate eluent, and a cation column (CS12A) and precolumn (CG12A) that works under continuous flow of methanesulfonic acid with eluent regeneration. Samples for analyzing the contents of trace and minor elements were collected in low density polyetilene bottles (Nalgene) that had previously been rinsed with 5% nitric acid (HNO 3) solution in the laboratory and then washed several times with the same water to be sampled before filling the bottle in the field. Samples were filtered with 0.45-μm-pore filters and acidified with ultrapure HNO3 (65%). The silicon (Si) content in aqueous solutions was determined by inductively coupled plasma optical emission spectrometry (ICPOES Ultima2, Jobin Yvon) at wavelength of 251.611 nm with the quantitative method using calibration solutions in the range 0.5–50 mg/L and by calculating the weighted regression curve constructed with 10 calibration points. Three replicate measurements were performed for each sample and the calibration standard, which indicated that the precision was always within ± 5%. Reference materials (SPSSW-1 and SPSSW-2, Spectrapure Standards) were analyzed during the analysis session to evaluate the accuracy, which was also always within ± 5%. The contents of lithium, boron (B), aluminum (Al), selenium (Se), Br, and iodine (I) were determined by ICP mass spectrometry (ICP-MS 7500ce, Agilent) with the quantitative method using calibration solutions in the range 0–100 μg/L and by calculating the weighted regression curve constructed with 12 calibration points. Five replicates were performed for each sample and the calibration standard, which indicated that the precision was always within ± 10%. Different reference materials (TM-24.3 and TM-61.2, Environment Canada; SLRS-5, National

Temp (°C)

Na

K

Mg

Ca

18.0 18.2 18.2 17.4 16.9 15.5 17.7 13.9 17.9 19.3 19.4 18.4 18.0 17.5 15.9 18.4 16.5 15.7 18.7 16.8 14.7 16.2 15.8 12.7 15.1 17.6 17.9 15.6 11.4 16.9 10.9

5.66 × 100 3.79 × 100 7.06 × 100 1.07 × 101 5.68 × 100 6.53 × 100 1.05 × 101 4.68 × 100 3.82 × 100 7.90 × 100 1.19 × 101 5.76 × 100 4.67 × 100 4.75 × 100 4.91 × 100 4.52 × 100 4.60 × 100 4.19 × 100 4.59 × 100 4.83 × 100 4.56 × 100 3.92 × 100 2.44 × 100 3.52 × 100 3.27 × 100 3.86 × 100 1.92 × 100 2.88 × 100 3.52 × 100 3.50 × 100 1.45 × 100

7.10 × 10−1 5.79 × 10−1 4.50 × 10−1 7.26 × 10−1 8.29 × 10−1 5.32 × 10−1 1.16 × 100 5.49 × 10−1 6.15 × 10−1 4.87 × 10−1 1.24 × 100 7.58 × 10−1 5.59 × 10−1 3.20 × 10−1 4.92 × 10−1 4.33 × 10−1 4.94 × 10−1 4.89 × 10−1 6.71 × 10−1 5.11 × 10−1 4.97 × 10−1 4.67 × 10−1 2.65 × 10−1 3.38 × 10−1 3.79 × 10−1 6.24 × 10−1 2.06 × 10−1 2.47 × 10−1 3.38 × 10−1 5.48 × 10−1 2.36 × 10−1

3.99 × 100 2.93 × 100 6.29 × 100 5.45 × 100 4.76 × 100 4.56 × 100 4.40 × 100 3.65 × 100 2.81 × 100 5.99 × 100 4.78 × 100 3.34 × 100 3.92 × 100 1.46 × 100 2.96 × 100 1.85 × 100 3.58 × 100 3.17 × 100 3.79 × 100 2.76 × 100 2.91 × 100 3.32 × 100 8.05 × 10−1 1.30 × 100 1.70 × 100 2.61 × 100 1.27 × 100 8.30 × 10−1 1.30 × 100 1.85 × 100 6.95 × 10−1

1.87 × 100 2.45 × 100 2.96 × 100 8.27 × 10−1 1.98 × 100 1.38 × 100 2.11 × 100 1.15 × 100 1.96 × 100 3.21 × 100 2.21 × 100 2.01 × 100 1.87 × 100 3.21 × 10−1 1.29 × 100 5.30 × 10−1 1.91 × 100 1.63 × 100 1.49 × 100 1.31 × 100 1.23 × 100 1.88 × 100 3.26 × 10−1 4.03 × 10−1 6.35 × 10−1 9.85 × 10−1 2.97 × 10−1 2.65 × 10−1 3.92 × 10−1 1.09 × 100 4.94 × 10−1

Research Council Canada; and SPSSW-1 and SPSSW-2, Spectrapure Standards) were analyzed during the analysis session to evaluate the accuracy, which was always within ± 15%. The I was determined during a different analysis session in which 0.5% ammonia solution was used to rinse the ICP-MS introduction system between samples to minimize memory effects (Al-Ammar et al., 2001), whereas 2% HNO3 was used for all of the other elements. The Cl and F were determined using ion chromatography to detect chloride (Cl−) and F−. Considering that Cl− and F− represent almost the total Cl and F contents in the compositional range of the sampled waters, we use the latter naming convention below except when referring to anionic species. The Br and I were determined on ICP-MS as total Br and total I. 4. Results The analytical results are given in Table 1. All of the samples are characterized by high alkalinity, ranging between 45% and 94% of the total anion equivalents. After alkalinity, chloride and sulfate are the most abundant anions. The Fig. 2 shows that Cl is very strongly correlated with Br. The other anions are only weakly correlated with Cl, even if the samples with the highest Cl contents also show high contents of − SO− 4 and F (Table 1). The dissolved cations in all of the samples show often the same relative contributions: sodium ions (Na+) N magnesium ions (Mg2 +) N calcium ions (Ca2 +) N potassium ions (K+). The pH values range between 6.03 and 7.46. The redox potential (Eh) indicates very different redox conditions, from reducing (Eh = −83 mV) at the Romito site to oxidizing at the Faro site (Eh = 630 mV). The groundwater temperature ranges between 10.9°C and 19.4°C, with a mean value of 16.6°C. The samples from Ilice and S. Giacomo exhibit the lowest temperatures (10.9°C and 11.4°C, respectively), and are significantly outside the range defined by the standard deviation (± 1σ). These two samples also showed the highest temporal variability of the oxygen-isotope composition, since they belong to shallower perched aquifers with fast-acting hydrological circuits (Ferrara, 1975), and

M. Liotta et al. / Chemical Geology 433 (2016) 68–80

71

Table 1 Chemical composition of groundwater. The concentrations are expressed in mmol/L except alkalinity which is expressed in meq/L. Cl

Alkalinity

Br

NO3

SO4

I

Li

B

Al

Si

Se

2.96 × 100 1.80 × 100 1.48 × 100 3.91 × 100 2.82 × 100 2.03 × 100 5.74 × 100 1.30 × 100 1.88 × 100 2.21 × 100 6.32 × 100 2.36 × 100 1.15 × 100 2.22 × 100 1.34 × 100 2.93 × 100 1.06 × 100 1.04 × 100 1.98 × 100 1.37 × 100 1.00 × 100 9.64 × 10−1 8.51 × 10−1 9.69 × 10−1 1.92 × 100 1.86 × 100 5.70 × 10−1 1.36 × 100 3.03 × 10−1 1.50 × 100 3.11 × 10−1

1.28 × 101 7.10 × 100 2.27 × 101 1.82 × 101 1.45 × 101 1.61 × 101 1.28 × 101 1.18 × 101 6.20 × 100 2.31 × 101 1.34 × 101 1.01 × 101 1.38 × 101 4.40 × 100 1.17 × 101 4.70 × 100 1.38 × 101 1.18 × 101 1.29 × 101 1.11 × 101 1.17 × 101 1.22 × 101 3.20 × 100 4.80 × 100 5.40 × 100 8.30 × 100 3.20 × 100 2.60 × 100 6.80 × 100 6.40 × 100 3.50 × 100

5.51 × 10−3 3.44 × 10−3 3.05 × 10−3 8.62 × 10−3 5.38 × 10−3 3.80 × 10−3 1.10 × 10−2 2.44 × 10−3 3.52 × 10−3 4.32 × 10−3 1.13 × 10−2 4.34 × 10−3 2.46 × 10−3 4.73 × 10−3 2.66 × 10−3 6.16 × 10−3 2.44 × 10−3 2.40 × 10−3 4.90 × 10−3 2.76 × 10−3 1.83 × 10−3 1.82 × 10−3 1.74 × 10−3 1.90 × 10−3 3.81 × 10−3 4.06 × 10−3 1.04 × 10−3 2.51 × 10−3 5.81 × 10−4 2.77 × 10−3 4.94 × 10−4

6.11 × 10−1 1.25 × 100 1.38 × 10−1 7.59 × 10−2 5.34 × 10−1 2.50 × 10−2 9.84 × 10−2 2.75 × 10−1 1.19 × 100 2.48 × 10−2 9.28 × 10−2 5.92 × 10−1 1.90 × 10−1 2.36 × 10−1 1.57 × 10−1 5.22 × 10−1 1.79 × 10−1 1.95 × 10−1 9.71 × 10−2 1.73 × 10−1 1.64 × 10−1 1.49 × 10−1 1.34 × 10−1 1.27 × 10−2 4.67 × 10−1 1.32 × 10−1 9.95 × 10−2 4.59 × 10−1 7.70 × 10−3 2.52 × 10−1 3.50 × 10−2

6.80 × 10−1 2.36 × 100 5.42 × 10−1 5.41 × 10−1 7.92 × 10−1 1.04 × 10−1 2.87 × 100 5.49 × 10−1 2.21 × 100 3.54 × 10−1 3.29 × 100 1.96 × 100 4.98 × 10−1 8.48 × 10−1 2.91 × 10−1 6.65 × 10−1 5.19 × 10−1 5.49 × 10−1 3.68 × 10−1 4.07 × 10−1 2.20 × 10−1 5.62 × 10−1 3.50 × 10−1 8.23 × 10−1 4.10 × 10−1 6.30 × 10−1 7.20 × 10−1 4.64 × 10−1 7.45 × 10−2 8.05 × 10−1 1.63 × 10−1

2.30 × 10−4 4.18 × 10−5 1.34 × 10−4 3.86 × 10−4 1.05 × 10−4 2.43 × 10−4 5.48 × 10−5 5.96 × 10−5 1.43 × 10−5 2.72 × 10−4 6.24 × 10−5 1.56 × 10−5 5.55 × 10−5 1.91 × 10−4 1.13 × 10−4 2.33 × 10−4 5.05 × 10−5 4.63 × 10−5 6.65 × 10−5 1.07 × 10−4 6.82 × 10−5 4.04 × 10−5 4.70 × 10−5 3.56 × 10−5 1.73 × 10−4 3.15 × 10−5 8.96 × 10−6 9.28 × 10−5 4.09 × 10−6 1.64 × 10−5 1.43 × 10−5

3.74 × 10−3 3.21 × 10−4 9.97 × 10−3 1.19 × 10−2 4.12 × 10−3 3.52 × 10−3 6.56 × 10−3 1.09 × 10−3 2.66 × 10−4 1.46 × 10−2 7.33 × 10−3 6.43 × 10−4 3.61 × 10−3 6.24 × 10−3 4.52 × 10−3 3.59 × 10−3 4.69 × 10−3 3.48 × 10−3 1.77 × 10−3 4.10 × 10−3 3.04 × 10−3 2.78 × 10−3 1.97 × 10−4 9.37 × 10−4 2.24 × 10−3 1.13 × 10−3 3.52 × 10−3 2.67 × 10−3 1.77 × 10−3 2.52 × 10−3 7.23 × 10−4

1.03 × 10−1 1.18 × 10−2 1.27 × 10−1 8.40 × 10−2 1.27 × 10−1 4.63 × 10−2 5.21 × 10−2 2.36 × 10−2 1.16 × 10−2 1.01 × 10−1 6.05 × 10−2 2.06 × 10−2 5.08 × 10−2 5.60 × 10−2 4.52 × 10−2 4.84 × 10−2 4.44 × 10−2 3.48 × 10−2 7.50 × 10−2 4.92 × 10−2 3.01 × 10−2 2.60 × 10−2 5.88 × 10−3 1.82 × 10−2 2.98 × 10−2 2.97 × 10−2 8.19 × 10−3 2.78 × 10−2 2.56 × 10−3 1.55 × 10−2 5.05 × 10−3

2.39 × 10−5 3.71 × 10−5 1.49 × 10−5 4.94 × 10−5 1.73 × 10−5 1.15 × 10−5 9.55 × 10−5 6.92 × 10−6 2.51 × 10−5 1.30 × 10−4 6.94 × 10−5 2.66 × 10−5 3.20 × 10−5 7.29 × 10−5 4.19 × 10−5 4.28 × 10−5 7.49 × 10−5 5.15 × 10−5 9.56 × 10−5 8.48 × 10−5 9.03 × 10−5 1.98 × 10−4 3.04 × 10−5 2.12 × 10−5 3.59 × 10−5 1.62 × 10−5 1.81 × 10−5 5.24 × 10−5 8.60 × 10−5 1.21 × 10−4 1.91 × 10−4

1.05 × 100 8.44 × 10−1 1.46 × 100 1.26 × 100 1.27 × 100 1.34 × 100 1.12 × 100 1.20 × 100 1.11 × 100 1.70 × 100 1.13 × 100 1.16 × 100 1.37 × 100 8.36 × 10−1 1.39 × 100 8.89 × 10−1 1.50 × 100 1.46 × 100 1.48 × 100 1.40 × 100 1.29 × 100 1.42 × 100 7.11 × 10−1 7.08 × 10−1 1.12 × 100 1.22 × 100 8.25 × 10−1 8.63 × 10−1 1.56 × 100 1.25 × 100 1.06 × 100

7.73 × 10−6 1.66 × 10−5 1.72 × 10−5 2.32 × 10−6 9.44 × 10−6 5.46 × 10−7 8.64 × 10−5 4.17 × 10−6 2.96 × 10−5 7.39 × 10−7 8.45 × 10−5 2.95 × 10−5 8.25 × 10−6 3.19 × 10−6 8.65 × 10−6 6.85 × 10−6 9.35 × 10−6 1.12 × 10−5 1.16 × 10−5 8.98 × 10−6 6.03 × 10−6 6.42 × 10−6 1.68 × 10−6 5.83 × 10−6 6.03 × 10−6 3.28 × 10−5 6.33 × 10−6 5.52 × 10−6 1.01 × 10−6 1.82 × 10−5 1.25 × 10−5

directly reveal compositional variations in meteoric input (D'Alessandro et al., 2004). The I concentrations range between 4 × 10−6 and 4 × 10−4 mmol/L, with a mean value of 1 × 10− 4 mmol/L. While the mean value falls within the range determined by Fehn (2012) (from 7 × 10− 5 to 2 × 10−4 mmol/L) for freshwater, several samples exhibit concentrations that are significantly lower. The sample from Solicchiata exhibits the highest I content and has a peculiar composition relative to all of the other samples. It has high Na and Cl contents but an intermediate concentration of S. 5. Discussion 5.1. Plume-derived Cl and Br The Cl and Br are usually considered conservative elements, and hence are often used as tracers of mixing processes involving seawater (e.g., Fehn and Snyder, 2003). The chemical composition of rain at Mt. Etna is strongly influenced by the volcanic plume. The upwind areas are characterized by an amount-weighted composition of about 0.14 mmol/L Cl (data from Aiuppa et al., 2001), and similar values have been observed in the rainwater at Mt. Vesuvius (Madonia and Liotta, 2010). For these sites, the most reasonable source of Cl is the sea aerosol. In contrast, the amount-weighted composition of rainwater in the summit area of Mt. Etna is indicative of a Cl concentration of 3.1–4.4 mmol/L (data from Aiuppa et al., 2001). Calabrese et al. (2011) measured up to 12 mmol/L Cl in rain samples from the summit area. In addition, since the closest area downwind of the volcanic edifice, Valle del Bove, is on the eastern flank (which is the most strongly influenced by the plume) and cannot be easily reached, precipitation samples have never been collected there. Therefore, previous investigations of the chemical composition of precipitation at Mt. Etna (Aiuppa et al., 2006; Calabrese et al., 2011) have not determined the composition of the rainwater that is the most likely to be influenced by the plume so that values they give could be underestimated. Anyhow, the available

data suggest that meteoric waters feeding the aquifer contain large amounts of plume-derived Cl. The strong positive correlation between Cl and Br in groundwater (Fig. 2) confirms very similar mobilities of these two halogens. Such a correlation can be derived from a mixing between two end-members or from different dilution levels of a constant source. The obtained Cl/ Br ratios are not far from that of seawater and this could induce to consider seawater as a possible mixing end-member. But a direct seawater intrusion can be ruled out for our sampling sites due to the large amount of precipitation and to the piezometric surface characterized by a high hydraulic gradient (Ferrara and Pappalardo, 2008). Based on the isotopic composition of circulating water, also D'Alessandro et al. (2004) excluded a possible seawater contribution. The possible contribution of sea aerosol can be estimated from Fig. 2. The bulk composition of sites not affected by the plume indicate that possible sea aerosol contribution of Cl just accounts for few percent of total dissolved Cl of groundwater. In addition, the average Cl/Br molar ratio differs significantly from that of seawater (Fig. 2). Another possible source of Cl and Br could be the “Salinelle” brines coming up from the sedimentary layers beneath the Etnean volcanic products in the southern flank of the volcano (Chiodini et al., 1996). Nevertheless, the Cl/Br molar ratio of “Salinelle” brines is higher than that of seawater [Cl/BrN 760 (Parello et al., 2001) versus about 650 in seawater (Davis et al., 1998)] and their contribution can also be ruled out. The most obvious and realistic source of Cl and Br is the persistent volcanic plume that strongly affects the bulk deposition. The Cl/Br molar ratio observed in groundwater is consistent with the range measured by Wittmer et al. (2014) in the volcanic plume of Mt. Etna (400bCl/Br b 700) while differs from previous measurement of Aiuppa et al. (2005) (Cl/Br≈1030; molar ratio). Different Cl/Br molar ratio observed in the volcanic plume can be ascribed to the different sampling methods used by Aiuppa et al. (2005) and Wittmer et al. (2014) as well as on changes in volcanic activity. Since Wittmer et al. (2014) developed a very efficient active alkaline trap to determine acidic gas ratios in volcanic plumes and three different methods have been used and compared, we are more confident in the results of Wittmer et al. (2014).

72

M. Liotta et al. / Chemical Geology 433 (2016) 68–80

Groundwater and the plume thus exhibit similar Cl/Br molar ratios. This is reasonable when taking into account that Cl and Br are mainly emitted as halogen acids during degassing (Villemant et al., 2005). The HCl and HBr are both strong acids that readily dissolve when interacting with hydrometeors forming around Mt. Etna. This implies that during degassing and/or during in-plume processes, Cl and Br also exhibit the same behavior, or at least the processes that could selectively involve only one of these halogens (e.g BrO formation) has a negligible effect on the total contribution to groundwater. Concerning their solubility in magmas, Schilling et al. (1980) observed that Cl and Br are highly incompatible and show very similar behaviors. Accordingly, Wang et al. (2014) suggested that Cl and Br are not strongly fractionated from each other during magmatic processes such as partial mantle melting, fractional crystallization, and degassing. Interestingly, at La Soufrière volcano (Guadeloupe, Lesser Antilles; closed conduit volcano), where the halogen contents of thermal waters are interpreted as a retarded record of magma degassing pulses dispersed into the hydrothermal system, the Cl/Br molar ratio ranges between 560-1130 (Villemant et al., 2005) (the authors gave mass ratio 250-500; values inferred from Fig. 3b in their paper). Although Mt. Etna and La Soufrière have different magmatism (basaltic and andesitic respectively), the derived Cl/Br molar ratios partially overlap. 5.2. Water–rock interaction Basaltic rocks are formed under thermodynamic conditions very different from posteruptive ones. As a consequence, many rock-forming primary minerals and amorphous glasses are unstable at surface conditions of pressure, temperature, and oxygen fraction, and react to form more-stable secondary minerals (e.g., metal oxides and clay minerals). The chemical weathering of basaltic rocks is the main underlying process. The weathering rate of the main constituents of basaltic rock is well known: glass ≈ olivine N plagioclase N pyroxene (Eggleton et al., 1987; Gislason et al., 1996). According to the Brønsted's concept of acids and bases, basaltic rocks can be considered bases since they are proton acceptors (Gislason and Eugster, 1987). In fact, protons are consumed and cations are released during chemical weathering. At Mt. Etna the weathering of volcanic rocks starts when acidic bulk deposition interacts with outcropping rocks and ashes. When protons coming from the dissociation of volcanic acidic gases have been neutralized by their interaction with volcanic rocks, the weathering processes are further promoted by magmatic CO2 coming up through the volcanic edifice and dissolving in the aquifers hosted in active volcanic edifices. After CO2 hydration, dissociation of carbonic acid provides the protons necessary for the chemical weathering of rocks (i.e., CDW). These two different weathering steps occur under very different conditions. The former occurs in a highly acidic environment (pH b 4) and the reaction rates depend strongly on the pH, while the latter usually occurs under slightly acidic conditions since the pH has been already neutralized by the interaction with volcanic rocks. The CDW increases the alkalinity of groundwater. The alkalinity is mainly determined by the amount of dissolved HCO− 3 (Berner and Rao, 1997). According to Stumm and Morgan (1996), alkalinity is the difference between the concentration of total conservative cations and total anions derived from strong acids. Consequently, alkalinity can also be expressed by the equivalent of protons needed to titrate all anions of weak acids to their respective acids. Due to the huge amount of CO2 degassing from soil at Mt. Etna (Camarda et al., 2012), HCO− 3 represents the dominant alkalinity component, and the pH values do not allow significant dissociation of H 4SiO 4 (pKa = 9.8) that could provide H3SiO− 4 as a further possible component contributing to the total alkalinity. Several reactions have been proposed to describe mineral and glass alterations in acid environments (e.g., Berner and Rao, 1997; Gislason

and Oelkers, 2003; Gudbrandsson et al., 2014; Chemtob and Rossman, 2014). Basalt weathering produces dissolved Ca, Mg, Na, K, iron (Fe), Al, and Si. In oxidizing environments, Fe and Al rapidly precipitate as oxides and hydroxides, while Si tends to form amorphous silica regardless of redox conditions (Chemtob and Rossman, 2014; Chemtob et al., 2015). These secondary phases do not involve carbon. The Ca and Mg can also precipitate as carbonates; in this case, 1 mol of carbon is also removed for each mole of Ca and Mg removed, thus keeping the ratio between alkalinity and Ca+Mg constant. The progressive precipitation of alkaline earth elements could produce a relative enrichment in alkalis. Especially in high-pH environments exposed to CO2, the precipitation of carbonate phases can remove most alkaline earth metals. In contrast, Na is considered the most mobile major element under low temperature CDW (Gysi and Stefánsson, 2012). Consequently, it is usually used as a reference for normalizing the relative mobility of elements in basalt aquifers (e.g. Aiuppa et al., 2000). However, this can only be done if atmospheric soluble salts and/or plume-derived elements (PDE) do not contribute significantly to the total amount of Na dissolved in water. The rainwater at Mt. Etna is characterized by large amounts of Na mainly due to the plume contribution (Calabrese et al., 2011), so the above-described common approach could lead to incorrect interpretations of results. Saturation indexes for groundwater at Mt. Etna computed using the PHREEQC software package (Parkhurst and Appelo, 1999) indicate that the samples from Faro, Valcorrente, Musmeci, A. Rossa, and Solicchiata are oversaturated with respect to dolomite and close to saturation with respect to calcite and aragonite. This indicates that some secondary carbonate minerals can also form. In fact, calcite, aragonite, and dolomite have been described in travertine outcropping in the southern part of Mt. Etna (D'Alessandro et al., 2007). When carbonates precipitate, the amount of cation equivalents removed corresponds to the equivalent alkalinity consumed, thus keeping constant the ratio between CDW elements and alkalinity. Since the weathering of rocks starts when acid rain infiltrates, it is desirable to distinguish between CDW and plume-derived weathering (PDW). As shown in Fig. 3, most of the samples exhibit an excess of cations that are not derived from CDW. Most of the samples deviate from the 1:1 line in Fig. 3, especially those from the eastern flank. This implies that CDW cannot be the only source of metals dissolved in groundwater. Instead, they can also derive from the chemical weathering of rocks by acid rain, that is PDW, and they can already be present as dissolved ions in the bulk deposition before water–rock interaction occurs. Further sources also cannot be excluded, such as anthropogenic input and sea aerosol. The Fig. 4 shows the concentration of K, Na, Mg, Ca, Si, Al, B, and Se versus the concentration of Cl and the total alkalinity. While Mg is very strongly correlated with alkalinity, many other elements such as Na, K, and B seem to be correlated to different extents with both Cl and alkalinity. 5.3. CDW versus PDW+PDE After we had determined that the main source of Cl is the plume and that the alkalinity is mainly controlled by CDW, we wanted to distinguish the relative contributions of CDW and PDW for each element. This can be achieved by simultaneously comparing the concentration of each element with that of Cl and the alkalinity. However, acid rain also transports large amounts of PDE. As a consequence, it is more accurate to examine the relative contributions of CDW and PDW+PDE. In order to quantify the relative contribution of each process/source, we performed multiple regression with respect to Cl and alkalinity, being representative of PDW + PDE and CDW, respectively. The results for the descriptive statistics are given in Table 2. Na, Mg, and Br show the highest coefficients of determination, as do the sum of all major cations. The K, Ca, and B exhibit lower coefficients of determination (0.78, 0.59, and 0.62, respectively), but they are still high enough to indicate that

M. Liotta et al. / Chemical Geology 433 (2016) 68–80

73

Fig. 1. On the left: Simplified geo-hydrological map of Mt. Etna and wind-rose inset. Wind-rose data are from the European Centre for Medium-Range Weather Forecasts (period 2007–2009), geological map modified from Branca et al. (2011), and main drainage axes (blue lines) are from Ferrara and Pappalardo (2008). The sampling sites and the identification numbers (Id as in Tab. 1) are plotted. On the right: schematic geological sketch of the area modified from Branca et al. (2011). Cold groundwater is confined within the volcanic edifice due to the impermeable sedimentary basement. Possible hydrothermal fluids circulate at depth but Aiuppa et al. (2000) ruled out any contribution to the cold groundwater.

PDW+PDE and CDW are the main sources of these elements. In contrast, F, I, SO42−, and nitrate (NO− 3 ) are not reliably correlated with either Cl or alkalinity. This implies that other sources and/or processes contribute and/or modify their abundances in groundwater. Focusing on those elements that exhibit the highest coefficients of determination, we can attempt to compute the relative contributions of PDW+PDE and CDW for each element. Since such a computation is based on the values predicted by multiple regression, we also consider residuals since they indicate how far the observed values are from the predicted ones. Table 3 reports the relative fractions of PDW + PDE and CDW and the residuals for Na, K, Ca, Mg, Br, and B. Of course, such

Fig. 2. Binary diagram of Cl versus Br. All of the samples fall within the range determined for plume emissions by Wittmer et al. (2014). Seawater ratio is from Davis et al. (1998). The concentrations of Cl in the bulk deposition are from Aiuppa et al. (2001) the maximum value is from Calabrese et al. (2011).

an approach is not effective when the residuals are large. Conversely, residuals are useful for identifying those sites that are affected by sources and/or processes different from PDW+ PDE and CDW. When considering the sum of all cations, all of the samples give very low residuals, thus confirming that PDW+PDE and CDW are the main mechanisms responsible for their chemical compositions. The samples from S. Leonardello and Faro have the largest residuals, especially for Mg and Ca, indicating a possible different source of dissolved ions. These samples also exhibit the highest NO− 3 concentrations (Table 1). Bearing in mind that the sampling location is characterized by intensive agricultural activities, these sites are probably affected by anthropogenic inputs of chemical compounds (e.g., fertilizer). PDW + PDE never contributes more than 52% to the sum of cation equivalents. Most of

Fig. 3. Sum of the cation equivalents versus total alkalinity. CDW should produce the same amount of cation and alkalinity equivalents (see text for details). Groundwater from the eastern flank shows the highest deviation from the 1:1 line. For comparison, samples from another basaltic aquifer (Koh et al., 2016) hosted in an inactive volcano (Jeju Island, Korea) are plotted. With the exception of the most diluted samples, probably affected by sea aerosol, water circulating in the inactive volcano perfectly match on the 1:1 line.

74

M. Liotta et al. / Chemical Geology 433 (2016) 68–80

Fig. 4. Concentrations of the studied elements (Na, K, Mg, Ca, Si, Al, B and Se) versus Cl and total alkalinity. Mg is highly correlated with alkalinity. Other elements such as Na, K, and B seem to be correlated to different extents with both Cl and alkalinity.

the samples with high PDW+PDE contributions are from the eastern flank (downwind) of the volcano edifice. The Primoti and S26 samples have the highest Cl concentration, and hence are probably those most influenced by the plume. Their location, exactly downwind of the plume and at the outlet of Valle del Bove, corroborates this finding. The relative fractions of cations clearly indicate that Na and K mainly derive from the plume while Mg and Ca mainly derive from CDW.

Table 2 Results for the descriptive statistics obtained by multiple regression analysis (95% confidence interval). Data input are mEq/L values except for I and B, which are presented as mmol/L values due to their speciation not being determined. Cat., cations. Cases with coeffincient of determination N0.9 are marked in bold. Parameter

R2

Intercept

Cl coefficient

Cat. Na+ K+ Mg2+ Ca2+ F− Br− NO− 3 SO2− 4 I B

0.962 0.931 0.783 0.962 0.594 0.250 0.981 0.088 0.562 0.255 0.615

8.51 × 10−1 4.54 × 10−1 1.86 × 10−1 –2.07 × 10−2 2.32 × 10−1 2.20 × 10−2 6.99 × 10−5 3.85 × 10−1 7.13 × 10−1 –1.01 × 10−6 –1.14 × 10−2

2.04 × 100 1.27 × 100 1.38 × 10−1 4.98 × 10−1 1.33 × 10−1 2.56 × 10−3 1.87 × 10−3 3.63 × 10−2 9.40 × 10−1 1.61 × 10−5 5.47 × 10−3

Alkalinity coefficient 9.33 × 10−1 2.06 × 10−1 8.70 × 10−3 5.01 × 10−1 2.17 × 10−1 –9.57 × 10−5 1.36 × 10−5 –1.72 × 10−2 –8.35 × 10−2 6.50 × 10−6 4.36 × 10−3

The samples from the southern part of the edifice show higher relative contributions of CDW to the sum of cations. For samples P5, Ercolino, A. Grassa, A. Rossa, Raffo (1 and 2), Ardizzone, Romito, S59, and S65, the multiple regression analyses of Na and Mg often predicted values very close to the observed ones. Conversely, the residuals for K and Ca are often higher than 10%. In order to determine if circulating waters reflect the chemical composition of volcanic rocks, it is necessary to consider that the degree of crystallization can be responsible for very different relative abundances of rocks-forming minerals. However, a large amount of groundmass is always present, and often represents the highest volume percentage (Tanguy and Clocchiatti, 1984; Métrich and Clocchiatti, 1989; Tanguy et al., 1997). Since the rate of weathering favors the dissolution of glass and olivine rather than plagioclase and pyroxene (Eggleton et al., 1987), we compared dissolved ions with the molar ratios of the glass matrix that represents the most abundant and most reactive phase. The glass matrix usually differs from the whole-rock composition. For example, D'Oriano et al. (2013) observed that a sample related to the January 5, 1990 fire-fountain episode exhibits the bulk rock composition of trachybasalt, whereas groundmass glass exhibits a phonotephritic composition. Similarly, Del Carlo et al. (2012) analyzed both whole rock and glass of Mt. Moio, and found that the glass of pyroclasts shows more-evolved compositions (ranging from hawaiitic to mugearitic) than rocks ranging between basalt and trachybasalt. Since K is an incompatible element with respect to forming minerals

Table 3 Relative residuals obtained by multiple linear regression, and computed relative fractions of PDW+PDE and CDW for each element. Cases with relative residuals b10 are marked in bold. Sample

Na

K

Mg

Ca

Br

B

res

f(PDW+PDE)

f(CDW)

res

f(PDW+PDE)

f(CDW)

res

f(PDW+PDE)

f(CDW)

res

f(PDW+PDE)

f(CDW)

res

f(PDW+PDE)

f(CDW)

res

f(PDW+PDE)

f(CDW)

res

f(PDW+PDE)

f(CDW)

3 1 33 36 −13 −9 14 −12 −1 −11 0 −15 5 −6 −7 −4 −14 −7 −1 2 −3 4 4 −5 −1 4 −4 −4 2 −7 1

51 50 40 36 58 52 34 53 31 37 34 44 28 33 32 34 16 9 30 19 16 15 15 22 17 13 21 20 16 25 14

49 50 60 64 42 48 66 47 69 63 66 56 72 67 68 66 84 91 70 81 84 85 85 78 83 87 79 80 84 75 86

6 1 −7 −10 −12 13 4 6 32 11 −5 −18 5 −15 16 −17 −8 57 −19 3 10 −6 −2 3 −2 1 8 7 0 −18 −1

74 74 65 61 79 76 59 76 56 62 59 69 52 58 57 59 35 22 55 41 35 33 34 44 37 29 43 41 35 49 32

26 26 35 39 21 24 41 24 44 38 41 31 48 42 43 41 65 78 45 59 65 67 66 56 63 71 57 59 65 51 68

6 6 23 17 −31 −40 27 −38 −6 −20 22 −24 −29 21 −18 1 −9 18 18 17 17 10 21 −12 −30 −23 8 4 13 18 9

88 88 83 80 91 89 79 89 76 81 79 85 74 78 77 79 58 41 75 64 57 56 57 67 60 51 66 64 58 71 55

12 12 17 20 9 11 21 11 24 19 21 15 26 22 23 21 42 59 25 36 43 44 43 33 40 49 34 36 42 29 45

−3 −5 40 32 −3 −12 7 −15 −9 −20 −6 −7 36 3 −1 1 −26 −27 10 12 −8 1 5 1 −5 4 −11 −9 −1 2 −4

32 31 23 20 38 33 19 34 17 21 19 26 15 18 18 19 8 4 16 10 8 7 8 11 9 6 11 10 8 13 7

68 69 77 80 62 67 81 66 83 79 81 74 85 82 82 81 92 96 84 90 92 93 92 89 91 94 89 90 92 87 93

11 12 114 143 −35 −57 46 −46 −43 −37 19 −23 −41 −14 −65 10 −5 −55 5 −23 −15 25 10 −31 16 10 −7 −13 11 −10 14

22 21 16 13 28 24 12 24 11 14 13 18 10 12 12 12 5 3 11 6 5 5 5 7 6 4 7 7 5 9 4

78 79 84 87 72 76 88 76 89 86 87 82 90 88 88 88 95 97 89 94 95 95 95 93 94 96 93 93 95 91 96

−6 0 −4 −2 10 11 −6 −5 −3 2 −6 2 −12 11 13 −5 −29 −20 −3 −8 −13 −11 2 −7 −4 −3 −1 −3 11 24 9

98 98 98 97 99 99 97 99 97 97 97 98 96 97 97 97 92 86 96 94 92 92 92 95 93 90 94 94 92 95 91

2 2 2 3 1 1 3 1 3 3 3 2 4 3 3 3 8 14 4 6 8 8 8 5 7 10 6 6 8 5 9

−26 −31 −55 −60 93 181 −55 277 23 −18 −37 32 44 −15 −6 70 −9 −87 88 −50 −33 −45 −8 −34 0 33 11 −4 −24 35 −19

37 36 28 24 44 39 23 40 20 25 23 31 18 22 21 23 10 5 20 12 10 9 9 14 11 8 13 13 10 16 9

63 64 72 76 56 61 77 60 80 75 77 69 82 78 79 77 90 95 80 88 90 91 91 86 89 92 87 87 90 84 91

M. Liotta et al. / Chemical Geology 433 (2016) 68–80

PRIMOTI S26 S. LEONARDELLO FARO ELLERA MASARAC. BAGLIO RANIERI ROCCA CAMPANA PEDARA P31 S. MARIA MURI ANTICHI PIANO ELISI SOLICCHIATA MUSMECI ILICE S. GIACOMO VALCOR. S65 CHERUBINO S59 P5 DIFESA ERCOLINO ROMITO A.GRASSA −S63BIS A.ROSSA S. VITO ARDIZZONE RAFFO2 SERAFICA RAFFO1

Cat

75

76

M. Liotta et al. / Chemical Geology 433 (2016) 68–80

in magma, it is usually present at higher levels in the glass matrix (Métrich and Clocchiatti, 1989; Métrich et al., 2004; Del Carlo et al., 2012; D'Oriano et al., 2013). While the charge balance imposes that the cation equivalent is always closely related to Cl and alkalinity, each cation may undergo exchange processes that change its abundance, thus losing its pristine relationships with Cl and alkalinity. Gay and Le Maitre (1961) suggested that the mechanism underlying the formation of iddingsite from olivine involves the diffusion of hydrogen ions into the structure and results in the release of Mg, Fe2+, and Si from their sites, allowing their replacement by Fe3 +, Al, and Ca [iddingsite has been observed by Sturiale (1968) in pillow lavas outcropping at Aci Castello]. In addition, clay mineral exchange processes further modify the chemical composition of the waters. In light of this, we can compare the Mg equivalents with those of the other major cations to check if they reflect the ratios for the rock samples (Fig. 5). With respect to the isochemical dissolution of the glass matrix, the Mg/(Na+K+Ca) ratio is always higher in groundwater than in the glass matrix and the bulk deposition; we therefore find an excess of Mg. As discussed above, olivines experience a similar rate of weathering as the glass matrix and can contribute to the total amount of dissolved Mg. In addition, the formation of iddingsite also provides Mg at the expense of Ca. Fig. 5 indicates that the less mineralized samples fall within the range determined for the glass matrix, thus suggesting that secondary processes play a minor role. In contrast, more mineralized waters with deeper circuits shows a higher excess of Mg. Those samples more influenced by the volcanic plume (Primoti and S26) also fall within the range determined for the glass matrix, but they also receive large amount a semi-volatile elements such as Na and K that shift their position to the right in Fig. 5. 5.4. Boron B is slightly positively correlated with both alkalinity and Cl (Fig. 4). Linear multiple regression yielded high residuals. This could depend either on the non-conservative behavior of the element or on the presence of several sources. Despite the uncertainty of the results for the descriptive statistics, we can appreciate that the estimated fraction of plume-derived B is about 40% at the Primoti and S26 sites, while it never exceeds 15% at those sites where CDW contributes about 80% to the sum of cations (Table 3). In high-temperature hydrothermal systems, B is positively correlated with temperature and most of the B is derived from the rock (Arnórsson and Andrésdóttir, 1995). In contrast, at Mt. Etna the plume can significantly contribute to the total amount. In order to understand how much the chemical weathering of rock can contribute to the total amount of dissolved B, Mg/B and Na/B

Fig. 5. Comparison between Mg and Na+K+Ca equivalents in groundwater and rock samples. The same ratio is given for the bulk deposition.

molar ratios in dissolved water should be compared with the values of the glass matrix, which is more easily weathered, but unfortunately these data are not yet available in the literature. We therefore performed a rough comparison with the molar ratios in rock. Tonarini et al. (2001) analyzed 16 rock samples in the temporal range from prehistoric alkaline lavas to the recent activity (in 1998) for both the chemical and isotopic compositions of B. They found concentrations within a restricted range (7.4–11.3 ppm), with an average of 8.7 ppm. The average Mg content of Etnean volcanic rocks derived from a large data set published by Armienti et al. (2012) is about 5.5% magnesium oxide. Consideration of both of these average values yields an Mg/B molar ratio of about 1700. Analogously, we can derive an average Na/B molar ratio of about 740. Fig. 6 shows binary diagrams of Mg versus B and Na versus B. The B content is always higher than those of Na and Mg. This implies that chemical weathering of the host rocks cannot be the only mechanism responsible for the dissolved B content, and it confirms a significant contribution from the volcanic plume. Fogg and Duce (1985) suggested that volcanic emissions represent an important source of the total atmospheric B. More recently, Calabrese et al. (2011) observed that the B content at Mt. Etna is strongly enriched in rainwater and in the plume with respect to the rock, thus confirming that B is easily volatilized. 5.5. Origin and mobility of volatile elements in groundwater After H2O and CO2, S, Cl, and F represent the most abundant volatile elements emitted as gaseous species from the craters; together they account for almost 100% of the total released gases. On the other hand, minor amounts of I and Br are also discharged (Wittmer et al., 2014), and they can provide useful information on degassing mechanisms as well as on their mobility. Despite the high abundance of H2O and CO2 in the plume, their relative contributions to groundwater via meteoric recharge are difficult to determine due to their relatively high background level in the atmosphere and their chemical behavior. When gases leave the magma they are rapidly diluted in the atmosphere, and under these conditions the relative contribution of volcanogenic H2O and CO2 is negligible with respect to the air background. As a consequence, precipitation feeding the groundwater does not transport significant amounts of H2O and carbon that originate from the plume. In contrast, the other volatile elements can significantly affect the local precipitation. The plume contribution of Cl and Br is discussed in Section 5; here we focus our attention on S, F, and I. When considering the logarithmic binary diagram of SO42− versus Cl (Fig. 7), groundwater samples do not show a reliable correlation and all but one of the samples have an S/Cl ratio higher than that for seawater; the only sample with an S/Cl ratio identical to that of seawater is the Romito sample. However, it should be remembered that under reducing conditions, SO42− in groundwater can be reduced to hydrogen sulfide (H2S) that easily escapes from water. Since the Romito sample is the most-reducing one and the typical rotten-eggs odor of H2S is evident at the spring outlet, we cannot exclude that loss of S has occurred. All

Fig. 6. Binary diagrams of B versus Mg (a) and Na (b).

M. Liotta et al. / Chemical Geology 433 (2016) 68–80

77

Fig. 7. Binary diagram of SO4 versus Cl. The plume range, bulk deposition, and seawater ratios are also plotted. Bulk deposition values are from Aiuppa et al. (2001, 2006).

of the other samples fall between the ranges of bulk deposition and the plume, with the S26 and Primoti samples exactly falling within the range determined by Aiuppa et al. (2001, 2006) for those sites affected by plume emissions. This clearly suggests that the plume also contributes to the S dissolved in groundwater. Despite this, no clear trends can be recognized in Fig. 7, which is due to the complex evolution of S species in the volcanic plume and their interaction with rainwater. S mainly exsolves as SO2. The SO2 lifetime in the atmosphere is a matter of debate and its estimations range in the order of 1-2 days (e.g. Beirle et al., 2014) up to weeks (Von Glasow et al., 2009). With respect to HBr and HCl that readily dissociate in water, SO2 requires longer time to be oxidized. Once sulfur is converted to sulfuric acid, it should easily dissociate in water as sulfate. The limiting-reaction rate is then the oxidation of SO2 that is responsible for the lower SO4/Cl values found in precipitation with respect to S/Cl in plume emissions. An especially noteworthy finding was that the samples with the highest SO4 contents are from the eastern flank of Mt. Etna, east of Valle del Bove. Since the meteoric recharge of areas not affected by the plume usually exhibit very low S contents compared to those observed in groundwater, we believe that the volcanic plume contributes markedly to the total S content in groundwater. Conversely, we cannot exclude that some sites are affected by anthropogenic sources of SO4. This is probably the case for the S. Leonardello, Faro, and Baglio sites; they also exhibit the highest NO− 3 concentrations. Based on the multiple linear regression analyses discussed above, these samples also give high residuals for several cations. After the infiltration of meteoric water, S species and their abundances can change depending on the redox conditions. Under reducing conditions, S species can be converted to H2S that easily escapes from water in its gaseous form, thus lowering the total S content. With respect to the other volatile elements, F is not correlated with Cl (Fig. 9). Its concentration ranges between 8.5 × 10− 3 and 3.85 × 10−2 mmol/L, with an average of 2.65 × 10−2 mmol/L. This compositional range falls below the bulk deposition in the summit area. No mixing trend can be defined considering the bulk deposition not affected by the plume and that affected by the plume (Fig. 8). This means that, after deposition, F is affected by processes that change its pristine concentration. D'Alessandro et al. (2012) demonstrated that Etnean volcanic soils can trap up to 70% of the original F− dissolved in water. The high content of amorphous phases of Al (allophane and imogolite), clay minerals, and organic bound Al enhance the F−-adsorption properties of Etnean soils. Neverthless, some sectors of the volcanic edifice are not covered by well-developed volcanic soils; even if we cannot exclude

Fig. 8. Binary diagram of F versus Cl. The plume range, bulk deposition, and seawater ratios are also plotted. Bulk deposition values are from Aiuppa et al. (2001).

that not-outcropping paleosols can play some role in F removal, we should also take into account other possible processes. Aiuppa et al. (2003) observed that fluorite precipitation is probably not an Flimiting process since all of the analyzed groundwaters are undersaturated in F. This is consistent with our data that produce negative saturation indices for fluorite when processed using the PHREEQC software package (Parkhurst and Appelo, 1999). Nevertheless, the nearconstant concentration of F suggests that the involved processes buffer its content in the analyzed groundwaters. (See Fig. 9.) The I concentrations in Etnean groundwater cover three orders of magnitude. Unfortunately, previous data on the bulk deposition are not available for I, and so we compared our results with data from the literature worldwide. In nonvolcanic environments the I concentration is always strongly enriched with respect to chlorine (100–10000

Fig. 9. Binary diagram of I versus Cl. The plume range, the aerosols-rain range, and seawater ratio are also plotted.

78

M. Liotta et al. / Chemical Geology 433 (2016) 68–80

times) in both rain and aerosols compared to ocean water (Duce et al., 1966; Duce et al., 1967; Winchester and Duce, 1967; Woodcock et al., 1971; Sadasivan and Anand, 1979; Sturges and Barrie, 1988; Murphy et al., 1997; Vogt et al., 1999; Baker et al., 2000; Baker, 2004; Gilfedder et al., 2007). Several authors have also observed that organically bound I and iodide are the dominant I species (Baker et al., 2000; Baker, 2004, 2005; Gilfedder et al., 2007). Gilfedder et al. (2007) observed that the I concentrations in snow samples from the northern Black Forest (Germany) decrease exponentially with altitude, suggesting that orographic lifting and subsequent precipitation greatly influences I concentrations in snow. These findings reflect the complex geochemical behavior of I. Since the I/Cl ratio for the bulk deposition at Mt. Etna has not been determined yet, we cannot state that groundwater reflects the same I/Cl of the bulk deposition. Nevertheless, we observe that most of the circulating waters at Mt. Etna have I/Cl ratios that fall within the range determined for the plume by Wittmer et al. (2014). This range partially overlaps the wide range for worldwide aerosols and rain. It should be noted that some samples from the eastern flank of the volcano that exhibit large amounts of PDE fall below this range. All of these samples are characterized by high S contents. Even though I sources and mobility at Mt. Etna need to be investigated further (e.g. to explain high content of I at Solicchiata), the correlation between Cl and I suggests that the plume emissions may contribute to the total amount found in circulating waters. 5.6. Selenium The Se content in the Etnean groundwaters is lower than 2 × 10−5 mmol/L for most of the samples, and it is weakly correlated with Cl (Fig. 4). It is noteworthy that the S26 and Primoti samples exhibit Se concentrations that are fourfold higher than those at the other sites. As discussed above, these samples are the most influenced by the plume. According to our findings, these waters infiltrate under highly acidic conditions. Floor et al. (2011) observed that Se in the soils of Mt. Etna is highly mobilized by acid rain, and Calabrese et al. (2011) reported that it is among the most-enriched elements in the Etnean

plume emissions. The higher Se concentrations found in the easternflank groundwater also provide evidence for a contribution of volcanic emissions via meteoric recharge. 6. Conclusions Several processes underlie the chemical composition of the groundwater at Mt. Etna (Fig. 10), the most important of which are direct input through bulk deposition, early interaction of infiltrating water with volcanic rock, and CDW. Other processes such as ion exchange, iddingsite formation, and carbonate precipitation can also play roles, but only to minor extents. The detailed description of the processes occurring and the estimation of their relative contributions have yielded novel results in this study. The most relevant of these are as follows: – The very strong correlation between Cl and Br allows a ratio to be defined that clearly reflects that of the plume and differs from that of the seawater and brines within the sediments below Mt. Etna. Thus, the persistent plume is the main source of Cl and Br dissolved in water. The almost homogeneous Cl/Br ratio in groundwater directly derives from plume-affected acid rain that infiltrate towards the aquifer. Cl and Br exhibit similar behavior during degassing. Our findings suggest that, after degassing, they still behave similarly during interaction with meteoric water. – Acidic bulk depositions carry significant amounts of dissolved elements and are responsible for the weathering of the outcropping rocks. – CDW of basaltic rocks produces high alkalinity values that, under the observed pH range, are directly related to the released cation equivalents. – Since Cl mainly derives from plume emissions and alkalinity can be ascribed to the dissociation of carbonic acid (H2CO3) after the hydration of CO2, Cl and alkalinity can be used to compute the relative contributions of PDW+PDE and CDW for Na, K, Mg, Ca, and B. – Volatile elements impress their signature on circulating waters to different extents. S reflects the bulk deposition composition,

Fig. 10. Scheme of the processes occurring at Mt. Etna. Acid bulk deposition carries significant amount of PDE. Early weathering occurs during the first stage of infiltration of meteoric water (PDW). The weathering of rocks is further promoted by CO2 coming up through the volcanic edifice and dissolving in the water.

M. Liotta et al. / Chemical Geology 433 (2016) 68–80

while it differs from the plume one. F undergoes processes that decrease the pristine concentration of the bulk deposition. I, despite its complex geochemical behavior, often reflects the plume composition. – The content of dissolved B cannot be explained solely in terms of water-rock interaction and a plume contribution via meteoric recharge is responsible for a significant fraction of the total amount. – The high content of Se found in the sites that are most influenced by the plume confirms the finding of Floor et al. (2011) that Se is mobilized when volcano-derived acid rain interacts with poorly developed soils close to the crater and that this process might influence the chemical composition of groundwater. A comprehensive understanding of the processes described herein makes it possible to evaluate how the volcanic plume affects the groundwater composition. Our findings provide an overarching framework to assess how changes in the plume emissions as well as the CO2 input in the aquifer are reflected on the chemical composition of groundwater. Finally, we give new insights on the mobility of plumederived halogens in cold groundwater of a persistently degassing basalt volcano. Acknowledgments We are very grateful to all of the staff of Istituto Nazionale di Geofisica e Vulcanologia, Sezione di Palermo for their valuable support in the field and in the laboratories. Horizontal wind component data were obtained from the ECMWF (European Centre for Medium-Range Weather Forecasts) MARS archive under license from the Italian Air Force Meteorological Service. The insightful comments of an anonymous reviewer and of Benoît Villemant are warmly acknowledged as well as the editorial handling of Jerome Gaillardet. References Aiuppa, A., Allard, P., D'Alessandro, W., Michel, A., Parello, F., Treuil, M., Valenza, M., 2000. Mobility and fluxes of major, minor and trace metals during basalt weathering and groundwater transport at Mt. Etna volcano (Sicily). Geochim. Cosmochim. Acta 64, 1827–1841. Aiuppa, A., Bonfanti, P., Brusca, L., D'Alessandro, W., Federico, C., Parello, F., 2001. Evaluation of the environmental impact of volcanic emissions: insights from the chemistry of rainwater in the Mt. Etna area (Sicily). Appl. Geochem. 16, 985–1000. Aiuppa, A., Bellomo, S., Brusca, L., D'Alessandro, W., Federico, C., 2003. Natural and anthropogenic factors affecting groundwater quality of an active volcano (Mt. Etna, Italy). Appl. Geochem. 18, 863–882. Aiuppa, A., Federico, C., Franco, A., Giudice, G., Gurrieri, S., Inguaggiato, S., Liuzzo, M., McGonigle, A.J.S., Valenza, M., 2005. Emission of bromine and iodine from Mount Etna volcano. Geochem. Geophys. Geosyst. 6, Q08008. http://dx.doi.org/10.1029/ 2005GC000965. Aiuppa, A., Bellomo, S., Brusca, L., D'Alessandro, W., Di Paola, R., Longo, M., 2006. Majorion bulk deposition around an active volcano (Mt. Etna, Italy). Bull. Volcanol. 68, 255–265. Al-Ammar, A., Reitznerová, E., Ramon, M., Barnes, R.M., 2001. Thorium and iodine memory effects in inductively-coupled plasma mass spectrometry. J. Anal. Chem. 370, 479–482. Allard, P., Carbonnelle, J., Dajlevic, D., Le Bronec, J., Morel, P., Robe, M.C., Maurenas, J.M., Faivre-Pierret, R., Martin, D., Sabroux, J.C., Zettwoog, P., 1991. Eruptive and diffuse emissions of CO2 from Mount Etna. Nature 351, 387–391. Allen, A.G., Mather, T.A., McGonigle, A.J.S., Aiuppa, A., Delmelle, P., Davison, B., Bobrowski, N., Oppenheimer, C., Pyle, D.M., Inguaggiato, S., 2006. Sources, size distribution, and downwind grounding of aerosols from Mount Etna. J. Geophys. Res. 111, D10302. Armienti, P., Perinelli, C., Putirka, K.D., 2012. A New model to estimate deep-level magma ascent rates, with applications to Mt. Etna (Sicily, Italy). J. Petrol. 54, 795–813. Arnórsson, S., Andrésdóttir, A., 1995. Processes controlling the distribution of boronand chlorine in natural waters in Iceland. Geochim. Cosmochim. Acta 59 (20), 4125–4146. Bagnato, E., Aiuppa, A., Andronico, D., Cristalli, A., Liotta, M., Brusca, L., Miraglia, L., 2011. Leachate analyses of volcanic ashes from Stromboli volcano: a proxy for the volcanic gas plume composition? J. Geophys. Res. 116, D17204. Baker, A.R., 2004. Inorganic iodine speciation in tropical Atlantic aerosol. Geophys. Res. Lett. 31, L23S02. Baker, A.R., 2005. Marine aerosol iodine chemistry: the importance of soluble organic iodine. Environ. Chem. 2, 295–298. Baker, A.R., Thompson, D., Campos, M.L.A.M., Parryb, S.J., Jickellsa, T.D., 2000. Iodine concentration and availability in atmospheric aerosol. Atmos. Environ. 34, 4331–4336.

79

Bazan, G., Brullo, S., Raimondo, F.M., Schicchi, R., 2010. Le serie di vegetazione della Regione Sicilia. La vegetazione d'Italia. Palombi Ed., Roma, pp. 429–470. Beirle, S., Hörmann, C., Penning de Vries, M., Dörner, S., Kern, C., Wagner, T., 2014. Estimating the volcanic emission rate and atmospheric lifetime of SO2 from space: a case study for Kīlauea volcano, Hawai`i. Atmos. Chem. Phys. 14, 8309–8322. Bellomo, S., Aiuppa, A., D'Alessandro, W., Parello, F., 2007. Environmental impact of magmatic fluorine emission in the Mt. Etna area. J. Volcanol. Geotherm. Res. 165, 87–101. Bergametti, G., Martin, D., Carbonnelle, J., Faivre-Pierret, R., Vie Le Sage, R., 1984. A mesoscale study of the elemental composition of aerosols emitted from Mt. Etna Volcano. Bull. Volcanol. 47, 1107–1114. Berner, R.A., Rao, J.-L., 1997. Alkalinity buildup during silicate weathering under a snow cover. Aquat. Geochem. 2, 301–312. Branca, S., Coltelli, M., Groppelli, G., 2011. Geological evolution of a complex basaltic stratovolcano: Mount Etna, Italy. Ital. J. Geosci. 130, 306–317. Brusca, L., Aiuppa, A., D'Alessandro, W., Parello, F., Allard, P., Michel, A., 2001. Geochemical mapping of magmatic gas–water–rock interactions in the aquifer of Mount Etna volcano. J. Volcanol. Geotherm. Res. 108, 199–218. Calabrese, S., D'Alessandro, W., 2015. Characterization of the Etna volcanic emissions through an active biomonitoring technique (moss-bags): part 2 — morphological and mineralogical features. Chemosphere 119, 1456–1464. Calabrese, S., Aiuppa, A., Allard, P., Bagnato, E., Bellomo, S., Brusca, L., D'Alessandro, W., Parello, F., 2011. Atmospheric sources and sinks of volcanogenic elements in a basaltic volcano (Etna, Italy). Geochim. Cosmochim. Acta 75, 7401–7425. Calabrese, S., Scaglione, S., Milazzo, S., D'Alessandro, W., Bobrowski, N., Giuffrida, G.B., Tedesco, D., Parello, F., Yalire, M., 2014. Passive degassing at Nyiragongo (D.R. Congo) and Etna (Italy) volcanoes. Ann. Geophys. 57, 1–6. Camarda, M., De Gregorio, S., Gurrieri, S., 2012. Magma-ascent processes during 2005–2009 at Mt Etna inferred by soil CO2 emissions in peripheral areas of the volcano. Chem. Geol. 330-331, 218–227. Caracausi, A., Italiano, F., Nuccio, P.M., Paonita, A., Rizzo, A., 2003. Evidence of deep magma degassing and ascent by geochemistry of peripheral gas emissions at Mt. Etna (Italy): assessment of the magmatic reservoir pressure. J. Geophys. Res. 108, 2463–2484. Chemtob, S.M., Rossman, G.R., 2014. Timescales and mechanisms of formation of amorphous silica coatings on fresh basalts at Kilauea Volcano, Hawai'i. J. Volcanol. Geotherm. Res. 286, 41–54. Chemtob, S.M., Rossman, G.R., Young, E.D., Ziegler, K., Moynier, F., Eiler, J.M., Hurowitz, J.A., 2015. Silicon isotope systematics of acidic weathering of fresh basalts, Kilauea Volcano, Hawai'i. Geochim. Cosmochim. Acta 169, 63–81. Chester, D.K., Duncan, A.M., 1985. Mount Etna: The Anatomy Of A Volcano. Stanford University Press. Chiodini, G., D'Alessandro, W., Parello, F., 1996. Geochemistry of gases and water discharged by the mud volcanoes at Paternò, Mt. Etna (Italy). Bull. Volcanol. 58, 51–58. Chiodini, G., Caliro, S., Cardellini, C., Granieri, D., Avino, R., Baldini, A., Donnini, M., Minopoli, C., 2010. Long-term variations of the Campi Flegrei, Italy, volcanic system as revealed by the monitoring of hydrothermal activity. J. Geophys. Res. 115, B03205. http://dx.doi.org/10.1029/2008JB006258. Corsaro, R.A., Pompilio, M., 2004. Dynamics of Magmas at Mount Etna. Mt. Etna: Volcano Laboratory, Geophys. Monogr. Ser. 143. AGU, Washington, D. C., pp. 91–110. Crider, J.G., Frank, D., Malone, S.D., Poland, M.P., Werner, C., Caplan-Auerback, J., 2011. Magma at depth: a retrospective analysis of the 1975 unrest at Mount Baker, Washington, USA. Bull. Volcanol. 73, 175–189. D'Alessandro, W., De Gregorio, S., Dongarrà, G., Gurrieri, S., Parello, F., Parisi, B., 1997a. Chemical and isotopic characterization of the gases of Mount Etna (Italy). J. Volcanol. Geotherm. Res. 78, 65–76. D'Alessandro, W., Giammanco, S., Parello, F., Valenza, M., 1997b. CO2 output and δ13C(CO2) from Mount Etna as indicators of degassing of shallow asthenosphere. Bull. Volcanol. 58, 455–458. D'Alessandro, W., Federico, C., Longo, M., Parello, F., 2004. Oxygen isotope composition of natural waters in the Mt. Etna area. J. Hydrol. 296, 282–299. D'Alessandro, W., Giammanco, S., Bellomo, S., Parello, F., 2007. Geochemistry and mineralogy of travertine deposits of the SW flank of Mt. Etna (Italy): relationships with past volcanic and degassing activity. J. Volcanol. Geotherm. Res. 165, 64–70. D'Alessandro, W., Bellomo, S., Bonfanti, P., Brusca, L., Longo, M., 2011. Salinity variations in the water resources fed by the Etnean volcanic aquifers (Sicily, Italy): natural vs. anthropogenic causes. Environ. Monit. Assess. 173, 431–446. D'Alessandro, W., Bellomo, S., Parello, F., 2012. Fluorine adsorption by volcanic soils at Mt. Etna, Italy. Appl. Geochem. 27, 1179–1188. Davis, S.N., Whittemore, D.O., Fabryka-Martin, J., 1998. Uses of chloride/bromide ratios in studies of potable water. Ground Water 36, 338–350. Del Carlo, P., Branca, S., De Beni, E., Lo Castro, M.D., Wijbrans, J.R., 2012. The Mt. Moio eruption (Etna): stratigraphy, petrochemistry and 40Ar/39Ar age determination with inferences on the relationship between structural setting and magma intrusion. J. Volcanol. Geotherm. Res. 241–242, 49–60. Delmelle, P., Villiéras, F., Pelletier, M., 2005. Surface area, porosity and water adsorption properties of fine volcanic ash particles. Bull. Volcanol. 67, 160–169. Delmelle, P., Lambert, M., Dufrêne, Y., Gerin, P., Óskarsson, N., 2007. Gas/aerosol–ash interaction in volcanic plumes: new insights from surface analyses of fine ash particles. Earth Planet. Sci. Lett. 259, 159–170. D'Oriano, C., Pompilio, M., Bertagnini, A., Cioni, R., Pichavant, M., 2013. Effects of experimental reheating at different temperatures and redox conditions of natural basaltic ash. Contrib. Mineral. Petrol. 165, 863–883. Duce, R.A., Winchester, J.W., Van Nahl, T.W., 1966. Iodine, bromine, and chlorine in winter aerosols and snow from Barrow, Alaska. Tellus 18, 238–248. Duce, R.A., Woodcock, A.H., Moyers, J.L., 1967. Variation Of Ratios With Size Among Particles In Tropical Oceanic Air Tellus 19, 369–379.

80

M. Liotta et al. / Chemical Geology 433 (2016) 68–80

Durbin, C.S., Henderson-Sellers, A., 1981. Meteorological importance of the volcanic activity of Mount Etna. Weather 36, 284–291. Eggleton, R., Foudoulis, C., Varkevisser, D., 1987. Weathering of basalt: changes in rockchemistry and mineralogy. Clay Clay Miner. 35 (3), 161–169. Favalli, M., Mazzarini, F., Pareschi, M.T., Boschi, E., 2004. Role of local wind circulation in plume monitoring at Mt. Etna volcano (Sicily): insights from a mesoscale numerical model. Geophys. Res. Lett. 31, L09105. Fehn, U., 2012. Tracing crustal fluids: applications of natural 129I and 36Cl. Ann. Rev. Earth Planet. Sci. 40, 45–67. Fehn, U., Snyder, G.T., 2003. Origin of iodine and 129I in volcanic and geothermal fluids from the North Island of New Zealand: IMPLICATIONS for subduction zone processes. Econ. Geol. Spec. Publ. 10, 159–170. Ferrara, V., 1975. Idrogeologia del versante orientale dell'Etna. Atti 3° Conv. Intern. sulle Acque Sotterranee, Palermo 1-3 Nov.1975 91-144. Ferrara, V., Pappalardo, G., 2008. La carta idrogeologica del massiccio vulcanico dell'Etna come utili strumento per la gestione razionale delle risorse idriche sotterranee. Ital. J. Eng. Geol. Environ. 1, 77–89. Floor, G., Calabrese, S., Roman-Ross, G., D´Alessandro, W., Aiuppa, A., 2011. Selenium mobilization in soils due to volcanic derived acid-rain: an example from Mt Etna volcano, Sicily. Chem. Geol. 289, 235–244. Fogg, T.R., Duce, R.A., 1985. Boron in the troposphere: distribution and fluxes. J. Geophys. Res. 90, 3781–3796. Gat, J.R., Carmi, I., 1970. Evolution of the isotopic composition of atmospheric waters in the Mediterranean Sea area. J. Geophys. Res. 75, 3039–3048. Gat, J.R., Klein, B., Kushnir, Y., Roether, W., Wernli, H., Yam, R., Shemesh, A., 2003. Isotope composition of air moisture over the Mediterranean Sea: An index of the air-sea interaction pattern. Tellus 55B, 953–965. Gay, P., Le Maitre, R.W., 1961. Some Observations on Iddingsite. Am. Mineral. 46, 92–111. Gerlach, T.M., 1991. Etna's greenhouse pump. Nature 315, 352–353. Giggenbach, W.F., 1987. Redox processes governing the chemistry of fumarolic gas discharges from White Island, New Zealand. Appl. Geochem. 2, 143–161. Giggenbach, W.F., 1988. Geothermal solute equilibria. Derivation of Na–K–Mg–Ca geoindicators. Geochim. Cosmochim. Acta 52, 2749–2765. Gilfedder, B.S., Petri, M., Biester, H., 2007. Iodine and bromine speciation in snow and the effect of orographically induced precipitation. Atmos. Chem. Phys. 7, 2661–2669. Gislason, S.R., Eugster, H.P., 1987. Meteoric water–basalt interactions: II A field study in northeastern Iceland. Geochim. Cosmochim. Acta 51, 2841–2855. Gislason, S.R., Oelkers, E.H., 2003. Mechanism, rates, and consequences of basaltic glass dissolution: II. An experimental study of the dissolution rates of basaltic glass as a function of pH and temperature. Geochim. Cosmochim. Acta 67 (20), 3817–3832. Gislason, S.R., Arnorsson, S., Armannsson, H., 1996. Chemical weathering of basalt in Southwest Iceland; effects of runoff, age of rocks and vegetative/glacial cover. Am. J. Sci. 296 (8), 837–907. Grassa, F., Inguaggiato, S., Liotta, M., 2008. Fluid geochemistry of Stromboli. In: Calvari, S., Inguaggiato, S., Puglisi, G., Ripepe, M., Rosi, M. (Eds.), The Stromboli Volcano: An Integrated Study Of The 2002–2003 Eruption. American Geophysical Union Mon 182, pp. 49–64. Gudbrandsson, S., Wolff-Boenisch, D., Gislason, S.R., Oelkers, E.H., 2014. Experimental determination of plagioclase dissolution rates as a function of its composition and pH at 22 C. Geochim. Cosmochim. Acta 139, 154–172. Gysi, A.P., Stefánsson, A., 2012. CO2–water–basalt interaction. Low-temperature experiments and implications for CO2 sequestration into basalts. Geochim. Cosmochim. Acta 81, 129–152. James, P., Chester, D.K., Duncan, A.M., 2016. Development and spatial distribution of soils on an active volcano: Mt Etna, Sicily. Catena 137, 277–297. Jones, M.T., Gislason, S.R., 2008. Rapid releases of metal salts and nutrients following the deposition of volcanic ash into aqueous environments. Geochim. Cosmochim. Acta 72 (15), 3661–3680. Koh, D.-C., Chae, G.-T., Ryu, J.-S., Lee, S.G., Ko, K.-S., 2016. Occurrence and mobility of major and trace elements in groundwater from pristine volcanic aquifers in Jeju Island, Korea. Appl. Geochem. 65, 87–102. Lentini, F., 1982. The geology of Mt. Etna basement. Mem. Soc. Geol. Ital. 23, 7–25. Liotta, M., Paonita, A., Caracausi, A., Martelli, M., Rizzo, A., Favara, R., 2010. Hydrothermal processes governing the geochemistry of the crater fumaroles at Mount Etna volcano (Italy). Chem. Geol. 278, 92–104. Liotta, M., Rizzo, A., Paonita, A., Caracausi, A., Martelli, M., 2012. Sulfur isotopic compositions of fumarolic and plume gases at Mount Etna (Italy) and inferences on their magmatic source. Geochem. Geophys. Geosyst. 13, Q05015. Liotta, M., Grassa, F., D'Alessandro, W., Favara, R., Gagliano Candela, E., Pisciotta, A., Scaletta, C., 2013. Isotopic composition of precipitation and groundwater in Sicily, Italy. Appl. Geochem. 34, 199–206. Lowenstern, J.B., Smith, R.B., Hill, D.P., 2006. Monitoring super-volcanoes: geophysical and geochemical signals at Yellowstone and other large caldera systems. Phil. Trans. R. Soc. A 364, 2055–2072. http://dx.doi.org/10.1098/rsta.2006.1813. Madonia, P., Liotta, M., 2010. Chemical composition of precipitation at Mt. Vesuvius and Vulcano Island, Italy: volcanological and environmental implications. Environ. Earth Sci. 61, 159–171. Martelli, M., Caracausi, A., Paonita, A., Rizzo, A., 2008. Geochemical variations of air-free crater fumaroles at Mt Etna: new inferences for forecasting shallow volcanic activity. Geophys. Res. Lett. 35, L21302. http://dx.doi.org/10.1029/2008GL035118. Martin, R.S., Mather, T.A., Pyle, D.M., Power, M., Allen, A.G., Aiuppa, A., Horwell, C.J., Ward, E.P.W., 2008. Composition-resolved size distributions of volcanic aerosols in the Mt. Etna plumes. J. Geophys. Res. 113, D17211.

Martin, R.S., Witt, M.L.I., Sawyer, G.M., Thomas, H.E., Watt, S.F.L., Bagnato, E., Calabrese, S., Aiuppa, A., Delmelle, P., Pyle, D.M., Mather, T.A., 2012. Bioindication of volcanic mercury (Hg) deposition around Mt. Etna (Sicily). Chem. Geol. 310-311, 12–22. Métrich, N., Clocchiatti, R., 1989. Melt inclusion investigation of the volatile behavior in historic alkali basaltic magmas of Etna. Bull. Volcanol. 51, 185–198. Métrich, N., Allard, P., Spilliaert, N., Andronico, D., Burton, M., 2004. 2001 flank eruption of the alkali and volatile-rich primitive basalt responsible for Mount Etna’s evolution in the last three decades, Earth Planet. Sci. Lett. 228, 1–17. http://dx.doi.org/10.1016/j. epsl.2004.09.036. Moretti, R., Arienzo, I., Civetta, L., Orsi, G., Papale, P., 2013. Multiple magma degassing sources at an explosive volcano. Earth Planet. Sci. Lett. 367, 95–104. Murphy, D.M., Thomson, D.S., Middlebrook, A.M., 1997. Bromine, iodine, and chlorine in single aerosol particles at Cape Grim. Geophys. Res. Lett. 24, 3197–3200. Notcutt, G., Davies, F., 1989. Accumulation of volcanogenic fluoride by vegetation: Mt. Etna, Sicily. J. Volcanol. Geotherm. Res. 39, 329–333. Paonita, A., Caracausi, A., Iacono-Marziano, G., Martelli, M., Rizzo, A., 2012. Geochemical evidence for mixing between fluids exsolved at different depths in the magmatic system of Mt Etna (Italy). Geochim. Cosmochim. Acta 84, 380–394. Parello, F., D'Alessandro, W., Aiuppa, A., Federico, C., 2001. Cartografia geochimica degli acquiferi etnei. Gruppo Naz. per la Difesa dalle Catastrofi Idrogeologiche del C.N.R. — pubbl. n. 2190. Officine Grafiche Riunite, Palermo. Parkhurst, D.L., Appelo, C.A.J., 1999. User’s guide to PHREEQC (Version 2)—A Computer Program For Speciation, Batch-Reaction, One-Dimensional Transport, And Inverse Geochemical Calculations. U.S. Department of the Interior. USGS Water-Resources Investigations Report 99-4259. USGS, Reston, Virginia. Rizzo, A.L., Caracausi, A., Liotta, M., Paonita, A., Barnes, J.D., Corsaro, R.A., Martelli, M., 2013. Chlorine isotope composition of volcanic gases and rocks at Mount Etna (Italy) and inferences on the local mantle source. Earth Planet. Sci. Lett. 371–372, 134–142. Rizzo, A.L., Jost, H.J., Caracausi, A., Paonita, A., Liotta, M., Martelli, M., 2014. Real-time measurements of the concentration and isotope composition of atmospheric and volcanic CO2 at Mount Etna (Italy). Geophys. Res. Lett. 41, 2382–2389. Sadasivan, S., Anand, S.J.S., 1979. Chlorine, bromine and iodine in monsoon rains in India. Tellus 31, 290–294. Schilling, J.G., Bergeron, M.B., Evans, R., Smith, J.V., 1980. Halogens in the mantle beneath the North Atlantic. Philos. Trans. R. Soc. Lond. A 297, 147–178. Sorey, M.L., McConnell, V.S., Roeloff, E., 2003. Summary of recent research in Long Valley Caldera, California. J. Volcanol. Geotherm. Res. 127, 165–173. Stumm, W., Morgan, J.J., 1996. Aquatic Chemistry, Chemical Equilibria and Rates in Natural Waters. third ed JohnWilley and Sons. Sturges, W.T., Barrie, L.A., 1988. Chlorine, bromine, and iodine in Arctic aerosols. Atmos. Environ. 22, 1179–1194. Sturiale, C., 1968. Le formazioni eruttive submarine a Nord di Catania. Rend. Soc. Ital. Mineral. Petrol. 24, 313–346. Tanguy, J.C., Clocchiatti, R., 1984. The Etnean lavas, 1977–1983: petrology and mineralogy. Bull. Volcanol. 47, 879–894. Tanguy, J.C., Condomines, M., Kieffer, G., 1997. Evolution of the Mount Etna magma: constraints on the resent feeding system and eruptive mechanism. J. Volcanol. Geotherm. Res. 75, 221–250. Todesco, M., 2009. Signals from the Campi Flegrei hydrothermal system: the role of a “magmatic” source of fluids. J. Geophys. Res. 114, B05201. http://dx.doi.org/10. 1029/2008JB006134. Tonarini, S., Armienti, P., D'Orazio, M., Innocenti, F., 2001. Subduction-like fluids in the genesis of Mt. Etna magmas: evidence from boron isotopes and fluid mobile elements. Earth Planet. Sci. Lett. 192, 471–483. Villemant, B., Hammouya, G., Michel, A., Semet, M.P., Komorowski, J.C., Boudon, G., Cheminée, J.L., 2005. The memory of volcanic waters: shallow depth magma degassing revealed by long-term monitoring of hydrothermal springs at la Soufrière volcano (Guadeloupe, Lesser Antilles). Earth Planet. Sci. Lett. 237, 710–728. Villemant, B., Komorowski, J.C., Dessert, C., Michel, A., Crispi, O., Hammouya, G., Beauducel, F., De Chabalier, J.B., 2014. Evidence for a new shallow magma intrusion at La Soufrière of Guadeloupe (Lesser Antilles) Insights from long-term geochemical monitoring of halogen-rich hydrothermal fluids. J. Volcanol. Geotherm. Res. 285, 247–277. Vogt, R., Sander, R., Glasow, R.V., Crutzen, P.J., 1999. Iodine chemistry and its role in halogen activation and ozone loss in the marine boundary layer: a model study. J. Atmos. Chem. 32, 375–395. Von Glasow, R., Bobrowski, N., Kern, C., 2009. The effects of volcanic eruptions on atmospheric chemistry. Chem. Geol. 263, 131–142. Wang, L.X., Marks, M.A.W., Keller, J., Markl, G., 2014. Halogen variations in alkaline rocks from the Upper Rhine Graben (SW Germany): Insights into F, Cl and Br behavior during magmatic processes. Chem. Geol. 380, 133–144. Winchester, J.W., Duce, R.A., 1967. The global distribution of iodine, bromine, and chlorine in marine aerosols. Naturwissenschaften 54, 110–113. Wittmer, J., Bobrowski, N., Liotta, M., Giuffrida, G., Calabrese, S., Platt, U., 2014. Active alkaline traps to determine acidic-gas ratios in volcanic plumes: sampling technique and analytical methods. Geochem. Geophys. Geosyst. 15, 2797–2820. Woodcock, A.H., Duce, R.A., Moyers, J.L., 1971. Salt particles and raindrops in Hawaii. J. Atmos. Sci. 28, 1252–1257. Zhu, M., Wadge, G., Holley, R., James, I.N., Clark, P.A., Wang, C.G., Woodage, M.J., 2008. Structure of the precipitable water field over Mount Etna. Tellus A 60A, 679–687.