The crater lake of Copahue volcano (Argentina): geochemical and thermal changes between 1995 and 2015

Mariano R. Agusto, Alberto Caselli, Romina Daga, Johan Varekamp, Alcira Trinelli, María Dos Santos Afonso, María Laura Velez, Pablo Euillades and Sergio Ribeiro Guevara
Geological Society, London, Special Publications, 437, 107-130, 23 December 2016, https://doi.org/10.1144/SP437.16

Abstract

The crater lake and associated hydrothermal features of Copahue volcano have been studied intensively over the last 20 years (1995–2015). The geochemical and isotopic compositions of the waters provide insights into the processes occurring in the volcanic–hydrothermal system, the crater lake and the thermal springs. Variations in the temperature and chemical composition of the waters reveal fundamental changes in the system that precede and accompany the magmatic and phreatic eruptive events at Copahue. A conceptual model of the summit system was developed involving the intrusion of slivers of magma in the hot acidic hydrothermal cell, the saturation of waters with secondary minerals leading to reduced permeability, the blocking of fluid pathways by liquid sulphur during heating events and the transport of gas from the magma through the ductile–brittle transition into the hydrothermal system. Geophysical data were integrated with the chemical data to provide new insights into the behaviour of the deep magmatic system that feeds the volcano edifice. Multidisciplinary studies were used to identify precursory signals of eruptive activity to give an early warning of pending volcanic hazards. Several geochemical ratios in river water were identified as potential indicators of upcoming volcanic activity, which could be used in co-operation with the community and local authorities to deal with these volcanic hazards.

Crater lakes are a relatively common feature in volcanic systems (Varekamp et al. 2000; Christenson et al. 2015). The physical and chemical features of a crater lake strongly depend on the contributions of the magma-derived fluids to the underlying host system. Changes in lake chemistry may therefore be precursory signals of newly increased magmatic or volcanic activity.

Crater lakes also present their own volcanic hazard because the interactions between the lake water and the uprising magma may produce major phreatomagmatic Surtseyan eruptions that threaten the surrounding areas through ash fallout, lake expulsion or lahars (Brown et al. 1989; Cronin et al. 1997; Christenson 2000; Tassi et al. 2005; Christenson et al. 2007, 2010; Rouwet & Morrissey 2015). Sometimes the presence of a sealing zone at the interface between the lake bottom and the underlying magmatic–hydrothermal system may lead to a build-up of pressure (Ohba et al. 2008), possibly followed by limnic explosions. Common sealing minerals in ultra-acidic crater lakes include kaolinite, natroalunite, anhydrite, gypsum, pyrite, barite, anglesite, celestite and amorphous silica expulsed during phreatic eruptions (Christenson & Wood 1993; Delmelle & Bernard 1994; Varekamp et al. 2004; Christenson et al. 2010). A molten pool of sulphur may also act as a seal at the lake bottom; its presence is indicated by sulphur spherules floating on the lake surface (Takano & Watanuki 1989; Delmelle & Bernard 2015). The occurrence of a sealing zone may affect the mass input and pressure of the injected superheated vapour and, consequently, the vigour of phreatic eruptions (Morrissey et al. 2010).

Copahue volcano (37.53° S, 71.10° W, 2997 m a.s.l.) is located in the southwestern part of Caviahue Caldera (Fig. 1), a pull-apart basin in the Patagonian Andean range (Ramos & Folguera 2000; Melnick et al. 2006; Rojas Vera et al. 2010; Sruoga & Consoli 2011). Copahue has been one of the most active volcanoes in Patagonia during the last 25 years and two important tourist villages (Copahue and Caviahue) are located between 5 and 9 km from the active crater (Fig. 1). The volcano hosts an acidic brine-like crater lake with a diameter of c. 300 m (Fig. 2); sulphur slicks periodically float on its surface. Acidic hot springs on the eastern flank, Vertiente 1 and 2 (V1 and V2), located c. 200 m below the crater rim, flow into the Upper Río Agrio (URA), which discharges into Lago Caviahue, a volcanically acidified glacial lake (Varekamp et al. 2001, 2009; Gammons et al. 2005; Parker et al. 2008; Pedrozo et al. 2008; Varekamp 2008). Fluids in the crater lake, hot springs and URA waters derive largely from the volcanic–hydrothermal system (VHS). Downstream of Lago Caviahue, the Lower Río Agrio is progressively diluted with meteoric waters by several tributaries (Fig. 1). Representative tributaries of the acidic river and lake system include the Río Dulce, Pucón Mahuida and Trolope (snowmelt waters). The area NNE of the volcano edifice hosts several hydrothermal emissions of steam-heated waters, which do not seem to be related to direct volcanic degassing of the Copahue conduit, but are structurally controlled and fed by a separate geothermal aquifer (Mas et al. 2000; Agusto et al. 2013b; Tassi et al. 2016).

Fig. 1.
Fig. 1.

Location map of Copahue volcano and Caviahue caldera showing sampling sites of the volcanic–hydrothermal system (VHS) (crater lake (CRL), Vertientes V1 and V2 hot springs, Upper Río Agrio (URA)), streams of snowmelt waters (SNM; Río Dulce, Pucón Mahuida and Río Trolope) and the steam-heated waters (STH) from the geothermal areas. For more details about the sampling point areas, see Agusto (2011). The inset image shows the summit area of Copahue volcano edifice with an emission column from the crater during December 2012 and the springs on the eastern flank (modified from Caselli et al. 2016).

Fig. 2.
Fig. 2.

(a) Photograph and data of Copahue crater lake. (b) Photograph taken from Copahue crater lake showing the Vertientes (V1 and V2) hot springs, the Upper Río Agrio and Lago Caviahue (modified from Agusto et al. 2012).

Although many eruptions have been reported for Copahue volcano during the last century, detailed descriptions of volcanic events only began to be recorded from c. 1993. The cycle in the 1990s started in July 1992 with a phreatic eruption and continued until September 1995 (Delpino & Bermúdez 1993; Bermúdez & Delpino 1995). A major phreatomagmatic event took place in July 2000 (Varekamp et al. 2001, 2009; Delpino & Bermúdez 2002; Naranjo & Polanco 2004). From July to September 2004, the normally hot (30–40°C) crater lake froze over (Caselli et al. 2005; Agusto 2011; Agusto et al. 2012) and then returned to its usual state and remained relatively constant until late 2011, when the discharge rate of fluids from the Copahue summit started to increase. During 2012 the water temperature of the crater lake increased and the water level decreased (Agusto et al. 2013a; Caselli et al. 2013; Agusto & Varekamp 2016), with a phreatic (phreatomagmatic) event in July; a major phreatomagmatic–magmatic eruption finally occurred in December 2012 (Petrinovic et al. 2014; Caselli et al. 2016) with juvenile pumice and ash dispersed over >100 km from the vent. The instability of the VHS and the intermittent restoration of the crater lake continued over the next few years.

This paper reviews the variations in temperature and chemical composition of the Copahue crater lake, the flank Vertiente hot springs and the URA during the last 25 years. A conceptual model is described to explain the observed changes in the hydrological regime of the Copahue summit system and strategies for monitoring and volcanic surveillance are outlined.

Chemical and thermal trends: observations

Chemical and field data have been collected from the Copahue summit area since the late 1990s and data summaries are available in Ouimette (2000), Varekamp et al. (2004, 2009), Gammons et al. (2005), Agusto (2011), Agusto et al. (2012), Kading (2011), Alexander (2014), Varekamp (2015), Agusto & Varekamp (2016), Rodríguez et al. (2016).

The water temperature of the crater lake varied from 21 to 54°C during the period 1997–2000 (Varekamp et al. 2001). The Vertientes hot spring temperatures two years prior to the 2000 eruption were between 63 and 83°C and up to 75°C during the July 2000 events (Varekamp et al. 2001, 2004). However, the temperature of the crater lake water prior to the July 2000 eruption was extremely low: 8°C in January and 5°C in April 2000 (Pedrozo et al. 2008). No information on the lake water temperature is available for the period 2000–03. During May 2004, the lake water temperature was 13.8°C and showed a grey–green colour with sulphur patches floating on the lake surface (Fig. 3a). In July 2004, c. 80% of the lake surface was frozen (Fig. 3b) (Caselli et al. 2005; Agusto 2011; Agusto et al. 2012). During the following months the frozen portion of the lake surface reduced to c. 40%, with yellow patches of sulphur floating on the non-frozen part (Fig. 3c). Freezing of an active crater lake is an extremely unusual feature and had never before been observed at Copahue crater lake. This 2004 freezing event, defined as a negative thermal anomaly (NTA), was not followed by an eruption. By the end of 2004 the crater lake had returned to its usual appearance and during the following years (2005–10) it had temperatures between 30 and 40°C (Fig. 3d, e). During the crater lake NTA, the Vertientes hot springs had the highest temperatures measured since 2003: 81°C in V1 and 69°C in V2. After the NTA, the V1 and V2 temperatures remained relatively constant (TV1=60–70°C; TV2=40–50°C) (Agusto et al. 2012). The hot springs that feed the URA do not affect the temperature of the river, which varies between 2°C in winter and 15°C in summer (Fig. 4). The crater lake and hot spring water temperatures are less sensitive to seasonal temperature changes than the URA (Fig. 4), the waters of which are modified by the contribution of large amounts of snowmelt water from tributary streams during the southern spring–summer period (October–December) and small meteoric water inputs during the low rainfall of the austral summer–autumn period (December–April). It was not always possible to measurement the temperature at the crater lake and at V1 and V2 during winter (2004–09) to document this absence of seasonality for the summit system. During 2004–05 the weather was not anomalous in this region (AIC 2007), thus the 2004 lake NTA cannot be ascribed to climatic conditions.

Fig. 3.
Fig. 3.

(a–d) Photographs of the acid crater lake of Copahue volcano and changes in temperature from May 2004 to December 2005. (e) Detail of sampling point on Copahue crater lake (modified from Agusto et al. 2012).

Fig. 4.
Fig. 4.

Temporal variations in water temperatures of the crater lake (CRL), V1, V2 and Upper Río Agrio (URA) from 2003 to 2012 (modified from Agusto & Varekamp 2016). NTA, negative thermal anomaly.

The 2004 NTA was caused by a decrease in the heat input into the crater lake. An active crater lake can only freeze when the input of magmatic fluid becomes too low to counterbalance the ambient temperature of <0°C during winter. Steam-driven eruptions at Lake Yugama, Kusatsu-Shirane Volcano (Japan) in 1989 and 1996 breached the ice sheet of the lake surface (Ohba et al. 2008).

During 2012 the crater lake temperature increased to 63°C in March and the water level decreased. Two weeks after a phreatic explosion event in July 2012, the temperature of the crater lake was similar (60°C), while the water level continued to fall (Agusto et al. 2013a). During the major 2012 eruption in December (Caselli et al. 2016), the diameter of the crater lake reduced to c. 20 m.

The crater lake water usually has a very low pH (<1) and a high specific conductivity (up to 102 mS cm−1). The crater lake waters had high concentrations of SO42−, Cl− and F−, up to 66 000, 9100 and 950 mg l−1, respectively, prior to the 2000 eruption (Varekamp et al. 2009) and up to 42 000, 18 000 and 2000 mg l−1, respectively, prior to the 2012 eruption (Agusto et al. 2013a). Element concentrations in the V1 and V2 hot springs were also high, but were not identical to the crater lake compositions (Varekamp 2008). Hot spring concentrations of SO42−, Cl− and F− of up to 45 400, 6800 and 500 mg l−1, respectively, were found prior to the 2000 eruption and up to 28 000, 8500 and 600 mg l−1, respectively, prior to the 2012 eruption, with pH values at times well below 1. The URA waters showed a decrease in the specific conductivity from the source (V1 and V2) to Lago Caviahue from 284 to 1.05 mS cm−1 as a result of the input of several tributary streams fed largely by snowmelt (Agusto et al. 2012). These tributaries, e.g. the Pucón Mahuida, had a higher pH and lower specific conductivity (pH 6.3 and specific conductivity 0.03 mS cm−1). The Río Dulce stream largely consists of snowmelt water (pH 6.65, specific conductivity 0.01 mS cm−1) and discharges into Lago Caviahue (CVL), which has pH values ranging from 2.2 to >3 and a lower specific conductivity (1.06 mS cm−1). Previous work has documented a similar pattern of decreases in the concentration of major and trace elements downstream in the URA (Ouimette 2000; Gammons et al. 2005). The behaviour of these species encompasses the changes in this acidic hydrological system from hyperacidic (pH 1–2) to a pH between 2.5 and 3 and ultimately a pH of 6 near Loncopué village (>40 km downstream).

The crater lake, V1, V2 and URA waters (VHS) have a SO42−(Cl−)–Ca2+(Mg2+) composition (Fig. 5), with high contents of Fe and Al, which is clearly different from those of the steam-heated and snowmelt waters (Fig. 5). The steam-heated waters have a HCO3−–Ca2+ and SO42−–Ca2+(Na++K+) composition, while the snowmelt waters have intermediate compositions (Agusto 2011). The VHS waters have Σ anions/salinity molar ratios of 0.8–0.9 and Σ cations/salinity molar ratios of 0.1–0.2, where the anions are SO42−, Cl−, F−, Br− and the cations are Na+, K+, Ca2+ and Mg2+. We used the total anion content (TAC) as an approximation for the total salinity in the following discussions.

Fig. 5.
Fig. 5.

(a) SO42−–Cl−–HCO3− ternary diagram showing the volcanic–hydrothermal system (VHS) water compositions over the SO42−–Cl− axis involving crater lake (CRL) waters with higher Cl− contents, steam-heated waters (STH) waters from geothermal areas near the SO42− and HCO3− corners and snowmelt (SNM) waters showing mixed compositions. (b) The Mg–Ca–(Na+K) ternary diagram shows VHS waters enriched in Mg2+ with respect to the STH waters.

The volcanic elements derived from the input of magmatic gases into the hydrothermal cell (SO2, HCl, HF and minor elements such as As and B) and CO2 probably pass through this acidic system (Agusto et al. 2013b; Tamburello et al. 2015). The S : Cl ratios ranged from 1 to 9 (Fig. 6). Sample values with higher S : Cl ratios may have suffered HCl loss at higher temperatures with associated fractionation of HCl into the vapour phase (Giggenbach 1996; Arnórsson 2000; Symonds et al. 2001).

Fig. 6.
Fig. 6.

The SO42−–Cl− binary diagram showing a trend in dilution of the volcanic–hydrothermal system (VHS) waters from the highest crater lake (CRL) contents to the composition of snowmelt (SNM) waters and steam-heated (STH) waters depleted in Cl− by scrub processes in the geothermal reservoir (Agusto et al. 2013a).

The rock-forming elements (RFE) derive from water–rock interactions between very hot, acidic fluids and volcanic rocks at depth. At temperatures >200°C and pH values near 0, rock dissolution is almost complete (near congruent). Silica does not completely dissolve in the hydrothermal fluids and precipitates while rock dissolution is ongoing, as shown by Varekamp (2015) through isosol diagrams. The 1999 crater lake fluids had c. 15 g rock dissolved per litre where c. 8 g of hydrothermal silica were retained in the solid system.

Saturation indexes on the Vertientes hot springs (V) and crater lake fluids indicate that hydrothermal silica, anhydrite and possibly some Cu sulphides were almost always saturated in the fluids. During some periods the fluids became more concentrated and jarosite and alunite were predicted to be stable phases (Ouimette 2000; Fazio et al. 2008; Varekamp et al. 2009). The latter phases led to reduced permeability in the system (Agusto & Varekamp 2016), resulting in lower element concentrations in the URA and Lake Caviahue.

Isotope composition: δD and δ18O of water, δ34S, 129I and δ7Li

The stable isotope compositions of springs, rivers and lakes at Copahue show a wide range of δD and δ18O values, ranging from nearly pure meteoric waters to highly evolved waters (Ouimette 2000; Varekamp et al. 2001, 2004; Agusto 2011). A full data compilation is given in Agusto & Varekamp (2016). The snowmelt waters show the lowest isotopic values, with δ18O values ranging between −10.8 and −12.1‰ V-SMOW and δD values between −77 and −88‰ V-SMOW (meteoric end-member). The δ18O values for acidic waters ranged between −12.1 and +5.9‰ V-SMOW and the δD values between −90 and −22.5‰ V-SMOW (URA and crater lake, respectively).

The δ18O–δD binary diagram (Fig. 7a) shows the Local Meteoric Water Line (LMWL, δD‰=7.8×δ18O‰−0.21‰) according to the data from the regional precipitation network in Mendoza and the field that corresponds to magmatic fluids (andesitic water) as defined by Taran et al. (1989) and Giggenbach & Corrales Soto (1992), with δ18O and δD values of +10±3 and −20±10‰, respectively. The LMWL differs slightly from the global meteoric water line (Craig 1961) as a result of local atmospheric fractionation processes.

Fig. 7.
Fig. 7.

Stable isotopes of water. (a) δ18O–δD diagram with crater lake (CRL) and V fluids on mixing (dashed) line and Upper Río Agrio (URA) fluids near meteoric water (MW). The Local Meteoric Water Line (LMWL) is also shown (modified from Agusto & Varekamp 2016). (b) Evaporation (dashed line) and mixing (solid line) for CRL and V samples in a Cl−–δ18O diagram (modified from Varekamp 2004). The inset shows the initial mixing of a V-type fluid with true MW (stippled trend), which is moved along the solid arrow to the CRL water array through evaporation.

The crater lake and V water samples plot on a line with a lower slope than the LMWL, which trends towards the andesitic water field (dashed line in Fig. 7a), with the expression: δD‰=3.6×δ18O‰−41.65‰. The isotopic compositions of the crater lake waters are strongly influenced by evaporation and mixing with snowmelt, whereas the V waters are binary mixtures between magmatic water and some meteoric component. Evaporation at c. 30°C creates a trend that coincides with the overall mixing trend between meteoric and magmatic end-members, which is fortuitous. Some crater lake samples approach the isotopic composition of andesitic water (Fig. 7a), but this is as much an effect of evaporation at elevated temperatures as the mixing with a Vertiente component that carries andesitic water.

The Online Isotopes in Precipitation Calculator (OIPC) website (www.waterisotopes.org; Bowen & Revenaugh 2003) provides values for local meteoric waters at Caviahue and Copahue in good agreement with the empirical LMWL (Caviahue, δD=−70‰, δ18O=−9.9‰; Copahue, δD=−90‰, δ18O=−12.6‰). The intersection of the dashed VHS line with the LMWL gives a value similar to the OIPC value for mean local meteoric waters in Caviahue and close to those defined by Panarello (2002) (−80±5‰ for δD; −11.2±0.4‰ for δ18O) as the local meteoric end-member. Meteoric waters sampled in the area (Río Trolope, Río Dulce and the stream Pucón Mahuida, −82±5‰ for δD; −11.3±0.5‰ for δ18O) and of Lake Caviahue and Río Agrio (−86±5‰ for δD; −11.0±0.8‰ for δ18O) plot on the LMWL roughly at the same point as this meteoric water end-member (Agusto & Varekamp 2016).

The VHS line is the result of the evaporation of acidic waters at temperatures above ambient (Varekamp & Kreulen 2000) and the mixing of meteoric and andesitic waters. The V fluids are most likely to be binary mixtures between magmatic fluids (andesitic waters, Fig. 7a) and glacial meltwater that has infiltrated into the volcano superstructure. These meteoric waters become dilute geothermal fluids, which were sampled in the geothermal fluid pipeline near Copahue village (samples GT 1-1, 1-2 and 2-3 with δD=−72 to −81‰ and δ18O=−7.0‰; Varekamp, unpublished data) and show an offset in δ18O of 2–3‰ as a result of water–rock interactions at some elevated temperature (Craig 1963). The V fluids are probably binary mixtures between such geothermal fluids with a strong meteoric signature and the magmatic andesitic waters, with up to c. 60% of magmatic water in the most extreme fluids (Varekamp et al. 2004). The crater lake composition is a mixture between a V-type component from the hydrothermal reservoir and glacial meltwater (pure meteoric end-member), with added evaporation at elevated temperatures.

A Cl v. δ18O diagram (Fig. 7b) shows the two separate trends for the crater lake and V samples more clearly: a mixing line between magmatic water and geothermal water for the V samples (solid line) and a combined mixing and evaporation line (dashed) for the crater lake samples. The inset of Figure 7b shows the initial mixing of a V-type fluid with true meteoric water (stippled trend), which is then moved along the solid arrow to the crater lake water array through evaporation. Exact simulations of these processes are not possible because we do not know the exact composition of the V component as a lake input for each crater lake sample. In addition, the evaporation of Cl-rich waters will also lead to HCl loss (Varekamp & Kreulen 2000; Rouwet & Ohba 2015; Varekamp 2015) and so the evaporative concentration of Cl may be partially offset by HCl escape. The most magmatic V waters provide the closest estimate of the isotopic composition of the primary VHS reservoir fluids, which consist of c. 60% magmatic fluids. The other mineralized waters are either impacted by mixing with V-type fluids and evaporation processes, either at ambient or at elevated temperatures (Agusto & Varekamp 2016).

The sulphur isotopic composition of fluids at Copahue has been measured in several samples at different times. The δ34S for dissolved sulphate in V, the crater lake, URA, CVL and elemental sulphur samples from the slicks on the crater lake, from the crater lake wall and from crater lake sediments were collected and analysed. In general, the δ34S in bisulphate or sulphate from the crater lake waters is high, ranging from +13.7 to +21.2‰ over time (Agusto & Varekamp 2016). Elemental sulphur is much lighter, ranging from −8.2 to −11.4‰. The calculated temperatures of the δ34S system for the VHS range from 250 to 280°C (Varekamp et al. 2004) and probably provide the closest estimate to the true volcanic–hydrothermal reservoir temperature. The δ34S in HSO4− decreases downstream in the URA and CVL, although the effect is close to the data precision. More data are needed to establish the exact reason, but the trend is consistent over the years. This may be the result of the addition of light sulphur through the dilute tributaries of the URA.

Some more exotic isotope species and their elemental values were also analysed in the Copahue hydrothermal fluids, such as 129I and δ7Li, but the number of analyses is very limited. The isotope 129Iodine is cosmogenic with a T1/2=16.5 Ma. After rain-out from the atmosphere into the ocean, the iodine is largely incorporated into organic matter and then ends up in marine sediments. It survives the subduction process and may then be released again during sediment devolatilization in the arc magma generation zone. Measurements of 129I were reported for crater lake, V and CVL as well as local non-volcanic lakes (Fehn et al. 2002). V and crater lake fluid values show excess amounts of 129I, suggesting the presence of a volatile component derived from the degassing of subducted sediments in the magmatic fluids. The presence of a volatile component derived from subducted organic material was also suggested by Agusto et al. (2013b) and Tassi et al. (2016) based on the measurements of 13C CO2 and 15N N2 in fumarolic gas samples from geothermal areas of Copahue.

The crater lake and V fluids have δ7Li values of +8.4 and +3.4‰, respectively (referenced to NIST 8545). 7Li is preferentially removed from the rock matrix during incomplete water–rock interactions, creating higher δ7Li values in the fluid relative to the protolith. Isotopic fractionation of Li has been discussed by Wunder et al. (2006). The value in the crater lake fluids is higher than the mean value in arc rocks of +4±1.5‰ (White 2013) and thus may suggest a less complete degree of rock digestion below the lake. The V fluids may stem from a part of the hydrothermal system where more complete rock digestion takes place (Agusto & Varekamp 2016). Most importantly, these isotopes contribute to the hypothesis that the crater lake and the hot springs draw fluids from spatially diverse sections of the underlying overall hydrothermal system.

Changes in the volcanic–hydrothermal system during eruptive events

1992–95 eruptions

Several volcanic events have been reported for Copahue volcano since the eighteenth century; however, documentation was considerably improved during the 1992 eruptive cycle. The eruptions started during July 1992 and continued with several explosions in 1993 and major eruptions in December 1994 and September 1995. An outstanding feature was the ejection of sealing material from the conduit and floor of crater lake, such as hydrothermal silica and liquid native sulphur (Delpino & Bermúdez 1993, 1995). Unfortunately, no geochemical data on the fluids was collected during this period.

From 1997 to 2000, the Cl− concentrations in the crater lake initially decreased, but during 1999 the concentrations and lake water temperatures increased considerably (Varekamp et al. 2001). By early 2000 the temperature of the crater lake was extremely low (5–8°C), while the Vertientes springs remained at high temperatures (70–75°C). The low crater lake temperature may have been a precursor to the pending 2000 eruption, a result of small intrusions into the hydrothermal reservoir that limited the permeability and flow input into the lake.

2000 eruption

During July 2000 a magmatic eruption took place (VEI 1-2), starting with phreatomagmatic events and continuous degassing between eruptive phases until October 2000. Explosions ejected incandescent bombs and dark ash with chilled sulphur fragments over an area of 50 km (GVN 2000a, b); the emission of sealing material was also reported (Varekamp et al. 2001). The Vertientes hot springs and crater lake showed low pH values with high concentrations of the RFE (e.g. Al, K, Ca, Ti and P) before the event. After the eruption, a significant decrease in these elements was registered, whereas no clear trend was identified for the Fe concentrations. An increase was observed for the pH values in V fluids, whereas the crater lake remained very acidic. An increase in salinity : Cl− was also observed together with higher SO42− : Cl− (Fig. 8). After the eruptive period, the concentrations of Cl− and F− gradually increased with respect to SO42− (Varekamp et al. 2009), possibly consistent with the entry of more degassed magma into the shallow part of the magmatic system in the aftermath of the eruption. The saturation indices of the fluid system showed a higher number of saturated minerals of S, SO42− (e.g. alunite) and silica (Varekamp et al. 2004) in the syn- and post-eruptive fluids.

Fig. 8.
Fig. 8.

Temporal variations of SO42− : Cl− ratios in crater lake (CRL), V and Upper Río Agrio (URA) samples, showing a strong increase in values during 2000 eruption and the 2004 negative thermal anomaly (NTA) (modified from Agusto & Varekamp 2016).

The RFE/Cl− values and the degree of neutralization (Varekamp et al. 2000) of the fluids declined in the crater lake and hot springs during the three years prior to the 2000 eruption, suggesting an increased input of volcanic gas into the system. In addition, increased water–rock values in the hydrothermal system may have occurred as a result of long-term rock dissolution because the rock protolith became covered with liquid sulphur and/or solid silica phases, slowing the water–rock reactions (Varekamp et al. 2001). The degree of neutralization was c. 50%, but then decreased towards lower values by late 1999, also signalling enhanced volcanic gas inputs from rising degassing magma. The rare earth element (REE) patterns in the fluids during this period were enriched in light REEs (LREEs) relative to the rock.

The July 2000 hot spring fluids show a dramatic change: the RFE : Cl− ratios increased, and the RFE ratios in the fluids were different from those in the older Copahue rocks as well as in the 2000 magma (Varekamp et al. 2009). It is hypothesized that the intrusion of fresh magma slivers into the hydrothermal system in 2000 led to higher RFE : Cl− values in the fluids and the water compositions became especially enriched in Mg2+. During the 2000 eruption, the hot spring fluids became more concentrated, the REE pattern became closer to that of the bulk rock and the elemental fluxes of Mg, Na, Fe and Al from the hydrothermal system into the ambient environment increased strongly. The composition of these fluids resulted from the incongruent dissolution of the newly intruded magma with saturation of alunite, anhydrite and silica phases at temperatures ≥150°C (Varekamp et al. 2009). The retention of these secondary phases may explain why the K and Ca concentrations in the fluids increased less than those of Fe, Mg and Na.

In the years following the 2000 eruption, strong variations in the element ratios occurred and element fluxes from the hydrothermal system decreased to small fractions of those prior to the eruption. As a result, Lake Caviahue became much more dilute and its pH steadily increased (Varekamp 2008). The REE patterns in the V fluids became heavily fractionated in both the heavy REEs as well as the LREEs. The LREE depletion is probably caused by the retention of LREEs in secondary precipitates such as alunite (Varekamp 2015). The negative Eu anomaly in the pre-eruptive fluids decreased in magnitude during the 2000 eruption, but in late 2004 the Eu/Eu* value increased.

2004 NTA event

In May 2004 the temperature of the crater lake waters decreased to 13°C and by July 2004 c. 80% of the lake surface was frozen, the first time that this was documented (Fig. 3). The V hot springs displayed higher temperatures, however: 81°C for V1 and 69°C for V2. During the NTA, the TAC : SO42− ratio indicated initially a relative SO42− enrichment followed by a relative increase in Cl− and F− (Agusto et al. 2012). The V2 spring showed an increase in TAC, whereas the TACs of the crater lake and V1 showed the lowest values at that time (Fig. 9). The temperature and TAC decrease in the crater lake and the simultaneous increase in the V2 spring suggest that a decrease in the mass and heat flux occurred in the shallow part of the underlying hydrothermal system (Agusto 2011; Agusto & Varekamp 2016). The crater lake and V1 spring seem to be directly affected by the input of a hot fluid rich in SO42− and Cl−, whereas the V2 spring seems to have an independent fluid input for this period. Simultaneously, the URA water shows a period (stage 1) with a lower TAC and a second period (stage 2) with a higher TAC contemporaneous with increases in the TAC in the crater lake and the V1 spring (Fig. 9). During both stages, the URA waters displayed seasonal behaviour controlled by dilution processes. They showed the highest TAC in the dry summer, a lower TAC in correspondence with snowfall in winter and the lowest TAC during the snowmelt spring–summer period. Note that during the first stage, the URA water composition showed a relatively constant seasonal pattern, which then became disturbed during the NTA interval (Agusto et al. 2012). This behaviour was observed with more detail for any specific RFE and volcanic element in the URA (Fig. 10).

Fig. 9.
Fig. 9.

Temporal variations in the total anion content (TAC) for crater lake (CRL), V1, V2 and Upper Río Agrio (URA) samples (modified from Agusto et al. 2012).

Fig. 10.
Fig. 10.

Temporal variations in Upper Río Agrio (URA) waters. (a) Rock-forming elements (RFE), elements in mg L−1 and (b) volcanic elements (VE), anions in mg L−1. Note the regular seasonal pattern disturbed during 2004 negative thermal anomaly (NTA) (modified from Agusto 2011).

According to these changes in composition, the total mass flow (tonnes per month) of RFE and volcanic elements in URA waters shows variations over time (Fig. 11), with strong increments in element fluxes during the 2000 eruption and the 2004 NTA (Agusto & Varekamp 2016). Access to the summit area and sampling crater lake and V waters during the winter period was difficult; however, the composition of the URA provides essential information on the magmatic–hydrothermal system and can provide an efficient way of monitoring the volcano at this specific location. High-frequency monitoring of SO42− : Cl− and Mg2+ : Cl− ratios in the URA (near Caviahue village, Fig. 1) could provide a first-order volcanic risk indicator of new activity at Copahue volcano (Agusto 2011; Agusto & Varekamp 2016).

Fig. 11.
Fig. 11.

Trend in the total mass fluxes of volcanic elements (VE) and rock-forming elements (RFE) in Upper Río Agrio (URA) waters showing the strong pulse during the 2000 eruption and a smaller scale repeat during the 2004 negative thermal anomaly (NTA) (modified from Agusto & Varekamp 2016).

The trend of ratios of the ionic groups (anions and cations) with respect to the major magmatic species (SO42−, Cl−, F−) allows the identification of changes, filtering the effects of seasonality and dilution by rainfall, variations in the degassing regime of a magmatic system and changes in the water–rock interaction processes (Giggenbach & Glover 1975; Rowe et al. 1992a, b; Christenson & Wood 1993; Delmelle & Bernard 1994). A negative excursion in the anions : SO42− (by mass) ratio was observed during the NTA interval (Fig. 12a) for all members of the system (crater lake, V springs and URA), indicating a relative increase in SO42− compared with the other anions in the waters. After the NTA interval, positive peaks occurred and the whole system stabilized with the anion : SO42− ratio at c. 1.3–1.4. The positive peaks after the NTA interval indicate a relative increase in the remaining significant anions (Cl− and F−). This alternation of peaks in ratios indicates that, during the NTA, SO42− enrichment initially occurred followed by a relative increase in Cl− and F−. As observed during the 2000 eruption, the order of enrichment of magmatic species (anions) is consistent with the order of degassing of these species from a rising melt or degassing fluid (Giggenbach 1996; Symonds et al. 2001; Aiuppa et al. 2002).

Fig. 12.
Fig. 12.

Temporal variations in (a) anions : SO42− ratios and (b) SO42− ratios for crater lake (CRL), Vertiente 1 (V1), Vertiente (V2) and Upper Río Agrio (URA) samples (modified from Agusto 2011).

Saturation indices of the lake water during the NTA increased for gypsum, quartz, cristobalite and amorphous silica (Fazio et al. 2008), probably as a result of the arrival of a very concentrated deep fluid that cooled in the shallow part of the hydrothermal system. The hydrothermal inputs into the lake most likely ceased temporarily during the NTA as a result of partial sealing of the upper conduit and crater floor below the lake due to mineral precipitation in the fractures (Agusto et al. 2012). The permeability of the upper reaches of the hydrothermal system may also have been influenced by the increased viscosity of the molten sulphur present there. The viscosity of liquid sulphur increases at temperatures of 150–250°C (Oppenheimer 1992; Takano et al. 1994b) and a viscous cap of sulphur may have partly isolated the lake from uprising magmatic fluids (Takano et al. 1994a, b; Delmelle et al. 2000; Christenson et al. 2010), which would be an additional explanation for the frozen lake at that time.

A negative surface deformation was also observed from early 2004 on the Copahue volcanic edifice, with a maximum subsidence rate of 2 cm a−1 (Velez et al. 2011). Analytical models indicated that the source of deformation was located at c. 4 km depth below the volcanic edifice, with an estimated decrease in volume of 0.0015 km3 a−1 (Velez et al. 2016). Deflation processes on volcanoes are sometimes associated with the release of magmatic brines from the plastic to the brittle zone (Fournier 2006), resulting in strong degassing through the conduit towards the surface. Such a process could have caused a release of hot and concentrated deep fluids into the shallow Copahue VHS, leading to: (1) an increased TAC in the V springs and URA; (2) secondary mineral precipitation in fractures below the crater lake; (3) an increased temperature of the molten sulphur increasing the viscosity; and (4) decreased permeability in the upper part of the system, ultimately leading to the freezing over of the lake. The increased element fluxes from V during 2004 (Varekamp et al. 2009) may support the contention that a small magmatic intrusion could have also been involved (a failed eruption?), although the deflation of the volcanic edifice is not satisfactorily explained by that scenario.

By the end of 2004 the icy surface of the crater melted again (Fig. 3) and during the following years (2005–10) it had temperatures between 30 and 40°C, whereas the outlet temperature of the two hot springs had decreased and remained relatively constant (TV1=60–70°C; TV2=40–50°C). Between 2005 and 2011, the crater lake showed no significant change in temperature or water level. However, the concentrations of anions related to the input of magmatic gases progressively increased (Fig. 9).

The anomalous behaviour of the system during the NTA interval (2004) is associated with a disturbance of only the shallow section of the hydrothermal environment because no eruptive event occurred. High values of the cations : SO42− (by mass) ratio were observed prior to the NTA and a marked decrease occurred during the NTA interval, stabilizing towards the end of the sampling period (Fig. 12b). The relatively high RFE concentrations suggest increased primary rock leaching before the NTA. During the NTA interval, the cations : SO42− ratios decreased as a result of a marked increase in SO42− and/or the precipitation of secondary minerals, which involved both the leached material and fluids coming from deep environments. Towards the end of the high-frequency sampling period (2009–10), despite a significant increase in the TAC (stage 2 in Fig. 9), the element ratios remained constant (Fig. 12), indicating stabilization of the system.

2012 eruption

Phreatic and phreatomagmatic eruptions occurred in 2012 after 12 years of solfataric activity. A column of water vapour and acidic gases was observed c. 200–300 m above the crater in November–December 2011, indicating an intense fluid discharge rate from the volcano summit. By March 2012, the waters from the hot springs and crater lake showed (1) the highest acidity (pH<0), (2) the highest temperature (c. 65°C), (3) the highest contents of volcanic element species (SO42−, Cl−, F−) since the 2000 eruption (Fig. 9) and (4) a significant decrease in the crater lake water level, probably due to enhanced evaporation.

A phreatic eruption occurred on 17 July 2012, manifested by vertical jets 10 m above the crater lake level. On 19 July 2012, a phreato(magmatic?) eruption occurred with the expulsion of pyroclastic and sulphur-rich material (Fig. 13). During the following months, fumarolic activity and intense bubbling continued and the crater lake waters remained at high temperatures (c. 60°), with high acidity (pH<0) and extremely high electrical conductivity (Agusto et al. 2013a). The lake water level continued to decrease until it became a boiling pool only c. 20 m in diameter just before the December 2012 eruption.

Fig. 13.
Fig. 13.

(a) Phreatic explosion registered on 17 July 2012 at Copahue crater lake. (b) Pyroclastic material ejected during 19 July phreatic event, with evidence of liquid sulphur (modified from Caselli et al. 2016).

Samples obtained from the 19 July 2012 eruptive event correspond to a coarse ash fraction (0.062–2 mm) composed of sulphur particles (23%), glass shards (21%), scoriaceous fragments (4%), abundant whitish irregular argillaceous fragments (19%) and an important proportion of lithics, crystals and crystal fragments (33%) (Caselli et al. 2016). Sulphur particles with a general greenish aspect (with slight variations between yellowish green and greyish green) correspond to the coarser fraction of the collected material (size from <0.5 up to 3–4 mm). They also show distinctive morphologies of globular or spherical particles with inner vesicles, like more or less deformed ‘droplets’, or elongated shapes (Fig. 14a, b), resembling particles erupted as molten material. Under a scanning electron microscope, the external surfaces of these particles show numerous small crystals up to 30 µm in size (Fig. 14c–h) that seem to have been expelled as molten material.

Fig. 14.
Fig. 14.

Pyroclastic sulphur particles from July 2012 eruption. (a) Sulphur particles isolated from the ejected tephra. (b) Detail of the same sulphur particles showing some spherical or drop-like shapes, resembling particles ejected as molten material. (c) Subspherical sulphur particle with irregular surface and (d) surface detail showing the small crystals forming the fragment. (e) Highly irregular sulphur particles with subrounded surfaces, which (f) seem to be formed in a plastic/semi-molten state. (g) Irregular sulphur particles and (h) surface detail showing the small crystals forming the fragment. (a, b) Photographs under binocular magnifying glass; (c–h) images from scanning electron microscopy (Grupo de Caracterización de Materiales, Centro Atómico Bariloche) (modified from Caselli et al. 2016).

A new phreatomagmatic eruption occurred on 22 December 2012 (Caselli et al. 2016) and the erupted pyroclastic material (Fig. 15a, b) contained chilled liquid sulphur clods, suggesting the presence of molten sulphur at the lake bottom or below in the hydrothermal system. The crater lake disappeared after these eruptive events, the hot springs were covered by a pyroclastic surge and the course of the URA was modified. High-temperature fumaroles and liquid sulphur ponds formed inside the crater (Fig. 15c). The maximum measured fumarole temperature (420°C) was close to the local boiling temperature of liquid sulphur (440°C). Material emitted during this event included basaltic–andesitic volcanic bombs and highly vesiculated volcanic fragments with an identical composition.

Fig. 15.
Fig. 15.

The 22 December 2012 eruption. (a) White convective eruptive column reaching c. 3000 m high and dark grey plume observed during the initial phreatomagmatic phase. Detail shows the grey column emitting incandescent pyroclastic material in a ballistic projection (modified from Caselli et al. 2016). (b) Pyroclastic block with evidence of liquid sulphur ejected during 22 December 2012 eruption. (c) The crater lake disappeared during the eruption and left the crater fumaroles exposed; photograph from 30 December 2012.

Two months later the crater lake began to reform and two water pools with intense gas bubbling were observed in March–April 2013. During the two years prior to the 2012 eruption, the frequency of sampling monitoring had decreased and few chemical water data are available from 2011 and 2012.

Conceptual model of the summit Copahue volcanic–hydrothermal system

The available geochemical data for waters from the last two decades of (discontinuous) work at Copahue provide several new insights into the behaviour of the VHS and allow the construction of a conceptual model (Agusto et al. 2012; Agusto & Varekamp 2016). The VHS is disturbed by stringers of new magma prior to and during eruptions. Consequently, an increase in pressure and temperature is registered in the system, leading to a higher flow rate input of the V springs and more voluminous inputs into the crater lake. This is observed in the concentrations of the volcanic element species such as SO42− and Cl−. The stringers of new magma result in a fresh rock protolith that reacts with the hot acidic fluids, which then become enriched in RFEs such as K–Al–Fe–Mg as well as Ti and Mn. From these fluids, secondary mineral saturation is reached, involving sulphate-bearing minerals such as jarosite, alunite, anhydrite/gypsum and hydrothermal silica (Ouimette 2000; Varekamp et al. 2004; Fazio et al. 2008; Varekamp 2015). The precipitation of these minerals reduces the porosity and permeability of the upper part of the system and the occurrence of liquid sulphur fills the pore spaces and vents. An increase in temperature could increase the viscosity of liquid sulphur and could favour a reduced fluid influx into the summit hydrothermal system. According to the geochemical modelling performed in Kawah Ijen and Ruapehu volcanoes, the alteration assemblage of predominantly liquid elemental sulphur with anhydrite, alunite and quartz readily develops in the hydrothermal envelope enclosing the conduit (Delmelle & Bernard 1994, 2015; Christenson et al. 2010). Delpino & Bermúdez (1993) reported the occurrence of pyroclastic sulphur indicating temperatures >120°C and suggested the existence of a molten sulphur pool at the bottom of the crater lake (Varekamp et al. 2001).

A sulphur pool could also have favoured the decrease in permeability during the 2004 NTA. Possible mechanisms for the reduced permeability are mineral precipitation and an increase in liquid sulphur viscosity that led to partial sealing of the upper portion of the conduits. This reduction in permeability after an intrusive event or eruptive period leads to strongly reduced fluxes of many elements, as observed after the 2000 eruption and after the 2004 NTA (Varekamp et al. 2009; Agusto et al. 2012). The latter was not associated with an eruption, but was probably a result of an input of deep magmatic fluids and/or a small magmatic intrusion. This event caused the precipitation of minerals in the high part of the system, cutting off the pathways to the crater lake, which then froze over (Caselli et al. 2005; Agusto 2011; Agusto et al. 2012). The flow-rates of the V hot springs may have first increased (increased element fluxes), but then also decreased when further mineral precipitation reduced the permeability in the system. The sealing of fractures that vent into the crater lake decreased the intensity of convection in the lake (Fig. 16). The lake may have had incipient stratification, with warmer concentrated bottom waters affected by hydrothermal inflow, and cooler, but more dilute, surface waters exposed to the low atmospheric temperatures. The combined effect of a decreased heat input into the lake with enhanced stratification may have led to the frozen lake water surface during the NTA in July 2004 (Fig. 16).

Fig. 16.
Fig. 16.

Conceptual model for fluid circulation patterns at the Copahue crater. (a) Upper volcanic–hydrothermal system affected by permeability changes, an open fracture system and free movement of fluids prior to the negative thermal anomaly. (b) Decreased permeability by fracture sealing affecting the crater lake (CRL) and Vertiente 1 (V1) during the negative thermal anomaly, generating the lake freezing (modified from Agusto et al. 2012).

A decrease in permeability as a result of the obstruction of fractures would lead to a gradual increase in pressure in the shallow VHS. Decompression can occur through (1) a sudden pressure release through a phreatic explosion (the 1992 and 1995 phreatic eruptions) or (2) a more gradual pressure release by changes in the fluid circulation in the summit system (post-NTA period). The return to ‘normal’ degassing then re-establishes the convective mechanism in the lake, leading to thermal and compositional homogenization, manifested in the crater lake by a more normal appearance (Fig. 14a). The recovery of the system occurs with the ongoing flushing of magmatic volatiles through the system and exhaustion of the protolith, with the secondary minerals starting to re-dissolve again (enhanced TAC and Al–K–Fe fluxes) as seen in 2008–09, creating new permeability and enhanced fluid flow (Fig. 9).

When the pressure build-up in the shallow VHS reaches a critical value and the decompression process does not take place by fluid drainage through the flank springs, phreatic explosions are necessary to relieve the excess pressure in the summit VHS (Agusto 2011), as occurred in 1992–95. This hypothesis is consistent with the composition of pyroclastic material ejected during the 1992 and 1995 phreatic explosions (Delpino & Bermúdez 1993, 1995) and the hydrothermally altered fragments involved in 2000 and 2013 eruptions (Delpino & Bermúdez 2002; Caselli et al. 2016).

Correlation with geophysical methods

To better understand the relationship between the different components of the VHS and the origin of the anomalous deep fluid flows, we aimed to approve the conceptual model through seismic (Ibáñez et al. 2008) and deformation surveys of Copahue volcano (Velez et al. 2011). According to Fournier (2006), the continuous emission of magmatic fluids (gases and brines) from the magma chamber to the plastic zone accumulates in lenticular forms at the top, resulting in a deformation process that can be recorded at the surface. However, fluid escapes through a deep self-sealing zone (carapace) are common. This fluid flow can range from a slight diffusion of gases and/or brines through minor fractures, to situations that generate large explosive events (Giggenbach 1997). It is considered that an increase in fluid pressure can generate an effective fracturing of the carapace area, resulting in a strong degassing of lenticular fluid accumulations in the plastic zone, not necessarily accompanied by an injection of magma into the overlying system. Depending on the intensity of fracturing, this can be recorded as seismic events. Degassing results in variations in the geochemistry of the shallow VHS, associated with deflation by decompression (Fournier 2006).

Ibáñez et al. (2008) registered tremors and volcano-tectonic events within the Caviahue depression, identifying peak seismic activity in early 2004. The volcano-tectonic events are related to the activity of minor fractures tens of metres in size at about 3 km depth. The tremor-type events, however, are related to hydrothermal activity as the signals are generated by fluid motion along fractures.

Deformation processes since 2004 indicate deflation of the volcanic edifice (Velez et al. 2011) from the centre towards the northeastern sector of Copahue volcano (Fig. 17). The maximum deformation shows a subsidence of 2 cm a−1 above the volcano edifice. The authors determined that the deformation source was located at about 4 km depth below the NE flank of the volcanic edifice, with an estimated volume change of −0.0015 km3 a−1. The deformation process responds to the release of the fluids accumulated in a lens from the brittle–ductile boundary that surrounds the magmatic zone. The seismic activity recorded inside the caldera is related to the fracturing of this limit, followed by decompression due to the leaking of fluids. Thus we suggest that the NTA time interval was preceded by the increase of seismic activity in January 2004 and accompanied by deflation of the volcano (Fig. 17). The mechanism of steady fluid leakage from the plastic (magmatic) zone affects the shallow VHS. The increase in seismic events in January 2004, related to fracturing at the carapace depth (Ibáñez et al. 2008; Velez et al. 2011), suggests that there could have been an increase in fluid pressure in the plastic zone. This could have generated anomalous brecciation and decompression and a higher magmatic flux through the fracture system of the fragile zone towards the main conduit. This could have caused a release of hyperconcentrated fluids into the shallow VHS, leading to: (1) an increased TAC in the hot springs and Río Agrio; (2) secondary mineral precipitation in the crater lake (Varekamp et al. 2004; Fazio et al. 2008); and (3) decreased permeability by fracture sealing leading to freezing of the lake.

Fig. 17.
Fig. 17.

Mean deformation velocity maps and time series extracted from summit area obtained for the period 2002–07 and between October 2011 and April 2012 (pre-eruptive uplift period). Both cases were obtained by DInSAR processing ENVISAT ASAR images. NTA, negative thermal anomaly.

A different scenario was observed previous to the 2012 eruption. Deformation measurements using ENVISAT images indicate an uplift process located above the volcanic edifice from early 2011 (Velez et al. 2016). A cumulative inflation of about 5 cm was observed from October 2011 until April 2012. Measurements performed with COSMO-Skymed images between December 2011 and January 2013 show a cumulative positive deformation of 17 cm (Velez et al. 2016). Modelling based on these data suggest the presence of a magmatic reservoir at 7.5 and 9 km depth. Thus a volume change at depth resulting from a new magmatic input would be responsible for the deformation measured at the surface. A progressive increase in the magnitude and quantity of seismic signals was also registered at Copahue after the Maule earthquake (Chile, 27 February 2010, MW 8.8), with a significant increase by 2012 (Fig. 18). According to several researchers (Linde & Sacks 1998; Hill et al. 2002; Watt et al. 2009), there is a close relationship between the occurrence of great earthquakes and andesitic volcanic eruptions. On the basis of numerical modelling, Walter & Amelung (2007) suggested that earthquakes of great magnitude induce volumetric expansion in active volcanic areas. Sudden decompression due to overpressure of the magmatic chamber and/or its plumbing system could initiate processes that lead to an eruption. The increased seismic activity registered for Copahue volcano and surrounding areas during 2010 could be a sign that the Maule earthquake was probably also affecting this eruptive centre (Agusto et al. 2013b; Caselli et al. 2016).

Fig. 18.
Fig. 18.

Temporal scheme for Copahue volcano since 2002 showing the correlation between the main eruptive events with seismic activity and surface deformation observed and changes registered at the crater lake. VT, volcano-tectonic.

Conclusions

The last eruptive cycles of Copahue volcano took place from the active crater lake from July 1992 to September 1995 and from July to October 2000; the most recent eruptive cycle began in 2012 and is still ongoing in 2016. The Copahue crater lake is a typical example of a hyperacidic lake, the water composition of which serves as a direct sensor for changes in the underlying VHS and possible eruptive activity. The chemical composition of waters from the VHS and crater lake are distinct from those from the geothermal areas and meteoric waters (steam-heated and snowmelt waters, respectively) and the isotopic compositions provide valuable information about the origin of fluids that feed VHS waters. The available compiled geochemical data provide several new insights into the characteristics and behaviour of this system. The VHS underlying the active crater consists of glacial meltwater acidified by magmatic gases and this mixture dissolves the local host rocks. The eruptive events eject pyroclastic material with abundant liquid sulphur and hydrothermal silica, jarosite, alunite and anhydrite/gypsum fragments, proving the precipitation of these phases in the VHS, which led to certain periods with reduced permeability in the system.

The VHS waters showed an increase in concentration, particularly for the acidic volcanic species (SO42−, Cl− and F−) as a result of ascending magmatic fluids prior to and during the eruptive events. The RFE concentrations also increased, followed in time by a strong decrease in concentrations and element fluxes after the eruptive periods, especially for the elements K, Al and S. Although direct access and sampling of the summit area is difficult during winter time, the URA water composition has been shown to be sensitive to compositional changes in the magmatic–hydrothermal system. This then provides an efficient way for volcano monitoring from Caviahue village. Multidisciplinary research performed at Copahue volcano during the last two decades has allowed us to determine the relationship between geochemical variations, seismicity and surface deformation. These projects provided a baseline for understanding the behaviour during the quiescent phase, the post-2000 eruption and the detection of anomalous conditions during 2004, 2011 and 2012 and contributed to improved volcanic surveillance techniques.

Acknowledgments

We thank the Universidad de Buenos Aires (UBACyT 01-W172, UBACyT 20020120300077BA and UBACyT 20020150200230BA) and ANPCyT (PICT-2015-3110) for funding. Ana Monasterio (Ente de Termas del Neuquén), Instituto IDEAN (UBA-CONICET), Adrian Arias (Defensa Civil Caviahue-Copahue), Christian Díaz and ‘Caniche’ are warmly thanked for their assistance in the field. The authors are grateful to the constructive discussions with Franco Tassi, Bruno Capaccioni, Dmitri Rouwet, Orlando Vaselli, Giovanni Chiodini, Sergio Calabrese, Giancarlo Tamburello and Christopher Gammons, which have improved our knowledge of this system. Special thanks to Pablo Forte, Clara Lamberti, Alejandra Gaviria Reyes, Caterina Liccioli, Joaquín Llano, Juliana Szentiványi and Victoria Nogués from Grupo de Estudio y Seguimiento de Volcanes Activos for their indispensable help. We thank Corentin Cuadron, Takeshi Obha and an anonymous reviewer for their comments.

References

  1. ↵
    Agusto M. 2011. Estudio geoquímico de los fluidos volcánicos e hidrotermales del Complejo Volcánico Copahue Caviahue y su aplicación para tareas de seguimiento. PhD thesis, Universidad de Buenos Aires.
  2. ↵
    Agusto M. & Varekamp J.C. 2016. The Copahue volcano-hydrothermal system and applications for volcanic surveillance. In: Tassi F., Vaselli O. & Caselli A. (eds) Copahue Volcano. Springer, Heidelberg, 199–238.
  3. ↵
    Agusto M., Tassi F. et al. 2012. Seguimiento geoquímico de las aguas ácidas del sistema volcán Copahue – Río Agrio: posible aplicación para la identificación de precursores eruptivos. Revista de la Asociación Geológica Argentina, 69, 481–495.
    OpenUrl
  4. ↵
    Agusto M., Tassi F. et al. 2013a. Thermal and chemical changes in the crater lake of Copahue volcano (Argentina) prior to the December 2012 phreatomagmatic eruption. In: Forecasting Volcanic Activity – Reading and translating the messages of nature for society. IAVCEI 2013 Scientific Assembly, 20–24 July, 2013, Kagoshima, Japan. Acta, 1019, 4A2_3I-O11.
  5. ↵
    Agusto M., Tassi F. et al. 2013b. Gas geochemistry of the magmatic-hydrothermal fluid reservoir in the Copahue–Caviahue Volcanic Complex (Argentina). Journal of Volcanology and Geothermal Research, 257, 44–56.
    OpenUrl
  6. ↵
    AIC 2007. Autoridad Interjurisdiccional de Cuencas de los Ríos Limay, Neuquén y Negro. Datos meteorológicos de la región de Caviahue entre los años 2003 y 2007 Cipolletti, Río Negro, www.aic.gob.ar/aic/default.aspx#v
  7. ↵
    Aiuppa A., Federico C. et al. 2002. S, Cl and F degassing as an indicator of volcanic dynamics: the 2001 eruption of Mount Etna. Geophysical Research Letters, 29, https://doi.org/10.1029/2002GL015032
  8. ↵
    Alexander E.W. 2014. Aqueous geochemistry of an active magmato-hydrothermal system: Copahue Volcano, Río Agrio, and Lake Caviahue, Neuquén, Argentina. BA thesis, Wesleyan University.
  9. ↵
    Arnórsson S. 2000. Isotopic and Chemical Techniques in Geothermal Exploration, Development and Use. Sampling Methods, Data Handling and Interpretation. International Atomic Energy Agency, Vienna.
  10. ↵
    Bermúdez A. & Delpino D. 1995. Mapa de los peligros potenciales en el área del Volcán Copahue – sector Argentino. Volcanic Hazard Map. Geological Survey of the Province of Neuquen, Argentina.
  11. ↵
    Bowen G.J. & Revenaugh J. 2003. Interpolating the isotopic composition of modern meteoric precipitation. Water Resources Research, 39, 1299, https://doi.org/10.1029/2003WR002086
    OpenUrlCrossRef
  12. ↵
    Brown G., Rymer H. et al. 1989. Energy budget analysis for Poás crater lake: implications for predicting volcanic activity. Nature, 339, 370–373.
    OpenUrlCrossRef
  13. ↵
    Caselli A., Agusto M. & Fazio A. 2005. Cambios térmicos y geoquímicos del lago cratérico del volcán Copahue (Neuquén): posibles variaciones cíclicas del sistema volcánico. XVI° Congreso Geológico Argentino, 20–23 September, 2005, La Plata, Argentina. Acta, 1, 751–756.
  14. ↵
    Caselli A., Velez M.L. et al. 2013. Erupción del volcán Copahue (Argentina): evolución, productos e impacto social y ambiental. Foro Internacional sobre Peligros Geológicos, 16–18 August 2013, Arequipa, Peru, Acta, 104–109.
  15. ↵
    Caselli A., Agusto M. et al. 2016. The 2012 eruption. In: Tassi F., Vaselli O. & Caselli A. (eds) Copahue Volcano. Springer, Heidelberg, 61–80.
  16. ↵
    Christenson B.W. 2000. Geochemistry of fluids associated with the 1995–1996 eruption of Mt. Ruapehu, New Zealand: signatures and processes in the magmatic hydrothermal system. Journal of Volcanology and Geothermal Research, 97, 1–30.
    OpenUrlCrossRefWeb of Science
  17. ↵
    Christenson B.W. & Wood C.P. 1993. Evolution of a vent-hosted hydrothermal system beneath Ruapehu Crater Lake, New Zealand. Bulletin of Volcanology, 55, 547–565.
    OpenUrlCrossRefWeb of Science
  18. ↵
    Christenson B.W., Werner C.A. et al. 2007. Hazards from hydrothermally sealed volcanic conduits. EOS, 88, 53–55.
    OpenUrl
  19. ↵
    Christenson B.W., Reyes A.G. et al. 2010. Cyclic processes and factors leading to phreatic eruption events: insights from the 25 September 2007 eruption through Ruapehu Crater Lake, New Zealand. Journal of Volcanology and Geothermal Research, 191, 15–32.
    OpenUrl
  20. ↵
    Christenson B.W., Nemeth K. et al. 2015. Volcanic lakes. In: Rouwet D., Christenson B.W., Tassi F. & Vandemeulebrouck J. (eds) Volcanic Lakes. Springer, Heidelberg, 1–20.
  21. ↵
    Craig H. 1961. Isotopic variations in meteoric waters. Science, 133, 1702–1703.
    OpenUrlAbstract/FREE Full Text
  22. ↵
    Craig H. 1963. The isotope geochemistry of water and carbon in geothermal areas: nuclear geology in geothermal areas. In: Tongiorgi E. (ed.) Consiglio Nazionale della Richerche. Laboratorio de Geologia Nucleare, Spoleto, Pisa, 17–53.
  23. ↵
    Cronin S.J., Hodgson K.A. et al. 1997. 1995 Ruapehu lahars in relation to the late Holocene lahars of Whangaehu River, New Zealand. New Zealand Journal of Geology and Geophysics, 40, 507–520.
    OpenUrlWeb of Science
  24. ↵
    Delmelle P. & Bernard A. 1994. Geochemistry, mineralogy, and chemical modeling of the acid crater lake of Kawah Ijen Volcano, Indonesia. Geochimica et Cosmochimica Acta, 58, 2445–2460.
    OpenUrlCrossRefWeb of Science
  25. ↵
    Delmelle P. & Bernard A. 2015. The remarkable chemistry of sulfur in hyper-acid crater lakes: a scientific tribute to Bokuichiro Takano and Minoru Kusakabe. In: Rouwet D., Christenson B.W., Tassi F. & Vandemeulebrouck J. (eds) Volcanic Lakes. Springer, Heidelberg, 239–259.
  26. ↵
    Delmelle P., Bernard A. et al. 2000. Geochemistry of the magmatic hydrothermal system of Kawah Ijen volcano, east Java, Indonesia. Journal of Volcanology and Geothermal Research, 97, 31–53.
    OpenUrlCrossRefWeb of Science
  27. ↵
    Delpino D. & Bermúdez A. 1993. La actividad del Volcan Copahue durante 1992: erupcion con emisiones de azufre piroclastico, Provincia del Neuquen, Argentina. XII Congreso Geologico Argentino y II Congreso de Exploracion de Hidrocarburos, 19–22 August 1993, Mendoza, Argentina. Acta, 292–301.
  28. ↵
    Delpino D. & Bermúdez A. 1995. Eruptions of pyroclastic sulfur at crater lake of Copahue Volcano, Argentina. International Union of Geodesy and Geophysics, XXI General Assembly, 2–14 July 1995, Boulder, USA. Acta, 128.
  29. ↵
    Delpino D. & Bermúdez A. 2002. La erupción del volcán Copahue del año 2000. Impacto social y al medio natural. Provincia del Neuquen, Argentina. XV Congreso Geologico Argentino, 20–26 April 2002, El Calafate, Argentina. Acta, 377–382.
  30. ↵
    Fazio A., Agusto M., Farías S. & Caselli A. 2008. Evaluación de posibles fases minerales en equilibrio en el sistema volcánico Copahue (Neuquén) y su vinculación con parámetros químicos. XVII Congreso Geológico Argentino, 7–10 October 2008, San Salvador de Jujuy, Argentina. Acta, 1343–1344.
  31. ↵
    Fehn U., Snyder G. & Varekamp J.C. 2002. Detection of recycled marine sediment components in crater lake fluids using 129I. Journal of Volcanology and Geothermal Research, 115, 451–460.
    OpenUrlCrossRefWeb of Science
  32. ↵
    Fournier R.O. 2006. Hydrothermal systems and volcano geochemistry. In: Dzurisin D. (ed.) Volcano Deformation. Springer, Heidelberg, 153–194.
  33. ↵
    Gammons C., Wood S. et al. 2005. Hydrogeochemistry and rare earth element behavior in a volcanically acidified watershed in Patagonia, Argentina. Chemical Geology, 222, 249–267.
    OpenUrlCrossRefWeb of Science
  34. ↵
    Giggenbach W.F. 1996. Chemical composition of volcanic gases. In: Scarpa W.F. & Tilling R. (eds) Monitoring and Mitigation of Volcano Hazards. Springer, Berlin, 222–256.
  35. ↵
    Giggenbach W.F. 1997. The origin and evolution of fluids in magmatic-hydrothermal systems. In: Barnes H.L. (ed.) Geochemistry of Hydrothermal Ore Deposits. Springer, Berlin, 737–796.
  36. ↵
    Giggenbach W.F. & Corrales Soto R. 1992. Isotopic and chemical composition of water and steam discharges from volcanic-magmatic-hydrothermal systems of the Guanacaste geothermal province, Costa Rica. Applied Geochemistry, 7, 309–332.
    OpenUrlWeb of Science
  37. ↵
    Giggenbach W.F. & Glover R.B. 1975. The use of chemical indicators in the surveillance of volcanic activity affecting the crater lake on Mt. Ruapehu, New Zealand. Bulletin of Volcanology, 39, 70–81.
    OpenUrl
  38. ↵
    GVN 2000a. Frequent ash explosions and acidic mudflows starting on July 1. Bulletin of Global Volcano Network, 25, 6.
    OpenUrl
  39. ↵
    GVN 2000b. Continued ash explosions and tremor during August–October. Bulletin of Global Volcano Network, 25, 9.
    OpenUrl
  40. ↵
    Hill D.P., Dzurisin D. et al. 2002. Response Plan for Volcano Hazards in the Long Valley Caldera and Mono Craters Region. US Geological Survey Bulletin 2185.
  41. ↵
    Ibáñez J.M., Del Pezzo E. et al. 2008. Volcanic tremor and local earthquakes at Copahue volcanic complex, Southern Andes, Argentina. Journal of Volcanology and Geothermal Research, 174, 284–298.
    OpenUrl
  42. ↵
    Kading T.J. 2011. Natural pollutant attenuation by Schwertmannite at Copahue volcano, Argentina. MA thesis, Wesleyan University.
  43. ↵
    Linde A.T. & Sacks L.S. 1998. Triggering of volcanic eruptions. Nature, 395, 888–890.
    OpenUrlCrossRefWeb of Science
  44. ↵
    Mas L., Mas G. & Bengochea L. 2000. Heatflow of Copahue geothermal field, its relation with tectonic scheme. In: Proceedings of World Geothermal Congress, 1–10 June 2000, Kyushu-Tohoku, Japan. Acta, 1419–1424.
  45. ↵
    Melnick D., Rosenau M., Folguera A. & Echtler H. 2006. Neogene tectonic evolution of the Neuquen Andes western flank (37–39°S). In: Kay S.M. & Ramos V.A. (eds) Evolution of an Andean Margin: a Tectonic and Magmatic View from the Andes to the Neuquen Basin (35°–39°S lat). Geological Society of America, Special Papers, 407, 73–95.
    OpenUrlAbstract/FREE Full Text
  46. ↵
    Morrissey M., Gisler G., Weaver R. & Gittings M. 2010. Numerical model of crater lake eruptions. Bulletin of Volcanology, 72, 1169–1178.
    OpenUrl
  47. ↵
    Naranjo J.A. & Polanco E. 2004. The 2000 AD eruption of Copahue Volcano, Southern Andes. Revista Geológica de Chile, 31, 279–292.
    OpenUrlWeb of Science
  48. ↵
    Ohba T., Hirabayashi J. & Nogami K. 2008. Temporal changes in the chemistry of lake water within Yugama Crater, Kusatsu-Shirane Volcano, Japan: implications for the evolution of the magmatic-hydrothermal system. Journal of Volcanology and Geothermal Research, 178, 131–144.
    OpenUrl
  49. ↵
    Oppenheimer C. 1992. Sulphur eruptions at Volcán Poás, Costa Rica. Journal of Volcanology and Geothermal Research, 49, 1–21.
    OpenUrlCrossRefWeb of Science
  50. ↵
    Ouimette A.P. 2000. Hydrothermal processes at an active volcano, Copahue, Argentina. MA thesis, Wesleyan University.
  51. ↵
    Panarello H.O. 2002. Características isotópicas y termodinámicas de reservorio del campo geotérmico Copahue-Caviahue, provincia del Neuquén. Revista de la Asociación Geológica Argentina, 57, 182–194.
    OpenUrl
  52. ↵
    Parker S., Gammons C., Pedrozo F. & Wood S. 2008. Diel changes in metal concentrations in a geogenically acidic river: Rio Agrio, Argentina. Journal of Volcanology and Geothermal Research, 178, 213–223.
    OpenUrl
  53. ↵
    Pedrozo F., Temporetti P., Beamud G. & Diaz M. 2008. Volcanic nutrient inputs and trophic state of Lake Caviahue, Patagonia, Argentina. Journal of Volcanology and Geothermal Research, 178, 205–212.
    OpenUrl
  54. ↵
    Petrinovic I., Villarosa G. et al. 2014. La erupción del 22 de diciembre de 2012 del volcán Copahue, Neuquén, Argentina: Caracterización del ciclo eruptivo y sus productos. Revista de la Asociación Geológica Argentina, 71, 161–173.
    OpenUrl
  55. ↵
    Ramos V.A. & Folguera A. 2000. Tectonic evolution of the Andes of Neuquén: constraints derived from the magmatic arc and foreland deformation. In: Veiga G.D., Spalletti L.A., Howell J.A. & Schwartz E. (eds) The Neuquén Basin, Argentina: a Case Study in Sequence Stratigraphy and Basin Dynamics. Geological Society, London, Special Publications, 252, 15–35, https://doi.org/10.1144/GSL.SP.2005.252.01.02
    OpenUrl
  56. ↵
    Rodríguez A., Varekamp J.C., van Bergen M.J., Kading T.J., Oonk P., Gammons C.G. & Gilmore M. 2016. Acid rivers and lakes at Caviahue-Copahue volcano as potential terrestrial analogues for aqueous paleo-environments on Mars. In: Tassi F., Vaselli O. & Caselli A. (eds) Copahue Volcano. Springer, Heidelberg, 141–174.
  57. ↵
    Rojas Vera E.A., Folguera A. et al. 2010. Neogene to Quaternary extensional reactivation of a fold and thrust belt: the Agrio belt in the Southern Central Andes and its relation to the Loncopué trough (38°–39°S). Tectonophysics, 92, 279–294.
    OpenUrl
  58. ↵
    Rouwet D. & Morrissey M.M. 2015. Mechanisms of crater lake breaching eruptions. In: Rouwet D., Christenson B.W., Tassi F. & Vandemeulebrouck J. (eds) Volcanic Lakes. Springer, Heidelberg, 73–92.
  59. ↵
    Rouwet D. & Ohba T. 2015. Isotope fractionation and HCl partitioning during evaporative degassing from active crater lakes. In: Rouwet D., Christenson B.W., Tassi F. & Vandemeulebrouck J. (eds) Volcanic Lakes. Springer, Heidelberg, 179–200.
  60. ↵
    Rowe G.L. Jr.. Brantley S.L. et al. 1992a. Fluid–volcano interaction in an active stratovolcano: the crater lake system of Poás volcano, Costa Rica. Journal of Volcanology and Geothermal Research, 49, 23–51.
    OpenUrlCrossRefWeb of Science
  61. ↵
    Rowe G.L. Jr.. Ohsawa S. et al. 1992b. Using crater lake chemistry to predict volcanic activity at Poás volcano, Costa Rica. Bulletin of Volcanology, 54, 494–503.
    OpenUrlCrossRefWeb of Science
  62. ↵
    Sruoga P. & Consoli V. 2011. Volcán Copahue. In: Leanza H., Arregui C., Carbone O., Danieli J. & Vallés J. (eds) Relatorio XVIII Congreso Geológico Argentino, 2–6 May 2011, Neuquén, Argentina. Asociación Geológica Argentina, Neuquén, 613–620.
  63. ↵
    Symonds R., Gerlach T. & Reed M. 2001. Magmatic gas scrubbing: implications for volcano monitoring. Journal of Volcanology and Geothermal Research, 108, 303–341.
    OpenUrlCrossRefWeb of Science
  64. ↵
    Takano B. & Watanuki K. 1989. Monitoring of volcanic eruptions at Yugama crater lake by aqueous sulfur oxyanions. Journal of Volcanology and Geothermal Research, 40, 71–87.
    OpenUrl
  65. ↵
    Takano B., Ohsawa S. & Glover R. 1994a. Surveillance of Ruapehu Crater Lake, New Zealand, by aqueous polythionates. Journal of Volcanology and Geothermal Research, 60, 29–57.
    OpenUrl
  66. ↵
    Takano B., Saitoh H. & Takano E. 1994b. Geochemical implications of subaqueous molten sulfur at Yugama crater lake, Kusatsu-Shirane volcano, Japan. Geochemical Journal, 28, 199–216.
    OpenUrlWeb of Science
  67. ↵
    Tamburello G., Agusto M.R. et al. 2015. Intense magmatic degassing through the lake of Copahue volcano, 2013–2014. Journal of Geophysical Research: Solid Earth, 120, 6071–6084.
    OpenUrl
  68. ↵
    Taran Y., Pokrovsky B. & Dubik Y. 1989. Isotopic composition and origin of water from andesitic magmas. Doklady (Translations) Academy of Sciences USSR, 304, 440–443.
    OpenUrl
  69. ↵
    Tassi F., Vaselli O. et al. 2005. The hydrothermal-volcanic system of Rincon de la Vieja volcano (Costa Rica): a combined (inorganic and organic) geochemical approach to understanding the origin of the fluid discharges and its possible application to volcanic surveillance. Journal of Volcanology and Geothermal Research, 148, 315–333.
    OpenUrlCrossRefWeb of Science
  70. ↵
    Tassi F., Agusto M., Vaselli O. & Chiodini G. 2016. Geochemistry of the magmatic-hydrothermal fluid reservoir of Copahue volcano (Argentina): insights from the chemical and isotopic features of fumarolic discharges. In: Tassi F., Vaselli O. & Caselli A. (eds) Copahue Volcano. Springer, Heidelberg, 119–140.
  71. ↵
    Varekamp J.C. 2004. Copahue Volcano: a modern terrestrial analog for the opportunity landing site. Eos, 85, 401–407.
    OpenUrl
  72. ↵
    Varekamp J.C. 2008. The acidification of glacial Lake Caviahue, Province of Neuquen, Argentina. Journal of Volcanology and Geothermal Research, 178, 184–196.
    OpenUrl
  73. ↵
    Varekamp J.C. 2015. The chemical composition and evolution of volcanic lakes. In: Rouwet D., Christenson B.W., Tassi F. & Vandemeulebrouck J. (eds) Volcanic Lakes. Springer, Heidelberg, 93–123.
  74. ↵
    Varekamp J.C. & Kreulen R. 2000. The stable isotope geochemistry of volcanic lakes: examples from Indonesia. Journal of Volcanology and Geothermal Research, 97, 309–327.
    OpenUrlCrossRefWeb of Science
  75. ↵
    Varekamp J.C., Pasternack G. & Rowe G. 2000. Volcanic lake systematics. II. Chemical constraints. Journal of Volcanology and Geothermal Research, 97, 161–179.
    OpenUrlCrossRefWeb of Science
  76. ↵
    Varekamp J.C., Ouimette A. et al. 2001. Hydrothermal element fluxes from Copahue, Argentina: a ‘beehive’ volcano in turmoil. Geology, 29, 1059–1062.
    OpenUrlAbstract/FREE Full Text
  77. ↵
    Varekamp J.C., Ouimette A. & Kreulen R. 2004. The magmato-hydrothermal system of Copahue volcano, Argentina. In: Wanty R.B. & Seal R.B. II. (eds) Water–Rock Interaction. Vol. 1. Balkema, Leiden, 215–218.
    OpenUrl
  78. ↵
    Varekamp J.C., Ouimette A.P., Hermanm S.W., Flynn K.S., Bermúdez A. & Delpino D. 2009. Naturally acid waters from Copahue volcano, Argentina. Applied Geochemistry, 24, 208–220.
    OpenUrlCrossRefWeb of Science
  79. ↵
    Velez M.L., Euillades P. et al. 2011. Deformation of Copahue volcano: inversion of InSAR data using a genetic algorithm. Journal of Volcanology and Geothermal Research, 202, 117–126.
    OpenUrlCrossRefWeb of Science
  80. ↵
    Velez M.L., Euillades P., Blanco M. & Euillades L. 2016. Ground deformation between 2002 and 2013 from InSAR observations. In: Tassi F., Vaselli O. & Caselli A. (eds) Copahue Volcano. Springer, Heidelberg, 175–198.
  81. ↵
    Walter T.R. & Amelung F. 2007. Volcanic eruptions following M9 megathrust earthquakes: implications for the Sumatra-Andaman volcanoes. Geology, 35, 539–542.
    OpenUrlAbstract/FREE Full Text
  82. ↵
    Watt S.F., Pyle D.M. & Mather T.A. 2009. The influence of great earthquakes on volcanic eruption rate along the Chilean subduction zone. Earth and Planetary Science Letters, 277, 399–407.
    OpenUrlCrossRefWeb of Science
  83. ↵
    White W.M. 2013. Geochemistry. Blackwell-Wiley, Oxford.
  84. ↵
    Wunder B., Meixner A., Romer R.L. & Heinrich W. 2006. Temperature-dependent isotopic fractionation of lithium between clinopyroxene and high-pressure hydrous fluids. Contributions to Mineralogy and Petrology, 151, 112–120.
    OpenUrlCrossRefWeb of Science
View Abstract
Back to top
Alerts
Citation tools
Permissions
View PDF
Email to
Print
Download PPT

Related Articles

Similar Articles

Cited By...