Source identification of volcanic ashes by geochemical analysis of well preserved lacustrine tephras in Nahuel Huapi National Park

This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

Author's personal copy

Source identification of volcanic ashes by geochemical analysis of wellpreserved lacustrine tephras in Nahuel Huapi National Park

Romina Daga a,�, Sergio Ribeiro Guevara a, Marıa Lidia Sanchez b, Marıa Arribere a

a Laboratorio de Analisis por Activacion Neutronica, Centro Atomico Bariloche, 8400 Bariloche, Argentinab Universidad Nacional de Rıo Cuarto, 5800 Rıo Cuarto, Argentina

a r t i c l e i n f o

Article history:

Received 24 December 2007

Received in revised form

17 March 2008

Accepted 17 March 2008

Keywords:

Geochemistry

Tephra

Tephrochronology

Neutron activation analysis210Pb dating137Cs dating

Northern Patagonia Andean range

a b s t r a c t

Well preserved volcanic ashes produced in recent events, recovered from sedimentary sequences

extracted from three lakes belonging to Nahuel Huapi National Park, Northern Patagonia, were

geochemically characterized in order to reveal patterns that allow the identification of the source. Two

water bodies are situated in the direct impact area of volcanoes Calbuco and Puyehue-Cordon Caulle,

while the third, lake Moreno, is situated in-between. The sedimentary sequences were dated by 210Pb

and 137Cs techniques; the elemental composition was determined by Instrumental Neutron Activation

Analysis. Distinctive patterns were found out when comparing glass shards, and also white pumice but

in a lesser degree, of tephra layers extracted from lake Ilon, related mostly to volcano Calbuco events,

and lake Nahuel Huapi (Brazo Rincon site), associated to Puyehue-Cordon Caulle eruptions. The

geochemical parameters that showed decisive differences were SiO2 and Na2O+K2O contents, the Eu

anomaly, Rare Earth element ratios; the concentration of incompatible elements Cs, Rb, Th, Hf, Ta and Zr,

and the compatible elements Cr and V. The six upper tephra layers extracted from lake Moreno showed

geochemical patterns that allowed clear association with Calbuco and Puyehue-Cordon Caulle sources.

These results set up the base for tephrochronological applications in historical periods in Nahuel Huapi

National Park area.

& 2008 Elsevier Ltd. All rights reserved.

1. Introduction

Tephrochronology, the association of well preserved volcanicash layers with events of known date (Haberle and Lumley, 1998;Ortega-Guerrero and Newton, 1998; Shane, 2000; Chambers et al.,2004; Wastegard, 2005; Fontaine et al., 2007; McHenry et al., inpress), is a powerful tool to date sedimentary sequences speciallyin Patagonia, where the 210Pb dating technique used for recentrecords is limited due to low 210Pb fluxes (Ribeiro Guevara et al.,2003; Arnaud et al., 2006). The Northern Patagonia Andean Range(401150 to 411250 South latitude, and 711 to 721450 West longitude;Fig. 1) is an ideal region for tephrochronological applications sinceit is located in the Southern Volcanic Zone (SVZ) of the Andesmountains, with numerous active volcanoes, namely Llaima,Villarrica, Carran-Los Venados complex, Rininahue, Puyehue-Cordon Caulle complex, Puntiagudo, Osorno, and Calbuco, withfrequent volcanic eruptions registered in historical descriptions ofthe zone since the 18th century. At the same time, the highfrequency of volcanic events demands an accurate characteriza-tion of the volcanic products generated by each source, as well as

the evaluation of their spatial distribution. At present, a thoroughand complete geochemical and morphological characterization ofthe volcanic ashes produced in recent events by the sourcesbelonging to Chilean regions has not been achieved, therefore notallowing clear tephrochronological correlations, nor an evaluationof their spatial distribution.

Tephrochronological correlations have been carried out inlakes belonging to the Nahuel Huapi National Park, NorthernPatagonia (Fig. 1) related to environmental applications, consider-ing the study region impacted principally by the Puyehue-CordonCaulle source. (Massaferro et al., 2005; Chapron et al., 2006). Butevidence of a second source in a sedimentary sequence collectedfrom a branch of lake Nahuel Huapi, in the region of directincidence of Puyehue-Cordon Caulle source (Fig. 1), obtained in aprevious work (Daga et al., 2006) put into discussion thisassumption. The present work studies tephra layers from sedi-ment sequences, dated by 210Pb and 137Cs techniques (Joshi andShukla, 1991; Robbins and Herche, 1993; Ribeiro Guevara andArribere, 2002; Ribeiro Guevara et al., 2003), collected from twolakes belonging to the Nahuel Huapi National Park. The waterbodies selected may receive, according to their geographicalsituation and considering that in this area the predominant winddirection is from northwest towards southwest (Fig. 1), contribu-tions from the events of Calbuco and Osorno volcanoes,

ARTICLE IN PRESS

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/apradiso

Applied Radiation and Isotopes

0969-8043/$ - see front matter & 2008 Elsevier Ltd. All rights reserved.

doi:10.1016/j.apradiso.2008.03.009

� Corresponding author. Tel.: +54 2944 445106; fax: +54 2944 445299.

E-mail address: [email protected] (R. Daga).

Applied Radiation and Isotopes 66 (2008) 1325– 1336

Author's personal copy

alternative sources to Puyehue-Cordon Caulle, in the case of thehigh altitude lake Ilon, and the combination of the three potentialsources in the case of lake Moreno (Fig. 1).

In the SVZ, the Nazca plate subduct under South Americanplate with a currently active volcanic arc extending from 331S to461S. Cordon Caulle volcanic complex (2236 m above see level;Fig. 1), part of Puyehue-Cordon Caulle system, strikes in anorthwest–southeast direction between the Cordillera Nevadacaldera at NW, and the Puyehue stratovolcano at SE (Gerlach et al.,1988). Cordon Caulle volcanic complex includes various fissurevents with aligned domes and pyroclastic cones (Lara et al., 2004),with abundance of silicic magma types, feature unique among themostly bimodal centres along the SVZ where basalts widelypredominate over more silica-rich rocks. The most recenteruptions (1921–1922 and 1960) consist of rhyodacitic andrhyolitic mainly. Osorno volcano (2652 m above see level; Fig. 1)is a Late Pleistocene to recent composite stratovolcano, mainlybasaltic with minor dacitic material. The prehistoric and historicactivity is related to the main cone and younger parasitic cones onthe eastern and western flank, with basaltic products exclusively(Lopez-Escobar et al., 1992). Calbuco volcano (2003 m above seelevel; Fig. 1) is a Late Pleistocene–Holocene active compositestratovolcano, unique by having erupted andesite mainly over its150,000 yr. The most recent activity consists of a plinian eruptionin 1893–1894, and a young historic dome-cone and lava flows,developed during 1917, 1929 and 1961 eruptions (Lopez-Escobaret al., 1992, 1995).

Tephra layers are composed of pumice and scoria particles,crystals or crystal fragments, blocky juvenile clasts, glass shards

and lithic fragments (Westgate and Gorton, 1981; Ortega-Guerrero and Newton, 1998; Mc Phie et al., 1993). Volcanic glassand crystals provide the most valuable information to unravel thestyle and source of the volcanic event. Crystal fragment popula-tions may reflect variations according to size or density duringtransport and deposition especially in fallout deposits. For thisreason, the study was focused in the geochemical composition ofvolcanic glass shards, according to the methodology developed inprevious work (Daga et al., 2006), considering also that the glassygeochemistry reflects the composition of melt at the time oferuption (Ortega-Guerrero and Newton, 1998; Saminger et al.,2000).

Since a few decades ago several works (Gordon et al., 1968;Borchardt et al., 1972; Westgate and Gorton, 1981; Saminger et al.,2000; Steinhauser et al., 2006, 2007) have shown that thegeochemical characterization by Instrumental Neutron ActivationAnalysis (INAA) of the different volcanic products distinguished ineach tephra layer identified in the sequences, is the principalanalytical tool to associate the volcanic products with thepotential sources, in this case to define which might actuallyhave impacted Nahuel Huapi National Park. The geochemicalcharacterization of well preserved volcanic ash layers allows thedetermination of the age and sources of volcaniclastic sediments(Lacasse et al., 1998; Werner et al., 1998; Eastwood et al., 1999;Bichler et al., 2004; Horz et al., 2004), the ashes transport patterns(Lacasse et al., 1998), and the spatial distribution of primary airfalland secondary reworked tephra layers (Boygle, 1999). In generalterms, the study of ash layers enables the interpretation ofvolcanic processes (Clift and Fitton, 1998).

ARTICLE IN PRESS

Fig. 1. The study region: Nahuel Huapi National Park and the potential volcanic ash sources to the park. Scheme of the sedimentary sequences studied and sampling sites:

lakes Ilon, Moreno, and Nahuel Huapi, Brazo Rincon.

R. Daga et al. / Applied Radiation and Isotopes 66 (2008) 1325–13361326

Author's personal copy

2. Experimental

Sedimentary sequences were extracted with a messengeractivated gravity type corer from lake Ilon and lake MorenoOeste, belonging to Nahuel Huapi National Park. This is the largestArgentine National Park, comprising a drainage basin thatincludes three major river systems, 13 lakes of more than10 km2, and several hundred small lakes and ponds (Fig. 1). Allwatersheds within the park are glacial in origin and many of thelakes have been formed by the damming effect of the frontalmoraines left behind by retreating glaciers (Quiros, 1988). Thepark is included in the SVZ of the Andes Mountains, which isregarded to be the product of an oceanic–continental subductionzone with a currently active volcanic arc extending from 331S to461S. This arc includes several volcanic centres that were activesince Mioceno to present (Gerlach et al., 1988; Hickey et al., 1984;Hildreth and Moorbath, 1988; Tormey et al., 1991; Lopez-Escobaret al., 1995; Petit-Breuilh Sepulveda, 2004). Lake Ilon (411110 Southlatitude, and 711440 West longitude) is a high altitude water body(1350 m above see level), 45 m maximum depth situated in thesouthern region of the park, near to the highest peak in the region,Tronador mountain (Fig. 1). Lake Moreno Oeste (41150 Southlatitude, 711330 West longitude, 758 m above see level), thewestern part of lake Moreno, has a surface area of 6 km2 and themaximum depth is 90 m (Fig. 1).

The sediment cores were cut open, visually inspected, and sub-sampled every 1 cm. Visual inspection included tephra layersidentification by their granulometric properties and colour, clearlydistinctive from other core layers. Two sediment cores wereextracted from different sites of the Llao Llao bay in lake MorenoOeste. One core (MO) was analysed completely, including 210Pband 137Cs dating, while the second (MOR) was visually correlatedwith MO, analysing selected tephra layers considered replicates(Table 1). The tephra layers selected for analysis in lake Ilonsequence were those that showed a sedimentation patterncompatible with primary airfall deposits (Table 1). Each sedimentlayer sub-sampled was freeze-dried until constant weight. 210Pb,226Ra, and 137Cs specific activity of upper core layers weremeasured by high-resolution gamma-ray spectrometry, allowingcore dating (Joshi and Shukla, 1991; Ribeiro Guevara et al., 2003).

After 210Pb, 226Ra, and 137Cs measurements, tephra layers wereconditioned for geochemical characterization following thedefinitions and procedures described in a previous work (Dagaet al., 2006). The materials that were not associated with primary

volcanic stuff were carefully removed from volcanic ash particlesto avoid analytical interferences. Organic matter was eliminatedfrom volcanic ash samples by means of diluted H2O2 (6%), vacuumfiltered and dried at 40 1C. After that, volcanic layers were sieved,and the fraction bigger than 500 mm was studied under binocularmagnifying glass, in order to identify volcanic ash primarycomponents, setting them off for analysis. The material obtainedafter separation procedures was cleaned out of silty-clay materialswith distilled water in an ultrasonic bath at 30 1C for 75 min anddried out at 40 1C. The study was performed on the fraction biggerthan 500 mm since the composition variability with particle sizewas studied in a previous work (Daga et al., 2006), demonstratingthat this fraction provides the primary components representativeof each tephra. This fraction, due to the big size, allows clearidentification when extracting the particles. Brown, grey, andwhite pumice shards, scoria fragments, and glass shards (platyand cuspate or Y-shaped shards) were isolated. The volcanic ashcomponents distinguished in each case are shown in Table 1.Finally, the samples were placed in plastic vials, weighted andlabelled for analysis.

Even though several attributes of tephra layers allow theircharacterization and correlation with volcanic events (i.e. coloura-tion, granulometric parameters, type of juvenile magmaticcomponents, lithic content, and the weathering degree), theparameters that do not depend on the spatial distribution of thevolcanic products give more reliable data for tephrochronologicalpurposes. This is the case of the geochemical properties of glassshards and primary phenocrysts. The mineral assemblage isconditioned by sedimentary fractionating and differential settling,both during atmospheric transport and superficial sedimentation.Therefore, the glass shards are particularly suitable for thecharacterization of the source since they reflect the magmacomposition at the time of eruption, and they can be recorded indistal settings like ‘‘cryptotephras’’ keeping the original finger-print (Westgate and Gorton, 1981; Chambers et al., 2004; Gehrelset al., 2006). Glass shards composition is the key data analysed inthe present work for source characterization, complemented withwhite pumice composition since this material provided alsosignificant results. Brownish pumice and scoria fragments werenot considered. Brownish pumice showed significant variations incomposition with grain size (Daga et al., 2006), while thecomposition of scoria fragments did not show significantvariations to allow source characterization. However, thesematerials will be analysed in future research because they give

ARTICLE IN PRESS

Table 1Volcanic ash components identified in tephra layers from sedimentary sequences extracted in lake Ilon and lake Moreno

Core depth (cm) Estimated date of the eventa Primary components analysedb

Lake Ilon

Tephra IL1 12–14 White pumice, glass shards, and scoriae

Tephra IL2 27–28 – Glass shards

Tephra IL3 55–59 – White pumice, glass shards, and scoriae

Lake Moreno Oeste

Tephra MO1 from core MO 2–3 1960 White pumice, brown pumice, and scoriae

Tephra MO1 from core MOR 2–3 – White pumice

Tephra MO2 from core MO 5–7 1920 Glass shards and scoriae

Tephra MO3 from core MO 9–12 1860 White pumice, brown pumice, and scoriae

Tephra MO3 from core MOR 9–12 – White pumice and scoriae

Tephra MO4 from core MO 14–17 1775 Glass shards and scoriae

Tephra MO5 from core MO 20–23 – White pumice, brown pumice, glass shards, and scoriae

Tephra MO6 from core MO 24–27 – White pumice, glass shards, and scoriae

Tephra MO7 from core MO 31–34 – White pumice, brown pumice, grey pumice, and scoriae

Tephra MO8 from core MO 38–40 – White pumice, glass shards, and scoriae

a Estimation by extrapolation of the sedimentation rate determined in upper layers by 210Pb and 137Cs dating.b Primary components extracted from the fraction bigger than 500mm.

R. Daga et al. / Applied Radiation and Isotopes 66 (2008) 1325–1336 1327

Author's personal copy

information of several processes in the conduit and during magmaevolution.

The geochemical characterization was performed by INAA. Thesamples were irradiated in the RA-6 nuclear reactor, CentroAtomico Bariloche, inside plastic vials. Sample masses rangedfrom 10 to 60 mg. Two irradiations were performed for eachsample. First, a short-term irradiation in a predominantly thermalflux (fthffi8�1011 n cm�2 s�1) for 2–10 min. After decay times ofabout 10, 15, and 30 min, three gamma-ray spectra of sampleswere collected for 5 min, 10 min, and 2 h respectively. A secondirradiation was performed in the reactor core (fthffi7�1012 ncm�2 s�1, fepiffi2�1011 n cm�2 s�1 and ffffi2�1012 n cm�2 s�1),for 24 h. Samples were changed to non-irradiated vials after thelong-term irradiation. Three to five gamma-ray spectra werecollected after decay times longer than 7 days, with appropriatedcounting time for each case. An intrinsic HPGe n-type detector,12.3% relative efficiency and 1.8 keV resolution at 1.33 MeV, and a4096-channel analyser were used for the gamma-ray counting,and the spectra were analysed by using the GAMANAL routineincluded in the GANAAS package, distributed by IAEA. Theabsolute parametric method was used to determine the elementalconcentrations, using nuclear constants taken from current tables(Mughabghab et al., 1981; Firestone and Shirley, 1996; De Corte,2003; Mughabghab, 2003; Tuli, 2005). Thermal and epithermalneutron fluxes were determined in long-term irradiation by (n,g)reactions of the pair Co–Au, using high purity wires of pure Co and0.112% Au–Al. In short-term irradiations, the thermal neutronfluxes were determined by (n,g) reactions of Mn, using high puritywires of 81.2% Mn–Cu alloy. Corrections for spectral interferenceswere performed when necessary. Corrections due to contributionsof 235U fission products, and 27Al(n,p)27Mg, 54Fe(n,a)51Cr, and176Yb(n,g)177Yb-177Lu reactions were also included. The elementsdetermined were major Al, Ca, Fe, Mg, Mn, Na, K, and Ti, RareEarths La, Ce, Nd, Sm, Eu, Tb, Tm, Yb, Lu, and other relevant traceelements Sb, As, Ba, Br, Cs, Zn, Co, Cr, Hf, Sc, Sr, Ta, Th, U, and V.Standard Reference Materials (SRM) IAEA ES 405 Trace elementsand methylmercury in Estuarine Sediment, and IAEA SL Trace andminor elements in lake sediment, were analysed for qualitycontrol; the results show a good agreement with the certifiedvalues (Table 2). Good agreement was also obtained in replicates.

3. Results and discussion

3.1. Dating of sedimentary sequences

137Cs specific activity and total 210Pb together with 226Ra,representing supported 210Pb, specific activity profiles are shownin Figs. 2 and 3 for lake Moreno and lake Ilon respectively. Total210Pb specific activity profile in lake Moreno sequence has anapproximately exponential behaviour, with significant unsup-ported 210Pb until 1 g cm�2 depth, allowing dating hence. Thesedimentation rate determined by using CRS model was13.371.0 mg cm�2 yr�1 (0.058470.0044 cm yr�1; Ribeiro Guevaraet al., 2003). 137Cs dating, by correlation of the specific activityprofile with the historical fallout sequence in the study region(Ribeiro Guevara and Arribere, 2002), is in good agreement with210Pb dating. The dates of recent volcanic events registered in thesediment sequences presented in Table 1 were obtained byextrapolation with sediment mass by using the sedimentationrate determined by 210Pb dating, discounting volcanic ashes fromsediments by estimating the fraction in each layer from theanalysis under binocular microscope. It is necessary to emphasizethat this is a rough estimation, assuming that there was no changein sedimentation rate before 1900, that could involve significantdeviation from real dates particularly for older events.

The behaviour of total 210Pb specific activity profile in lake Ilonis not exponential, with a sharp increase at 1.5 g cm�2 (8 cm)depth (Fig. 3). The authors did not find a convincing explanationfor this behaviour since there is no evidence of biologicaldisturbances, and the sedimentology of upper layers showsphysical disturbance from 0 to 1.2 g cm�2 (7 cm) depth, butdownwards a regular sedimentation pattern was observed. Theelemental composition of bulk sediment, determined by INAA, ofupper 8 cm of the sequence does not show significant variations.Indicators of terrigenous inputs Al and Fe/Ca ratio do not showsignificant fluctuations, as well as Fe and Mn contents, that wouldsuggest effects due to peculiar sedimentation processes or later

ARTICLE IN PRESS

Table 2Elemental analysis of standard reference materials IAEA ES 405 trace elements in

estuarine sediment, and IAEA SL trace and minor elements in lake sediment

SRM IAEA ES 405 SRM IAEA SL

Measured Reported Measured Reported

Ti (wt%) 0.47770.096 – 0.4870.12 0.51770.037

Fe (wt%) 3.5370.20 3.7470.07 6.3270.35 6.7470.17

Mn (wt%) 0.055670.0020 0.049570.0011 0.38470.013 0.34670.016

Mg (wt%) 1.2570.20 1.2370.09 0.5070.32 –

Na (wt%) 1.84670.075 – 0.17070.011 0.1770.01

K (wt%) 2.5770.49 2.4970.72 1.4870.23 1.4570.21

Rb (mg g�1) 165.279.1 – 117.277.2 113711

Ba (mg g�1) 427748 – 739768 639753

Cs (mg g�1) 12.3670.71 12.572.1 6.6970.39 7.070.9

Br (mg g�1) 89.875.9 85725 5.2470.41 –

Hf (mg g�1) 6.1970.28 5.8070.87 4.9170.23 4.270.6

Ta (mg g�1) 1.53770.097 – 1.16270.071 1.5870.58

Th (mg g�1) 13.6370.67 14.372.1 14.270.68 1471

U (mg g�1) 3.7070.57 3.071.2 3.8170.44 4.0270.33

Sb (mg g�1) 2.1170.24 1.8170.19 1.1570.12 1.3170.12

Co (mg g�1) 14.3370.47 13.770.7 20.4670.65 19.871.5

V (mg g�1) 98711 9579 198716 170715

Cr (mg g�1) 85.173.9 8474 113.175.0 10479

Sc (mg g�1) 13.9970.42 13.572.0 17.4070.52 17.371.1

Zn (mg g�1) 300717 27977 219713 223710

As (mg g�1) 23.171.5 23.670.7 29.071.8 27.672.9

La (mg g�1) 39.171.2 40.477.3 51.871.5 52.673.1

Ce (mg g�1) 82.474.5 – 110.776.0 117717

Sm (mg g�1) 7.3070.44 6.8670.36 9.1870.54 9.2570.51

Eu (mg g�1) 1.33270.058 1.2570.36 1.91970.075 1.670.5

Tb (mg g�1) 0.95170.059 0.9370.43 1.16670.068 1.4070.46

Dy (mg g�1) 6.1170.54 – 7.3870.81 7.572.2

Yb (mg g�1) 3.0670.31 3.0470.85 3.6170.34 3.4270.65

Lu (mg g�1) 0.36270.039 0.4770.19 0.45870.047 0.5470.13

-4

-3

-2

-1

0

0 100 200 0 20 40specific activity (Bq.kg-1)

dept

h (g

.cm

-2)

total 210Pb226Ra 137Cs

Fig. 2. Lake Moreno. 137Cs, and total 210Pb and 226Ra specific activity profiles of the

sedimentary sequence.

R. Daga et al. / Applied Radiation and Isotopes 66 (2008) 1325–13361328

Author's personal copy

diagenetic processes associated to changes in redox potentials,generating mobilization after deposition (Ribeiro Guevara et al.,2003). Only Br, element found in association with organic matterin previous research ((Ribeiro Guevara et al., 2005), showsvariations with similar behaviour as total 210Pb specific activity.In any case, it is not possible to date the sequence by 210Pb.Although 210Pb anomalous behaviour could be due to unknownprocesses or events interfering the sequence dating, 137Cs activitycorrespond to sediments deposited after 1964 (Ribeiro Guevaraand Arribere, 2002), and therefore tephra IL1 correspond to anevent that occurred before this date, but close (Fig. 3). Thisobservation is in agreement with the event registered in 1961 ofCalbuco volcano (Table 3). Therefore, considering the geographicalproximity (Fig. 1) the source of tephra IL1 would be the Calbucoeruption at 1961, and tephras IL2 and IL3 would correspond also tovolcano Calbuco events, since there is no relevant event at Osornovolcano until 1836 (Table 3).

3.2. Geochemical tracers

The elemental composition of the primary componentsisolated from the three tephra layers studied in the sedimentary

sequence extracted from lake Ilon is shown in Table 4. The whitepumice samples isolated from IL1 and IL3 tephra layers, haveidentical elemental composition, considering uncertainties, aswell as scoria samples extracted from IL1 and IL3.

The glass shards samples, extracted from IL1, IL2, and IL3tephra layers, show similar concentration for most elements,except for the light and medium REE La, Ce, Sm, and Eu, the traceelements Cs, Hf, Ta, and Th, and the major Ti, that are slightlyhigher in tephra IL3 (concentration in the column ‘maximum’ inTable 4 correspond to tephra IL3 for these elements) but similarfor tephras IL1 and IL2 (concentration in the column ‘minimum’ inTable 4 correspond to tephra IL2). Cr also shows significantdifferences; it is higher in IL1 respect to IL2, and also higher in IL2respect to IL3.

The similitude in the elemental composition of the primarycomponents of tephras IL1, IL2, and IL3 allows the assumption of aunique volcanic source. According to 137Cs dating correlated withthe historical records, Calbuco volcano would be considered thesource of tephras IL1, IL2, and IL3.

Geochemical indicators will be studied in the primarycomponents of the tephra layers taken out from lake Ilon andlake Moreno, comparing with Puyehue-Cordon Caulle resultsobtained in previous work (Daga et al., 2006), in order tocharacterize each source, and to assess their impact in lakeMoreno. Since there is no significant difference in elementalcomposition among the three tephra layers analysed in lake Ilon,no significant difference is expected in geochemical indicatorseither. Therefore, the three volcanic events registered in lake Ilonhave the same geochemical characterization, allowing its identi-fication in other sedimentary sequences.

3.2.1. Total alkali–silica contents

The contents of total alkali (Na2O+K2O) and silica (SiO2) areused in the classification scheme for volcanic rocks. The silicaconcentration defines the level of basicity/acidity of the rocks, andtotal alkali content is used to define the magma series. Theseinformation is useful to evaluate the silica saturation grade inrocks, as well as the systematic variation of the others majorelements (Wilson, 1989; Rollinson, 1993; Wade et al., 2005). Silicacontents and Na2O+K2O concentrations are useful parameters tocharacterize volcanic products; Table 5 shows the concentrationsdetermined in the primary components of the tephra layersstudied in lakes Ilon and Moreno. Silica concentrations weredetermined following the methodology carried out by Daga et al.(2006). Table 5 also shows the SiO2 and Na2O+K2O concentrationsmeasured in tephra layers from a sedimentary sequence extractedfrom lake Nahuel Huapi, Brazo Rincon site (Fig. 1), consideredassociated to Puyehue-Cordon Caulle activity, and volcanic ashescollected manually after the 1960 volcanic event of this source(Daga et al., 2006).

The SiO2 and Na2O+K2O contents are similar when consideringeach primary components of the three tephra layers in BrazoRincon sequence which were associated to Puyehue-CordonCaulle events, and also for the volcanic ashes collected manuallyafter the 1960 eruption (Daga et al., 2006) (Table 5), defininghence a SiO2 and Na2O+K2O composition of glass shards and whitepumice typical of Puyehue-Cordon Caulle source. SiO2 andNa2O+K2O contents are also similar in the glass shards and whitepumice isolated from the three tephra layers of lake Ilon sequence(Table 5), clearly distinctive from Puyehue-Cordon Caulle. SiO2

concentrations are lower in lake Ilon respect to Brazo Rincon inwhite pumice, 62.470.2–67.370.8 wt% (average on the differenttephra layers considered in both sequences, the uncertainty is thestandard deviation of the average; Table 5), and particularly inglass shards, 50.570.9–66.671.2 wt%. A similar pattern is

ARTICLE IN PRESS

-5

-4

-3

-2

-1

0

0

tephra IL1

0

tephra IL1

137Cs137Csdetection limit

specific activity (Bk.kg-1)

Total 210Pb226Ra

60504030201030025020015010050

dept

h (g

.cm

-2)

Fig. 3. Lake Ilon. 137Cs, and total 210Pb and 226Ra specific activity profiles of the

sedimentary sequence.

Table 3Historical records of volcanic events in Calbuco, Osorno, and Puyehue-Cordon

Caulle volcanoes (Petit-Breuilh Sepulveda, 2004)

Sourcea Dateb

Calbuco 1893– 1895

1917

1929– 1932

1961

Osorno 1790– 1791

1834– 1836

Puyehue-Cordon Caulle 1759

1893

1919– 1922

1929– 1930

1960

a See Fig. 1.b Only the events with medium and high intensity are included; events

delayed by one or two years are considered the same since they cannot be

distinguish in a sedimentary sequence.

R. Daga et al. / Applied Radiation and Isotopes 66 (2008) 1325–1336 1329

Author's personal copyARTICLE IN PRESS

Table 4Elemental composition of volcanic ashes extracted from lake Ilon sedimentary sequence

White pumice Glass shards Scoria

Minimum Maximum Minimum Maximum Minimum Maximum

Ti (wt%) 0.45170.075 0.55670.094 0.60870.096 0.9270.14 0.77270.085 0.8070.10

Al (wt%) 8.5770.38 8.6370.37 8.6770.37 9.4170.39 8.8070.38 9.2070.41

Fe (wt%) 4.2670.25 4.3670.27 6.8070.40 8.0970.57 6.9570.45 7.1070.40

Mn (mg g�1) 1013737 1047737 1377746 1539752 1393748 1409749

Mg (wt%) 1.0470.19 1.0470.20 3.2770.32 3.3770.33 2.1570.27 2.5270.27

Ca (wt%) 3.9070.32 3.9470.40 6.4270.59 7.1270.58 5.8870.43 6.0070.42

Na (wt%) 3.3370.14 3.3570.14 2.52570.098 2.63070.098 2.7270.12 2.9770.12

K (wt%) 1.1470.17 1.4070.27 0.52370.098 0.6370.34 0.5670.17 0.9170.29

Rb (mg g�1) 48.873.9 55.575.2 19.873.6 29.074.1 27.173.2 30.073.0

Sr (mg g�1) 348755 355744 409761 431750 381755 450774

Ba (mg g�1) 397744 399743 164726 214733 211729 255731

Cs (mg g�1) 2.7470.23 2.7670.20 1.0870.11 1.5570.14 1.5570.17 1.7870.14

Br (mg g�1) 1.7370.28 1.9770.48 0.7570.12 1.0470.39 1.0070.18 1.0170.28

Zr (mg g�1) 239729 258737 135746 172750 148737 207753

Hf (mg g�1) 5.4270.32 5.5670.37 2.0770.15 2.9570.18 3.0970.13 3.3470.18

Ta (mg g�1) 0.33370.034 0.34670.047 0.09270.018 0.17970.023 0.16970.017 0.22070.032

Th (mg g�1) 5.1470.31 5.2370.40 1.0770.097 1.7370.14 1.7370.11 2.1270.18

U (mg g�1) 1.6670.30 1.9770.38 0.33370.083 0.49370.097 0.5070.15 0.7070.10

Sb (mg g�1) 0.55570.074 0.56770.093 0.26470.056 0.35870.071 0.40270.058 0.44570.085

Co (mg g�1) 11.3770.52 11.9570.62 33.471.5 38.072.0 25.771.3 28.4270.94

V (mg g�1) 93710 97714 330721 351721 291718 319721

Cr (mg g�1) 2.2070.42 3.6370.70 81.875.1 110.278.4 19.071.3 50.572.4

Sc (mg g�1) 15.8870.52 17.1070.60 37.371.4 39.472.0 31.271.4 34.9270.98

Zn (mg g�1) 91.876.8 100.578.3 106.478.1 129710 121.977.6 123.178.2

As (mg g�1) 9.9170.76 10.6870.94 5.5070.44 6.7370.69 7.4070.74 8.1270.59

La (mg g�1) 18.3070.62 18.4470.74 6.1370.22 9.2270.36 10.1170.43 10.3970.41

Ce (mg g�1) 42.673.1 44.073.7 15.571.2 23.271.7 23.771.3 26.171.6

Nd (mg g�1) 22.771.9 24.172.4 13.071.6 14.871.4 15.971.6 17.271.4

Sm (mg g�1) 5.8170.38 5.9870.39 3.2070.22 4.2070.30 4.3970.29 4.4170.37

Eu (mg g�1) 1.27670.064 1.28670.073 0.99070.054 1.19970.063 1.21570.070 1.28870.054

Tb (mg g�1) 0.94070.066 1.00870.077 0.65370.083 0.76270.072 0.76770.077 0.79570.074

Dy (mg g�1) 6.0670.58 6.0670.59 3.6670.40 4.7470.54 4.8770.47 5.1070.49

Tm (mg g�1) 0.45170.059 0.45670.051 0.34370.043 0.38370.044 0.37470.048 0.57570.058

Yb (mg g�1) 3.6370.36 3.6570.34 2.1370.19 2.6670.21 2.6570.24 2.8370.25

Lu (mg g�1) 0.47670.039 0.50070.045 0.30170.029 0.35870.029 0.3767.033 0.40470.029

Concentration range of primary components white pumice, glass shards, and scoriae in the three tephra layers analysed.

Table 5SiO2 and Na2O+K2O contents in the primary components of tephra layersa from sedimentary sequences extracted from lake Nahuel Huapi, Brazo Rincon (Daga et al., 2006),

lake Moreno Oeste, and lake Ilon, and volcanic ashes collected manually after the 1960 event of Puyehue-Cordon Caulle (MUS)

SiO2 (wt%) Na2O+K2O (wt%)

White pumice Glass shards White pumice Glass shards

Puyehue-Cordon Caulle source

MUS volcanic ashes 68.3 – 8.77 –

Lake Nahuel Huapi tephra BR1 67.0 65.3 8.74 8.32

Lake Nahuel Huapi tephra BR2 66.6 67.0 8.44 7.98

Lake Nahuel Huapi tephra BR4 67.3 67.6 7.71 8.14

Average (standard deviation) 67.3 (0.8) 66.6 (1.2) 8.42 (0.49) 8.15 (0.17)

Lake Ilon

Tephra IL1 62.3 50.8 5.89 4.14

Tephra IL2 – 51.2 – 4.03

Tephra IL3 62.5 49.4 6.18 4.30

Average (standard deviation) 62.4 (0.2) 50.5 (0.9) 6.03 (0.20) 4.16 (0.14)

Lake Moreno Oeste

Tephra MO1 67.3 – 8.58 –

Tephra MO2 – 52.2 – 3.74

Tephra MO3 61.1 – 6.40 –

Tephra MO4 – 51.0 – 3.68

Tephra MO5 67.6 67.3 8.64 7.88

Tephra MO6 60.5 53.5 6.79 3.99

Tephra MO7 58.8 – 5.43 –

Tephra MO8 61.3 49.5 6.01 4.05

a Fraction bigger than 500 mm.

R. Daga et al. / Applied Radiation and Isotopes 66 (2008) 1325–13361330

Author's personal copy

observed in Na2O+K2O concentrations, with lower values in lakeIlon; 6.0370.20–8.4270.49 wt% in white pumice and 4.1670.14to 8.15.170.17 wt% in glass shards (Table 5). Therefore, accordingto these patterns, two different compositions regarding SiO2 andNa2O+K2O contents can be clearly distinguished: Puyehue-CordonCaulle type and Ilon type, that could be associated to Calbucovolcano. SiO2 and Na2O+K2O concentrations in glass shards andwhite pumice from Brazo Rincon tephras are similar, but in Ilon asignificant increase in white pumice is observed (Table 5); this isanother distinctive feature between deposits in both lakes.

The SiO2 and Na2O+K2O contents of white pumice and glassshards from the tephra layers in lake Moreno Oeste sequenceshow different composition patterns. Two groups were identified,MO1 and MO5 with SiO2 and Na2O+K2O contents similar to thoseassociated to Puyehue-Cordon Caulle system, and MO2, MO3,MO4, MO6, MO7, and MO8 showing Ilon type patterns (Table 5).

3.2.2. Chromium and vanadium contents

The trace elements are indicative of processes occurring in thesolid–liquid equilibrium during the evolution of the volcanicevent towards the final state. Therefore, their analysis in thedifferent volcanic products is important for source characteriza-tion since their final concentrations reflect its behaviour duringthe complex eruption processes that depend on the volcanonature and on the event conditions. Co, Cr, Sc, and V aretransitional metals, compatible with mineral phases that crystal-lize since the first stages of the magma evolution, that decreasetheir concentration with the increase of magmatic differentiation.Even though a similar geochemical behaviour is expected due totheir similar chemical properties, geological processes can takeadvantage of subtle chemical differences resulting in variationswith the magma evolution. The ratios between these elementsvary according the mineral phases that are crystallizing(Rollinson, 1993). Variations in Cr concentrations of primarycomponents were observed in different events of the same source(Nakada et al., 2005).

Clear differences in Cr and V concentrations can be distin-guished between Brazo Rincon and lake Ilon tephras (Table 6). V inwhite pumice, and Cr and V in glass shards show decisivedifferences to allow source identification. Cr concentrations inglass shards are much higher in lake Ilon tephras (98715mg g�1,average) respect to recent events recorded in Brazo Rincon(1.7370.36, average), as well as V: 340711 mg g�1 in lake Ilonand 17712 in Brazo Rincon (average in both cases; Table 6). Vconcentrations in white pumice are also higher in lake Ilontephras (9573mg g�1, average) when compared with Brazo Rincon(16.874.9 mg g�1, average) (Table 6). Co and Sc concentrations,elements with similar chemical properties as Cr and V, and hencesimilar behaviour during magmatic fractioning, are also higher inglass shards and white pumice from lake Ilon (Table 4) respect toBrazo Rincon. Therefore, Cr and V concentrations allow sourceidentification of recent events in Nahuel Huapi National Park.

Cr and V concentrations in glass shards and white pumice fromBrazo Rincon tephras are similar, while in lake Ilon a significantincrease in glass shard was measured (34 times in Cr concentra-tions, and 3.5 times in V; Table 6) with respect to white pumice,resulting in another distinctive feature between tephra layers ofboth lakes sources.

When comparing Cr and V concentrations in glass shards andwhite pumice of lake Moreno tephra layers with the compositionpatterns of those of Brazo Rincon and lake Ilon tephras, clearassociation arise. Cr and V concentrations in glass shards allow theassociation of MO1 and MO5 tephras with Brazo Rincon, and MO2,MO3, MO4, MO6, MO7 and MO8 with lake Ilon (Table 6). It isnotable that Cr in glass shards of tephra MO4 is 2 folds higher

than lake Ilon contents, while tephra MO6 shows a value belowIlon pattern for this material. Similar association is obtainedanalysing V contents in white pumice, with the exception oftephra MO8 that range in-between both patterns (Table 6).

3.2.3. Rare earth element patterns

The analysis of rare earth elements (REE) contents is apowerful geochemical tool (Wilson, 1989; Rollinson, 1993; Cliftand Fitton, 1998; Wade et al., 2005). REE are not easilyfractionated during sedimentation. They are the least solubletrace elements, relatively immobile in superficial and aqueousenvironments. Therefore, sedimentary REE patterns may providean index associated to provenance composition. REE patterns arecontrolled by the REE chemistry in an igneous rock, but subtlevariations in the REE properties make them sensitive to crystal–melt equilibrium which have taken place during its evolution,making them particularly suitable for geochemical studies(Rollinson, 1993; McLennan, 1989). The REE are generallyincompatible in most petrogenetically important silicate minerals,but can be compatible in a few major minerals, particularly inhighly evolved systems (Hanson, 1989; McKay, 1989). Thefractionation degree of a REE pattern can be expressed by theratio of the concentration of a light REE (La or Ce) to middle REE(Eu or Sm) or to a heavy REE (Yb or Lu), normalized by Chondrite(Boynton, 1984). The ratios [La]/[Lu] and [La]/[Sm], normalized byChondrite (Boynton, 1984), are shown in Table 7. The Rare EarthElement patterns of lake Ilon tephra layers are shown in Fig. 4. Nosignificant difference can be distinguished in the patternsbetween the three tephra layers when comparing each primarycomponent, but significant differences can be observed betweenwhite pumice and glass shards.

3.2.4. Europium anomaly

An exception to the systematic variation of REE, the well-understood europium anomaly, is a powerful geochemical tool to

ARTICLE IN PRESS

Table 6Cr and V concentration in primary components of volcanic ashes from sedimentary

sequences extracted from lake Nahuel Huapi, Brazo Rincon (Daga et al., 2006), lake

Moreno Oeste, and lake Ilon, and volcanic ashes collected manually after the 1960

event of Puyehue-Cordon Caulle (MUS)

White pumicea (mg g�1) Glass shardsb (mg g�1)

Cr V Cr V

Puyehue-Cordon Caulle source

MUS ashes 1.9170.83 15.576.3 – –

Tephra BR1 1.3470.45 13.074.1 1.3770.80 4.771.9

Tephra BR2 1.5970.80 23.974.8 2.0970.95 18.575.5

Tephra BR4 2.1670.69 14.874.4 1.7270.93 28.577.2

Average (standard deviation) 1.75 (0.36) 16.8 (4.9) 1.73 (0.36) 17 (12)

Lake Ilon

Tephra IL1 3.6370.70 97714 110.278.4 351721

Tephra IL2 – – 100.675.4 330721

Tephra IL3 2.2070.42 93710 81.875.1 340720

Average (standard deviation) 2.9 (1.0) 95 (3) 98 (15) 340 (11)

Lake Moreno Oeste

Tephra MO1 2.571.0 19710 – –

Tephra MO2 – – 87.974.4 340719

Tephra MO3 5.2770.78 94.079.6 – –

Tephra MO4 – – 199.879.1 322719

Tephra MO5 2.2670.76 20.076.8 2.971.8 26.3070.57

Tephra MO6 1.9770.94 104714 64.273.5 326719

Tephra MO7 2.9970.57 94711 – –

Tephra MO8 2.7870.77 48.778.4 89.574.3 263717

a Fraction bigger than 500mm.b Concentrations of replicate are not coincident; both values are shown.

R. Daga et al. / Applied Radiation and Isotopes 66 (2008) 1325–1336 1331

Author's personal copy

study igneous systems. The plotted position of Eu sometimes liesout of the trend defined by the other REE on a Chondritenormalized diagram, defining a Eu anomaly controlled principallyby feldspars (McKay, 1989; Rollinson, 1993; Wilson, 1989). The Euanomaly is quantified by comparing the measured concentration,normalized by Chondrite (EuN), with an average value, Eu*,obtained by interpolation between the normalized concentrationsof Sm and Gd. The ratio EuN/Eu* greater than 1 indicates a positiveanomaly; if it is less than 1 the Eu anomaly is negative. Taylor andMcLennan recommend the use of a geometric average (Taylor andMcLennan, 1988). Tb normalized concentration (TbN) is consid-ered in present work to compute Eu*, by geometric average withnormalized Sm concentrations (SmN), using formula (1). Euanomalies of the tephra layers of Brazo Rincon, lake Ilon, andlake Moreno are shown in Table 7:

Eu� ¼ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiSm2

N TbN3

q(1)

The geochemical parameters that have significant differencesbetween Brazo Rincon and lake Ilon tephras to allow sourceidentification, are the Eu anomaly, light to middle, and light toheavy REE ratio of glass shards. The variations in white pumicebetween both patterns even significant are not high enough for aclear source distinction (Table 7). These geochemical parameters ofglass shards from lake Moreno tephra layers show clear grouping;MO2, MO4, and MO6 are lake Ilon type, while MO5 is Brazo Rincontype. Light and middle to heavy REE ratios in tephra MO8 range in-between lake Ilon and Brazo Rincon patterns, while the Eu anomalycorrespond to lake Ilon but it is significantly higher. Therefore, inthe case of tephra MO8 is not possible an association consideringthese geochemical parameters. It is noteworthy an increase in Euanomaly of white pumice in tephras MO6, MO7, and MO8 respectto lake Ilon and Brazo Rincon patterns.

3.2.5. Hafnium, tantalum, and zirconium contents

These elements are small highly charged cations with similarcharacteristics as lanthanides, Sc, and Th. They do not usuallyenter in the structures of rock-forming minerals, except in someaccessory phases. Hf, Ta, and Zr tend to increase in the melt phase;therefore these element ratios remain almost constant throughany fractional crystallization processes (Rollinson, 1993). SimilarHf, Ta, and Zr contents in white pumice and glass shards wereobtained for the three tephra layers extracted from lake Ilonsequence; it is the same situation for the three tephra layersassociated to recent events of Puyehue-Cordon Caulle source(Daga et al., 2006). Unambiguous grouping of Hf, Ta, and Zrconcentrations can be observed particularly in glass shards and ina lesser degree in white pumice, which allows source identifica-tion (Table 8). Hf, Ta, and Zr contents in glass shards and whitepumice from Brazo Rincon tephras are similar, but in lake Ilonsignificant differences between both materials can be observed(Table 8).

The comparison of Hf, Ta, and Zr concentrations in whitepumice and glass shards from the tephra layers extracted from

ARTICLE IN PRESS

Table 7Geochemical tracers from rare earth elements: Eu anomaly, ratio light to middle REE (LaN/SmN), and ratio light to heavy REE (LaN/LuN), normalized by Chondrite

Eu anomaly Light to middle REE (LaN/SmN) Light to heavy REE (LaN/LuN)

White pumicea Glass shardsa White pumicea Glass shardsa White pumicea Glass shardsa

Puyehue-Cordon Caulle sourceb

MUS ashes 0.554 – 2.23 – 4.13 –

Tephra BR1 0.574 0.554 2.32 2.37 4.09 4.11

Tephra BR2 0.642 0.629 2.29 2.27 4.11 4.03

Tephra BR4 0.648 0.620 2.39 2.22 4.17 4.04

Average (SDc) 0.605 (0.048) 0.601 (0.041) 2.31 (0.07) 2.29 (0.08) 4.12 (0.03) 4.06 (0.04)

Lake Ilon

Tephra IL1 0.645 0.867 1.94 1.24 3.83 2.42

Tephra IL2 – 0.870 – 1.20 – 2.06

Tephra IL3 0.667 0.835 1.98 1.38 3.99 2.68

Average (SDc) 0.656 (0.016) 0.857 (0.019) 1.96 (0.03) 1.28 (0.09) 3.91 (0.12) 2.39 (0.31)

Lake Moreno Oeste

Tephra MO1 0.577 – 2.26 – 4.22 –

Tephra MO2 – 0.951 – 1.31 – 2.07

Tephra MO3 0.679 – 1.96 – 3.77 –

Tephra MO4 – 0.829 – 1.33 – 2.29

Tephra MO5 0.556 0.589 2.23 2.22 4.11 4.10

Tephra MO6 0.851 0.838 1.77 1.54 3.46 2.74

Tephra MO7 0.853 – 1.38 – 2.65 –

Tephra MO8 0.899 1.005 1.69 1.88 3.40 3.39

a Fraction bigger than 500 mm.b Daga et al., 2006.c Standard deviation of the average.

10

100

Con

cent

ratio

n no

rmal

ized

by

Con

drite

LuYbTmDyTbEuSmNdCeLa

IL1 white pumice IL3 white pumice IL1 glass shards IL2 glass shards IL3 glass shards IL1 scoriae IL3 scoriae

Fig. 4. Lake Ilon. Chondrite-normalized rare earth element patterns (Boynton,

1984) of white pumice shards, glass shards, and scoriae isolated from the three

tephra layers studied.

R. Daga et al. / Applied Radiation and Isotopes 66 (2008) 1325–13361332

Author's personal copy

lake Moreno shows that tephras MO1 and MO5 correspond toBrazo Rincon pattern while tephras MO2, MO3, MO4, MO6, MO7,and MO8 correspond to lake Ilon pattern (Table 8).

3.2.6. Caesium, rubidium, and thorium contents

Incompatible element concentrations are particularly sensitiveto igneous processes. The more incompatible an element is, themore sensitive it is to degrees of partial melting. Incompatibleelements in volcanic ashes are hence suitable geochemicalindications to characterize volcanic events. Cs and Rb are lowfield strength ions, large cations of small charge with almostidentical ionic radii and charge. They are a pair of highlyincompatible elements whose bulk partition coefficients aresimilar and tend to concentrate in the liquid phase, increasingtheir concentration with the magmatic differentiation. Th is alarge ion lithophyle element, which is also concentrated in theliquid phase remaining incompatible until the liquid evolves todacitic and rhyolitic melts.

Table 9 shows the Cs, Rb, and Th concentrations in whitepumice and glass shards of lakes Ilon and Moreno tephra layers, aswell as volcano ashes associated to Puyehue-Cordon Caulle source(Daga et al., 2006). White pumice and glass shards belonging toIlon tephra layers have similar Cs, Rb, and Th contents, as well asvolcano ashes from Puyehue-Cordon Caulle, but having significantdifferences between sources to allow identification, particularly inglass shards. Cs, Rb, and Th contents in glass shards and whitepumice from Brazo Rincon tephras are similar, but in lake Ilonsignificant differences between both materials can be observed(Table 9).

As well as the other geochemical tracer studied, Cs, Rb, and Thconcentrations allow source identification in tephra layers fromlake Moreno sediment sequence. Tephras MO1 and MO5 are BrazoRincon type and tephras MO2, MO3, MO4, MO6, MO7, and MO8are lake Ilon type. Rb and Th contents in white pumice of tephrasMO6, MO7, and MO8 are below the range of lake Ilon pattern(Table 9).

Summarizing, very good agreement between the differentgeochemical tools applied for the characterizations of volcanicashes was found. Similar behaviour was observed in all para-meters considered, allowing the identification of two kinds ofgeochemical signature: one corresponding to the Brazo Rincontephra layers and volcanic ashes collected manually after the 1960volcanic event of Puyehue-Cordon Caulle, and other correspond-ing to lake Ilon tephra layers. Very good agreement was observedbetween three layers identified in the Brazo Rincon sequenceassociated to Puyehue-Cordon Caulle source and the volcanicashes collected manually, as well as between the three tephralayers identified in lake Ilon sequence, that could be associatedmostly to volcano Calbuco.

The geochemical characterization of Brazo Rincon tephras andvolcanic ashes collected manually showed very good agreementbetween white pumice and glass shards. On the other hand,significant differences were observed when comparing whitepumice and glass shards from lake Ilon (Tables 5–9). This isanother distinctive feature between both tephra patterns.

Even low, higher variability in some geochemical tracer can beobserved in glass shards from lake Ilon compared to white pumiceand scoria fragments (Table 4). Glass shards composition in lakeIlon tephras reflects a less evolved melt, with lower SiO2 and alkalicontents (Table 5), also lower concentrations of incompatibleelements (i.e. Hf, Ta, Zr, Cs, Rb and Th, and REE ratios), and higherconcentrations of compatible elements Cr, V, Sc and Co(Cr concentrations show high variability in most cases studied)than white pumice (Tables 6–9).

Previously, we stated that considering the geographicalproximity and 137Cs dating, the tephras extracted from lake Ilonwould correspond to volcano Calbuco events. In a previous work,the tephras from lake Nahuel Huapi, Brazo Rincon, wereassociated to events from Puyehue-Cordon Caulle system (Dagaet al., 2006). The rock composition from both sources correspond-ing to eruptions occurred in historic times can be found in theliterature (Gerlach et al., 1988; Lopez-Escobar et al., 1995). Thepatterns of incompatible elements, normalized by Chondrite

ARTICLE IN PRESS

Table 8Hf, Ta, and Zr concentration in primary components of volcanic ashes from sedimentary sequences extracted from lake Nahuel Huapi, Brazo Rincon (Daga et al., 2006), lake

Moreno Oeste, and lake Ilon, and volcanic ashes collected manually after the 1960 event of Puyehue-Cordon Caulle (MUS)

White pumicea (mg g�1) Glass shardsa (mg g�1)

Hf Ta Zr Hf Ta Zr

Puyehue-Cordon Caulle source

MUS ashes 9.2070.47 0.57670.045 421751 – – –

Tephra BR1 9.5770.44 0.55770.041 425737 10.2370.53 0.60170.048 422753

Tephra BR2 8.8170.40 0.55070.043 374755 9.0370.45 0.56270.052 399761

Tephra BR4 8.2170.43 0.50170.039 381749 8.4470.43 0.51970.056 470768

Average (SDb) 8.95 (0.58) 0.546 (0.032) 400 (26) 9.23 (0.91) 0.561 (0.041) 430 (36)

Lake Ilon

Tephra IL1 5.5670.37 0.34670.047 258737 2.2170.21 0.15870.030 154757

Tephra IL2 – – – 2.0770.15 0.09270.018 135746

Tephra IL3 5.4270.32 0.33370.034 239729 2.9570.18 0.17970.023 172750

Average (SDb) 5.49 (0.10) 0.340 (0.009) 249 (13) 2.41 (0.47) 0.143 (0.045) 154 (19)

Lake Moreno Oeste

Tephra MO1 9.6670.40 0.58170.036 451717 – – –

Tephra MO2 – – – 2.0770.14 0.08670.018 144758

Tephra MO3 5.5770.19 0.33170.018 256713 – – –

Tephra MO4 – – – 2.6270.17 0.12870.027 133763

Tephra MO5 9.4170.22 0.59070.016 419723 9.4570.23 0.60670.008 430719

Tephra MO6 4.6870.26 0.26970.026 251747 3.1970.21 0.16870.024 171760

Tephra MO7 3.4670.18 0.24070.020 154744 – – –

Tephra MO8 3.5370.26 0.26770.031 187748 2.0970.14 0.11870.019 152761

a Fraction bigger than 500 mm.b Standard deviation of the average.

R. Daga et al. / Applied Radiation and Isotopes 66 (2008) 1325–1336 1333

Author's personal copy

(Thompson, 1984), are shown in Fig. 5, with the shaded arearepresenting the range of concentrations for each volcanic system.The glass shards compositions from lake Ilon and Brazo Rincontephras are also shown in Fig. 5. There is a clear differentiationbetween lake Ilon and Brazo Rincon patterns, in agreement withprevious analysis of different geochemical parameters. But it isnoteworthy the coincidence of lake Ilon pattern with the rock datafrom Calbuco volcano, and Brazo Rincon pattern with Puyehue-Cordon Caulle, not only in shape but in absolute values (Fig. 5),confirming that the glass shards extracted from lake Ilon tephrascan be associated to volcano Calbuco source, while the glass

shards from Brazo Rincon were generated by Puyehue-CordonCaulle volcanic events.

In Brazo Rincon, white pumice fragments show similarcomposition as glass shards. But in lake Ilon, white pumicecomposition range in-between glass shards and Brazo Rinconpumice or glass shards. The different behaviour of white pumicecompositions compared to glass shards could be explained by adifferential evolution of processes occurring in the magmachamber and conduit at time of eruption or by the contributionof both sources to the same tephra layer. Considering that theeruptive historic records show that there was explosive volcanicactivity of both volcanoes simultaneously (Table 3), it is possibleto support the hypothesis that in the same tephra layers there arerecords of both sources. The composition of white pumicefragments from Brazo Rincon coincides with published rock datafrom Puyehue-Cordon Caulle system, but in lake Ilon whitepumice composition varies in-between Calbuco and Puyehue-Cordon Caulle rock patterns (Fig. 5), supporting the hypothesis ofwhite pumice contributions from both sources in lake Ilon.Further research on spatial distribution of the volcanic productsfrom both sources is on going, which would clarify this point.

The characterization of the volcanic ashes from lake Morenosequence show unambiguous association of MO1 and MO5 tephralayers with Puyehue-Cordon Caulle source, and MO2, MO3, MO4,MO6, and MO7 have lake Ilon patterns, that could be associated tovolcano Calbuco source. Some geochemical parameters showvalues in tephra MO8 that neither correspond to Puyehue-CordonCaulle source nor to lake Ilon patterns, particularly in whitepumice, but most of glass shards show associations to lake Ilon.Therefore, we can associate tephra MO8 to lake Ilon patterns.

4. Conclusions

The primary components white pumice, glass shards, andscoriae, isolated from the three tephra layers extracted from thesediment sequence sampled from lake Ilon, that correspond to

ARTICLE IN PRESS

Rb

10

100

Sample/Chondrite

YbTiSmZrHfNdSrCeLaTaThBa

Fig. 5. Lake Ilon and lake Nahuel Huapi, Brazo Rincon (Daga et al., 2006)

incompatible elements composition, normalized to Chondrite (Thompson, 1984),

of glass shards primary component. In shadow, the composition of rock samples

collected from Calbuco and Puyehue-Cordon Caulle volcanoes (Gerlach et al., 1988;

Lopez-Escobar et al., 1995).

Table 9Cs, Rb, and Th concentration in primary components of volcanic ashes from sedimentary sequences extracted from lake Nahuel Huapi, Brazo Rincon (Daga et al., 2006), lake

Moreno Oeste, and lake Ilon, and volcanic ashes collected manually after the 1960 event of Puyehue-Cordon Caulle (MUS)

White pumicea (mg g�1) Glass shardsa (mg g�1)

Cs Rb Th Cs Rb Th

Puyehue-Cordon Caulle source

MUS ashes 4.7570.32 76.975.5 8.7470.51 – – –

Tephra BR1 4.6470.27 75.974.6 8.8470.49 4.9770.34 78.475.3 9.4770.56

Tephra BR2 4.4470.30 70.074.8 8.1370.43 4.417 0.30 72.075.1 8.2670.47

Tephra BR4 3.8770.26 60.374.0 7.3370.43 4.2670.29 68.375.3 7.8470.46

Average (SDb) 4.43 (0.39) 70.8 (7.6) 8.26 (0.70) 4.55 (0.38) 72.9 (5.1) 8.52 (0.85)

Lake Ilon

Tephra IL1 2.7470.23 55.575.2 5.2370.40 1.2770.16 20.074.4 1.1970.11

Tephra IL2 – – – 1.0870.11 19.873.6 1.0770.10

Tephra IL3 2.7670.20 48.873.9 5.1470.31 1.5570.14 29.074.1 1.7370.14

Average (SDb) 2.75 (0.01) 52.2 (4.7) 5.19 (0.06) 1.30 (0.24) 22.9 (5.3) 1.33 (0.35)

Lake Moreno Oeste

Tephra MO1 4.9270.06 7872 8.9770.28 – – –

Tephra MO2 – – – 1.1670.13 21.275.1 1.2470.11

Tephra MO3 2.9170.15 51.171.4 5.2770.23 – – –

Tephra MO4 – – – 1.3370.15 25.473.3 1.3970.11

Tephra MO5 4.6270.18 74.372.0 8.7970.19 4.5270.03 76.671.3 8.5270.11

Tephra MO6 2.2870.18 34.372.9 3.5370.21 1.5770.15 28.174.7 2.0070.13

Tephra MO7 1.8370.14 29.973.2 1.8370.12 – – –

Tephra MO8 2.1270.17 31.673.8 1.9970.17 0.87870.088 24.974.2 1.9170.13

a Fraction bigger than 500 mm.b Standard deviation of the average.

R. Daga et al. / Applied Radiation and Isotopes 66 (2008) 1325–13361334

Author's personal copy

primary airfall of different volcanic events, have similar composi-tion. The source of these events would be volcano Calbuco,according to the geographical situation and the association of137Cs dating with historical records. Therefore, the geochemicalparameters obtained from the major and trace elements char-acterize volcano Calbuco source in recent events. This character-ization is clearly distinctive from the Puyehue-Cordon Caullesource analysed in previous work. The distinctive geochemicalparameters are SiO2 and Na2O+K2O concentrations in whitepumice and glass shards; Cr and V concentrations in glass shards,and V in white pumice; the Eu anomaly, light to middle and lightto heavy REE ratios in glass shards; Hf, Ta, and Zr concentrations inglass shards and white pumice; and Cs, Rb, and Th concentrationsin white pumice and glass shards. Another distinctive featurebetween both sources is that the geochemical parameters aresimilar comparing glass shards and white pumice from BrazoRincon tephras, but they are notably different comparing glassshards and white pumice from lake Ilon.

The pattern of incompatible elements contents in glass shardsfrom lake Ilon coincides with published rock data from volcanoCalbuco, while the pattern of glass shards from Brazo Rincon,clearly distinctive from lake Ilon, coincide with rock data fromPuyehue-Cordon Caulle, confirming source association in bothcases.

The geochemical characterization allowed the source identifi-cation of the tephra layers extracted from lake Moreno, with aconsistent association of all the geochemical parameters studied.MO1 and MO5, dated in 1960 and 1620 respectively, can beassociated to Puyehue-Cordon Caulle source; MO2, MO3, andMO4 (corresponding to 1920, 1860, and 1775 dates, respectively),MO6, MO7, and MO7 can be associated to Calbuco source. MO7and MO8 tephra layers show differences to lake Ilon patternsparticularly in white pumice.

Acknowledgments

We thank Andrea Rizzo and Ricardo Sanchez for sediment coresampling, Juan Otamendi for his valuable comments, and reactorRA-6 operation staff for their assistance in sample analysis. Thisproject has been partially funded by the Secretarıa de Ciencia,Tecnica y Posgrado, UNCuyo, project Cod. 06/C175.

References

Arnaud, F., Magand, O., Chapron, E., Bertrand, S., Boes, X., Charlet, F., Melieres, M.A.,2006. Radionuclide dating (210Pb, 137Cs, 241Am) of recent lake sediments in ahighly active geodynamic setting (Lakes Puyehue and Icalma—Chilean LakeDistrict). Sci. Total Environ. 366 (2–3), 837–850.

Bichler, M., Duma, B., Huber, H., 2004. Application of INAA to reveal the chemicalevolution of selected volcanic eruptiva from Santorini, Greece. J. Radioanal.Nucl. Chem. 262 (1), 57–65.

Borchardt, G.A., Aruscavage, P.J., Millard, H.T., 1972. Correlation of the Bishop ash, aPleistocene marker bed, using instrumental neutron activation analysis. J.Sediment. Petrol. 42 (2), 301–306.

Boygle, J., 1999. Variability of tephra in lake and catchment sediments, Svınavatn,Iceland. Global Planet. Change 21, 129–149.

Boynton, W.V., 1984. Geochemistry of the rare earth elements: meteorite studies.In: Rollinson, H. (Ed.), Using Geochemical Data: Evaluation, Presentation,Interpretation. Wiley, New York, USA, p. 134.

Chambers, F.M., Daniell, J.R.G., Hunt, J.B., Molloy, K., O’Connell, M., 2004.Tephrostratigraphy of An Loch Mor, Inis Oırr, western Ireland: implicationsfor Holocene tephrochronology in the northeastern Atlantic region. Holocene14 (5), 703–720.

Chapron, E., Ariztegui, D., Mulsow, S., Villarosa, G., Pino, M., Outes, V., Juvigine, E.,Crivelli, E., 2006. Impact of 1960 major subduction earthquake in NorthernPatagonia (Chile, Argentina). Quat. Int. 158, 58–71.

Clift, P., Fitton, G., 1998. Trace and rare earth element chemistry of volcanic ashesfrom sites 918 and 919: implications for Icelandic volcanism. In: Proceedings ofthe Ocean Drilling Program, Scientific Results, vol. 152, pp. 67–84.

Daga, R., Ribeiro Guevara, S., Sanchez, M.L., Arribere, M., 2006. Geochemicalcharacterization of volcanic ashes from recent events in Northern PatagoniaAndean range by INAA. J. Radioanal. Nucl. Chem. 270 (3), 677–694.

De Corte, F., 2003. The updated NAA nuclear data library derived from the Y2K k0-database. J. Radioanal. Nucl. Chem. 257 (3), 493–499.

Eastwood, W., Pearce, N., Westgate, J., Perkins, W., Lamb, H., Roberts, N., 1999.Geochemistry of Santorini tephra in lake sediments from Southwest Turkey.Global Planet. Change 21, 17–29.

Firestone, R.B., Shirley, V., 1996. Table of Isotopes. Wiley, New York, USA.Fontaine, C.S., Kaufman, D.S., Anderson, R.S., Werner, A., Waythomas, C.F., Brown,

T.A., 2007. Late quaternary distal tephra-fall deposits in lacustrine sediments,Kenai Peninsula, Alaska. Quat. Res. 68, 64–78.

Gehrels, M.J., Lowe, D.J., Hazell, Z.J., Newnham, R.M., 2006. A continuous 5300-yrHolocenet cryptotephrostratigraphic record from northern New Zealand andimplications for tephrochronology and volcanic hazard assessment. Holocene16 (2), 173–187.

Gerlach, D., Frey, F., Moreno-Roa, H., Lopez-Escobar, L., 1988. Recent volcanism inthe Puyehue—Cordon Caulle region, Southern Andes, Chile (40?51S):petrogenesis of evolved lavas. J. Petrol. 29 (2), 333–382.

Gordon, G.E., Randle, K., Goles, G.G., Corliss, J.B., Beeson, M.H., Oxley, S.S., 1968.Instrumental activation analysis of standard rocks with high-resolution g-raydetectors. Geochim. Cosmochim. Acta 32, 369–396.

Haberle, S., Lumley, S., 1998. Age and origin of tephras recorded in postglacial lakesediments to the west of the southern Andes, 441S to 471S. J. Volcanol.Geotherm. Res. 84, 239–256.

Hanson, G.N., 1989. An approach to trace element modelling using a simpleigneous system an example. In: Lipin, A., McKay, B. (Eds.), Geochemistry andMineralogy on Rare Earth Elements, pp. 79–97.

Hickey, R.L., Gerlach, D.C., Frey, F.A., 1984. Geochemical variations in volcanic rocksfrom central south Chile (33–421 S). In: Harmon, R.S., Barreiro, B. (Eds.),Andean Magmatism: Chemical and Isotopic Constraints. Shiva PublicationsLtd., pp. 72–92.

Hildreth, W., Moorbath, S., 1988. Crustal contributions to arc magmatism in theAndes of Central Chile. Contrib. Mineral. Petrol. 98, 455–489.

Horz, K., Worthington, T., Winn, K., Stoffers, P., 2004. Late quaternary tephra in theNew Ireland Basin, Papua New Guinea. J. Volcanol. Geotherm. Res. 132, 73–95.

Joshi, S.R., Shukla, B.S., 1991. Ab initio derivation of formulations for 210Pb dating ofsediments. J. Radioanal. Nucl. Chem. 148 (1), 73–79.

Lacasse, C., Werner, R., Paterne, M., Sigurdsson, H., Carey, S., Pinte, G., 1998.Long-range transport of Icelandic tephra to the Irminger Basin, Site 919.In: Proceedings of the Ocean Drilling Program, Scientific Results, 152,pp. 51–65.

Lara, L.E., Naranjo, J.A., Moreno, H., 2004. Rhyodacitic fissure eruption in SouthernAndes (Cordon Caulle; 40.5 8S) after the 1960 (Mw: 9.5) Chilean earthquake: astructural interpretation. J. Volcanol. Geotherm. Res. 138, 127–138.

Lopez-Escobar, L., Parada, M., Moreno, H., 1992. A contribution to the petrogenesisof Osorno and Calbuco volcanoes, Southern Andes (411000–411300S): compara-tive study. Rev. Geol. Chile 19 (2), 211–226.

Lopez-Escobar, L., Parada, M., Hickey-Vargas, R., Frey, F., Kempton, P., Moreno, H.,1995. Calbuco Volcano and minor eruptive centers distributed along theLiquine-Ofqui Fault Zone, Chile (41–421 S): contrasting origin of andesitic andbasaltic magma in the Southern Volcanic Zone of the Andes. Contrib. Mineral.Petrol. 119, 345–361.

Massaferro, J., Ribeiro Guevara, S., Rizzo, A., Arribere, M., 2005. Short-termenvironmental changes in Lake Morenito (41 1S, 71 1W, Patagonia, Argentina)from the analysis of sub-fossil chironomids. Aquat. Conserv.: Marine Fresh-water Ecosyst. 15 (1), 23–30.

McHenry, L.J., Mollel, D.F., Swisher, C.C., Compositional and textural correlationsbetween Olduvai Gorge Bed I tephra and volcanic sources in the NgorongoroVolcanic Highlands, Tanzania. Quat. Int., in press.

McKay, G.A., 1989. Partitioning of rare earth elements between major silicateminerals and basaltic melts. In: Lipin, A., McKay, B. (Eds.), Geochemistry andmineralogy on Rare Earth Elements, pp. 45–77.

McLennan, S.M., 1989. Rare earth elements in sedimentary rocks: influence ofprovenance and sedimentary processes. In: Lipin, A., McKay, B. (Eds.),Geochemistry and Mineralogy on Rare Earth Elements, pp. 169–200.

Mc Phie, J., Doyle, M., Allen, R., 1993. Volcanic Textures: A Guide to theInterpretation of Textures in Volcanic Rocks. Centre for Ore Deposit andExploration Studies, University of Tasmania, pp. 198.

Mughabghab, S.F., 2003. Thermal neutron capture cross sections resonanceintegrals and g-factors. IAEA INDC(NDS)-440.

Mughabghab, S.F., Divadeenam, M., Holden, N.E., 1981. Neutron Cross Sections, vol.1 and 2. Academic Press, New York, USA.

Nakada, S., Matsushima, T., Yoshimoto, M., Sugimoto, T., Kato, T., Watanabe, T.,Chong, R., Camacho, J., 2005. Geological aspects of the 2003–2004 eruption ofAnatahan Volcano, Northern Mariana Islands. J. Volcanol. Geotherm. Res. 146,226–240.

Ortega-Guerrero, B., Newton, A.J., 1998. Geochemical characterization of latepleistocene and holocene tephra layers from the basin of Mexico, CentralMexico. Quat. Res. 50, 90–106.

Petit-Breuilh Sepulveda, M.E., 2004. La historia eruptiva de los volcaneshispanoamericanos (siglos XVI al XX). Serie Casa de los Volcanes No. 8. Ed.Servicio de Publicaciones Exmo. Cabildo Insular de Lanzarote, Huelva, Spain.

Quiros, R., 1988. Relationships between air temperature, depth, nutrients,and chlorophyll in 103 Agentinian lakes. Verh. Internat. Ver. Limnol. 23,647–658.

ARTICLE IN PRESS

R. Daga et al. / Applied Radiation and Isotopes 66 (2008) 1325–1336 1335

Author's personal copy

Ribeiro Guevara, S., Arribere, M., 2002. 137Cs dating of sedimentary cores fromlakes of Nahuel Huapi National Park, Patagonia, Argentina: historical recordsand profile measurements. J. Radioanal. Nucl. Chem. 252 (1), 37–45.

Ribeiro Guevara, S., Rizzo, A., Sanchez, R., Arribere, M., 2003. 210Pb fluxes insediment layers sampled from Northern Patagonia lakes. J. Radioanal. Nucl.Chem. 258 (3), 583–595.

Ribeiro Guevara, S., Rizzo, A., Sanchez, R., Arribere, M., 2005. Heavy metal inputs inNorthern Patagonia lakes from short sediment cores analysis. J. Radioanal.Nucl. Chem. 265 (3), 481–493.

Robbins, J.A., Herche, L.R., 1993. Models and uncertainty in 210Pb dating ofsediments. Radiochem. Limnol. 25, 217–222.

Rollinson, H., 1993. Using Geochemical Data: Evaluation, Presentation, Interpreta-tion. Wiley, New York, USA.

Saminger, S., Peltz, M., Bichler, M., 2000. South Aegean volcanic glass: separationand analysis by INAA and EPMA. J. Radioanal. Nucl. Chem. 245 (2), 375–383.

Shane, P., 2000. Tephrochronology: a New Zealand case study. Earth-Sci. Rev. 49,223–259.

Steinhauser, G., Sterba, J.H., Bichler, M., Huber, H., 2006. Neutron activationanalysis of Mediterranean volcanic rocks, an analytical database for archae-ological stratigraphy. Appl. Geochem. 21, 1362–1375.

Steinhauser, G., Sterba, J.H., Bichler, M., 2007. Chemical fingerprints of pumice fromCappadocia (Turkey) and Kos (Greece) for archaeological applications. Appl.Radiat. Isotopes 65, 488–503.

Thompson, R.N., 1984. British tertiary volcanic province. In: Rollinson, H. (Ed.),Using Geochemical Data: Evaluation, Presentation, Interpretation. Wiley, NewYork, USA, pp. 143–161.

Tormey, D., Hickey-Vargas, R., Frey, F., Lopez-Escobar, L., 1991. Recent lavas from theAndean volcanic front (33–421S); Interpretations of along-arc compositionalvariations. In: Russell Harmon, Carlos Rapela (Eds.), Andean Magmatism andits tectonic setting, Geological Society of America. Special Paper 265,pp. 57–77.

Taylor, S.R., McLennan, S.M., 1988. The significance of the rare earths ingeochemistry and cosmochemistry. In: Gschneidner, A., Eyring, B. (Eds.),Handbook on the Physics and Chemistry of Rare Earth, Vol. 11, pp. 485–579.

Tuli, K.J., 2005. Nuclear Wallet Cards. National Brookhaven Laboratory, USA.Wade, J.A., Plank, T., Stern, R.J., Tollstrup, D.L., Gill, J.B., O’Leary, J.C., Eiler, J.M.,

Moore, R.B., Woodhead, J.D., Trusdell, F., Fischer, T.P., Hilton, D.R., 2005. TheMay 2003 eruption of Anatahan volcano, Mariana Islands: geochemicalevolution of a silicic island-arc volcano. J. Volcanol. Geotherm. Res. 146,139–170.

Wastegard, S., 2005. Late quaternary tephrochronology of Sweden: a review. Quat.Int. 130 (1), 49–62.

Werner, R., van den Bogaard, P., Lacasse, C., Schmincke, H., 1998. Chemicalcomposition, age, and sources of volcaniclastic sediments from sites 917 and918. In: Proceedings of the Ocean Drilling Program, Scientific Results, 152,pp. 93–113.

Westgate, J.A., Gorton, M.P., 1981. Correlation techniques in tephra studies. In: Self,A., Sparks, B. (Eds.), Tephra Studies, pp. 73–94.

Wilson, M., 1989. Igneous Petrogenesis. Chapman & Hall, London.

ARTICLE IN PRESS

R. Daga et al. / Applied Radiation and Isotopes 66 (2008) 1325–13361336