University of Groningen

New radiometric and petrological constraints on the evolution of the Pichincha volcaniccomplex (Ecuador)Robin, Claude; Samaniego, Pablo; Le Pennec, Jean-Luc; Fornari, Michel; Mothes, Patricia;van der Plicht, Johannes; Stix, J.Published in:Bulletin of Volcanology

DOI:10.1007/s00445-010-0389-0

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2010

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Robin, C., Samaniego, P., Le Pennec, J-L., Fornari, M., Mothes, P., van der Plicht, J., & Stix, J. (Ed.)(2010). New radiometric and petrological constraints on the evolution of the Pichincha volcanic complex(Ecuador). Bulletin of Volcanology, 72(9), 1109-1129. DOI: 10.1007/s00445-010-0389-0

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 10-02-2018

RESEARCH ARTICLE

New radiometric and petrological constraintson the evolution of the Pichincha volcaniccomplex (Ecuador)

Claude Robin & Pablo Samaniego &

Jean-Luc Le Pennec & Michel Fornari &Patricia Mothes & Johannes van der Plicht

Received: 20 May 2009 /Accepted: 18 June 2010 /Published online: 18 July 2010# Springer-Verlag 2010

Abstract Fieldwork, radiometric (40Ar/39Ar and 14C) agesand whole-rock geochemistry allow a reconstruction oferuptive stages at the active, mainly dacitic, PichinchaVolcanic Complex (PVC), whose eruptions have repeatedlythreatened Quito, most recently from 1999 to 2001. Afterthe emplacement of basal lavas dated at ∼1100 to 900 ka,the eruptive activity of the old Rucu Pichincha volcano

lasted from ∼850 ka to ∼150 ka before present (BP) andresulted in a 15×20 km-wide edifice, which comprisesthree main building stages: (1) A lower stratocone (LowerRucu, ∼160 km3 in volume) developed from ∼850 to600 ka; (2) This edifice was capped by a steeper-sided andless voluminous cone (the Upper Rucu, 40–50 km3), thehistory of which started 450–430 ka ago and ended around250 ka with a sector collapse; (3) A smaller (8–10 km3) butmore explosive edifice grew in the avalanche amphitheatreand ended Rucu Pichincha's history about 150 ka ago. TheGuagua Pichincha volcano (GGP) was developed from60 ka on the western flank of Rucu with four growth stagesseparated by major catastrophic events. (1) From ∼60 to47 ka, a basal effusive stratocone developed, terminatingwith a large ash-and-pumice flow event. (2) This basalvolcano was followed by a long-lasting dome buildingstage and related explosive episodes, the latter occurringbetween 28–30 and 22–23 ka. These first two stagesformed the main GGP (∼30 km3), a large part of whichwas removed by a major collapse 11 ka BP. (3) Sustainedexplosive activity and viscous lava extrusions gave rise to anew edifice, Toaza (4–5 km3 in volume), which in turncollapsed around 4 ka BP. (4) The ensuing amphitheatrewas partly filled by the ∼1-km3 Cristal dome, which is thehistorically active centre of the Pichincha complex. Theaverage output rate for the whole PVC is 0.29 km3/ka.Nevertheless, the chronostratigraphic resolution weobtained for Lower Rucu Pichincha and for the two mainedifices of Guagua Pichincha (main GGP and Toaza), leadsto eruptive rates of 0.60–0.65 km3/ka during theseconstruction stages. These output rates are compared tothose of other mainly dacitic volcanoes from continentalarcs. Our study also supports an overall SiO2 and large-ionlithophile elements enrichment as the PVC develops. Inparticular, distinctive geochemical signatures indicate the

Editorial responsibility: J. Stix

C. Robin : P. Samaniego : J.-L. Le PennecLaboratoire Magmas et Volcans, Clermont Université,Université Blaise Pascal,BP 10448,63000 Clermont-Ferrand, France

C. Robin : P. Samaniego : J.-L. Le PennecCNRS UMR 6524, IRD R163,5 rue Kessler,63038 Clermont-Ferrand cedex, France

C. Robin : P. Samaniego : P. MothesInstituto Geofísico, Escuela Politécnica Nacional,Ap. 17-01-2759,Quito, Ecuador

M. FornariIRD, UMR Géosciences Azur, Université Nice-Sophia Antipolis,Parc Valrose,06108 Nice, France

J. van der PlichtCenter for Isotope Research, University of Groningen,Nijenborgh 4,9747 AG Groningen, Netherlands

C. Robin (*) : P. Samaniego (*)Laboratoire Magmas et Volcans, Université Blaise Pascal,5 rue Kessler,63038 Clermont-Ferrand cedx, Francee-mail: claude.robin@ird.fre-mail: pablo.samaniego@ird.fr

Bull Volcanol (2010) 72:1109–1129DOI 10.1007/s00445-010-0389-0

involvement of a new magma batch at the transitionbetween Rucu and Guagua. At the GGP, the samephenomenon occurs at each major collapse event markingthe onset of the ensuing magmatic stage. Since the 11-ka-BP collapse event, this magmatic behaviour has led toincreasingly explosive activity. Four explosive cycles ofbetween 100 and 200 years long have taken place at theCristal dome in the past 3.7 ka, and repose intervalsbetween these cycles have tended to decrease with time. Asa consequence, we suggest that the 1999–2001 eruptiveperiod may have initiated a new eruptive cycle that mightpose a future hazard to Quito (∼2 million inhabitants).

Keywords Explosive volcanism . 40Ar–39Argeochronology . 14C geochronology . Pichincha volcano .

Eruptive rate . Eruption frequency

Introduction

The eruptive activity of volcanoes with a dominantlydacitic composition comprises dome-building periods withconcomitant block-and-ash flow eruptions (e.g. Mt. Pelée,Unzen), violent open-conduit Plinian phases and directedblast events (e.g. El Chichón, Pinatubo), as well as sectorcollapse of all sizes (e.g. Shiveluch, St. Helens). Suchvolcanoes are found mainly in oceanic arcs and activecontinental plate margins, as in the cordilleras of thewestern Americas. In the Ecuadorian Andes such volcanoesoccur in urbanized regions (Fig. 1a, b). In AD 1660, the lastnotable eruption of Guagua Pichincha volcano severelyaffected Ecuador's capital, Quito, with heavy tephra fallsand rain-triggered debris flows.

Phreatic activity resumed in 1981 at this volcano,followed by a magmatic-phreatomagmatic eruption lastingfrom September 1999 to March 2001, which includedvulcanian-subplinian explosions alternating with small-volume dacite-dome extrusions (Garcia-Aristizabal et al.2007, Fig. 2a, c). However, despite its importance in termsof potential volcanic hazards, the eruptive chronology ofGuagua Pichincha and that of the associated RucuPichincha volcanic complex (PVC) have been littleinvestigated. Geotermica Italiana (1989) and Barberi etal. (1992) dated the onset of the volcanism at ∼1 Ma andidentified a large sector collapse in late Pleistocene times.They also focused on the Holocene activity, which wasfurther investigated by Robin et al. (2008). These previousstudies, however, provide few constraints on the growth ofthe full PVC, its eruptive stages, magma production andpetrological characteristics.

Geochronological results combined with field investiga-tions and geochemical data allow us to reconstruct the maineruptive stages of the PVC since middle Pleistocene times,

with emphasis on the active Guagua Pichincha volcano. Weare thus able to identify the magmatic evolution whichtakes place over this period. Finally, we compare latePleistocene-Holocene eruptive return rates and styles topre-Columbian and historical eruptions for hazard assess-ment purposes.

Geological setting

The Ecuadorian segment of the Northern Andean VolcanicZone results from subduction of the oceanic Nazca Platebeneath the South American continental lithosphere. InEcuador, the volcanic arc is 120-km wide and is dividedinto three zones: the volcanic front along the WesternCordillera, the main arc in the Eastern Cordillera, and theback-arc in the upper Amazonian regions (Fig. 1a). Thevolcanoes from the Eastern Cordillera and back-arc regionsare mainly andesitic in composition, for example Antisana(Bourdon et al. 2002), Cotopaxi (Hall and Mothes 2008a),Tungurahua (Hall et al. 1999; Le Pennec et al. 2008),Sangay (Monzier et al. 1999) and El Reventador (Samaniegoet al. 2008). In contrast, the 250-km-long volcanic frontconsists of mainly dacitic centres including the dormantvolcanoes of Pululahua (Papale and Rosi 1993), Atacazo(Hidalgo et al. 2008), Illiniza (Hidalgo et al. 2007) andQuilotoa (Hall and Mothes 2008b), and the historicallyactive Pichincha volcano. This volcanic front sits above an∼50-km-thick continental crust and the slab surface is at100–120-km depth (Guillier et al. 2001).

The large twin-peaked PVC comprises the old, erodedRucu Pichincha (hereafter referred to as "Rucu"; 4,627 masl) and the younger Guagua Pichincha (the "baby"Pichincha hereafter referred to as "GGP"; 4,776 m asl).The active lava dome (0.167°S–78.610°W) is nested in aprominent avalanche amphitheatre located only 12 km westof the colonial centre of Quito. This urban area extendsalong a NS-oriented basin (the Quito basin) at 2,800–2,900-m elevation between the Western Cordillera and theInter-Andean valley (Fig. 1b). The eastern margin of theQuito basin is an active, westward-dipping reverse faultsystem (Ego et al. 1996), whose curved geometry, as seenin plain view in Fig. 1b, probably results from the spreadingof the PVC. Legrand et al. (2002) showed the link betweenthe 1998 and 2001 seismicity in the area and eruptiveactivity at the GGP. The western suburbs of Quito extendonto the slopes of the volcanic complex (Figs. 2b, 3a).

The PVC is asymmetric in shape with a basal elevation at∼2,900 m asl to the east and ∼2,500 m asl to the west; whereas,the eastern part of the complex overlies volcaniclasticsequences in theQuito basin, the southwest, west and northwestsides exhibit rugged topography (Fig. 3b, c) controlled bySSW-NNE-trending tectonic lineaments affecting the Western

1110 Bull Volcanol (2010) 72:1109–1129

Cordillera basement. The latter consists of Cretaceousoceanic-plateau basalts and related ultramafic rocks locallyoverlain by Mesozoic to Cenozoic volcano-sedimentarysequences (Hughes and Pilatasig 2002).

Methodology

Fieldwork and sampling were carried out with the aim ofidentifying the main volcanic units and fingerprinting theirgeochemistry. Steep topography and dense forest covergenerally result in patchy and disconnected exposures, butthe GGP tephra deposits are extensively exposed alongpathways, notably the Lloa-Rio Cristal and Lloa-Refugeroads (Fig. 4). Forty key stratigraphic sections wereinvestigated for this study, with 18 presented below.

The Pleistocene chronology is based on 12 new40Ar–39Ar age determinations obtained on whole-rocksamples at the Geochronology Laboratory at GéosciencesAzur, University of Nice (France). The criteria used fordefining a “plateau” age include: (1) at least 60% of total

39Ar is released in the plateau; (2) at least three successivestep-heating fractions are analysed in the plateau; (3) theintegrated age (weighted average of apparent ages ofindividual fractions) agrees with each apparent age of theplateau within a two sigma error. Ages were obtained frommeasured isotope ratios corrected for mass fractionation,system blanks, and interfering isotopes produced duringirradiation. Isochron ages were calculated using a reverseisochron diagram of 36Ar/40Ar vs. 39Ar/40Ar, using themethod of least squares. Results are presented in Table 1.

Twelve new radiocarbon ages were obtained fromsamples of organic matter in pyroclastic deposits andpalaeosols. Determinations were carried out at the Centrefor Isotope Research, Groningen University (Netherlands),using pre-treatment and analytical procedures described inMook and Streurman (1983). Results in Table 2 are givenas conventional 14C ages corrected for 13C fractionation,with one sigma confidence level.

Major and trace element analysis of 232 whole-rocksamples was carried out by ICP-AES at the LaboratoireDomaines Océaniques, Université de Bretagne Occidentale

Fig. 1 a The Ecuadorian volcanic arc. Main volcanoes: (1) Pululahua;(2) Atacazo-Ninahuilca; (3) Cayambe; (4) El Reventador; (5) Chacana;(6) Antisana; (7) Cotopaxi; (8) Tungurahua; (9) Chimborazo; (10)Sangay. b Digital elevation model of the PVC and the reverse Quitofault system. Note the fault flexures and faults (yellow lines) east ofPichincha and Atacazo volcanic complexes. c Geodynamic setting of

the Ecuadorian arc, including main oceanic features. Andes Cordilleradefined by a 2,000-m contour line. Trench is defined by a toothed lineand active volcanoes by open triangles. Black arrow is the direction ofsubduction. GSC Galápagos Spreading Centre, GFZ Grijalva FractureZone, PVC Pichincha Volcanic Complex

Bull Volcanol (2010) 72:1109–1129 1111

(Brest, France). Here we have selected major and traceelement analyses to correlate and characterize the volcanicunits identified in stratigraphic successions.

Chronostratigraphy

Global structure of the volcanic complex

The PVC comprises two large, andesitic to daciticcomposite volcanoes, which were built over lavas from ofMiddle Pleistocene age. The earlier Rucu Pichinchavolcano was built up in three stages. A lower stratoconeand an upper steeper-sided cone form the main volume ofRucu (Table 3). After a large sector collapse, a smalleredifice grew in the avalanche amphitheatre and ended RucuPichincha's history. Rucu samples are mostly andesitic (59-63 wt.% SiO2) with scarce basic andesites (55–57 wt.%SiO2) and contain a mineral assemblage of plagioclase,ortho- and clinopyroxene, Fe-Ti oxides, and scarce amphi-bole. Four growth stages separated by major collapse eventsmark the development of Guagua Pichincha since the latePleistocene: (1) a basal effusive stratocone; (2) long-lastingphases of dome building and related explosive activity;these two early stages formed the main volume of GuaguaPichincha (main GGP), which collapsed at the Pleistocene–Holocene transition; (3) a smaller edifice, Toaza, which inturn collapsed around 4 ka before present (BP); and (4) Theensuing amphitheatre is currently filled by the active Cristaldome. The magmatic suite of GGP (58–66 wt.% SiO2) ismore differentiated than that of Rucu and significant

differences in the composition of the rocks are notedbetween the successive edifices (Table 3).

El Cinto basal lava flows (>900 ka)

The arc-shaped El Cinto ridge located southeast of thecomplex between the Rio Cinto valley and Quito (Fig. 3b)consists of an ∼300-m-thick lava series of acid andesite todacitic compositions (62.0-63.5 wt.% SiO2). A lava nearthe base is 40Ar–39Ar dated at 1100±10 ka (PICH 111C,Table 1), while a dacite sample from the Ungüi dome,which caps the series, yields a plateau age of 910±7 ka(PICH 21A). The shape of the ridge, largely dissected byerosion and faulting, and its distinct composition suggest along-lived edifice of significant size and Lower Pleistoceneage, now buried below Rucu volcano.

Rucu Pichincha volcano (∼850 to ∼150 ka)

Rucu Pichincha is a slightly elongated (15–20-km-wide)effusive stratovolcano. The eastern and northeastern slopesare gently inclined (8–13°) and dissected by deep radialglacial valleys such as Rio Pichán, Quebrada Rumihurco,Quebrada Rumipamba, and Quebrada Rumihurcu (Fig. 3a).Our age constraints and the geomorphologic relationshipsbetween pre- and post-erosion lavas point to three structuralunits.

Lower Rucu edifice (∼850–600 ka) Monotonous lavasequences with sparse inter-layered breccias on the easternflank locally reach 1,000 m in thickness, as in the

Fig. 2 a View of the Cristaldome complex, the active centreof Guagua Pichincha, as seen in1999. b View of the PVC fromQuito. c Vulcanian to sub-plinian explosion from October7, 1999, showing a plumereaching 12 km above the crater(photography by P. Zway)

1112 Bull Volcanol (2010) 72:1109–1129

Rumipamba valley, giving Rucu's edifice its conical shape.Similar flows on the southwestern flank comprise thebuttresses of the volcanic complex (e.g. Loma Pucara,Fig. 3b). Two 40Ar–39Ar ages of 850±9 and 850±10 kahave been obtained for andesite blocks from the lowernorthern flank (PICH 70B and 71, Table 1), and twoconsistent ages of 638±15 ka and 608±10 ka (samplesPICH 107B and 56) are obtained for andesite lavas fromdivergent Rumipamba and Rumihurcu valleys on theeastern and northeastern flanks, respectively.

Upper Rucu edifice (∼430–250 ka) After a repose periodand severe erosion, lavas flowed 10–15 km eastward from

the presumed vent and infilled the drainage channelsincised in the Rumipamba and Rumihurcu lava sequences(Fig. 5). An andesite sample dated at 409±10 ka (PICH57, Table 1) reveals that this new lava cone is notablyyounger than Lower Rucu. Higher on the edifice, above4,100 m asl, the lower part of a 300- to 500-m-thick seriesof subglacial breccias exhibit andesitic compositions(57.3–59.1 wt.% SiO2) and a similar age (PICH 63C;436±18 ka). The youngest products from this cone arerepresented by acid andesites (62.0–63.0 wt.% SiO2); asummit breccia sample is dated at 260±6 ka (PICH 51),while a lava block from the northern slope yields an age of270±5 ka (PICH 76).

Fig. 3 Panoramic views of thePVC. a View from the eastshowing the Rucu Pichincha(RP), the Quito basin, and theQuito reverse fault system. bSouthwest flank of the PVCshowing the Rucu and GuaguaPichincha edifices, and thepyroclastic Lloa fan. c Viewfrom the northwest, showing theCantera del Diablo outcrop andthe Rio Mindo valley

Bull Volcanol (2010) 72:1109–1129 1113

The depression that opens to the north (D1 structure,Fig. 5) and the nearby cliffs on the upper flanks of the Rucuvolcano probably resulted from intense glacial erosionduring the late Pleistocene times (Clapperton 1993; Clappertonet al. 1997). However, a 3-km-wide horseshoe-shapedstructure is preserved high in the Rumipamba valley and isinterpreted as the result of a flank failure event (RAA=Rucuavalanche amphitheatre, Fig. 5). The opening of this structureto the east is consistent with debris avalanche depositsdescribed in the eastern Quito basin (Villagómez et al.2002). Since late lavas of the Upper Rucu edifice wereinvolved in the collapse, this avalanche event can be dated ataround 250 ka.

We have estimated the size of the Lower and Upper Rucuedifices as follows. The projected surface of the base of theedifice, calculated by Geographic Information Systems (GIS)techniques, is 306 km2, and the pre-existing topographyinferred from field mapping of lava units points to areconstructed summit elevation of 4,800 m asl, after theconstruction of the Upper Rucu. Taking the unevenmorphology of the base into account (i.e. elevation from2,900 to 2,500 m asl), we base Rucu's morphology on that ofan ∼1,900–2,000-m-high regular cone truncated at 3,950 masl, and capped by a steeper-sided summit cone. Thiscalculation yields, excluding the El Cinto lavas, a bulkvolume of 200–210 km3. Taking into consideration theextent and thickness of Upper Rucu lavas to the north, south

and east, this volume can be divided into 160 km3 and 40–50 km3 for the Lower and Upper Rucu edifices, respectively.

Terminal Rucu edifice The uppermost part of Rucu consistsof lava flow relics and subglacial breccias which arecorrelated to a small, now deeply eroded, edifice nested inthe ∼250 ka collapse amphitheatre. Based on geochrono-logical constraints, we link this edifice to tephra depositswhich are exposed below 3,200 m asl on the eastern flankof the PVC near Lloa to the south and in the Quito basin(Fig. 6). The tephras generally consist of 20–40-m-thickfallout sequences interlayered with pyroclastic flow andlahar deposits and palaeosols. Geotermica Italiana (1989)divided these deposits into four main units: (1) Holocenedeposits; (2) “recent” deposits (from 11 to ∼50 ka BP); (3)intermediate deposits, aged from ∼50 to 100 ka and; (4)ancient deposits, of older age. The study related thisexplosive record to the activity of GGP. However, astratigraphic discrepancy arises between the intermediateand ancient GGP tephra deposits when compared to our39Ar-40Ar results of early GGP activity, dated here at 60–50 ka (see below). Importantly, the Plinian "Pifo layers"erupted from the Chacana caldera ∼40 km east of the PVC(Hall and Mothes 1997) are key stratigraphic markers inEcuador dated at ∼170 ka, and these occur in the tephrasequence (Fig. 6). Consequently, we infer that theterminal, post-avalanche Rucu centre emitted the interme-

Fig. 4 Digital elevation modelof the PVC showing locations ofsamples (opened circles), stud-ied sections (solid circles) andages obtained by 40Ar–39Ar and14C datings. RP Rucu Pichincha,GGP Guagua Pichincha, TToaza, Cr Cristal dome, C ElCinto, E La Esperanza, UUngui, SR San Ricardo, E LaEsperanza, P Panecillo; RCr RioCristal, RAA Rio AguaAzufrada, RB Rio Blanco, RMRio Mindo, RC Rio El Cinto,RPi Rio Pichan,QRc Quebrada Rumicucho, QRpQuebrada Rumipamba, QRuQuebrada Rumiurco,QCh Quebrada Chimborazo,QSi Quebrada Singuna, CdDCantera del Diablo, MR MinaRoxana quarry, HLP HaciendaLas Palmas, PHS Palmira HotSprings, PB Peña Blanca quarry,Re Refuge

1114 Bull Volcanol (2010) 72:1109–1129

Table 1 40Ar–39Ar ages for rocks from PVC

Samplenumber

Volcanic unit(Fig.5)

Rock type andlocation

SiO2

(wt.%)UTMcoordinatea

east–north

Age ±2σ 39Arb Isochroneage

Ri MSWD

PICH 1 Guagua Lava flow, summit 63.7 07674–99808 52±6 70 44±4 308.7±2.4 0.5

PICH 135 Guagua Lava flow, lower partof Cantera del Diablo

61.7 07682–99834 55±5 77 52±4 293.4±1.5 1.0

PICH 51 Rucu, upperstratocone

Upper Rucu breccias 63.2 07717–99836 260±6 57 262±10 292.7±1.9 3.8

PICH 76 Rucu, upperstratocone

Upper Rucu, lava flow,northern flank

62.0 07673–99937 270±5 64 264±10 296.9±1.1 5.9

PICH 57 Rucu, upperstratocone

Lava flow, Singunavalley

60.6 07761–99871 409±10 76 406±18 292.8±2.1 3.2

PICH 63C Rucu, upperstratocone

Summital breccias 58.8 07701–99824 436±18 61 440±18 293±1.8 5.2

PICH 56 Rucu, loweredifice

Lava flow, Rumiurcuvalley

61.2 07738–99851 608±10 73 621±18 288.4±3.4 3.8

PICH 107B Rucu, loweredifice

Lava flow, Rumipambavalley

60.6 07757–99807 638±15 73 656±22 287±1.7 3.1

PICH 70B Rucu, loweredifice

Lava flow, northernflank

55.3 07723–99900 850±90 83 882±166 292.6±6.0 6.6

PICH 71 Rucu, loweredifice

Lava flow, northernflank

58.5 07719–99902 850±10 23 881±22 288.3±2.2 7.0

PICH 21A El Cinto Ungui dome 63.2 07718–99744 910±70 88 864±48 295.1±3.6 0.3

PICH 111C El Cinto Lower lava flow 61.8 07757–99766 1,100±10 68 1 112±24 288.2±2.2 7.8

MSWD mean square of weighted deviations, i initial ratio (40 Ar/36 Ar)a Location samples are given to 100 m using the UTM metric grid (1956, Provisional South America, zone 17), which is shown on Instituto GeográficoMilitar maps.b Percentage of total 39 Ar included in the plateau age

Table 2 14C data for Guagua Pichincha volcano

Sample number Location Unit Dated material Lab Nber (GrN #) UTM coordinate Age BP ±1σ

PICH 132Ba Rio Cristal Cristal PF c.r. 25810 07578–99791 2,990±20

PICH 29Ba Rio Cristal Cristal PF c.r. 25507 07534–99822 3,549±30

PICH 72Aa Nono Cristal TF a.e. 25512 07706–99923 3,700±30

PICH 91C Rio Agua Azufrada Toaza AF c.r. 25806 07603–99769 9,820±60

PICH 66A Rio Agua Azufrada Toaza AF c.r. 25511 07602–99767 9,920±50

PICH 123 Rio Agua Azufrada Toaza AF c.r. 25807 07612–99766 10,130±120

PICH 91-1 Rio Agua Azufrada Toaza AF c.r. 25518 07603–99769 10,250±35

PICH 37C2B Lloa, refuge road Toaza AF c.r. 25805 07686–99787 10,720±120

PICH 37C2A Lloa, refuge road Toaza AF c.r. 25804 07686–99787 11,020±220

PICH 155A Hacienda Las Palmas HLP AF c.r. 30894 07617–99760 22,500±240

PICH 153A Hacienda Las Palmas HLP AF c.r. 30893 07614–99762 22,560±140

PICH 124C Peña Blanca quarry HLP AF c.r. 25808 07666–99735 22,820±110

PICH 23D Mina Roxana quarry LLoa fan a.e. 24855 07674–99726 30,320+1,080−950PICH 58D Q. Singuna Old GGP AF c.r. 25510 07768–99872 >44,400

PICH 58C Q. Singuna Old GGP AF c.r. 25509 07768–99872 47,500+2,800−2,100

Results were rounded to the nearest significant 10 years

a.e. alkali extract fraction dated, c.r. charred remains (charcoal) fraction dateda Data from Robin et al. (2008a)

Bull Volcanol (2010) 72:1109–1129 1115

diate deposits and possibly a fraction of the ancientdeposits, whose lower sequences might then correspondto the activity of the Upper Rucu cone.

The reconstructed size of the terminal edifice of Rucu isabout 600 m high with a 1.5–2-km-wide base. Its originalvolume is estimated as being 8–10 km3, including the lavaflows that extend to the east and northeast outside theavalanche amphitheatre (Fig. 5).

Guagua Pichincha volcano

GGP rests upon the western side of Rucu and experienced amajor flank failure, which removed a large portion of theedifices and left the C1 collapse scarp (Fig. 5). Remnantunits on the eastern, northeastern and northwestern flanksconsist of lava flows and tephra sequences whose distribu-

tion and inclination point to an original sub-symmetricalcone about 9 km in basal diameter. Prior to collapse, thiscone culminated at about 5,100 m asl. Erosion has locallyincised the volcanic pile right down to the basement,revealing an average thickness for post-Rucu sequences of300–400 m at ∼4 km from the vent, as seen at Cantera delDiablo ("Devil's quarry") on the northwest flank (Fig. 3c).Lava inclinations near the C1 scarp and discordantgeometry between GGP and Rucu products point to amaximum GGP lava thickness of ∼1,300 m. We estimatethe resulting GGP volume in two ways. On one hand weconsider the geometry of a cone with a 79-km2 ellipticalbase, as inferred from GIS-based estimates on the digitalelevation model. On the other hand, we segmented theedifice into 12 vertical prisms, whose basal surface couldbe easily measured on a map. The height of each prism wasestimated from field constraints in the thickness range given

Table 3 Generalized chronostratigraphy showing the main stages of the PVC

Lava flow sequences and domes 1100-900 ka

~ 850 ka

~250 ka

Dominant explosive activity

Rucu sector collapse Avalanche deposit in Quito basin

Edification of a basal edifice: mainly effusive

Summit dome complex and related pyroclastic deposits (Lloa fan)

~60 to 47 ka

Singuna Pyroclastic event

47 to 22 ka

Intense pyroclastic activity,

GGP sector collapse

Lava flows and domes

11 to 9.8 ka

Block-and-ash flow deposits

Toaza sector collapse

Dacitic extrusions and related

(four main cycles with

3.7 ka toPresent

Debris avalanche deposit DAD 2

PIC

HIN

CH

AG

UA

GU

A

Debris avalanche deposit (DAD3) ~4 ka

RU

CU

PIC

HIN

CH

A

VOLCANIC UNITS ERUPTIVE STYLES and DEPOSITS

AGE (ka)MAGMA

COMPOSITION

Blast events at ~22-23 ka

~11 ka

~47 ka

9.8 to 4 ka

- 150 ka

EL CINTO andLA ESPERANZA

New magma batch

New magma batch

New magma batch

61-66 wt.% SiO21.5-1.9 wt.% K2O

34-46 ppm Rb

61-66 wt.% SiO21.6-2.2 wt.% K2O

37-53 ppm Rb

59-65 wt.% SiO21.1-1.9 wt.% K2O

38-42 ppm Rb

55-63 wt.% SiO20.8-1.3 wt.% K2O

11-30 ppm Rb

58-63 wt.% SiO20.7-1.2 wt.% K2O

12-28 ppm Rb

CRISTAL

TOAZA

MAIN GUAGUAPICHINCHA

TERMINAL RUCU

UPPER RUCU

LOWER RUCU

ESTIMATEDVOLUME

(km3)

Mainly effusive activity

~ 600 ka

430-450 ka

160

Upper cone. Mainly effusive. 35 to 50

8 - 10

29 (cone)

ashflow and blast events

pyroclastic flow deposits

plinian episodes)

4 to 5

~1

Total : 31-32

Terminal cone

Basal stratocone

1116 Bull Volcanol (2010) 72:1109–1129

above. Calculations using both models yield consistent bulkGGP volumes of ∼28 km3 prior to collapse.

Early effusive cone (∼60 to 48 ka) Lava flow sequences,such as the one at Cantera del Diablo, form a basal cone ofandesitic composition (60.6–62.9 wt.% SiO2) while thenorth and east sides of the C1 scarp expose younger couléesand domes with an acid andesite composition (61.7–63.7 wt.% SiO2; “summit domes unit” in Fig. 5). The age

of this early edifice is constrained by 40Ar–39Ar dating oftwo lava samples and the 14C age of a voluminous pumiceflow deposit that marks the end of the effusive activity. Abasal lava from the lower part of Cantera del Diablosequence yields an age of 55±5 ka (PICH 135, Table 1),while an upper lava cut by the C1 scarp has been dated at52±6 ka (PICH 1). In addition, 15–20-m-thick pumice flowdeposits exposed northeast of the PVC (Singuna valley;PICH 58, Fig. 4) contain charcoal which was 14C dated at

Fig. 5 Sketch of the geologicalmap of the PVC

Bull Volcanol (2010) 72:1109–1129 1117

>44.4 ka and 47.5 +2.8/−2.1 ka BP (Table 2). Althoughtheir location suggests that they could have erupted fromRucu volcano, pumice mineralogy and geochemistry beardistinctive features of GGP magmatic suites, such ashornblende phenocrysts, dacitic compositions (63.9–64.3 wt.% SiO2) and incompatible element contents (seediscussion). Their age and features support a correlationwith GGP instead of Rucu and suggest the end of the earlyeffusive cone.

Major pyroclastic episodes related to summit domes Thelower slopes of early effusive GGP are mantled byconspicuous pyroclastic deposits, the main facies of whichare the following:

1. Block-and-ash flow deposits. To the south, the Lloa areahosts an ∼120-m-thick pyroclastic fan, which is cut by the

Chimborazo valley (Figs. 7, 8a, b). Between 2,800 and3,050 m, block-and-ash flow deposits form valley-ponded units, which were emplaced discordantly. Theolder and thickest sequence exposed at the Mina Roxanaquarry (sites PICH 9, 23 and 129) overlies a carbon-richpalaeosol dated at 30,320+1,080/−950 years BP, consis-tent with the 14C age of 28,450±1,200 years BPobtained by Geotermica Italiana (1989) in block-and-ash flow deposits at nearby Hacienda Las Palmas. TheLloa fan deposits and block-and-ash flow deposits foundin a similar position in the Rio Mindo valley on thenortheast flank of GGP, point to an active dome complexin the summit area at that time. The upper unit of theLloa fan consists of >10-m-thick pyroclastic flow andfall deposits interlayered with altered ash beds. In thePeña Blanca quarry (section PICH 124, Fig. 7) a 2–3-m-thick pumice-rich horizon yields a 14C age of 22,280±110 years BP on charcoal (Table 2). We estimate thevolume of the Lloa and Rio Mindo deposits as being2.5–3 km3.

2. Blast deposits. A notable ∼5-m-thick, coarse-grainedbreccia is exposed near the top of the Lloa pyroclastic fan(section PICH 43, Figs. 7, 8c). It contains juvenile blocksand bombs and non-juvenile clasts of acid andesite(62.0–63.0 wt.% SiO2). At sites PICH 22 and 44 (Figs. 7and 8d), a similar ∼1 m-thick deposit was described as a“lateral blast deposit” by Geotermica Italiana (1989).Our field observations, namely topographic mantling,fines-poor matrix and abundant angular clasts showingimbrication, agree with this interpretation. This blastlayer is covered by a 20–40-m thick block-and-ash flowdeposit that we correlate with the upper part of the ∼20–30-ka-old Mina Roxana deposits.

3. Surge deposits. Southwest of the avalanche amphi-theatre, at sites PICH 153, 154 and 155 (Hacienda LasPalmas, Figs. 7, 8e), a 2.5-m-thick charcoal-rich surgedeposit is dated at 22,500±240 and 22,560±140 yearsBP (samples PICH 153A and 155A, Table 2), an agesimilar to that obtained for pumice deposits at the PeñaBlanca quarry (22,280±110 years BP). Therefore, wecombine these surges and the upper pyroclastic flowand fall deposits of the Lloa fan into an intenseexplosive period around 22–23 ka BP.

4. Tephra fallout deposits. A distinctive unit of pumicefallout layers cover the Lloa fan and occurs on thesouthern slopes of the PVC. At the Mina Roxana quarrythe ∼22.5-ka pumice flow deposit (section PICH 156,Fig. 7) is covered by a 10–20-m-thick pumice falloutsequence, which is also well exposed around Lloa, nearHacienda Las Palmas, and along the track to El Cintoridge. The systematic thickness variation of tephralayers, as well as the geochemical characteristics of thepumice (see below) unambiguously point to an eruption

Pifo tephra

~ 170 ka

South Quito Section 22 (GI, 1989)

PICH 160 Avenida Oriental North-east Quito

El Cinto Section 40 (GI, 1989)

(cm)

50

30

300

400

100 15

300

150

200

50

200

200

150

15

200

30

155

15

800

100

soil (black or brown)

fine ash deposit pumice-and-lithicsfallout deposit

medium ash deposit

coarse ash deposit

Fig. 6 Stratigraphic sections of Rucu Pichincha tephras exposed inthe Quito basin, showing the relationship with the “Pifo” pumice fromChacana caldera (∼170 ka BP) and Cangahua deposits. Sections 22and 40 are from Geotermica Italiana (1989)

1118 Bull Volcanol (2010) 72:1109–1129

origin from the nearby Atacazo-Ninahuilca volcano(Hidalgo et al. 2008). We thus infer that GuaguaPichincha volcano was relatively quiet or had non-explosive to weakly explosive activity during this periodof time.

5. Debris avalanche deposits (DAD). DAD related toflank failure C1 are confined to the confluence of theRio Blanco and Rio El Cinto, where proximal outcropsare >50 m high (DAD1 unit in Fig. 5). Upstreams,between the Agua Azufrada and Blanco rivers, occur ashydrothermally altered breccias with reddish, yellowishand greenish-coloured clayey facies (Fig. 8f).

Toaza edifice A distinctive series of pyroclastic depositsoverlies DAD1. At site PICH 66, a 22-cm-thick amphibole-rich pumice fall layer is covered by a 60-cm-thick, fines-poor and lithic-rich blast deposit (Fig. 8g). These productsare, in turn, overlain by pinkish pumice flow layersinterbedded with decimetre-thick pumice fallout beds, insequences up to 6 m thick (sections PICH 66 and 91,Fig. 9). Beyond the avalanche amphitheatre, an 80-m-thickpumice flow deposit fills the Chimborazo valley incisedinto the Lloa fan overlying the Mina Roxana block-and-ashflow sequence (site PICH 41). Thick surge deposits from

this series are exposed along the Lloa-Refuge road (sitePICH 37, Figs. 8h, 9). Six 14C ages obtained on charredremains from these pumice flow and surge deposits revealthat intense explosive activity spanned about 1,000 yearsfrom 10.9 to 9.8 ka BP at the Pleistocene–Holocenetransition (Table 2).

A new 5×3-km-wide edifice, here named Toaza, grew inthe avalanche amphitheatre of GGP, consisting mostly ofviscous dacitic coulées and domes. Concomitant block-and-ash flows spread over the uneven topography of the 11-kaDAD1 and were concentrated in the Rio Agua Azufradavalley, where the resulting deposits reach 120 m inthickness. A younger DAD (DAD2) crops out in the RioCristal and Rio Cinto, and probably correlates with the 2-km-wide amphitheatre carved into the Toaza dome complex(Fig. 5). DAD2 contains hydrothermally altered lava blocksand locally grades into crudely stratified lahar deposits.

Present dome complex The active 1-km-wide dome complex,or "Cristal dome" (Figs. 2a, 5), sits on the floor of the west-trending amphitheatre and is mostly dacitic in composition(63.7–65.5 wt.% SiO2). It grew in the late Holocene timesand associated pyroclastic products occur chiefly to the west,notably in the Rio Cristal valley where DAD2 deposits are

Site 41 Site 43Site 44 Sites 23 and 129Roxana quarry

Site 124Peña Blanca quarry

PICH 23D= 30,320 BP

100 m

2900

2850

2950

3000

3050m asl

WNW ESE

PICH 124

PICH 41(Calderic event ≈ 10-11 ka BP)

PICH 9A-B

PICH 43

PICH 23& 129

PICH 44

reworkedPICH 23D

22,820 BPsrete

m 01

PICH156

PICH 153A= 22560 ± 140 BP

sretem 2

Site 156Roxana quarry

60

4010

110

25

80

22

50

42

80

20

185

20

100

> 20 m

[cm]

Site 154

Site 155Hda. Las PalmasSite 153

PICH 155A

= 22500 ± 240 BP

Lloa

12923

12441

9A-B

SAMPLING SITES

4344

156153 - 154 -

155

pumice flow deposit

altered ash (cangahua)w/ interlayered tephra

soil (black or brown)

block-and-ashflow deposit

fine ash deposit

lithic-rich deposit(blast)

pumice-and-lithicsfallout deposit

ash flow depositcoarse ash deposit

Lloatephra

deposits

Fig. 7 Stratigraphic sections of late Pleistocene pyroclastic deposits from the southern flank of Guagua Pichincha (Chimborazo valley andHacienda Las Palmas area)

Bull Volcanol (2010) 72:1109–1129 1119

blanketed by block-and-ash pumice flow and debris flowdeposits, including those of the 1999–2001 period. Radio-carbon ages obtained from palaeosols and charcoal indicatethat the correlated tephra fall and pyroclastic flow depositspreserved on the upper slopes of GGP (sections PICH 55, 97,35; Fig. 9) were emplaced over four explosive periods. Afterthe last sector collapse dated at ∼4 ka BP (1875–2140 yearcal BC), explosive cycles occurred at around 1310–1150 calBC, 1–140 cal AD, AD 820 to 1030, and from AD 1450 to1660 (Robin et al. 2008).

Discussion

Reconstruction of PVC development

Our geological investigation at Pichincha shows that thePVC developed as a result of successive magmatic bursts,which fed distinctive eruptive stages as summarized belowand in Fig. 10. A 1100 to 850-ka-old basal unit of lavas

forming the El Cinto ridge is identified south of the mainPVC. On geochemical grounds these lavas are correlatedwith the 600-m-thick La Esperanza lavas on the northernside of the volcanic complex (Figs. 11, 12), and both unitsare possibly contemporaneous with volcanic relics in Quitobasin, such as the Panecillo dome (Fig. 5). Low K2O andincompatible trace element contents in El Cinto–LaEsperanza lavas (Table 4; Fig. 12) suggest a magmaticsuite that resulted in a now essentially eroded and partiallyburied edifice below the Rucu.

The growth of the andesitic Rucu volcano lasted from ∼850to ∼150 ka. The effusive Lower Rucu volcano grew from 850to 600 ka (Fig. 10a), and a marked unconformity between theLower and Upper Rucu edifices suggests a prolonged reposeperiod. The activity of the Upper Rucu commenced at 430–450 ka and resulted in a cone that culminated at ∼5,000 masl, extending to the east above older Rucu lavas as well asto the north and the south (Fig. 10b). The eastern flankexperienced a major sector collapse at about 250 ka. From250 ka to ∼150 ka, post-avalanche activity constructed asmall volcanic edifice in the avalanche depression which

Fig. 8 a View of the Chimborazo valley. b Block-and-ash flowdeposits of the Lloa fan at the Mina Roxana locality (sections, Fig. 7).c Coarse blast deposit on top of the Mina Roxana sequence, site PICH43, in Fig. 7. The dacitic composition and the shape of juvenile blockssupport a dome collapse origin. d Blast deposit (site PICH 44, Fig. 7).

e Hacienda Las Palmas pyroclastic sequence (site PICH 154, Fig. 7). fDAD2 in the Rio Azufrado valley. g Blast deposit, 10–11 ka old (sitePICH 66, Fig. 9). h Pyroclastic flow deposit associated with the Toazaedifice at site PICH 37. i Late Holocene scoria fall deposits related toCristal dome activity

1120 Bull Volcanol (2010) 72:1109–1129

became more explosive, as testified by thick tephra deposits,which are exposed below 3,200 m asl on the eastern andsouthern flanks and in the Quito basin. Higher on thevolcano, these deposits were removed by erosion.

In our interpretation the GGP is younger than inferred inprevious studies and comprises four evolutionary stages.

1. Andesitic to dacitic effusive activity started at around60 ka and constructed a basal stratocone about 10-kmwide (Fig. 10c).

2. About 12,000 years later, the first major pyroclastic eventoccurred with emplacement of the Singuna pumice flows.The distribution of the deposits suggests that the GGPsummit was notably higher than the already eroded Rucupeak. Prominent dome-building periods followed in thesummit area and resulted in major dome collapse eventsdated at ∼30–27 ka and ∼23–22 ka BP, sometimesaccompanied by violent directed blasts that depositedlithic-rich layers. The major collapse, which destroyed thesummit area of GGP around 11 ka BP also involved the

western flank of Rucu, and the DAD (DAD1) is found onthe southwest. The volume of DAD1 remains poorlyconstrained, but its distribution and the size of theresulting amphitheatre (3×5 km) point to a huge collapse,which defines the end of the main GGP edifice (Fig. 10d).

3. The 11-ka-BP-sector collapse marked the onset of intensepyroclastic flow activity that waned ∼9.8 ka BP. Highlymobile pumice flows spilled over the amphitheatre walland spread across the southern sector. The activitycontinued in early Holocene times with lava domeextrusions that progressively constructed the Toazaedifice (Fig. 10e), ending with a collapse of the westernflank that resulted in a 2-km-wide, west-oriented amphi-theatre and concomitant DAD2 deposits. The flankfailure of Toaza is the youngest in the history of thePVC.

4. The composition of the dacitic block-and-ash flowdeposits overlying DAD2 and dated at ∼3.7 ka BP bearsstrong similarities to the Cristal dacites. Thus, we surmisethat Cristal dome activity most likely commenced around

1660 AD

PICH 97

Quilotoaash

PICH 35

1566-1582 AD

retem 1

800-

1000AD

s

s

PICH 55PICH 55 (suite)

reworked

ox.

ox.

Toaza

reworkedash-and-pumice

PICH 66A/B= 9920 ± 50 yr BP

sretem 2

PICH 66

PICH 91

PICH 91A/B= 10250 ± 35 yr BP

andon nearbyPICH 123

= 10,130 ± 120 yr BP

PICH 91C= 9820 ± 60 yr BP

soil (black or brown)

fine ash falloutdeposit

lithic-rich deposit(blast)

pumice-and-lithicsfallout deposit

ash flow deposit

coarse ash falloutor surge deposit (s)

s

s

s

s

~700 AD

ReworkedPululahua

tephra

Toaza edifice

Cristal

Cristal dome complex

Fig. 9 Stratigraphic sections of the Holocene deposits

Bull Volcanol (2010) 72:1109–1129 1121

4–3.7 ka BP (Fig. 10f), and the Toaza sector collapseprobably took place shortly before this. Finally, theCristal dome complex experienced four eruptive cyclesin late Holocene times, as discussed further below.

Possible additional collapse event at Guagua Pichincha Adistinct hydrothermally altered avalanche deposit (herenamed DAD0, Fig. 5) occurs south of the volcanic complex

and forms the uneven surface upon which the Mina Roxanapyroclastic sequence is deposited. DAD0 reaches up to 60–80 m in thickness and consists of intensely fractured lavablocks enclosed within a mauve-coloured, fine-grainedclayey and moderately indurated matrix, which alsoincludes crushed fragments from the basement. The sourceand the age of DAD0 are poorly constrained but, owing toits location, the deposit is neither correlated to a collapsefrom the Upper Rucu nor to C1 collapse from GGP. One

9980

775755 765

10 km

Lower Rucu(850 to 600 ka)

crater(~4800 m asl)

10 km

Upperand terminal Rucu

(450 to 250 Ka)

9980

775755 765

10 km

Basal Guagua& summit domes(~60 to 11 ka BP)

9980

775755 765

10 km

GGP sectorcollapse & debris

(11 Ka BP)

9980

775755 765

10 km

Toaza(~11 to 4 ka BP)

9990

9980

775755 765

10 km

Cristal(3.7 ka to Present)9990

9980

775755 765

(a) (b)

(c) (d)

(e) (f)

Fig. 10 Sketch diagrams show-ing the main development stagesof the PVC. a Lower Rucu, bUpper and terminal Rucu, cBasal Guagua Pichincha andsummit domes, d Basal GuaguaPichincha sector collapse anddebris avalanche, e Toazaedifice, f Cristal domes

1122 Bull Volcanol (2010) 72:1109–1129

Tab

le4

Geochem

ical

analyses

representativ

eof

mainvo

lcanic

units

ofPVC

Edifice

ElCinto

Rucu

MainGGP

Toaza

Cristal

Sam

ple

no.

PICH

111A

PICH

21B

PICH

70B

PICH

56PICH

63C

PICH

51PICH

135

PICH

1PICH

58A1

PICH

23A

PICH

44A

PICH

37-1

PICH

91A1

PICH

66D

PICH

48PICH

132A

PICH

35A1

PICH

78C1

PICH

78A

PICH

93

Volcanic

unit

Lava

flow

Ungui

dome

Lava

flow

Lava

flow

Sum

mit

breccia

Lava

flow

Lava

flow

Sum

mit

dome

Singuna

PF

Lloafan

PF

Blast

deposit

Surge

deposit

Plin

ian

deposit

Blast

deposit

Lava

dome

Cristal

BandA

PF

Plin

ian

deposit

Plin

ian

deposit

Plin

ian

deposit

Lava

dome

Age

(ka)

1,100

910

850

608

436

260

5552

47,5

∼30

∼30–

2310–11

10–11

10–11

<10

2,9

∼AD

1–140

∼AD

970

AD

1660

AD

1999

SiO

261.75

63.00

54.70

61.00

58.50

62.60

61.60

63.40

63.00

62.40

62.00

62.75

62.60

62.00

60.80

61.80

63.30

64.50

63.20

64.90

TiO

20.57

0.52

0.56

0.60

0.52

0.49

0.50

0.45

0.45

0.46

0.48

0.45

0.46

0.48

0.51

0.46

0.41

0.37

0.40

0.37

Al 2O3

16.90

16.55

15.40

16.70

16.20

16.20

15.96

16.50

16.00

16.7

16.37

16.35

16.15

16.5

16.85

16.30

16.30

16.25

16.35

16.30

Fe 2O3a

6.39

5.64

8.36

6.39

7.30

5.64

6.37

5.55

5.34

5.84

6.08

5.32

5.38

5.65

6.57

5.92

5.08

4.58

5.28

4.90

MnO

0.09

0.08

0.14

0.10

0.11

0.08

0.10

0.08

0.08

0.09

0.10

0.08

0.08

0.09

0.10

0.10

0.08

0.08

0.09

0.08

MgO

2.85

2.59

7.66

3.36

5.23

3.22

4.00

2.50

2.41

2.81

3.08

2.37

2.24

2.60

3.22

3.13

2.22

2.00

2.38

2.34

CaO

5.60

5.20

7.90

6.24

7.00

5.23

5.87

5.15

5.00

5.49

5.64

4.91

4.92

5.45

6.00

5.80

4.93

4.54

5.00

4.68

Na 2O

4.28

4.51

2.86

3.95

3.60

4.06

3.80

4.05

3.86

4.08

3.87

3.87

3.70

3.92

3.82

3.75

3.94

4.03

3.98

4.25

K2O

0.97

1.10

1.12

1.16

0.89

1.37

1.49

1.77

1.77

1.77

1.67

2.02

2.05

1.75

1.62

1.65

1.75

1.82

1.73

1.78

P2O5

0.12

0.12

0.13

0.15

0.12

0.13

0.14

0.13

0.13

0.14

0.14

0.15

0.16

0.15

0.14

0.15

0.14

0.14

0.13

0.12

LOI

0.43

0.35

0.52

0.17

0.08

0.31

0.10

0.25

1.89

0.05

0.09

1.75

1.85

0.76

0.18

0.70

1.55

1.63

1.51

0.26

Total

99.95

99.66

99.35

99.82

99.55

99.33

99.73

99.83

99.93

99.83

99.52

100.02

99.59

99.35

99.81

99.76

99.7

99.94

100.05

99.98

Sc

13.0

10.7

26.5

17.4

22.5

12.7

15.8

11.7

9.9

13.0

14.8

9.7

8.8

10.5

13.0

13.6

9.5

8.6

10.5

10.9

V92.0

120.0

194.0

156.0

179.0

130.0

145.0

128.0

120.0

133.0

142.0

120.0

116.0

125.0

148.0

129.0

105.0

101.0

118.0

110.0

Cr

53.0

82.0

300.0

98.0

220.0

128.0

161.0

57.0

56.0

64.0

70.0

45.0

41.0

54.0

65.0

67.0

35.0

34.0

42.0

40.0

Co

19.0

17.0

35.0

20.0

25.0

19.0

22.0

18.0

15.0

18.0

20.0

15.0

15.0

17.0

21.0

18.0

15.0

13.0

14.5

14.0

Ni

28.0

36.0

150.0

26.0

61.0

40.0

60.0

23.0

22.0

27.0

31.0

23.0

23.0

31.0

36.0

34.0

24.0

20.0

25.0

25.0

Rb

17.0

22.5

22.0

23.5

17.0

29.0

37.0

41.0

41.5

41.5

40.0

50.0

53.0

43.0

39.5

38.0

42.0

45.0

42.0

42.0

Sr

395.0

418.0

390.0

444.0

402.0

473.0

444.0

505.0

488.0

512.0

492.0

508.0

515.0

522.0

485.0

500.0

517.0

517.0

515.0

525.0

Y10.5

8.4

12.9

12.6

10.6

8.4

10.4

9.4

7.9

10.3

10.7

8.4

8.4

9.2

10.8

10.5

8.6

7.8

8.6

8.5

Zr

77.0

63.0

68.0

100.0

69.0

88.0

74.0

87.0

68.0

88.0

89.0

85.0

70.0

85.0

73.0

69.0

66.0

72.0

77.0

77.0

Nb

2.8

2.7

2.5

3.1

2.1

2.8

3.0

2.8

3.1

3.1

3.3

4.0

3.8

3.1

2.9

2.8

2.8

2.9

2.9

2.5

Ba

466.0

490.0

430.0

590.0

392.0

655.0

640.0

765.0

725.0

750.0

770.0

800.0

834.0

740.0

708.0

735.0

750.0

880.0

820.0

810.0

La

6.4

6.7

6.8

10.0

5.4

8.7

10.9

14.2

12.0

14.5

13.8

16.3

19.0

15.8

11.7

13.0

14.5

12.8

12.2

11.0

Ce

13.5

13.0

14.0

18.5

12.0

16.5

21.0

26.0

22.5

27.0

25.0

29.5

32.5

29.0

22.0

25.5

27.0

22.5

22.5

20.5

Nd

98.5

8.0

11.5

7.2

9.5

11.2

12.5

11.3

12.7

12.8

14.2

15.5

14.0

11.5

12.0

12.4

11.0

10.7

10.0

Sm

2.2

2.1

2.1

2.6

2.0

2.2

2.5

2.6

2.2

2.7

2.7

2.7

2.7

2.9

2.4

2.6

2.4

2.0

2.1

2.1

Eu

0.7

0.7

0.7

0.8

0.6

0.7

0.7

0.7

0.7

0.8

0.8

0.8

0.8

0.8

0.7

0.8

0.7

0.6

0.6

0.6

Gd

2.5

2.1

2.3

2.7

2.1

2.2

2.5

2.5

2.2

2.3

2.5

2.5

2.6

2.7

2.6

2.2

2.6

2.2

2.1

2.0

Dy

1.9

1.5

2.2

2.1

1.8

1.4

2.0

1.8

1.4

1.8

1.8

1.4

1.5

1.7

1.9

1.8

1.6

1.3

1.5

1.5

Er

1.0

0.8

1.3

1.2

1.1

0.8

1.0

0.9

0.6

1.0

1.1

0.8

0.8

0.8

1.0

1.0

0.8

0.8

0.8

0.9

Yb

0.9

0.7

1.3

1.1

1.0

0.8

0.9

0.9

0.6

0.9

1.0

0.7

0.7

0.8

0.9

1.0

0.8

0.7

0.8

0.8

Th

0.9

0.8

1.4

2.1

1.4

2.0

3.1

3.8

3.4

3.8

3.7

5.2

5.6

4.4

3.3

3.6

3.9

3.6

3.5

3.1

aTo

taliron

asFe 2O3

Bull Volcanol (2010) 72:1109–1129 1123

option is to assign DAD0 to an old flank failure event of thesouthwestern flank of the Lower Rucu, now concealedbelow GGP lavas. Alternatively, because the irregularDAD0 surface is directly covered by the relatively youngLloa pyroclastic fan (∼30 to 20 ka), a correlation with anearly collapse event at GGP might be considered, theresulting depression having been completely concealed bythe domes. This interpretation raises a possible correlationbetween DAD0 and the voluminous Singuna ash flowdeposits dated at ∼47 ka.

Relationship between structural development and magmachemistry

Overall geochemical behaviour in the PVC The silicacontent in PVC rocks globally increases with time, from theandesites of El Cinto and Rucu Pichincha (55–63 wt.% SiO2),

to more silica-rich compositions for the main GuaguaPichincha (59–65 wt.% SiO2), Toaza edifice and Cristaldomes (61–66 wt.% SiO2). In a SiO2 vs. K2O diagram(Fig. 11), Rucu and Guagua Pichincha samples define a broadmedium-K trend. In addition, at similar silica content theGuagua Pichincha samples show higher concentrations ofincompatible elements such as Th, the large-ion lithophileelements (LILE, e.g. Rb), and light rare earth elements(LREE, e.g. La) than Rucu lavas. When looked at in moredetail, there are no significant differences between the threesuccessive units of Rucu (El Cinto, Lower Rucu and UpperRucu Pichincha). In contrast, important differences areobserved between the samples of the main Guagua Pichincha,Toaza and Cristal units, the Toaza rocks showing highervalues for most incompatible elements than samples fromthe main Guagua Pichincha and Cristal. In Fig. 12, thebehaviour of key trace element abundances and ratiosplotted against age, reveal a progressive increase in LILE

0

1

2

3

K2O

wt.%

A D

LK

MK(a)

0

2

4

6

8

MgO

wt.% El Cinto

Lower Rucu

Main Guagua

ToazaCristal

Upper Rucu

(b)

0

10

20

30

40

50

Rb

ppm

(c)

LLoa TF

Atacazo

Th

ppm

(d)

0

1

2

3

4

5

La p

pm

55 60 65 70

SiO2 wt.%

0

4

8

12

16

20(e)

55 60 65 70

SiO2 wt.%

0

40

80

120

Ni p

pm

(f)

Fig. 11 Selected major andtrace elements for Pichinchasamples, plotted against silica. aSiO2 vs. K2O classification dia-gram (modified from Peccerilloand Taylor 1976). A andesites, Ddacites, LK low potassium, MKmedium potassium. b–f Selectedvariation diagrams for MgO, Rb,Th, La and Ni respectively. Thedotted line corresponds to thefield of the magmatic series ofthe neighbouring Atacazovolcano. Note that Lloa tephrafallout samples fall in the fieldfor Atacazo (Hidalgo et al.2008)

1124 Bull Volcanol (2010) 72:1109–1129

and LREE (e.g. Rb, La) with time or even a sharpaugmentation fromRucu to Guagua (e.g. Th).We also observean increase in several trace element ratios (e.g. Th/Nb, La/Yb)as well as a sharp decrease of the so-called “fluid-mobile” to“fluid-immobile” element ratios (e.g. Ba/Th).

In the La/Yb vs. MgO diagram (Fig. 13), a fractionalcrystallization trend is depicted. Nevertheless, fractionationof the observed mineral phases is unable to explain thegeochemical changes between the Rucu and GuaguaPichincha series. In contrast, the geochemical changespoint to a magma mixing process, whose end-membersare a primitive Rucu Pichincha andesite and a Toaza dacite.Geochemical (Schiano et al. 2010) and mineralogical data(Samaniego et al. 2010) both point to the predominance ofmagma mixing between a mafic, trace-element depleted,mantle-derived end-member, and a siliceous, trace-elementenriched end-member. The origin of these end-members,namely the genesis of the silica-rich dacites, was discussedby Samaniego et al. (2010). To summarize, two non-exclusive models have been proposed for explaining theoverall magmatic evolution recorded by Pichincha lavas;either the “maturation” of the magmatic system in a deep“hot zone” in the crust following the model of Annen et al.(2006) or the interaction of slab-derived silica-rich magmaswith the mantle-wedge.

Magma changes related to sector collapses Whatever theorigin of the geochemical changes recorded throughout thePichincha evolution, there is a strong link betweenthe changes in rock chemistry and the structural evolutionof the complex. In particular, the enrichment in mostincompatible trace elements in the andesites and dacitessubsequent to the 11-ka sector collapse clearly distinguishesthe Toaza rocks from those of previous episodes. On thebasis of petrogenetic models tested by Samaniego et al.(2010), we conclude that Toaza magmas represent a newbatch of magma whose geochemical characteristics departsignificantly from those of the main Guagua Pichinchalavas. Similarly, a geochemical change coincides with the

0.1 1 10 100 1000

Time (ka)

Th

ppm

55

60

65

70

SiO

2 w

t.%

Andesiticenclaves

C1C2

Rucu + EC

MainGGP

Toaza

Cristal

0

1

2

3

4

5

0

10

20

30

La/Y

b(a)

(b)

(c)

Fig. 12 Temporal geochemical variations of PVC magmas. Samesymbols as in Fig. 11 a SiO2, b Th, c La/Yb plotted against time BP

0 2 4 6 80

10

20

30

La/Y

b

MgO0 2 4 6 8

0

10

20

30

La/Y

b

MgO

Fractionalcrystallization

Magmamixing

(a) (b)Fig. 13 MgO vs. La/Yb dia-gram for a Pichincha samplesand b fields for Rucu, as well asthe three main volcanic units ofGuagua. Schematic bulk mixingand fractional crystallization(FC) models are also depicted.Note that FC models are unableto explain the high-La/Yb valuesof Toaza samples. Same symbolsas in Fig. 11

Bull Volcanol (2010) 72:1109–1129 1125

emplacement of the Cristal dome after the Toaza sectorcollapse. Thus, new magma batches appear to be associatedwith major sector collapses and in the initiation of newmagmatic cycles at Pichincha volcano. Similar behaviour isseen at nearby volcanoes in the Ecuadorian volcanic front,such as the Fuya-Fuya (Robin et al. 2009) and NevadoCayambe volcano (Samaniego et al. 2005).

Eruption rates

Considering the entire period of activity of the PVC, theeruptive rate is 0.29 km3/ka (Table 5). However, this estimatesuffers greatly from the averaging effect, which does not takeinto account the long repose times. Output rate variationshave been inferred for continental arc volcanoes, such asPuyehue-Cordon del Caule (Singer et al. 2008), Mt. Adams(Hildreth and Lanphere 1994), Katmai volcanic cluster(Hildreth et al. 2003b), and Mt. Mazama (Bacon andLanphere 2006), suggesting that many composite volcanoesgrow in spurts with peak output rates as high as 1–2 km3/ka

(Hildreth and Lanphere 1994; Davidson and de Silva 2000).As regards Rucu development, the eruptive rate we canestimate for Lower Rucu is 0.64 km3/ka (Table 5). At GGP,for a duration of about 50 ka (from ∼60 to 11 ka BP), theeruptive rate of the main Guagua Pichincha is similar(0.63 km3/ka), with a peak value of 2.2 km3/ka during theconstruction of the basal cone, as well as that of Toaza, from11 to 4 ka BP (0.64 km3/ka).

These similar values, calculated for edifices whosevolumes and lifespans are very different, allow us toconclude that there was an almost constant output rate atthe PVC during the main periods of eruptive activity.This eruptive rate (0.60–0.65 km3/ka) is similar to thatobtained at El Misti, another andesitic-to-dacitic volcanoin Peru (Thouret et al. 2001). Our estimates for the PVCare significantly higher than those for Mt. Baker (0.20–0.33 km3/ka; Hildreth et al. 2003a) and are notably higherthan the 0.04–0.20 km3/ka calculated at other continentalarc volcanoes such as Ollagüe, Llullaillaco, and Aucanquilchain the Central Andes or Ceboruco and Tancítaro in Mexico(see values and references in Table 5). To summarize, on one

Table 5 Magma eruptive rates at the PVC compared to the eruptive rates from other dacitic volcanoes in active continental margins

Volcano Compositionalrange

Arc Volume(km3)

Lifespan(ka)

Eruptive rate(km3/ka)a

Peak eruptiverates (km3/ka)

Reference

Mt. Baker Andesiticto dacitic

Cascades 20–33 100 0.20–0.33 Hildreth et al. (2003a)

Mt. Adams Basalticto dacitic

Cascades 230–400 940 0.24–0.42 1.5–2.5 Hildreth and Lanphere (1994)

Mt. Mazama/Crater Lake Andesitic todacitic

Cascades 176 420 0.42 0.8–2.5 Bacon and Lanphere (2006)

Katmai Basaltto rhyolitic

Alaska 70 100 0.5–1.2 Hildreth et al. (2003b)

Ollague Dacitic CentralAndes

85 1000 0.09 Feeley and Davidson (1994)

Llullaillaco Dacitic CentralAndes

50–60 1000–1500 0.04–0.06 Richards and Villeneuve (2001)

Aucanquilcha Dacitic CentralAndes

38 1000 0.04–0.16 Klemetti and Grunder (2008)

El Misti Andesiticto dacitic

CentralAndes

70 112 0.63 2.1 Thouret et al. (2001)

Ceboruco Daciticto rhyolitic

Mexico 81 800 0.10 0.60 Frey et al. (2004)

Tancitaro Andesitic Mexico 97 556 0.17 Ownby et al. (2007)

Nevado Cayambe Andesiticto dacitic

Ecuador 154 400 0.39 Samaniego et al. (2005)

Pichincha Volcanic Complex (PVC) Ecuador

Whole PVC Andesiticto dacitic

250 850 0.30 This study

Lower Rucu Pichincha Andesiticto dacitic

160 250 0.64 This study

Main Guagua Pichincha Andesiticto dacitic

31–32 50 0.63 2.2 (basal cone) This study

Toaza Dacitic 4–5 7 0.57–0.71 This study

(mean, 0.64)

a Deduced from estimated total volume and entire duration of the volcanic activity

1126 Bull Volcanol (2010) 72:1109–1129

hand, the Lower Rucu, Guagua and Toaza volcanoes showoutput rates of 1.5 to 3 times greater than the rates generallyinferred for other dacitic volcanoes in continental arcs. Onthe other hand, the average output rate of the PVC is of thesame order as the 0.4 km3/ka value obtained for the nearby,mostly dacitic Nevado Cayambe volcano over the last 400 ka(Samaniego et al. 2005). Though additional studies areneeded, we infer that the high density of volcanoes in theEcuadorian arc and these elevated eruption rates are relatedto the rapid subduction of the Nazca Plate beneath this partof the South American Plate (8–9 cm/year, Pennington 1981;Gutscher et al. 1999).

Late Holocene eruption frequency and hazards

Eruption frequency Although the morphology of the PVChas been continuously altered by glacial and fluvialerosion, the youngest horseshoe-shaped depressions, C1and C2, which open respectively to the southwest andwest, are still well preserved, as are Holocene deposits.Late Holocene activity at GGP includes four explosivecycles at the Cristal dome, the products of which werepreferentially directed in these directions (Robin et al.2008). Our 14C ages and the historical archives suggestthat the duration of the three youngest cycles spanned oneto two centuries with decreasing intervening reposeperiods. An apparent ∼1,200-year-repose period separatesthe first and second cycles; whereas, the followingquiescent intervals lasted ∼800 and ∼500 years (Fig. 14).The eruptions in 1999–2001 took place after 340 years ofquiescence and may therefore represent the beginning of anew eruptive cycle, which could last at least severaldecades.

The eruptive cycles at the Cristal dome all comprisedPlinian events, phases of dome growth and collapse, andsubordinate pumice and ash falls. The largest cycleoccurred in the tenth century AD and ranked at VEI 5;the recent AD 1450 cycle also culminated with a powerfulVEI 4 Plinian eruption in AD 1660 (Robin et al. 2008).This eruptive chronology sheds new light on the potential

volcanic hazards posed by Pichincha volcano on Quito.The unrest observed at the end of the twentieth centuryand the explosive–extrusive eruptions in 1999–2001could lead to a stronger Plinian phase as witnessed inAD 1660.

Plinian tephra falls: a major hazard A major collapse ofthe active Cristal dome is unlikely because its volume isstill small (1 km3 or less). Pyroclastic flows erupted fromthe dome complex are preferentially channelled westwardby the 300–700-m-deep avalanche amphitheatre carved intothe Toaza edifice and still further by the Rio Cristal valley(Fig. 5). These areas are uninhabited and concentrate mostof the block-and-ash flows and lahars which occurredduring the 1999–2001 eruptions. On the eastern, populatedside of PVC, we evaluate that the possibility of futurepyroclastic flows impacting on Quito is low because of theexistence of Rucu to the east of GGP, which acts as atopographic barrier. On the other hand, the potential forheavy tephra fall on Quito remains a major concern.Historical chronicles report severe impact on the city inAD 1660 while the light to moderate ash falls thatshowered Quito in 1999–2001 led to economic disturbancein the capital. This light tephra deposition caused completedisruption of flights at Quito international airport, whichwas closed for several days. The impact of a stronger, VEI3–5 Plinian-type eruption would inevitably be much morepronounced for at least two reasons. Firstly, Quito'spopulation has grown drastically since AD 1660, hencethe town is more vulnerable (D’Ercole and Metzger 2004).Secondly, any explosive eruption of significant size (VEI >2)would pose major threats to aviation safety in the Interandeancorridor, an aviation route that is increasingly used byinternational flights.

Conclusions

Our study provides new constraints on the developmentof the large PVC near Quito. The old, chiefly effusive

2000 AD

4th "Historic" cycle ~1450-1660 AD (1)

Third cycle~840-980 AD (1)

1980 AD

1999-2001 eruptive episode

Second cycle~ 0-140 AD (1)

1000 AD 0 1000 BC 2000 BC

Toaza Collapse event

Edification of Cristal dome

TOAZACRISTAL

3540+/-30 BP

First cycle(~1150-1300 BC)

(1) : From Robin et al. (2008)Period of explosive activity

Fig. 14 Temporal diagrammaticrepresentation of eruptive cyclesat Cristal dome complex

Bull Volcanol (2010) 72:1109–1129 1127

∼200-km3-Rucu volcano comprises three growth stages: (1)a Lower Rucu edifice (160 km3) developed from ∼850 ka to600 ka, (2) the activity of Upper Rucu commenced around450–430 ka BP and this new cone collapsed ∼250 ka, (3) ayounger cone constructed in the avalanche amphitheatre wasactive between 150 and 170 ka.

The location of the feeding system shifted westward whenGuagua Pichincha commenced activity at around 60 ka. Theinitial dominantly effusive activity changed at ∼47 ka BP witha major ash-and-pumice-flow event followed by the emplace-ment of wide summit extrusions. Explosive periods related todomes occurred at around 28–30 and 22–23 ka BP. At about11 ka BP, the southwestern flank of GGP collapsed, producinga large-volume debris avalanche. The activity of the Toazaedifice, which developed rapidly in the existing amphitheatre,comprised intense pyroclastic activity over 1,000 years,switching to dominantly extrusive eruptions that resulted ina new collapse at ∼4 ka BP. The west-oriented amphitheatrecarved into the Toaza rocks hosts the currently active Cristaldome complex.

Taking into account the whole period of PVC develop-ment, the average output rate is close to 0.30 km3/ka whilethe estimated output rates for the Lower Rucu Pichincha,main Guagua Pichincha and Toaza edifices are 0.63–0.64 km3/ka. These values, relatively high compared tothose of other dacitic centres in similar tectonic environ-ments, might be related to the rapid subduction of theNazca Plate beneath Ecuador.

This study also reveals a geochemical evolution relatedto the PVC development: (1) Over the entire PVC, themagmatic behaviour is characterized by an increase in silicacontent and most incompatible trace element abundancesand ratios, a characteristic not related to upper crustalprocess and instead probably related to deep processoccurring in the lower crust or mantle, (2) each collapseevent is linked to a new magmatic batch with a distinctivepetro-geochemical signature. These changes in compositionhave given rise to more explosive activity, with decreasingrepose periods for the last 4 ka. In this context the explosiveunrest witnessed in 1999–2001 after ∼340 years ofquiescence possibly marks the onset of a new magmaticcycle, which may lead to further tephra fall events overQuito in future decades.

Acknowledgements We warmly thank Joseph Cotten for carryingout the chemical analyses at the Laboratoire Domaines Océaniques,Université de Bretagne Occidentale (Brest, France). Constructivereviews by T. Sisson and S. de Silva of a previous version helped toimprove the manuscript. We deeply thank the suggestions and theeditorial handling of J. Stix. This contribution is part of anEcuadorian–French cooperation programme between the InstitutoGeofísico, Escuela Politécnica Nacional (IG-EPN), Quito, Ecuadorand the UR 031 “Volcanic processes and Hazards” of IRD (todayUMR 163 “Magmas et Volcans”).

References

Annen C, Blundy JD, Sparks RSJ (2006) The genesis of calc-alkalineintermediate and silicic magmas in deep crustal hot zones. JPetrol 47:505–539. doi:10.1093/petrology/egi084

Bacon CR, Lanphere MA (2006) Eruptive history and geochronologyof Mount Mazama and the Crater Lake region, Oregon. Geol SocAm Bull 118:1331–1359

Barberi F, Ghigliotti M, Macedonio G, Orellana H, Pareschi MT, RosiM (1992) Volcanic hazard assessment of Guagua Pichincha(Ecuador) based on past behaviour and numerical models. JVolcanol Geotherm Res 49:53–68

Bourdon E, Eissen JP, Monzier M, Robin C, Martin H, Cotten J, HallML (2002) Adakite-like Lavas from Antisana Volcano (Ecuador):evidence for slab melt metasomatism beneath the AndeanNorthern Volcanic Zone. J Petrol 43:199–217

Clapperton CM (1993) Glacier readvances in the Andes at 12,500–10,000 yr BP: implications for mechanism of late—glacialclimatic change. J Quat Sci 8:197–215

Clapperton CM, Hall M, Mothes P, Hole MJ, Still JW, Helmens KF,Kuhry P, Gemmel AMD (1997) A Younger Dryas icecap in theequatorial Andes. Quat Res 47:13–28

D’Ercole R, Metzger P (2004) La vulnerabilidad del Distrito Metropol-itano de Quito. Colección Quito Metropolitano 23, Quito

Davidson J, de Silva S (2000) Composite volcanoes. In: SigurdssonH, Houghton B, McNutt S, Rymer H, Stix J (eds) Encyclopediaof Volcanoes. Academic, San Diego, pp 663–681

Ego F, Sébrier M, Lavenu A, Yepes H, Egues A (1996) Quaternarystate of stress in the northern Andes and the restraining bendmodel for the Ecuadorian Andes. Tectonophysics 259:101–116

Feeley TC, Davidson JP (1994) Petrology of calc-alkaline lavas atVolcán Ollagüe and the origin of compositional diversity atCentral Andean Stratovolcanoes. J Petrol 35:1295–1340

Frey HM, Lange RA, Hall CM, Delgado-Granados H (2004) Magmaeruption rates constrained by 40Ar/39Ar chronology and GIS forthe Ceboruco-San Pedro volcanic field, western Mexico. GeolSoc Am Bull 116:259–276

Garcia-Aristizabal A, Kumagai H, Samaniego P, Mothes P, Yepes H,Monzier (2007) Seismic, petrologic, and geodetic analyses of the1999 dome-forming eruption of Guagua Pichincha volcano,Ecuador. J Volcanol Geotherm Res 161:333–351

Geotermica Italiana (1989) Mitigación del riesgo volcánico en el areametropolitana de Quito. Informe final vol. 2, Evolución geo-vulcanológica del Guagua Pichincha, Ed. Pisa, 105 p

Guillier B, Chatelain JL, Jaillard E, Yepes H, Poupinet G, Fels JF(2001) Seismological evidence on the geometry of the orogenicsystem in central-northern Ecuador (South America). GeophysRes Lett 28:3749–3752

Gutscher MA, Malavieille J, Lallemand S, Collot JY (1999) Tectonicsegmentation of the north Andean margin: impact of the CarnegieRidge collision. Earth Planet Sci Lett 168:255–270

Hall M, Mothes P (1997) El origen y edad de la Cangahua Superior,Valle de Tumbaco, Ecuador. In: Zebrowski C, Quantin P, TrujilloG (eds), III Simposio Internacional Suelos Volcánicos Endur-ecidos, Quito 19-28

Hall M, Mothes P (2008a) The rhyolitic-andesitic eruptive history ofCotopaxi Volcano, Ecuador. Bull Volcanol 70:675–702

Hall ML, Mothes PA (2008b) Quilotoa volcano—Ecuador: anoverview of the young dacitic volcanism in a lake-filled caldera.J Volcanol Geotherm Res 176:44–55

Hall ML, Robin C, Beate B, Mothes P, Monzier M (1999) TungurahuaVolcano, Ecuador: structure, eruptive history and hazards. JVolcanol Geotherm Res 91:1–21

Hidalgo S, Monzier M, Martin H, Chazot G, Eissen JP, Cotten J(2007) Adakitic magmas in the Ecuadorian Volcanic Front:

1128 Bull Volcanol (2010) 72:1109–1129

petrogenesis of the Iliniza Volcanic Complex (Ecuador). JVolcanol Geotherm Res 159:366–392

Hidalgo S, Monzier M, Almeida E, Chazot G, Eissen JP, van derPlicht J, Hall ML (2008) Late Pleistocene and Holocene activityof Atacazo–Ninahuilca Volcanic Complex (Ecuador). J VolcanolGeotherm Res 176:16–26

Hildreth W, Lanphere MA (1994) Potassium-argon geochronology ofa basalt-andesite-dacite arc system: the Mt. Adams volcanic field,Cascade Range of southern Washington. Geol Soc Am Bull106:1413–1429

Hildreth W, Fierstein J, Lanphere M (2003a) Eruptive history andgeochronology of the Mount Baker volcanic field, Washington.Geol Soc Am Bull 115:729–764

Hildreth W, Lanphere MA, Fierstein J (2003b) Geochronology anderuptive history of the Katmai volcanic cluster, Alaska Peninsula.Earth Planet Sci Lett 214:93–114

Hughes RA, Pilatasig LF (2002) Cretaceous and tertiary terraneaccretion in the Cordillera Occidental of the Andes of Ecuador.Tectonophysics 345:29–48

Klemetti EW, Grunder AL (2008) Volcanic evolution of VolcánAucanquilcha: a long-lived dacite volcano in the central Andes ofnorthern Chile. Bull Volcanol 70:633–650

Le Pennec JL, Jaya D, Samaniego P, Ramón P, Moreno Yánez S,Egred J, van der Plicht J (2008) The AD 1300–1700 eruptiveperiods at Tungurahua volcano, Ecuador, revealed by historicalnarratives, stratigraphy and radiocarbon dating. J VolcanolGeotherm Res 176:70–81

Legrand D, Calahorrano A, Guillier B, Rivera L, Ruiz M, Villagómez D,Yepes H (2002) Stress tensor analysis of the 1998-1999 tectonicswarm of northern Quito related to the volcanic swarm of GuaguaPichincha volcano, Ecuador. Tectonophysics 344:15–36

Monzier M, Robin C, Samaniego P, Hall ML, Cotten J,Mothes P, ArnaudN (1999) Sangay volcano, Ecuador: structural development, presentactivity and petrology. J Volcanol Geotherm Res 90:49–79

Mook WG, Streurman HJ (1983) Physical and chemical aspects ofradiocarbon dating. PACT Publications 8:31–55

Ownby S, Delgado Granados H, Lange RA, Hall CM (2007) VolcánTancítaro, Michoacán, Mexico, 40Ar/39Ar constraints on itshistory of sector collapse. J Volcanol Geotherm Res 161:1–14

Papale P, Rosi M (1993) A case of no-wind Plinian fallout atPululagua caldera (Ecuador): implications for models of clastdispersal. Bull Volcanol 55:523–535

Peccerillo P, Taylor SR (1976) Geochemistry of Eocene calc-alkalinevolcanic rocks from the Kastamonu area, northern Turkey.Contrib Mineral Petrol 58:63–81

Pennington WD (1981) Subduction of the eastern Panama Basin andseismotectonics of northwestern South America. J Geophys Res86:10753–10770

Richards JP, Villeneuve M (2001) The Llullaillaco volcano, northwestArgentina: construction by Pleistocene volcanism and destructionby sector collapse. J Volcanol Geotherm Res 105:77–105

Robin C, Samaniego P, Le Pennec JL, Mothes P, van der Plicht J(2008) Late Holocene phases of dome growth and Plinianactivity at Guagua Pichincha volcano (Ecuador). J VolcanolGeotherm Res 176:7–15

Robin C, Eissen JP, Samaniego P, Martin H, Cotten J (2009)Evolution of the late Pleistocene Mojanda—Fuya Fuya VolcanicComplex (Ecuador), by progressive adakitic involvement inmantle magma sources. Bull Volcanol 71:233–258. doi:10.1007/s00445-008-0219-9

Samaniego P, Martin H, Monzier M, Robin C, Fornari M, EissenJP, Cotten J (2005) Temporal evolution of magmatism atNorthern Volcanic Zone of the Andes: the geology andpetrology of Cayambe Volcanic Complex (Ecuador). J Petrol46:2225–2252

Samaniego P, Eissen JP, Le Pennec JL, Robin C, Hall ML, Mothes P,Chavrit D, Cotten J (2008) Pre-eruptive physical conditions of ElReventador volcano (Ecuador), deduced from the petrology ofthe 2002 and 2004–05 eruptions. J Volcanol Geotherm Res176:82–93

Samaniego P, Robin C, Chazot G, Bourdon E, Cotten J (2010)Evolving metasomatic agent in the northern Andean subductionzone, deduced from magma composition of the long-livedPichincha Volcanic Complex (Ecuador). Contrib Mineral Petrol.doi:10.1007/s00410-009-0475-5

Schiano P, Monzier M, Eissen JP, Martin H, Koga KT (2010) Simplemixing as the major control of the evolution of volcanic suites inthe Ecuadorian Andes. Contrib Mineral Petrol. doi:10.1007/s00410-009-0478-2

Singer BS, Jicha BR, Harper MA, Naranjo JA, Lara LE, Moreno-RoaH (2008) Eruptive history, geochronology, and magmaticevolution of the Puyehue-Cordón Caulle volcanic complex,Chile. Geol Soc Am Bull 120:599–618. doi:10.1130/B26276.1

Thouret JC, Finizola A, Fornari M, Legeley-Padovani A, Suni J,Frechen M (2001) Geology of El Misti volcano near the city ofArequipa, Peru. Geol Soc Am Bull 113:1593–1610

Villagómez D, Eguez A, Winkler W, Spikings R (2002) Plio-Quaternary sedimentary and tectonic evolution of the centralinter-Andean valley. 5th International symposium on Andeangeodynamics, Toulouse, France, Extended Abstracts 689–692

Bull Volcanol (2010) 72:1109–1129 1129