RESEARCH ARTICLE

Andesitic Plinian eruptions at Mt. Ruapehu:quantifying the uppermost limits of eruptive parameters

Natalia Pardo & Shane Cronin & Alan Palmer &

Jonathan Procter & Ian Smith

Received: 27 May 2011 /Accepted: 2 March 2012 /Published online: 31 March 2012# Springer-Verlag 2012

Abstract New tephro-stratigraphic studies of the TongariroVolcanic Centre (TgVC) on the North Island (New Zealand)allowed reconstruction of some of the largest, andesitic,explosive eruptions of Mt. Ruapehu. Large eruptions werecommon in the Late Pleistocene, before a transition tostrombolian-vulcanian and phreatomagmatic eruptive stylesthat have predominated over the past 10,000 years. Consid-ering this is the most active volcano in North Island of NewZealand and the uppermost hazard limits are unknown, weidentified and mapped the pyroclastic deposits correspondingto the five largest eruptions since ~27 ka. The selected erup-tive units are also characterised by distinctive lithofacies asso-ciations correlated to different behaviours of the eruptivecolumn. In addition, we clarify the source of the ~10–9.7 kaPahoka Tephra, identified by previous authors as the productof one of the largest eruptions of the TgVC. The most com-mon explosive eruptions taking place between ~13.6 and~10 ka cal years BP involved strongly oscillating, partiallycollapsing eruptive columns up to 37 km high, at mass dis-charge rates up to 6×108 kg/s and magnitudes of 4.9, ejectingminimum estimated volumes of 0.6 km3. Our results indicate

that this volcano (as well as the neighbouring andesitic Mt.Tongariro) can generate Plinian eruptions similar in magni-tude to the Chaitén 2008 and Askja 1875 events. Such erup-tions would mainly produce pyroclastic fallout covering aminimum area of 1,700 km2 ESE of the volcano, whereimportant touristic, agricultural and military activities arebased. As for the 1995/1996 eruption, our field data indicatethat complex wind patterns were critical in controlling thedispersion of the eruptive clouds, developing sheared, com-monly bilobate plumes.

Keywords Explosive volcanism . Eruptive parameters .

Isopach . Isopleths . Physical volcanology . Pyroclast

Introduction

Understanding the physical processes controlling Plinianeruptions at subduction-related composite volcanoes is crit-ical for estimating the associated maximum potential hazardfrom them (Jeanloz 2000). The term Plinian (Escher 1933;Walker and Croasdale 1971) traditionally refers to the ex-tremely energetic eruptive style characterised by large dis-persal areas and intermediate to high fragmentation index(Walker 1973; Wilson 1976; Rosi 1998; Cioni et al. 2000).During these eruptions, large volumes of pyroclasts (0.1–10 km3, corresponding to 1011–1013 kg) are ejected at highspeeds from the vent (100–400 m/s), at extreme mass dis-charge rates (106–108 kg/s) (Cioni et al. 2000). The resultingeruptive column reaches tropospheric to stratosphericheights (~30 km) and can be maintained for tens of hours(Wilson 1976).

Magma degassing processes, fragmentation depth andmechanisms, syn-and-inter-eruptive conduit geometry con-ditions, vent migration, physical and chemical changes in

Editorial responsibility: S. Nakada

Electronic supplementary material The online version of this article(doi:10.1007/s00445-012-0588-y) contains supplementary material,which is available to authorized users.

N. Pardo (*) : S. Cronin :A. Palmer : J. ProcterInstitute of Natural Resources, Massey University,Private Bag 11222,Palmerston North 4442, New Zealande-mail: npardov@gmail.com

I. SmithSchool of Environment, The University of Auckland,Private Bag 92019,Auckland, New Zealand

Bull Volcanol (2012) 74:1161–1185DOI 10.1007/s00445-012-0588-y

magma storage zones and magma rheology modificationsalong the conduit are all potential factors determining theeruptive behaviour of Plinian columns (Wilson et al. 1980;Papale and Dobran 1993; Varekamp 1993; Macedonio et al.1994; Cioni et al. 2003; Sulpizio 2005). The 27,097±957 cal years BP to ~10,000 year BP tephro-stratigraphicrecord of Mt. Ruapehu (New Zealand) provides a clearexample of the variability and complexity of Plinian stylesexperienced by a single andesitic volcano, where contrastinglithofacies associations are inferred to reflect variable mag-matic (i.e. vesiculation state, composition and volatile con-tent of the erupting magma, fragmentation mechanisms) andenvironmental conditions (i.e. conduit/vent geometry, in-flow of external water into the conduit), ultimately affectingthe steadiness of the eruptive column and tephra dispersion(Pardo et al. 2011). In this paper, we present new isopach andisopleth maps and quantify the main corresponding physicaleruptive parameters of the five eruptive units representing thelargest explosive eruptions known in the Late Pleistocenerecord of Mt. Ruapehu. These units were also selected torepresent contrasting lithofacies association types to studythe variability within Plinian eruptive styles. Our new dataallow us to establish the uppermost hazard limits and themaximum eruptive scenarios we could expect from the largestand most active andesitic volcano of New Zealand.

Geological and geographical setting

Mt. Ruapehu (i.e. Rua “pit”, pehu “to explode”; in Maorilanguage) is a 2,797 m andesitic–dacitic stratovolcano onthe Central Volcanic Plateau of North Island, within theactive Rangipo graben (Fig. 1), at the southern end of theTaupo Volcanic Zone (TVZ) (Cole 1978; Hackett 1985;Cole et al. 1986; Graham and Hackett 1987; Hackett andHoughton 1989; Graham et al. 1995; Wilson et al. 1995;Cronin et al. 1996a: Gamble et al. 2003; Villamor et al.2010). The main edifice is capped by small permanentglaciers and snowfields and comprises multiple overlappingcraters (Hackett and Houghton 1989; Cronin et al. 1996a;Kilgour et al. 2010). During the Holocene, eruptive activitywas focused at the youngest southernmost crater, which iscurrently filled with the acidic Crater Lake (Cole and Nairn1975; Christenson and Wood 1993). Holocene events weremainly phreatic to phreatomagmatic eruptions, producingbase surges and ballistic fallout, subplinian eruptive col-umns and lahars (Neall 1990; Cronin et al. 1996a, 1997;Neall et al. 1999; Kilgour et al. 2010). The edifice is asym-metrical, comprising lava-flows and rare domes, and recod-ing sector collapses (Palmer and Neall 1989; McClellandand Erwin 2003) and extensive lahars that contribute to thevolume of the eastern ring plain (Hackett and Houghton1989; Cronin and Neall 1997; Neall et al. 2001; Lecointreet al. 2004; Lube et al. 2009; Procter et al. 2010).

The lithostratigraphic record shows that major explosiveactivity of Mt. Ruapehu took place during the Late Pleisto-cene (Topping 1973), and the tephras produced during oneof the most active periods are grouped into the ~27 to ~10 kaBullot Formation (Donoghue 1991; Donoghue et al. 1995;Neall et al. 1995; Cronin et al. 1996a, b; Cronin and Neall1997). New stratigraphic data presented by Pardo et al.(2011) reveal at least 33 distinct eruptive units within thisformation, which they grouped into products of six eruptiveperiods that reflect important variations in eruptive behav-iour, particularly in column stability.

Around 10 % of New Zealand’s national economy isconcentrated in the Central North Island, and the surround-ings hold ~24 % of the country’s population (The Treasury2010). As demonstrated by the 1995/1996 eruptions, evenvery small tephra falls from this volcano may disperse ashbroadly over agricultural areas, along with towns and citiesand key infrastructure lifelines (Cronin et al. 1998, 2003;Johnston et al. 2000). Mt. Ruapehu is also located within theTongariro National Park, a UNESCO World Heritage areathat is one of the most visited tourist attractions in thecountry. In addition, up to ten thousand people can bepresent at any one time on the ski fields of Mt. Ruapehuduring the winter season (Kilgour et al. 2010). The populat-ed centres surrounding the park include Turangi (3,441inhabitants) to the north, Ohakune (1,101 inhabitants),Waiouru (1,383 inhabitants) and Taihape (1,788 inhabitants)to the South (Statistics New Zealand 2007, 2009).

Methodology

The best exposures of the Bullot Formation are found on theeastern flanks of the volcano and on the upper ring plain(Fig. 1) where 158 stratigraphic profiles were described. Wefocus on the five most widespread, thickest and coarsest-grained eruptive units, which also show contrasting lithofa-cies associations. The maximum thickness and the threemaximum axes of the largest five pumice and lithic clastsof the corresponding pyroclastic fall deposits were measuredto construct isopach and isopleth maps. Contours weredigitally drawn on a 20-m resolution digital elevation model(DEM), using ArcMap 9, to facilitate calculation of distan-ces, areas and circularity (expressed here as a shape factor,Sh0((4π×Area)/Perimeter2)). During each procedure, wesystematically compared our manual results to contoursderived by using automatic interpolations (e.g. Naturalneighbour/Kriging) generated within Surfer 8 (Golden Soft-ware) to better constrain the dispersal axis.

Volume calculations were made from both irregularlyshaped, whole-deposit isopachs and from the approximatedellipses following Pyle (1989) modified by Fierstein andNathenson (1992), Bonadonna et al. (1998) and Sulpizio(2005), to compare the effect of thickness extrapolation to

1162 Bull Volcanol (2012) 74:1161–1185

an infinite area. Long axes were measured from the vent andfollowing the distortion of the corresponding contour; max-imum short axes were measured in three different pointsperpendicular to the longest axis within an error of ±6 %.

The eruptive column height (Ht) was determined follow-ing the model of Carey and Sparks (1986), further comparedwith methods of Sparks (1986) and Sulpizio (2005). Themass-and-volume discharge rates (MDR and Q, respective-ly) were obtained using the model of Sparks (1986) andSparks et al. (1997). The eruptive magnitude was estimatedbased on Pyle (2000), by measuring the deposit density inthe field and calculating the total mass of the deposit (cf.Arana-Salinas et al. 2010).

In this study, the pumice textures are qualitatively de-scribed as finely vesicular if >90 % of vesicles are <2 mmand coarsely vesicular if larger. Vesicularity is classifiedaccording to Houghton and Wilson (1989). Average totalvesicularity and bulk density were obtained followingHoughton and Wilson (1989) by measuring 30 pumice clastsper lapilli fall bed within each eruptive Plinian phase withineach selected unit. Bulk clast volumes were obtained with anenvelope density analyzer (GeoPyc 1360-micromeritics) forlapilli clasts −3 to −4ϕ in size, previously cleaned in distilledwater and then dried. This instrument is based on the Archi-medes principle and uses a quasi-fluid displacement mediumcomposed ofmicrospheres having a high degree of flowability(DryFlo-micromeritics).

Pumice textures are referred to as: (1) frothy; highlyvesicular clasts dominated by subspherical vesicles <2 mmin diameter; (2) fluidal; highly to moderately vesicular clastswith strongly orientated, elongate (i.e. ellipsoidal) vesicles,commonly aligned with the longest axes of phenocrysts; (3)microvesicular; dense clasts with highly distorted, oftenaligned vesicles only visible in the microscope and usuallyshowing pinched shapes and angular terminations.

To evaluate the magma fragmentation mechanisms in-volved in the studied eruptives, clean juvenile ash grains,picked from the 3ϕ size fraction of the Plinian phase depos-its were imaged at 20 kV with a FEI Quanta 200 Environ-mental scanning electron microscope (SEM) at MasseyUniversity (New Zealand).

Mt. Ruapehu Plinian eruption lithofacies associations

Mt. Ruapehu’s eruptive behaviour has systematicallychanged since ~27 ka from older explosive eruptions char-acterised by steady eruptive columns, to younger eventscharacterised by unsteady, partially collapsing columns(Pardo et al. 2011). These stages were separated by aninterval between 17,625+425 cal years BP and shortly after13,635+165 cal years BP, when eruptive columns weresteady, but powerful and involved large proportions of ac-cessory and accidental lithic fragments. Each one of theseeruptive behaviours has been inferred by analysing pyroclastic

Fig. 1 a North Island of NewZealand tectonic setting(modified from Reyners et al.(2006) and Villamor et al.(2010)), showing theHikurangi-Kermadec subduc-tion margin. TVZ: Taupo volca-nic zone, with andesiticTongariro volcanic centre(TgVC) and the rhyolitic Oka-taina (OCC), Rotorua (RCC)and Taupo calderas (TCC); bTgVC comprises Mt. Ruapehuand Mt. Tongariro compositevolcanoes. SH: state highwaysconnect the urban centres(red squares). Study sites indi-cated by blue circles; mainreference type sectionslabelled as B

Bull Volcanol (2012) 74:1161–1185 1163

deposits with contrasting lithofacies associations, which Pardoet al. (2011) classified into three main types: Type 1 associa-tion comprises a few (<3) overlapping fall deposits character-ised by massive to normally graded, framework-supported,pumice lapilli beds and scarce porphyritic lithics. Usually, theinitial eruptive products include a basal concentration of freshandesitic lithics or angular to rounded dense pumice clasts.Juvenile pumice textures in the thickest, coarsest and mostwidespread beds representing the main eruptive phases arehighly vesicular, varying within the same stratigraphic levelfrom finely vesicular–frothy pumices, to coarsely vesicular–fluidal textures and raremicrovesicular (nearly dense), crystal-rich textures. Lithic content and lithological variability aresignificantly lower in comparison with the other lithofaciesassociations.

Type 2 association comprises one or a few, distinctivelylithic-rich, thick, ungraded fall deposits with abundant, multi-coloured, hydrothermally altered, aphanitic to finely porphy-ritic lithic clasts. Crude stratification is present across thedeposits, particularly away from the dispersal axis. The coars-est and thickest beds are usually underlain by a thin, massiveor laminated red, yellowish, purplish or olive grey, platycoarse ash to fine-ash bed, locally containing accretionarylapilli. The main part of an individual eruptive unit is typicallyrepresented by a single thick, coarse-grained bed consisting offramework-supported pumice fragments that are iron-stainedwith dark brown interiors, typically highly and finely vesicu-lar, microphenocryst-bearing, varying from frothy to micro-fluidal and microvesicular (nearly dense). Bombs showingbread-crust (as far as 15 km from the vent) or cauliflowerstructures (as far as 5 km from the vent) are common.

Type 3 association deposits comprises multiple, well-stratified fall deposits of contrasting grain-size interfingeredwith thin pyroclastic density current deposits. Lapilli fallbeds are commonly bounded by thin (<3 cm) fine ash bedscontaining abundant accretionary lapilli. Pumice texturesare the most heterogeneous among all lithofacies associa-tions, varying from incipiently vesicular to moderately andhighly vesicular, typically fibrous and colour-banded.

The new stratigraphic analysis presented by Pardo et al.(2011) shows the systematic change from older, typicallytype 1, to younger, typically type 3 eruptive units. Based ontheir lithofacies analysis, for this study, we chose the fol-lowing coarsest, thickest and most widely distributederuptive units within each lithofacies association type forfurther quantification: Eruptive unit IX (Mangatoetoenuieruptive unit) represents lithofacies association type 1; unitXXVI (Shawcroft Eruptive Unit) represents lithofacies as-sociation type 2 and eruptive units XXVII (OruamatuaEruptive Unit), XXIX (Akurangi Eruptive Unit) and XXXI(Okupata-Pourahu Eruptive Unit) together exemplify thelithofacies association type 3, which was the most common-ly produced between 13,635±165 and 10,000 cal years BP,

immediately before the transition to lower-magnitude Holo-cene eruptions.

The lowermost unit of the ~10 ka–9,700 years BP Pahoka-Mangamate Sequence (Nairn et al. 1998), known as thePahoka Tephra, was also mapped to clarify its source, whichhad been attributed variously toMt. Ruapehu,Mt. Tongariro ora vent located between the two, known as the “Saddle Cone”(Topping 1973; Donoghue 1991; Nairn et al. 1998).

Type 1: thickly bedded, lithic-poor association(Mangatoetoenui eruptive unit)

Field characteristics The Mangatoetoenui eruptive unit(Mgt) consists of two, framework-supported and well-sorted pyroclastic subunits (lower L-Mgt and upper U-Mgt) that mantle previous topography (Fig. 2), separatedby a volcaniclastic, cross-bedded sand deposit (IX-1d inFig. 2), which is very, well sorted, but varies considerablyin thickness and sedimentary structures laterally and distally.The L-Mgt consists of two fall beds (Fig. 2): The basal, thinlithic–lapilli deposit (IX-1a) was only identified to the eastand disappears beyond 10 km from the source. The overly-ing main, medium and coarse pumice lapilli bed (IX-b),gradually fines upwards to fine lapilli and very coarse ash(IX-1c) and is widely exposed over the entire study area.The U-Mgt comprises two massive beds, also mantlingprevious topography. The lower is the thickest and coarsest(IX-2a) and is capped by the thinner ash bed IX-2b (Fig. 2),which commonly grades into loess upward (Fig. 2).

The L-Mgt is mainly characterised by highly and coarse-ly vesicular pumice clasts (average vesicularity080 %; av-erage pumice density00.64 g/cm3. See Online resource 1)showing strong alignment of ellipsoidal vesicles and phe-nocrysts in a dark brown glass groundmass. The U-Mgtis characterised by microvesicular to dense, crystal-richclasts (total vesicularity064 %; average pumice density01.28 g/cm3), containing distorted, and in some instances,pinched vesicles, and entrapped lithics. Within any particularstratigraphic level, all the textural types from highly vesicular,frothy-like to fluidal and even microvesicular, dense juvenileclasts can be found. The phenocryst association is plagioclase(Pl)±orthopyroxene (Opx)±clinopyroxene (Cpx)>>Magne-tite (Mt), with most of the pyroxenes forming glomerocrysts.Juvenile pumice bulk composition ranges from basaltic an-desite to andesite (SiO2, 56.2–58.3 wt.%, normalised to drybasis), without showing significant variations with time(Online resource 2). The non-juvenile lithics are recognisedby their grey colour, microporphyritic texture and absence ofvesicles, containing variable amounts of Pl phenocrysts withina microlitic groundmass, very similar to the older andesiticlavas exposed on the slopes of the volcano reported by Gra-ham and Hackett (1987).

1164 Bull Volcanol (2012) 74:1161–1185

Interpretation TheMangatoetoenui eruptive unit represents atleast four eruptive phases. The eruption began with a conduit/vent clearing, probably phreatic explosion, producing themonolithologic lithic-rich bed IX-1a (e.g. Bursik 1993;Hammer et al. 1999; Wright et al. 2007). The restricted depos-its suggest that this opening phase was a laterally directedexplosion. Subsequently, a sustained eruptive column arose,producing the first main fall deposit (IX-1b), then graduallywaned over time as indicated by the normal grading (IX-1c).During a brief inter-phase hiatus, small, scour and fill channels

formed in and locally reworked previous deposits (IX-1d),before resumption of a buoyant eruptive plume deposited thevery coarse-grained and massive IX-2a fall bed of the U-Mgt.The uppermost ash-bed (IX-2b) was probably accumulatedfrom the dissipating cloud and further reworked by aeolianand fluvial processes. Beyond 17 km from the source, both L-and U-Mgt subunits merge into a single fall bed.

Isopach and isopleth maps Isopach and isopleth maps forindividual phases could not be constructed but were

Fig. 2 Lithofacies associationtype 1, represented by theMangatoetoenui Tephra. aStratigraphic position within theBullot Fm., above the 21,800±500 cal years BP, rhyoliticOkareka Tephra; b Exposure15 km from source showingnormally graded L-Mgt andmassive U-Mgt pumice lapillibeds, locally separated by syn-eruptive fluvial deposits (IX-1d); c Phases distinguished inproximal areas by contrastinggrain-sizes; d Composite strati-graphic profile. Relative pro-portion of juvenile glass (J),crystals (X) and lithics (L) aregiven for the main Pliniandeposits as volume percentbased on component analysis of300 grains within 1, 2, and 3ϕsize fractions

Bull Volcanol (2012) 74:1161–1185 1165

developed for the combined pyroclastic fall deposits. Thick-ness distributions (Online resource 3) show two clear depo-sitional lobes, the largest one directed towards the southeast(found as far as 35 km from source) and a more restrictedone to the northeast, deflected distally towards the north(Fig. 3a). The two lobes are also evident in both lithic andpumice clast isopleths (Fig. 4a, b). By tracing the axes ofapproximated ellipses extrapolated from field data, the most

probable vent for the Mgt eruptive unit is the North Crater ofMt. Ruapehu.

Type 2: massive-lithic rich (Shawcroft eruptive unit)

Field characteristics The Shawcroft eruptive unit (Sw)overlies the regional marker identified as the rhyolitic

Fig. 3 Isopach maps for: aMgt-Mangatoetoenui eruptiveunit (lithofacies association-type 1); b Sw-Shawcroft erup-tive unit (lithofaciesassociation-type 2); c Oru-Oruamatua eruptive unit; dAk-Akurangi eruptive unit; eOkp-Lower and Upper Okupatatephras(c–e: lithofaciesassociation-type 3); f U-Pk-Mt.Tongariro sourced UpperPahoka Tephra (N: currentNgauruhoe vent; R: current Mt.Ruapehu Crater Lake). Con-tours are labelled within whitesquares and shown in centi-metres (cm), drawn on a proxi-mal 5 m DEM combined with adistal 20 m DEM. In blacksquares, some of the localaverage field data values areshown (see Online resource 3).Upper right sub-quadrantsshow the contours interpretedfrom field data to illustrate thedispersion axes in relation tointermediate-distal urban areas(e.g. Napier, Hastings)

1166 Bull Volcanol (2012) 74:1161–1185

13,635±165 cal years BP Waiohau Tephra (Donoghue et al.1995) and is distinguished by a thick, multi-coloured, lithic-rich, massive, coarse lapilli and bomb-rich bed, defined as

“Shawcroft lapilli” by Donoghue et al. (1995), bounded atthe base and top by finer grained thin beds. The lowermost,lithic-rich, platy ash (Fig. 5), commonly with accretionary

Fig. 4 Isopleth maps showingthe distribution of lithic andpumice clast diameters inmillimetres: a–bMangatoetoenui; c–dShawcroft; e–f Oruamatua;g–h Akurangi; i–j Okupata;k–l U-Pahoka. Isopach tracedaxis extrapolated towards thecraters are shown in subfiguressuggesting North Crater as themost probable vent for mostunits, but not the youngestOkupata tephras, which origi-nated from a vent closer toCrater Lake, and the Pahokatephra which was produced byMt. Tongariro. NC: North cra-ter, CC: Central Crater, SC:South Crater, N: Mt. Ngauruhoe(see Online resources 4, 5 forcomplete field data set)

Bull Volcanol (2012) 74:1161–1185 1167

Fig. 4 (continued)

1168 Bull Volcanol (2012) 74:1161–1185

lapilli (XXVI-1a), is only exposed within 8 km of the vent.Bread-crust bombs in the Shawcroft lapilli (XXVI-1b) bed(Fig. 5c) are found on the eastern slopes of the volcano, upto 10 km from Crater Lake. At localities within 4 km of thevent, on the northern wall of the upper Whangaehu valley,there is an interbedded firm deposit (XXVI-1s bed, Fig. 5),which consists of a lower, yellow coarse ash bed, which ismatrix-supported, well-sorted, shows low-angle cross-lamination and contains abundant accretionary lapilli(Fig. 5d–f). The yellow ash bed is overlain by a dark grey,

firm, vesicular fine ash bed showing crude lamination. Bothof these yellow and grey beds vary laterally in thickness andshow impact sags (Fig. 5d–e).

Textural variability of juvenile pumice within a singlestratigraphic level is more restricted than in Mgt. Pumiceranges from dark brown, microvesicular to yellowish-brown, micro-fluidal clasts (average vesicularity063 %; av-erage pumice density01.25 g/cm3; Online resource 1), withfine (<1 mm) pyroxenes (Px), (Opx+Cpx+Pl)-glomeropor-phyres and non-juvenile, probably accessory andesitic

Fig. 5 Lithofacies association type 2, represented by the ShawcroftEruptive Unit (Sw); a Stratigraphic position above the 13,635±165 cal years BP, rhyolitic Waiohau Tephra (Wh); b zoom at 10 kmfrom the vent showing the deposits of individual phases; c typical lithicrich, coarse-grained lithofacies of the main phase (i.e. Shawcroft lap-illi) with bread-crust bombs up to 30 cm in diameter; d proximaloutcrop (5 km) showing cross-laminated pyroclastic surge deposits

(XXVI-1s) interbedded within the main lapilli fall deposits. Note theimpact-sag (sketched in e), under a ballistic clast, the crossed lamina-tion and accretionary lapilli (arrows) in f; g composite stratigraphicprofile. Relative proportion of juvenile glass (J), crystals (X) and lithics(L) are given for the main Plinian deposits as volume percent based oncomponent analysis of 300 grains within 1, 2, and 3ϕ size fractions

Bull Volcanol (2012) 74:1161–1185 1169

lithics. The mineral association for both types is Pl>Opx±Cpx>>Mt. Juvenile pumice bulk composition is basaltic–andesitic to andesitic (SiO2, 57.0– 58.4 wt.%, normalised todry basis; see Online resource 2). Non-juvenile lithic clastsvary from fresh grey andesites to multicoloured, hydrother-mally altered lithics and occasional metasedimentary lithics.

At intermediate and proximal locations, there are at leastfour thin fine ash beds on top of the Shawcroft lapilli(Fig. 5b), but at distal locations, it is directly covered by apoorly sorted, silty-sand, heterolithologic, crystal-rich depositof variable thickness, grading into loess at top (XXVI-e inFig. 5b, g).

Interpretation The Shawcroft lapilli is a widespread fall de-posit from a sustained eruptive column, preceded (XXVI-1a)and followed (XXVI-1c-d) by smaller, phreatomagmatic erup-tions that deposited the bounding thin, fine-grained, lithic-rich, platy ash beds. Although in most locations the mainphase (XXVI-1b) is represented as a single bed, there mayhave been two fall phases/pulses, separated by emplacementof the laminated pyroclastic surge deposits that crop out inproximal locations (Fig. 5f). The abundant lithic fragmentsand their diversity indicate intense erosion of the hydrother-mally altered regions of the conduit. Post-eruptive sedimenta-ry processes (XXVI-1e) deposited massive gravelly sand andparallel-bedded, silty-sand from high-discharge floods. Be-yond 15 km from the vent, only the Shawcroft lapilli ispreserved, immediately capped by post-eruptive volcaniclas-tic sediments or a weak paleosol.

Isopach and isopleth maps Isopach and isopleth maps of themain eruptive phase (XXVI-1b) (Fig. 3b, Online resource 3)show a main lobe towards the southeast and a short, sec-ondary lobe to the northeast. The two lobes are also evidentin both lithic and pumice clast isopleths (Fig. 4c, d). The mapssuggest that the vent probably located between the southernsector of North Crater and the Central Crater.

Type 3: thinly bedded (Oruamatua, Akurangiand Okupata-Pourahu eruptive units)

Lithofacies association type 3 deposits are the most abun-dant below the transition to the typically phreatomagmaticand phreatic deposits of the Holocene. The three thickest ofthese units are XXVII (here named Oruamatua eruptiveunit), XXIX (Akurangi eruptive unit) and XXXI (Okupata-Pourahu eruptive unit) in the tephro-stratigraphy defined byPardo et al. (2011).

Field characteristics The Oruamatua eruptive unit (Oru)consists of three pyroclastic subunits, here distinguished asLower (L-Oru), Middle (M-Oru) and Upper (U-Oru) (Figs. 6

and 7), with distinctive tephras separated by syn-eruptive,thin hyperconcentrated flow deposits. The L-Oru is lithic-rich and mantles older topography. The non-juvenile frag-ments comprise fresh, grey and brown, coarse porphyriticlava fragments, multi-coloured, hydrothermally altered por-phyritic to aphanitic clasts and occasional metasediments(Figs. 6c and 7). The M-Oru (Fig. 6d) shows importantfacies variations along the Upper Waikato stream (Fig. 1),where the clast-supported, well-sorted pumice lapilli bed iscompletely replaced by a massive, matrix-supported, verypoorly sorted deposit showing abrupt thickness variation(Fig. 7). The U-Oru (Figs. 6e and 7) mantles the M-Oru,and it is distinguished by the widespread, lowermostyellowish-brown platy ash bed (U-Oru3a in Fig. 7) contain-ing abundant, reversely graded accretionary lapilli. Lateralvariation of the main clast-supported pumice lapilli facies(U-Oru3b; Fig. 7) to poorly sorted, matrix-supported faciesis also evidenced along paleo-valleys.

The entire Oruamatua unit is distinguished from previousunits by having typically pale brown, crystal-rich (porphy-ritic), moderately to poorly microvesicular pumice clasts(average vesicularity058 %; average pumice density01.26 g/cm3; Online resource 1). Locally, and predominantlyin the M-Oru, there are pinkish-brown pumice clasts varyingfrom coarsely vesicular–fluidal to finely vesicular–fibroustexture and colour-banded, microvesicular clasts. Pheno-cryst content and size are large relative to other lithofaciesassociation types, and the general mineral assemblage isPl>>Opx>Cpx>>Mt. Juvenile pumice bulk compositionlies within the andesite field, showing a wider spread of silicacontent than do older units (SiO2, 57.33– 60.14 wt.%, nor-malised to dry basis; Online resource 2). At exposures >30 kmfrom the vent, individual subunits merge into a single, man-tling framework-supported fine lapilli bed.

The younger Akurangi eruptive unit (Figs. 6a and 7) isalso widely distributed and consists of four distinct pyro-clastic subunits very similar to Oru, locally separated bymatrix-supported, poorly sorted deposit with hetherolitho-logic granules and pebbles set in a lithic-crystal sand matrix(Figs. 3d and 7). The youngest Okupata-Pourahu eruptiveunit (Fig. 8) was described by Topping (1973) and Donog-hue et al. (1999). Pumice clasts are highly vesicular, typi-cally fibrous, often colour-banded and coarsely porphyritic.The phenocryst assemblage is Pl>Opx>Cpx>Mt. Our fielddata (Fig. 8c) indicate this unit consists of two main falldeposits: the Lower-Okupata (L-Okp) tephra (total vesicu-larity062 %; average pumice density01.15 g/cm3; Onlineresource 1), and the Upper-Okupata (U-Okp) tephra (averagevesicularity071 %; average pumice density00.92 g/cm3),locally separated by the Pourahu pyroclastic flow deposit(Donoghue et al. 1999), or by its correlative co-ignimbriteash (Ph-1d in Fig. 8b). Juvenile pumice bulk composition is themost variable of all the studied units, with most of the samples

1170 Bull Volcanol (2012) 74:1161–1185

plotting inside the basaltic–andesite and andesitic fields (SiO2,56.1–61.6 wt.%, normalised to dry basis; Online resource 2).Non-juvenile lithics include fresh, coarsely porphyritic greyandesites and red to orange, hydrothermally altered, porphy-ritic clasts. The pyroclastic deposits are capped by reworkedvolcaniclastic units and a paleosol (Fig. 8c). At distal loca-tions, the unit is condensed into a single mantling clast-supported bed, known as the Okupata Tephra (11,620±190 cal years BP, Topping 1973; Donoghue et al. 1995; Loweet al. 2008), which was emplaced at the onset of the earlyHolocene warming in Central North Island (Newnham andLowe 2000).

Interpretation The bedding in this type suggests pulsatingand unstable eruptive columns (c.f. Sieh and Bursik 1986;Bursik 1993; Cioni et al. 2003; Sulpizio et al. 2005). In thecase of the Oruamatua and Akurangi units, the main Plinianphases represented by the coarsest grained, thickest andwidely exposed fall beds were usually preceded by openingphreatomagmatic events producing thin, fine grained, accre-tionary lapilli-rich beds. The opening phases cleared theconduit and allowed subsequent decompression, drivingthe main Plinian phases (e.g. Arce et al. 2003; Rosi et al.2004; Gurioli et al. 2005; Hammer et al. 1999). Matrix-supported facies of variable thickness and limited distribution

were accumulated from pyroclastic density currents and indi-cate partial column collapse. Syn-eruptive and inter-eruptivereworked volcaniclastic deposits are intercalated with thepyroclastic beds.

Isopach and isopleth maps Because individual phases with-in each eruptive unit could not be discerned at medial-to-distal locations, isopach and isopleth maps were constructedfor the total thickness of fall beds of each unit (Fig. 3c, d, e).Field data (Online resource 3) suggest a main depositionallobe towards the southeast and a short lobe to the northeast.These lobes are also evident in both lithic and pumice clastisopleths (Fig. 4e, f). The most probable vent is the NorthCrater of Mt. Ruapehu. Isopach data of the Akurangi erup-tive unit can be explained by drawing a single lobe with adispersal axis towards the East (Fig. 3d); however, isoplethsindicate two lobes (Fig. 4g, h). This apparent discrepancymight reflect the effect of different wind patterns in a verti-cally stratified atmosphere with overlapping lobes produc-ing a composite deposit and the limited exposure in theeastern area. The dispersion axis of the deposits suggests avent located between central and north craters. Isopachmaps constructed with the merged thickness data of L-and-U-Okp tephras (Fig. 3e) show two lobes, one towardsthe north and the other one towards the south-east. Isopleth

Fig. 6 Bedded lithofaciesassociation type 3, Oruamatuaand Akurangi eruptive units: arelative stratigraphic positionabove Sw; b individual subunitsrepresenting different eruptivephases within the Oruamatuaeruptive: c Lithic-richLower-Oru, d Middle-Orushowing three bedsets indicatingthree main fallout phases sepa-rated by fine ash (oscillatingcolumns or wanderingplume effects) e Upper-Oru,partially reworked here (B15in Fig. 1)

Bull Volcanol (2012) 74:1161–1185 1171

data (Online resource 3) indicate a shift in the vent positionto the southern crater of Mt. Ruapehu (Fig. 4i, j).

The Pahoka Tephra

Field characteristics In addition to the units exposedabove, clearly from Mt. Ruapehu, the 10,000–9,800 yearsBP Pahoka Tephra (Topping 1974) was proposed to besourced from Mt. Tongariro (Topping 1974; Donoghue etal. (1995), before being re-assigned to a vent betweenMt. Ruapehu and Mt. Tongariro, beneath the “SaddleCone” lavas (Nairn et al. 1998). New field data showthat this eruption produced two subunits, here termedLower and Upper Pahoka (L-Pk and U-Pk, respectively;Fig. 9). The L-Pk comprises two thin, very fine-grainedand commonly eroded mantling beds, whereas the over-lying three beds are grouped within the U-Pk, which isthe only one that can be mapped well. The latter cor-responds to the Pahoka Tephra reported by Topping(1973) and is characterised by blocky-shaped, dark olive-grey and commonly colour-banded, microvesicular juvenilefragments (Pl>>>Cpx+Opx>Ol). Lithic clasts are darkgrey, vesicular to dense, coarsely porphyritic to aphanit-ic lava fragments, dark red and orange, hydrothermallyaltered, aphanitic lithics, with the majority coated with

ash and fine lapilli. At proximal locations, dense bombsup to 15 cm occur, showing chilled, cracked crusts (Fig. 9). Atdistal locations, the unit is condensed into a thinly laminated,compact, olive-grey ash.

Interpretation The unit represents at least five eruptivephases producing fall deposits of contrasting grain-size,the first two grouped within the L-Pk. The U-Pk2a bedindicates that an opening event ejected degassed microlite-rich, foamy magma, probably from a conduit plug or col-lapsed magma foam. This event led to further magma de-compression and development of the main Plinian columnaccumulating the U-Pk2b fall deposit. The uppermost U-Pk2c bed comprises ash accumulated from the dissipatingcloud and is overlain by post-eruptive hyperconcentratedflow deposits (U-Pk2d) in the E-NE area.

Isopach and isopleth maps The new mapping data indi-cate that the Upper Pahoka Tephra was sourced fromMt. Tongariro (proto-Ngauruhoe). Isopachs show a clearlobe towards the southeast (Fig. 3f), while isoplethsshow irregular proximal contours and regular mid-distal contours, supporting this source (Fig. 4k, l). Thereis no evidence supporting a source location between Mt.Ruapehu and Mt. Tongariro at the “Saddle Cone”. The

Fig. 7 Stratigraphic profile from the Oruamatua and Akurangi eruptive units showing lateral variation from fall to pyroclastic density current deposits(PDC)

1172 Bull Volcanol (2012) 74:1161–1185

Fig. 8 Uppermost studiedunits, comprising: a the lastknown Plinian deposit sourcedat Mt. Ruapehu (Okp-Ph) andthe first Plinian deposit of thePahoka-Mangamate explosiveperiod of Mt. Tongariro(Upper Pahoka Tephra); b detailof the two main fall depositsforming the Okupata-Pourahueruptive unit (L- and U-Okp),separated by a co-ignimbriteash at proximal locations (andin the same stratigraphic posi-tion as the pyroclastic flowdeposit named Pourahu mem-ber by Donoghue et al. 1999;Ph-1d bed); c composite strati-graphic profile (see legendFig. 5). Relative proportion ofjuvenile glass (J), crystals (X)and lithics (L) are given for themain Plinian deposits as vol-ume percent based on compo-nent analysis of 300 grainswithin 1, 2, and 3ϕ sizefractions

Fig. 9 Upper Pahoka Tephra asexposed a in proximal locations(<6 km from source) on thenortheastern slopes of Mt.Ruapehu. Note the dense,chilled bombs b typical facies atintermediate locations (13.5 kmfrom source), showing thedetailed textures representingthe phases described in the text

Bull Volcanol (2012) 74:1161–1185 1173

coarse bombs exposed on the north-eastern slopes ofMt. Ruapehu are enclosed within the same isopach/iso-pleths from Tongariro, including the extensive bomb fieldfound on Mangatepopo Ridge (Fig. 1).

Ash morphology

Ash particles (Fig. 10) indicate that the fragmenting magmahad variable vesicularity. The Mangatoetoenui unit (Fig. 10a),the Okupata tephra (Fig. 10e) and the Mt. Tongariro-sourced Pahoka Tephra (Fig. 10f) are dominated by highlyvesicular pumice with cuspate glass shards. Besides highly

vesicular clasts, poorly to non-vesicular, platy and blockyash grains with conchoidal fractures and v-shaped pits arepresent in the Shawcroft (Fig. 10b), Oruamatua (Fig. 10c)and Akurangi (Fig. 10d) units. There is high lithic content(up to 31 vol.% in the 3ϕ size fraction) and up to 24 vol.% ofthe shards (normalised to total juvenile content) are onlyweakly vesicular, but the combination of distinctive quench-ing cracks together with abundant stepped-surfaces that wouldunequivocally indicate thermo-hydraulic explosions (e.g.Büttner et al. 1999; Dellino et al. 2001) is lacking, suggestingthat magma–water interaction did not play a major role infragmenting the magma.

Fig. 10 SEM images ofjuvenile ash grains: a Mgtjuvenile, highly vesicular glassshards; b Sw poorly vesicularglass shards. Note theconchoidal fracture and sharpedges zoomed on theuppermost-right image; c Orucoarsely vesicular shards withthick vesicle walls around large,irregular vesicles, d Ak platy-shaped and poorly vesicularglass shards; e Okp fibrousglass shards; f Mt. Tongariro Pkglass shards

1174 Bull Volcanol (2012) 74:1161–1185

Eruptive parameters

The thickness of the studied fall units exponentially de-crease over distance (Fig. 11). Most of them show two linearsegments when Log10 thickness (Log10 T) is plotted againstdistance, expressed as the square root of the isopach area(A1/2), as is typical Plinian columns (Bonadonna et al.1998). Isopach geometries are mostly bilobate, except for

the Akurangi and Pahoka units (Fig. 3, Table 1). The Man-gatoetoenui, Okupata, Akurangi and Pahoka units are char-acterised by nearly elliptical isopachs with wide radialexpansion (aspect ratios >0.5 and shape factors >0.8),whereas the Shawcroft and Oruamatua units have elongateisopachs with narrower radial expansion (aspect ratios <0.5and shape factors <0.65). In general, NE lobes are shorterthan the main SE ones, except for Okupata tephras, which

Fig. 11 Whole-deposit isopach data plots for each eruption showing: athickness vs. isopach area; b log (T) vs distance expressed as (isopacharea)1/2. Individual eruptive units show two to three individual seg-ments with different slopes: c Mangatoetoenui (Mgt); d Shawcroft

(Sw); e Oruamatua (Oru); f Akurangi (Ak); g Combined Lower andUpper Okupata (Okp); hUpper Pahoka (U-Pk). Colours in c–h separatedifferent segments (S): proximal S0 (red), proximal–intermediate S1(blue) and in some cases, intermediate-distal S2 (yellow)

Bull Volcanol (2012) 74:1161–1185 1175

Table 1 Isopach data for individual units and resulting geometrical values calculated with ArcGis 9.0 for the whole deposit and for individuallobes. T: thickness, A: area, P: perimeter, Sh: shape factor, ASE: area of the southeast lobe, ANE: area of the northeast lobe

Whole deposit SE lobe NE lobe

T[cm]

A[km2]

A1/2 Sh ASE

[km2]ASE

1/2

[km]Longaxis

Shortaxis

Aspectratio

Sh ANE

[km2]ANE

1/2

[km]Longaxis

Shortaxis

Aspectratio

Sh

Mangatoetoenui eruptive unit (Mgt)

50 52 7.2 0.6 37 6.1 11.0 3.1 0.3 0.5 16 4.0 7.9 2.7 0.3 0.6

45 63 8.0 0.8 47 6.8 12.0 3.7 0.3 0.7 23 4.8 8.5 2.7 0.3 0.6

40 83 9.1 0.5 59 7.7 12.5 4.3 0.3 1.0 44 6.7 13.9 2.5 0.2 0.6

35 105 10.3 0.5 73 8.6 13.0 5.4 0.4 0.8 52 7.2 14.4 2.7 0.2 0.6

30 126 11.2 0.5 98 9.9 14.1 6.8 0.5 0.8 63 7.9 14.6 3.0 0.2 0.6

25 169 13.0 0.6 166 12.9 15.6 8.6 0.6 0.8 79 8.9 15.8 3.5 0.2 0.6

20 293 17.1 0.6 264 16.3 22.6 12.8 0.6 0.8 102 10.1 17.2 5.0 0.3 0.6

15 442 21.0 0.6 403 20.1 27.2 16.0 0.6 0.9 146 12.1 19.0 5.5 0.3 0.7

10 596 24.4 0.6 526 22.9 30.6 19.2 0.6 0.9 208 14.4 21.7 6.1 0.3 0.7

5 824 28.7 0.6 696 26.4 34.2 23.5 0.7 0.9 315 17.7 26.5 7.6 0.3 0.7

Shawcroft eruptive unit (Sw)

30 128 11.3 3.0 71 8.4 12.9 7.7 0.6 0.8 52 7.2 8.4 4.6 0.5 0.9

25 204 14.3 1.5 160 12.6 22.8 7.3 0.3 0.6 101 10.0 9.3 5.8 0.6 1.0

20 268 16.4 0.9 216 14.7 24.2 8.5 0.4 0.7 152 12.3 10.9 7.9 0.7 0.9

15 651 25.5 0.3 491 22.2 45.4 13.2 0.3 0.5 299 17.3 12.9 12.5 1.0 0.5

10 973 31.2 0.1 739 27.2 47.0 18.0 0.4 0.6 470 21.7 16.8 13.1 0.8 0.6

5 1764 42.0 0.0 1403 37.5 60.0 24.1 0.4 0.7 670 25.9 18.3 18.9 1.0 0.6

Oruamatua eruptive unit (Oru)

30 38 6.2 0.7 55 7.4 14.6 4.2 0.3 0.6 33 5.7 9.3 4.5 0.5 0.7

25 133 11.5 0.5 151 12.3 28.8 4.9 0.2 0.4 47 6.8 10.3 4.9 0.5 0.7

20 438 20.9 0.4 377 19.4 38.5 10.5 0.3 0.5 136 11.7 20.0 6.8 0.3 0.6

15 684 26.2 0.5 555 23.6 43.0 14.3 0.3 0.5 245 15.7 29.9 8.6 0.3 0.5

10 1009 31.8 0.5 827 28.8 52.1 19.8 0.4 0.6 385 19.6 30.4 10.9 0.4 0.6

5 1965 44.3 0.5 1742 41.7 52.4 20.7 0.4 0.6 555 23.6 32.0 14.9 0.5 0.7

Akurangi eruptive unit (Ak)

25 76 8.7 0.8 88 9.4 13.3 7.3 0.5 0.9

20 141 11.9 0.8 157 12.5 18.1 9.8 0.5 0.9

15 360 19.0 0.9 347 18.6 25.9 16.6 0.6 0.9

10 587 24.2 0.9 625 25.0 30.4 23.1 0.8 0.9

5 2684 51.8 0.9 2576 50.8 69.0 13.6 0.2 0.9

L- and U-Okupata eruptive unit (Okp)

20 243 15.6 0.7 43 6.6 10.5 5.7 0.5 0.9 240 15.5 23.3 13.0 0.6 0.8

15 396 19.9 0.5 173 13.2 19.1 10.5 0.5 0.9 349 18.7 25.4 18.0 0.7 0.9

10 839 29.0 0.5 394 19.8 26.3 17.1 0.7 0.9 552 23.5 28.1 24.0 0.9 0.9

5 1274 35.7 0.6 927 30.4 31.9 26.8 0.8 0.9 754 27.5 30.4 30.4 1.0 0.9

U-Pahoka eruptive unit (UPk)

30 53 7.3 0.8 7 11.0 6.2 0.6 0.8

25 144 12.0 0.7 12 19.3 8.4 0.4 0.7

20 402 20.0 0.9 20 24.2 18.9 0.8 0.9

15 673 25.9 0.9 26 31.2 25.1 0.8 0.9

10 1315 36.3 0.9 36 44.0 36.2 0.8 0.9

5 1727 41.6 0.9 42 51.9 40.5 0.8 0.9

1176 Bull Volcanol (2012) 74:1161–1185

have a large lobe dispersed towards the north. The thinningrate of the SE lobes is highest for the Mangatoetoenui (5 cmisopach, 37.5 km from source) and L- and U- Okupata units(5 cm isopach, 33.9 km from source) and lowest in theOruamatua eruptive unit (5 cm isopach, 76.5 km fromsource). The Shawcroft, Akurangi and the Upper Pahokaunits have intermediate values (5 cm isopach, 58.5, 68.5and 52.1 km from source, respectively).

Eruptive volumes

Due to limitations of exposure, individual units could onlybe traced reliably to the 5 cm isopach, out to ~80 km fromsource (Figs. 1 and 3). Therefore, volume calculations(Table 2) are minimum estimates, and significant volumesin distal areas are not accounted for. Depending on themethod applied, the thinning rate used (k total or k0-proxi-mal) and the isopach shape (i.e. working with individuallobes or with whole-deposit contours) total volumes differby between 8 % and 31 %. If working with separate lobes,the sum of the individual volumes shows no significantdifference relative to data taken from whole-deposit iso-pachs. However, to avoid possible overlapping when com-bining two lobes and based on irregular shapes of theerupted clouds as shown by the Chaitén 2008 eruption (Lara2009; Watt et al. 2009), we prefer not to approximate iso-pachs to perfect ellipses and restrict the discussion to theresults obtained from the irregular-shaped, whole-depositmaps. For the total data set (one line-segment only), vol-umes calculated following Sulpizio (2005) and the tradition-al Pyle method (1989, modified by Fierstein and Nathenson1992) are very similar (±4–10 %); if integrating multiplesegments (Bonadonna et al. 1998) and using the proximalthinning rate (k0) in the Sulpizio (2005) method, calculatedvolumes differ by 1 % to 16 %, except for Shawcroft andPahoka units, where volumes are 30 % and 57 % higher ifapplying the Sulpizio (2005) approach.

The calculated minimum erupted volumes for the select-ed units range between 0.2 and 1.2 km3 using the range ofmethodologies described above. Following the Sulpizio(2005) method and considering the uncertainties given bythe lack of distal isopach data, calculated volumes rangebetween 0.3 and 0.6 km3, assuming Vp/Vt<0.3 (Table 2).

Isopleths and classification of the eruptions

Lithic and pumice isopleths (Fig. 4) are irregularly shapedand suggest two directions of dispersion, similar to thecorresponding isopachs. The Mangatoetoenui unit (Fig. 4a,b) has a main lobe towards the SE (0.5 cm isopleths reach35 km) and a shorter lobe towards the NE. The Shawcroftcontours are more irregularly shaped (Fig. 4c, d), with prox-imal lithic isopleths toward the east, but more distal <3 cm

isopleths are deflected towards the southeast (0.5 cm isoplethsextend 42 km). The secondary Shawcroft NE-lobe is wellconstrained for pumice and lithic lapilli isopleths, and is bentin distal reaches to the east. The highly irregular shapes ofproximal lithic isopleths might reflect multiple vents. TheOruamatua eruptive unit (Fig. 4e, f) has a main lobe towardsthe SE (0.5 cm, 57 km from the vent) and a smaller lobetowards the NE, which is best reflected by pumice isopleths.For the Akurangi unit (Fig. 4g, h) and the Okupata tephras(Fig. 4i, j), the NE-lobe is straight and as large as the SE-lobe.

For the whole deposit, both isopleths and isopachs, allstudied units fit within the Pyle (1989) Plinian classification(Fig. 12a, Table 3). Estimated dispersal areas inside the 0.01Tmax isopach (c.f., Wilson 1976) are greater than 2,000 km2.

Column heights, mass discharge rates and eruptivemagnitude

A two-lobe geometry explains best fits the field data, so wecalculated the column height (HT) for each individual lobe.SE-lobes suggest relatively higher HT, varying from 21 to37 km by either Sulpizio (2005) or Carey and Sparks (1986)methods. On the other hand, the NE-lobes indicate columnsbetween 17.5 and 28 km. In general, column heights obtainedby all applied methods reached stratospheric levels.

When plotting cross-wind vs. down-wind maximumranges (Fig. 12b) as done by Carey and Sparks (1986), mostof the data imply strong winds were >>30 m/s, except duringdeposition of the Oruamatua eruptive unit (10 to 15 m/s).

Volume discharge rates (Q) calculated from column heightsobtained with the Sulpizio (2005) method are on the orderof ~104 m3/s, except for the Oruamatua and Akurangi units(Fig. 12c, d), which are up to one order of magnitude higher(~105 m3/s). Corresponding mass discharge rates (MDR) varybetween ~107 and 108 kg/s (d, Fig. 12c, d), as is characteristicof Plinian eruptions. When plotting HT vs. Q and vs. MDR(Fig. 12c, d), all data indicate magma eruption temperaturesbetween 600 °C and 1,000 °C, as expected (cf. Sparks 1986).Independently of the input data (whole-deposit or individuallobe) and the method used to obtain HT values, the Manga-toetoenui unit has the smallest HT and MDR values and theOruamatua the highest (Fig. 13).

The magnitudes M estimated by the Pyle (2000) methodare >4.9 (apart fromMangatoetoenuiM>4.4 andOkupata >4.6),similar to other Plinian eruptions (Table 4 and Fig. 12).

Discussion

Physical volcanology

A segmented distribution of tephra thickness over distancefor Plinian fall deposits was explained by Bonadonna et al.

Bull Volcanol (2012) 74:1161–1185 1177

1178 Bull Volcanol (2012) 74:1161–1185

Tab

le2

Eruptivevo

lumes

incubickilometresob

tained

usingmetho

dsfrom

differentauthors,consideringon

esing

lesegm

ent,multip

lesegm

ents,as

wellas

thedata

from

who

le-deposits

andas

obtained

from

individu

aldepo

sitio

nallobes

UnitPyle(198

9,mod

ifiedby

Fierstein

and

Nathenson

1992

,andBon

adon

naet

al.19

98)

Sulpizio(200

5)

Who

ledepo

sit

SELob

eNELob

eCom

bined

lobes

Who

ledepo

sit

usingkvs

Aip1/2

Wholedeposit

usingk 0

vsAip1/2

SE+NElobes

usingkvs

Aip1/2

SE+NElobes

usingk 0

vsAip1/2

Observed

BS12

[km]

Observed

BS23

[km]

Aip1/2

calc

(k)

(Sulpizio

2005)

Aip1/2

calc

(k0)

(Sulpizio

2005)

0.01

Tmax

(vol

1segm

ent)

D [km

2]

Walker

(197

3)1 seg

mult.

seg.

1 seg

mult.

seg.

1 seg

mult.

seg.

1seg

multVp/

Vt<

0.3

0.3<

Vp/Vt

<0.7

Vp/

Vt>

0.7

Vp/

Vt<

0.3

0.3<

Vp/Vt

<0.7

Vp/

Vt>

0.7

Vp/

Vt<

0.3

0.3<

Vp/Vt

<0.7

Vp/

Vt>

0.7

Vp/

Vt<

0.3

0.3<

Vp/Vt

<0.7

Vp/

Vt>

0.7

Mgt

0.2

0.20.2

0.2

0.1

0.1

0.3

0.3

0.3

1.9

1.5

0.2

0.4

1.10.3

0.8

2.3

0.4

1.2

3.5

12.5

21.1

31.4

25.4

1.0

2,25

4

Sw

0.5

0.30.3

–0.2

0.1

0.5

0.4

0.6

1.0

3.0

0.3

0.6

4.80.6

1.0

3.1

0.6

1.1

3.3

16.0

27.6

59.3

38.7

0.6

8,48

3

Oru

0.4

0.40.3

0.3

0.1

0.1

0.5

0.4

0.5

0.9

2.7

0.8

1.1

3.00.6

1.1

3.2

1.1

1.8

5.4

20.9

20.9

62.9

107.1

0.5

9,60

0

Ak

0.5

0.6

––

––

––

0.6

1.0

3.0

0.4

0.7

2.2

––

––

––

19.4

24.3

80.8

61.6

0.3

16,184

Okp

0.3

0.20.1

0.1

0.2

0.2

0.3

0.3

0.4

0.6

1.8

0.4

0.8

2.20.4

0.8

2.3

0.5

1.0

2.9

29.2

–45

.758

.40.6

4,94

3

UPk

0.4

0.4

––

––

––

0.5

0.9

2.8

1.2

1.9

5.7

––

––

––

29.3

–59

.396

.80.5

8,48

3

k:Slope

(thinn

ingrate)whenalldata

arefitwith

inasing

lecurve;

k 0:slop

eof

theprox

imal

segm

entwhenmultip

lesegm

entsareidentified(m

axim

umcase).Aip0ob

served

isop

acharea

atthe

inflexionpo

intw

henmultip

lesegm

entsareconsidered.V

p:prox

imalvo

lume.Vt:totalv

olum

e.Aip1/20distance

from

vent

expressedas

thesquare

root

oftheisop

acharea

inkilometres.The

suffix

“calc”

staysforcalculated

whenusingSulpizio(200

5)metho

d.BS 1

2:b

reak

inslop

ebetweenthefirsttwosegm

ents;B

S 23:b

reak

inslop

ebetweenthesecond

andthirdsegm

ents.D

:dispersalindex

obtained

byextrapolatingthearea

enclosed

with

inthe0.01

Tmax

isop

ach(W

alker19

73).Tmax0maxim

umthickn

ess.Eruptiveun

its:Mangatoetoenu

i(M

gt),Shawcroft(Sw),Oruam

atua

(Oru),

Aku

rang

i(A

k),Oku

pata

Tephras

(Okp

*)andUpp

erPahok

aTephra(U

Pk)

(1998) as the result of the deposition of particles havingdifferent Reynolds-numbers (Re). The pumice and lithicclasts accumulated during the main eruptive phases of eachselected unit correspond to high and intermediate Re

particles (coarse ash and lapilli), lifted to 15– 35 km in theatmosphere, which suggests that most of the particles wereincorporated in the turbulent-flow regime of the eruptivecloud and were accumulated according to their inertial

Fig. 12 Classification schemes for the studied eruptions: a isopachand isopleth data in the Pyle (1989) diagram lie within the Plinian field;b isopleth data in the Carey and Sparks (1986) diagram for columnheight and wind-speed, based on 0.8 cm-diameter lithic clasts data; c–d

Sparks (1986) diagram to determine mass discharge rates consideringcolumn heights obtained with the Carey and Sparks (1986) method (c)and Sulpizio (2005) method (d). Other eruption parameters are respec-tively plotted for comparison

Bull Volcanol (2012) 74:1161–1185 1179

settling velocities, independent of air viscosity (cf. Bona-donna et al. 1998). Hence, thinning rate is better describedby exponential laws than by power laws (Fig. 11).

Using the method of Pyle (1989, modified by Fierstein andNathenson 1992), integrating multiple segments usually pro-vides a smaller volume than working with a single segment.This might be due to the fact that most of our data lie withinproximal and medial regions, where the last segment usuallyhas a faster thinning rate than the most proximal segments(Fig. 10b). Data published for other eruptions (e.g. Hudson1991 and Cerro Negro 1971 in Sulpizio 2005) suggest that thisis a common feature for the first 50 km, before the depositbegins to display the typical exponential or power-law thin-ning trend with distance (c.f. Walker 1973; Pyle 1989;Fierstein and Nathenson 1992). This is likely due to particleaggregation, with accretionary lapilli beds reflecting the effi-ciency of high columns in promoting fine-particle aggregation(Watt et al. 2009). Alternatively, it may reflect poor preserva-tion of the complete fall sequence at every location and/orfluctuating winds during each eruptive phase. Most tephrasdescribed here are >5 cm thick and within the lapilli to verycoarse ash grade (1–64 mm in diameter), so the Sulpizio(2005) method, in which the break in slope is calculated fromthe proximal–intermediate dataset, is more appropriate. It isimpossible, however, to define the actual relationship betweenproximal (Vp) and total volume (Vt), and thus possible rangesare presented in Table 2. These data show that the Mangatoe-toenui unit is the smallest described (0.3 km3), with the othershaving volumes ~0.5–0.6 km3. Erupted volumes of the multi-bedded lithofacies association type-3 units (Oruamatua,Akurangi and Okupata) do not account for the associated co-Plinian pyroclastic density current deposits.

The asymmetrical and irregular shapes of isopleths reflectthe strong influence of the local wind pattern. The bilobateisopachs and isopleths (Figs. 3 and 4) suggest two predomi-nant wind directions (north-westerlies and south-westerlies),varying: (a) with time, so that pauses between eruptive pulses/phases cannot be distinguished, or (b) with altitude in theatmosphere, so that higher portions of the plume could havebeen affected by north-westerlies and the lower portion bysouth-westerlies. Both options are very likely in the NorthIsland, where wind direction and speed can significantlychange within a few hours (cf. Cronin et al. 1998; Turnerand Hurst 2001).

The deflection of the dispersal axis reconstructed fromboth lithic and pumice data of the Shawcroft Unit (Fig. 4a,c) and pumice data from the Oruamatua unit (Fig. 4f) sug-gests important cross-wind effects in these cases, causingbending of the plume. Deflections of the eruptive plume axisand multiple lobes have been recently documented fromfield data and simulated maps for other high-latitude volca-noes, such as Katla 1676 ±12 and ~3,600 years BP (Larsenet al. 2001), Askja 1875 (Carey et al. 2010) and HudsonT

able

3Geometricalparametersob

tained

from

who

le-depositisop

achandisop

lethsdata

Unit

Mangatoetoenu

iShawcroft

Oruam

atua

Aku

rang

i(L

+U)Oku

pata

Upp

erPahok

a

1Seg

S0

S1

S2

1Seg

S0

S1

S2

1Seg

S0

S1

1Seg

S0

S1

S2

1Seg

S0

S1

1Seg

S0

S1

bt4.0

3.2

6.2

2.7

7.8

5.0

8.7

6.1

8.3

14.5

6.5

4.0

3.2

6.2

2.7

6.0

7.7

3.8

7.8

13.0

4.9

bc(lith

ics)

6.5

6.5

6.5

6.5

5.6

5.6

5.6

5.6

9.8

9.8

9.8

5.8

5.8

5.8

5.8

9.8

9.8

9.8

6.5

6.5

6.5

bc/bt*

1.6

2.0

1.1

2.4

0.7

1.1

0.6

0.9

1.2

0.7

1.5

1.4

1.8

0.9

2.1

1.6

1.3

2.6

0.8

0.5

1.3

bc(pum

ice)

7.8

7.8

7.8

7.8

9.1

9.1

9.1

9.1

13.0

13.0

13.0

7.0

7.0

7.0

7.0

9.8

9.8

9.8

19.6

19.6

19.6

bc/bt**

1.9

2.4

1.3

2.9

1.2

1.8

1.0

1.5

1.6

0.9

2.0

1.7

2.2

1.1

2.6

1.6

1.3

2.6

2.5

1.5

4.0

Followingtheterm

inolog

yof

Pyle(198

9):bt0“thickness

half-distance”

isthedistance

atwhich

thethickn

essdecreasesto

onehalfof

itsmaxim

umvalue,which

describesthemorph

olog

yof

the

depo

sit;bc

0“clasthalf-distance”

isthedistance

atwhich

themaxim

umclastdiam

eter

halves

with

respectto

itsmaxim

umvalue,reflectin

gthecorrespo

ndingcolumnheight.H

ence,b

c/bt

givesan

estim

ationof

thefragmentatio

nindex.

1Seg:consideringalldata

ason

esing

lesegm

ent;S0

-2:individu

alsegm

entsas

separatedby

coloursin

Fig.11

*refers

tobc/btbasedon

lithicdata

**refers

tobc/btbasedon

pumicedata

1180 Bull Volcanol (2012) 74:1161–1185

1991 (Kratzmann et al. 2010); the complexity of tephradispersion, strong contour distortion, axis bending and theinfluence of shifts in the wind-direction over a short timeinterval (hours) were strongly corroborated by the eruptionof Chaitén (Chile) in 2008 (Lara 2009; Watt et al. 2009).

Our data are consistent with similar eruptions at otherandesitic–dacitic stratovolcanoes (Figs. 12 and 13). It isclear that the lack of distal data containing low-Reparticle information (cf. Bonadonna et al. 1998) andthe Late Pleistocene windy, poorly vegetated periglacial

Fig. 13 Comparison of eruptive parameters with others published forPlinian eruptions at andesitic volcanoes worldwide. Our data indicate:a increasing column heights with erupted volume as obtained from thewhole deposit of each unit and b with MDR; c–d eruptive intensity

(MDR) and column height vs magnitude (M0Log (mass of the depositin kilogrammes) −7), with higher intensities (c) and column heights (d)reached at larger magnitudes

Bull Volcanol (2012) 74:1161–1185 1181

conditions unfavourable for tephra preservation are cru-cial factors in causing significant underestimations oferupted volumes.

When plotting HT and Log10 MDR vs. erupted mag-nitude, expressed as (Log10 (total deposit mass) −7), aweak positive correlation indicates that larger columnheights and eruptive intensities correlate with largereruptive magnitudes (Fig. 13). Mangatoetoenui andOkupata eruptions were the least violent of those stud-ied, whereas Shawcroft, Oruamatua and Akurangi werethe most violent explosive eruptions known from Mt.Ruapehu. All of the most violent eruptions produceddeposits between 13,635±165 cal years BP and ~10 ka.These units show a characteristic basal thin, fine-ashbed linked to a phreatomagmatic opening phase, asinterpreted for lithofacies associations 2 and 3. Theyalso show the highest content of non-juvenile, highlyhydrothermally altered lithics in the deposits of thePlinian phases, indicating strong conduit erosion (e.g.

Wilson et al. 1980, Macedonio et al. 1994) during theeruption and probably the disruption of a pre-existinghydrothermal system (e.g. Varekamp 1993; Thouret etal. 2002). Juvenile shards in the 3ϕ size fraction (Fig. 10) inthe Shawcroft, Oruamatua and Akurangi units containpoorly vesicular, blocky shards showing conchoidal frac-tures and rare stepped surfaces (c.f., Buttner et al. 1999;Dellino et al. 2001), which occur together with the highlyvesicular shards. This implies some degree of magma–waterinteraction occurred, but its role in fragmentation is not wellunderstood.

Our study reveals that Mt. Ruapehu is capable of produc-ing events of magnitudes 4 to 5. The characterisation andquantification of Late Pleistocene events (the last one closeto ~11,620±190 cal years BP) indicate that the greatesthazard scenario expected for this volcano involves Pliniancolumns similar to those produced during the eruptions ofAskja 1875 (Carey et al. 2010) and Chaitén 2008 (Lara2009; Watt et al. 2009).

Table 4 Estimated eruptive parameters considering the volume (expressed as a minimum value vol*) calculated by using the method of Sulpizio(2005)

Unit Mgt Sw Oru Ak Okp U-Pk

Column heights(Ht) [km]

Vol* [km3] 0.3 0.6 0.5 0.6 0.4 0.5

Ht-Sparks (1986) 21.4 25.0 25.0 24.3 26.0 25.0

Ht-Sulpizio (2005) 22.8 31.1 32.0 36.2 27.4 31.1

HtSE: Sulpizio (2005) 22.0 29.0 37.2 35.3 24.8 31.1

HtNE: Sulpizio (2005) 18.9 23.4 23.4 – 21.3 –

HtSE: Carey and Sparks (1986) 21.0 25.0 32.0 21.0 22.0 21.5

HtNE: Carey and Sparks (1986) 17.5 20.0 23.0 17.5 28.0 –

Volume discharge rate(Q) [m3/s]

Q: Sparks (1986) 1.9E+04 3.4E+04 3.4E+04 3.1E+04 4.0E+04 3.4E+04

Q: Sulpizio (2005) 2.4E+04 8.0E+04 8.9E+04 1.4E+05 4.9E+04 8.0E+04

QSE: Sulpizio (2005) 2.1E+04 6.1E+04 1.6E+05 1.3E+05 3.3E+04 –

QNE: Sulpizio (2005) 1.2E+04 2.7E+04 2.7E+04 – 1.8E+04 –

QSE: Carey and Sparks (1986) 1.8E+04 3.4E+04 8.9E+04 1.8E+04 2.1E+04 1.9E+04

QNE: Carey and Sparks (1986) 8.7E+03 1.5E+04 2.5E+04 8.7E+03 5.3E+04 –

Mass discharge rate(MRD) [kg/s]

MDR: Sparks (1986) 6.0E+07 7.8E+07 7.5E+07 7.0E+07 8.0E+07 7.8E+07

MDR: Sulpizio (2005) 6.8E+07 3.5E+08 4.5E+08 6.0E+08 9.2E+07 3.5E+08

MDRSE: Sulpizio (2005) 6.8E+07 1.5E+08 6.0E+08 6.0E+08 7.5E+07 –

MDRNE: Sulpizio (2005) 5.5E+07 7.0E+07 7.0E+07 – 6.0E+07 –

MDRSE: Carey and Sparks (1986) 6.0E+07 7.8E+07 4.3E+08 7.2E+07 6.2E+07 6.0E+07

MDRNE: Carey and Sparks (1986) 5.5E+07 5.8E+07 7.0E+07 4.0E+07 9.8E+07 –

Eruptive magnitude(M)

Vol (m3) 3.0E+08 6.0E+08 5.0E+08 6.0E+08 4.0E+08 5.0E+08

Wet Deposit density (kg/m3) 990.2 1348.6 1449.6 1380 1156.7 1441

Total mass (kg) 2.97E+11 8.09E+11 7.25E+11 8.28E+11 4.63E+11 7.21E+11

M0(Log10mass)−7 4.5 4.9 4.9 4.9 4.7 4.9

Eruptive units: Mangatoetoenui (Mgt), Shawcroft (Sw), Oruamatua (Oru), Akurangi (Ak), Combined Lower and Upper Okupata tephras(Okp*) and Upper Pahoka Tephra (U-Pk). HtSE,NE: total column height for the southeast and northeast lobe, respectively. Q: volumedischarge rate; the suffix SE or NE stays for estimated Q based on the southeast or northeast lobe Ht data. MDR: mass discharge rate;the suffix SE or NE stays for estimated MDR based on the southeast or northeast lobe Ht data. M: eruptive magnitude calculated fromthe total mass of the deposit

1182 Bull Volcanol (2012) 74:1161–1185

Conclusions

We quantified five Plinian-style eruptions corresponding tothree contrasting lithofacies associations in the geologicalrecord of Mt. Ruapehu. These represent the most violentevents known for this volcano. Isopach and isopleth mapsindicate that the North Crater was the main active ventduring the Late Pleistocene, responsible for all eruptions,except for the last one (11,620±190 cal years BP), whichmay have its source near or at the South Crater. Eruptedvolumes varied from at least 0.3 to 0.6 km3; column heightsranged between 22 and 37 km, volume discharge ratesfrom ~104–105 m3/s, mass discharge rates from ~107–108 kg/s and estimated magnitudes (M0Log10 depositmass −7) from 4.4 and 4.9. All of these values are character-istic of Plinian eruptions, two to three orders of magnitudelarger than eruptions occurring over the past ~4,500 years (c.f.Donoghue 1991; Donoghue et al. 1995).

Tephra dispersal patterns were complex, the result of highto intermediate Re particles dispersed within the turbulentportion of an ash cloud subject to strong cross-winds andwandering plume effects.

Based on our results, the Oruamatua and Akurangi erup-tions were the strongest produced by Mt. Ruapehu in theLate Pleistocene and represent the worst scenario expectedfor this volcano. These units indicate that the most violenteruptions occurred when porphyritic magma bodies sudden-ly decompressed, involving high erosion of the conduitunder high mass-discharge rates, and producing unsteady,partially collapsing eruptive columns. Dilute and concen-trated pyroclastic density currents represent the greatesthazard down the main proximal catchments, and coarse-grained tephras (lapilli size) can fall over distances ~30 km,reaching the town of Waiouru.

Future mitigation strategies should consider the potentialillustrated in this study for high-intensity eruptions in com-bination with the low frequency of such large events and apopulation that has never experienced such an eruption.

Acknowledgements This study was supported by the New ZealandFoundation for Research Science and Technology Grant MAUX0401,“Living with Volcanic Risk” and the subsequent New Zealand NaturalHazards Research Platform, as well as the Tongariro Natural HistorySociety Memorial Award to NP. We thank H. Keys, J. Johnson (De-partment of Conservation) and the Range control staff of the NZNational Army camp at Waiouru for allowing access to the TongariroNational park and Army land. We also thank Dr. H. Wright (USGS,California), M. Brenna, G. Lube, A. Moebis, K. Németh, V. Neall, E.Phillips, B. Stewart and T. Wang (VRS-Massey University) for theirsupport in the field and discussions; M. Irwin was also extremelyhelpful with GIS-applications and Mr. Doug Hopcroft with SEM.Roberto Sulpizio, an anonymous reviewer, Setsuya Nakada and JamesWhite are gratefully thanked for their valuable comments, discussionand suggestions which significantly improved the quality of thismanuscript.

References

Adams NK, Houghton BF, Hildreth W (2006) Abrupt transitionsduring sustained explosive eruptions: examples from the 1912eruption of Novarupta, Alaska. Bull Volcanol 69:189–206

Arana-Salinas L, Siebe C, Macías JL (2010) Dynamics of the ca.4965 yr 14C BP “Ochre Pumice” Plinian eruption of Popocatépetlvolcano, México. J Volcanol Geotherm Res 192:212–231

Arce JL, Macías JL, Vázquez SL (2003) The 10.5 Ka Plinian eruptionof Nevado de Toluca, México: stratigraphy and hazard implica-tions. Geol Soc Am Bull 115(2):230–248

Arce JL, Cervantes KE, Macías JL, Mora JC (2005) The 12.1 kamiddle Toluca pumice: a dacitic Plinian-subplinian eruption ofNevado de Toluca in Central México:. J Volcanol Geotherm Res147:125–143

Biass S, Bonadonna C (2010) A quantitative uncertainty assessment oferuptive parameters derived from tephra deposits: the example oftwo large eruptions of Cotopaxi volcano, Ecuador. Bull Volcanol73:73–90

Bonadonna C, Ernst GGJ, Sparks RSJ (1998) Thickness variations andvolume estimates of tephra fall deposits: the importance of parti-cle Reynolds number. J Volcanol Geotherm Res 81:173–187

Bursik M (1993) Subplinian eruption mechanisms inferred from vola-tile and clast dispersal data. J Volcanol Geotherm Res 57:47–60

Büttner R, Dellino P, Zimanowski B (1999) Identifying modes ofmagma/water interaction from the surface features of ash par-ticles. Nature 401:688–690

Carey S, Sparks RSJ (1986) Quantitative models of the fall anddispersal of tephra from volcanic eruption columns. Bull Volcanol48:109–125

Carey S, Gardner J, Sigurdsson H (1995) The intensity and magnitudeof Holocene Plinian eruptions from Mount St. Helens volcano. JVolcanol Geotherm Res 66:185–202

Carey RJ, Houghton BF, Thordarsson T (2010) Abrupt shifts betweenwet and dry phases of the 1875 eruption of Askja Volcano:microscopic evidence for macroscopic dynamics. J Volcanol Geo-therm Res 184:256–270

Christenson BW, Wood CP (1993) Evolution of a vent-hosted hydro-thermal system beneath Ruapehu Crater Lake, New Zealand. BullVolcanol 55:547–565

Cioni R, Marianelli P, Stantacroce R, Sbrana A et al (2000) Plinian andsubplinian eruptions. In: Sigurdsson (ed) Encyclopaedia of volca-noes. Academic Press, San Diego, pp 477–494

Cioni R, Sulpizio R, Garruccio N (2003) Variability of the eruptiondynamics during a Subplinian event: greenish pumice eruption ofSomma-Vesuvius (Italy). J Volcanol Geotherm Res 124:89–114

Cole JW (1978) Andesites of Tongariro Volcanic Centre, North Island,New Zealand. J Volcanol Geotherm Res 3:121–153

Cole JW, Nairn IAI (1975) Catalogue of the active volcanoes of theworld 22: New Zealand. Naples, International Association ofVolcanology and Chemistry of the Earth’s Interior, p 156

Cole JW, Graham IJ, Hackett WR, Houghton BF (1986) Volcanologyand petrology of the Quaternary composite volcanoes of Tongar-iro Volcanic Centre, Taupo Volcanic Zone. In: Smith IEM (ed.),‘Late Cenozoic volcanism in New Zealand’. R Soc NZ Bull:222–250

Cronin SJ, Neall VE (1997) A late Quaternary stratigraphic frameworkfor the northeastern Ruapehu and eastern Tongariro ring plains,New Zealand. N Z J Geol Geophys 40:179–191

Cronin SJ, Neall VE, Palmer AS (1996a) Geological history of thenortheastern ring plain of Ruapehu volcano, New Zealand. QuatInt 34(36):21–28

Cronin SJ, Neall VE, Stewart RB, Palmer AS (1996b) A multiple-parameter approach to andesitic tephra correlation, Ruapehu vol-cano, New Zealand. J Volcanol Geotherm Res 72:199–215

Bull Volcanol (2012) 74:1161–1185 1183

Cronin SJ, Hodgson KA, Neall VE, Palmer AS, Lecointre JA (1997)1995 Ruapehu lahars in relation to the late Holocene lahars ofWhangaehu River, New Zealand. N Z J Geol Geophys 40:507–520

Cronin SJ, Hedley MJ, Neall VE, Smith G (1998) Agronomic impactof tephra fallout from 1995 and 1996 Ruapehu volcano eruptions,New Zealand. Environ Geol 34:21–30

Cronin SJ, Neall VE, Lecointre JA, Hedley MJ, Loganathan P (2003)Environmental hazards of fluoride in volcanic ash: a case studyfrom Ruapehu volcano, New Zealand. J Volcanol Geotherm Res121:271–291

Dellino P, Isaia R, La Volpe L, Orsi G (2001) Statistical analysis oftextural data from complex pyroclastic sequences: implicationsfor fragmentation processes of the Agnano-Monte Spina Tephra(4.1 ka), Phlegraean Fields, southern Italy. Bull Volcanol 63:443–461

Donoghue SL (1991) Late Quaternary volcanic stratigraphy of thesoutheastern sector of the Ruapehu ring plain, New Zealand.PhD thesis Massey University, Palmerston North, New Zealand.p 615

Donoghue SL, Neall VE, Palmer AS (1995) Stratigraphy and chronologyof late Quaternary andesitic tephra deposits, Tongariro VolcanicCentre, New Zealand. R Soc N Z 25(2):112–206

Donoghue SL, Palmer AS, McClelland EA, Hobson K, Stewart RB,Neall VE, Lecointre J, Price R (1999) The Taurewa eruptiveepisode: evidence for climactic eruptions at Ruapehu Volcano,New Zealand. Bull Volcanol 60:223–240

Escher BG (1933) On a classification of central eruptions according togas pressure of the magma and viscosity of the lava. LeidscheGeol Meded, Deel VI Afl I:45–48

Fierstein J, Nathenson M (1992) Another look at the calculation offallout tephra volumes. Bull Volcanol 54:156–167

Gamble JA, Price RC, Smith IEM, McIntosh WC, Dunbar NW (2003)40Ar/39Ar geochronology of magmatic activity, magma flux andhazards at Ruapehu volcano, Taupo Volcanic Zone, New Zealand.J Volcanol Geotherm Res 120:271–287

Graham IJ, Hackett WR (1987) Petrology of calc-alkaline lavas fromRuapehu Volcano and related vents, Taupo Volcanic Zone, NewZealand. J Petrol 28:531–567

Graham IJ, Cole JW, Briggs RM, Gamble JA, Smith IEM (1995)Petrology and petrogenesis of volcanic rocks from the Taupovolcanic zone: a review. J Volcanol Geotherm Res 68:59–87

Gurioli L, Houghton BF, Cashman KV, Cioni R (2005) Complexchanges in eruption dynamics during the 79 AD eruption ofVesuvius. Bull Volcanol 67:144–159

Hackett WR (1985) Geology and petrology of Ruapehu volcano andrelated vents. PhD. Thesis, Victoria University of Wellington, p312

Hackett WR, Houghton BF (1989) A facies model for a Quaternaryandesitic composite volcano, Ruapehu, New Zealand. Bull Volcanol51:51–68

Hammer JE, Cashman KV, Hoblitt RP, Newman S (1999) Degassing andmicrolite crystallization during pre-climactic events of the 1991eruption of Mt. Pinatubo, Philippines. Bull Volcanol 60:355–380

Houghton BF, Wilson CJN (1989) Vesicularity index for pyroclasticdeposits. Bull Volcanol 51:451–462

Jeanloz R et al (2000) Mantle of the Earth. In: Sigurdsson (ed) Ency-clopaedia of volcanoes. Academic Press, San Diego, pp 41–54

Johnston DM, Houghton BF, Neall VE, Ronan KR, Paton D (2000)Impacts of the 1945 and 1995–1996 Ruapehu eruptions, NewZealand: an example of increasing societal vulnerability. GSABull 112(5):720–726

Kilgour G, Manville V, Della Pasqua F, Graettinger A, Hodgson KA,Jolly GE (2010) The 25 September 2007 eruption of MountRuapehu, New Zealand: directed ballistics, surtseyan jets, andice-slurry lahars. J Volcanol Geotherm Res 191:1–14

Kratzmann DJ, Carey SN, Fero J, Scasso RA, Naranjo JA (2010)Simulations of tephra dispersal from the 1991 explosive eruptionsof Hudson volcano, Chile. J Volcanol Geotherm Res 190:337–352

Lara L (2009) The 2008 eruption of the Chaitén Volcano, Chile: apreliminary report. Andean Geol 36(1):125–129

Larsen G, Newton AJ, Dugmore AJ, Vilmundardóttir EG (2001) Geo-chemistry, dispersal, volumes and chronology of Holocene silicictephra layers from the Katla volcanic system, Iceland. J Quat Sci16(2):119–132

Lecointre J, Hodgson K, Neall V, Cronin SJ (2004) Lahar-triggeringmechanisms and hazard at Ruapehu Volcano, New Zealand. NatHazard 31:85–109

Lowe DJ, Shane PAR, Alloway BV, Newnham RM (2008) Finger-prints and age models for widespread New Zealand tephra markerbeds erupted since 30,000 years ago: a framework for NZ-INTIMATE. Quat Sci Rev 27:95–126

Lube G, Cronin SJ, Procter J (2009) Explaining the extreme mobilityof volcanic ice-slurry flows, Ruapehu volcano, New Zealand.Geology 37(1):15–18

Macedonio G, Dobran F, Neri A (1994) Erosion processes in volcanicconduits and application to the AD 79 eruption of Vesuvius. EarthPlanet Sci Lett 121(1–2):137–152

Macías JL, Arce JL, Mora JC, Espíndola JM, Saucedo R, Manetti P(2003) A 550-year-old Plinian eruption at El Chichón Volcano,Chiapas, Mexico: explosive volcanism linked to reheating of themagma reservoir. J Geophys Res 108(B12):2569

McClelland E, Erwin PS (2003) Was a dacite dome implicated in the9,500 B.P. collapse of Mt Ruapehu? A palaeomagnetic investiga-tion. Bull Volcanol 65:294–305

Nairn IA, Kobayashi T, Nakagawa M (1998) The 10 ka multiple ventpyroclastic eruption sequence at Tongariro Volcanic Centre,Taupo volcanic zone, New Zealand: part 1. Eruptive processesduring regional extension. J Volcanol Geotherm Res 86(1–4):19–44

Neall VE (1990) Natural Hazard assessment in New Zealand. ResearchSchool of Earth Sciences Victoria University of Wellington. NatHazard 90:24–28

Neall VE, Cronin SJ, Donoghue SL, Hodgson KA, Lecointre JA,Purves AM (1995) The Potential volcanic threat at Ruapehu.New Zealand Ministry of Civil Defense. Tephra 14(2):18–21

Neall VE, Houghton BF, Cronin SJ, Donoghue SL, Hodgson KA,Johnston DM, Lecointre JA, Mitchell AR (1999) Volcanic hazardsat Ruapehu volcano. Ministry of Civil Defence, Wellington, p 30,Volcanic hazards information series, no. 8

Neall VE, Cronin SJ, Donoghue SL, Hodgson KA, Lecointre JA,Palmer AS, Purves AM, Stewart RB (2001) Lahar hazards mapfor Ruapehu volcano. Institute of Natural Resources–MasseyUniversity, Soil and Earth Sciences Occasional Publication No. 1

Newnham RM, Lowe DJ (2000) Fine-resolution pollen record of lateglacial climate reversal from New Zealand. Geology 28:759–762

Palmer BA, Neall VE (1989) The Murimotu Formation, 9500 year olddeposits of a debris avalanche and associated lahars, MountRuapehu, New Zealand. NZ J Geol Geophys 32:477–486

Papale P, Dobran F (1993) Modelling of the ascent of magma duringthe Plinian eruption of Vesuvius in A.D. 79. J Volcanol GeothermRes 58:101–132

Pardo N, Cronin SJ, Palmer A, Németh K (2011) Reconstructing thelargest explosive eruptions of Mt. Ruapehu, New Zealand: lithos-tratigraphic tools to understand subplinian-Plinian eruptions atandesitic volcanoes. Bull Volcanol: 1–24. doi:10.1007/s00445-011-0555-z

Procter JN, Cronin SJ, Zernack AV (2010) Landscape and sedimentaryresponse to catastrophic debris avalanches, Western Taranaki,New Zealand. Sediment Geol 38(1):67–70

Pyle DM (1989) The thickness, volume and grainsize of tephra falldeposits. Bull Volcanol 51:1–15

1184 Bull Volcanol (2012) 74:1161–1185

Pyle DM (2000) Sizes of volcanic eruptions. In: Sigurdsson H et al.(eds.) Encyclopaedia of volcanoes. London, Academic Press,263–269

Reyners M, Eberhart-Phillips D, Graham S, Yuichi N (2006) Imagingsubduction from the trench to 300 km depth beneath the centralNorth Island, New Zealand, with Vp and Vp/Vs. Geophys J Int165:565–583

Rosi M (1998) Plinian eruption columns: particle transport and fallout.In: Freundt A, Rosi M (eds) From magma to tephra. Elsevier,Amsterdam, pp 139–172

Rosi M, Landi P, Polacci M, Di Muro A, Zandomeneghi D (2004) Roleof conduit shear on ascent of the crystal-rich magma feeding the800-year-BP Plinian eruption of Quilotoa volcano (Ecuador). BullVolcanol 66:307–321

Sable JE, Houghton BF, Wilson CJN, Carey RJ (2006) Complexproximal sedimentation from Plinian plumes: the example ofTarawera 1886. Bull Volcanol 69:89–103

Saucedo R, Macías JL, Gavilanes JC, Arce JL, Komorowski JC,Gardner JE, Valdez-Moreno G (2010) Eyewitness, stratigraphy,chemistry, and eruptive dynamics of the 1913 Plinian eruption ofVolcán de Colima, México. J Volcanol Geotherm Res 191:149–166

Sieh K, Bursik M (1986) Most recent eruption of the Mono Craters,eastern central California. J Geophys Res 91:12539–12571

Sparks RSJ (1986) The dimension and dynamics of volcanic eruptioncolumns. Bull Volcanol 48:13–15

Sparks RSJ, Bursik M, Carey SN, Gilbert JS, Glaze LS, Sigurdsson H,Woods AW (1997) Volcanic plumes. Wiley, Chichester, p 574

Statistics New Zealand (2007) Regional gross domestic product http://www.stats.govt.nz/reports/analytical-reports/regional-gross-domestic-product.aspx. Retrieved 18 February 2010)

Statistics New Zealand (2009) Subnational population estimates: at 30June 2009 http://www.stats.govt.nz/methods_and_services/access-data/tables/subnational-pop-estimates.aspx. Retrieved 2009-10-23

Sulpizio R (2005) Three empirical methods for the calculation of distalvolume of tephra-fall deposits. J Volcanol Geotherm Res 145(3–4):315–333

Sulpizio R,Mele D, Dellino P, La Volpe L (2005) A complex, subplinian-type eruption from low-viscosity, phonolitic to tephri-phonoliticmagma: the AD 472 (Pollena) eruption of Somma-Vesuvius, Italy.Bull Volcanol 76:743–767

The Treasury (2010) New Zealand Economic and Financial Overview2010. 60P. ISSN: 1173-2334. http://www.treasury.govt.nz/economy/overview/2010

Thorarinsson S, Sigvaldason GE (1972) The Hekla eruption of 1970.Bull Volcanol 36:1–20

Thouret JC, Juvigné E, Gourgaud A, Boivin P, Dávila J (2002) Recon-struction of the AD Huaynaputina eruption based on the correla-tion of geologic evidence with early Spanish Chronicles. JVolcanol Geotherm Res 115:529–570

Topping WW (1973) Tephrostratigraphy and chronology of late Qua-ternary eruptives from the Tongariro Volcanic Centre, New Zea-land. N Z J Geol Geophys 16:397–423

Topping WW (1974) Some aspects of quaternary history of Tongarirovolcanic centre. PhD thesis, Victoria University of Wellington,New Zealand, p 347

Turner R, Hurst T (2001) Factors influencing volcanic ash dispersalfrom the 1995 and 1996 Eruptions of Mount Ruapehu, NewZealand. Appl Meteor 40:56–69

Varekamp JC (1993) Some remarks on volcanic vent evolution duringPlinian eruptions. J Volcanol Geotherm Res 54:309–318

Villamor P, Van Dissen R, Alloway BV, Palmer AS, Litchfield N(2010) The Rangipo fault, Taupo rift, New Zealand: an exampleof temporal slip-rate and single-event displacement variability in avolcanic environment. GSA Bull 119:529–547

Walker GPL (1973) Explosive volcanic eruptions—a new classifica-tion scheme. Geologiche Rundshau 62:431–446

Walker GPL (1980) The Taupo pumice: product of the most powerfulknown (ultraplinian) eruption? J Volcanol Geotherm Res 8:69–94

Walker GPL, Croasdale R (1971) Two Plinian-type eruptions in theAzores. J Geol Soc Lond 127:17–55

Watt SFL, Pyle DM, Tamsin AM, Martin RS, Matthews NE (2009)Fallout and distribution of volcanic ash over Argentina followingthe May 2008 explosive eruption of Chaitén, Chile. J GeophysRes 114:B04207. doi:10.1029/2008JB006219

Wilson L (1976) Explosive volcanic eruptions III, Plinian eruptioncolumns. Geophys J R Astron Soc 45:543–556

Wilson L, Sparks RSJ, Walker GPL (1980) Explosive volcaniceruptions-IV. The control of magma properties and conduit ge-ometry on eruption column behaviour. Geophys J R Astron Soc63:117–148

Wilson CJN, Houghton BF, McWilliams MO, Lanphere MA, WeaverSD, Briggs RM (1995) Volcanic and structural evolution of Taupovolcanic zone, New Zealand: a review. J Volcanol Geotherm Res68:1–28

Wright HMN, Cashman KV, Rosi M, Cioni R (2007) Breadcrustbombs as indicators of Vulcanian eruption dynamics at GuaguaPichincha volcano, Ecuador. Bull Volcanol 69:281–300

Bull Volcanol (2012) 74:1161–1185 1185