UBC Faculty Research and Publications

Temporal Geochemical Variations in Lavas from Kilauea's Pu'u ‘O'o Eruption (1983-2010): Cyclic Variations… Weis, Dominique Nov 15, 2013

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52383-Weis_AGU_2013_10.1002ggge.20285.pdf [ 4.03MB ]
52383-ggge20285-sup-0001-suppinfo01.pdf [ 87.2kB ]
52383-ggge20285-sup-0002-suppinfo02.pdf [ 129.53kB ]
52383-ggge20285-sup-0003-suppinfo03.pdf [ 87.07kB ]
52383-ggge20285-sup-0004-suppinfo01.xlsx [ 110.13kB ]
52383-ggge20285-sup-0005-suppinfo02.xls [ 173kB ]
52383-ggge20285-sup-0006-suppinfo03.xlsx [ 16.94kB ]
Metadata
JSON: 52383-1.0075983.json
JSON-LD: 52383-1.0075983-ld.json
RDF/XML (Pretty): 52383-1.0075983-rdf.xml
RDF/JSON: 52383-1.0075983-rdf.json
Turtle: 52383-1.0075983-turtle.txt
N-Triples: 52383-1.0075983-rdf-ntriples.txt
Original Record: 52383-1.0075983-source.json
Full Text
52383-1.0075983-fulltext.txt
Citation
52383-1.0075983.ris

Full Text

Temporal geochemical variations in lavas fromK?lauea?s Pu?u ?O?o eruption (1983?2010): Cyclicvariations from melting of source heterogeneitiesAndrew R. GreeneDepartment of Natural Sciences, Hawai?i Pacific University, 45-045 Kamehameha Hwy, Kane?ohe, Hawaii,96744, USA (agreene@hpu.edu)Michael O. GarciaDepartment of Geology and Geophysics, University of Hawai?i, Honolulu, Hawaii, USAAaron J. PietruszkaDepartment of Geological Sciences, San Diego State University, San Diego, California, USANow at U. S. Geological Survey, Denver Federal Center, Denver, Colorado, USADominique WeisPacific Centre for Isotopic and Geochemical Research, Department of Earth and Ocean Sciences, University ofBritish Columbia, Vancouver, British Columbia, CanadaJared P. MarskeDepartment of Terrestrial Magnetism, Carnegie Institution, Washington, D.C., USAMichael J. VollingerRonald B. Gilmore XRF Lab, Department of Geosciences, University of Massachusetts, Amherst, Massachusetts, USAJohn EilerPlanetary and Geological Sciences Institute, California Institute of Technology, Pasadena, California, USA[1] Geochemical time series analysis of lavas from K?lauea?s ongoing Pu?u ?O?o eruption chroniclemantle and crustal processes during a single, prolonged (1983 to present) magmatic event, which hasshown nearly two-fold variation in lava effusion rates. Here we present an update of our ongoingmonitoring of the geochemical variations of Pu?u ?O?o lavas for the entire eruption through 2010. Oxygenisotope measurements on Pu?u ?O?o lavas show a remarkable range (18O values of 4.6?5.6%), which areinterpreted to reflect moderate levels of oxygen isotope exchange with or crustal contamination byhydrothermally altered K?lauea lavas, probably in the shallow reservoir under the Pu?u ?O?o vent. Thisprocess has not measurably affected ratios of radiogenic isotope or incompatible trace elements, whichare thought to vary due to mantle-derived changes in the composition of the parental magma delivered tothe volcano. High-precision Pb and Sr isotopic measurements were performed on lavas erupted at 6month intervals since 1983 to provide insights about melting dynamics and the compositional structure ofthe Hawaiian plume. The new results show systematic variations of Pb and Sr isotope ratios thatcontinued the long-term compositional trend for K?lauea until 1990. Afterward, Pb isotope ratios showtwo cycles with 10 year periods, whereas the Sr isotope ratios continued to increase until 2003 andthen shifted toward slightly less radiogenic values. The short-term periodicity of Pb isotope ratios mayreflect melt extraction from mantle with a fine-scale pattern of repeating source heterogeneities or strands,which are about 1?3 km in diameter. Over the last 30 years, Pu?u ?O?o lavas show 15% and 25% of the? 2013. American Geophysical Union. All Rights Reserved. 4849ArticleVolume 14, Number 1115 November 2013doi: 10.1002/ggge.20285ISSN: 1525-2027known isotopic variation for K?lauea and Mauna Kea, respectively. This observation illustrates that thedominant time scale of mantle-derived compositional variation for Hawaiian lavas is years to decades.Components: 13,235 words, 12 figures, 2 tables.Keywords: Hawaiian plume; tholeiitic volcanism; melt extraction; oceanic island.Index Terms: 3619 Magma genesis and partial melting: Mineralogy and Petrology; 3621 Mantle processes: Mineralogyand Petrology; 1025 Composition of the mantle: Geochemistry; 1037 Magma genesis and partial melting: Geochemistry;1038 Mantle processes: Geochemistry.Received 5 March 2013; Revised 9 September 2013; Accepted 7 October 2013; Published 15 November 2013.Greene, A. R., M. O. Garcia, A. J. Pietruszka, D. Weis, J. P. Marske, M. J. Vollinger, and J. Eiler (2013), Temporal geo-chemical variations in lavas from K?lauea?s Pu?u ?O?o eruption (1983?2010): Cyclic variations from melting of source het-erogeneities, Geochem. Geophys. Geosyst., 14, 4849?4873, doi:10.1002/ggge.20285.1. Introduction[2] K?lauea, on the Island of Hawai?i (Figure 1), isone of the most active and best-monitored volca-noes in the world [Heliker and Mattox, 2003;Wolfe et al., 1987]. The ongoing Pu?u ?O?o erup-tion on K?lauea?s East Rift Zone (Figure 1) hasbeen active nearly continuously for 30 years and isHawai?i?s longest and most voluminous (4 km3)historical eruption [Poland et al., 2012]. The con-tinuous petrologic and geochemical monitoring ofthe Pu?u ?O?o eruption [e.g., Garcia et al., 2000;Marske et al., 2008; Thornber, 2003] has allowedus to witness the dynamic changes in the meltingprocess and mantle source composition during asingle, long-lasting magmatic event. Extractionand transport of melt through open channels dur-ing the Pu?u ?O?o eruption has efficiently transmit-ted variations of melting in the heterogeneoussource to lavas erupted at the surface without sig-nificant pooling and homogenization, preservingshort-term isotopic and geochemical variations[Pietruszka et al., 2006].[3] The long-term geochemical variations (manythousands of years) of Hawaiian and other oceanisland basalts has been well documented due todetailed geochemical work on 3? km deep drillcore [e.g., Albare`de et al., 1997; Blichert-Toftet al., 2003; Bryce et al., 2005; Caroff et al.,1995; Rhodes et al., 2012]. These studieschronicle processes on millennium time scales butmiss potential short-term variations (<100 years),which may provide better insights into meltingand crustal processes. K?lauea?s historical (1823?1982) and prehistoric (AD 900?1400) summitlavas reveal rapid and systematic changes in Pb,Sr, Nd, O, and U-series isotope ratios on a timescale of decades to centuries [Garcia et al., 2003,2008; Marske et al., 2007; Pietruszka and Garcia,1999; Pietruszka et al., 2001]. The Pu?u ?O?oeruption (sampled from hourly to monthly) showscompositional change over hours (in rare cases)for major elements to a few years for isotope ratios[Garcia et al., 2000; Marske et al., 2008]. Thelong duration and vigorous activity (0.35  106m3 of lava erupted daily) of Pu?u ?O?o [e.g., Suttonet al., 2003] provides a rare opportunity to lookbeyond the shallow-level crustal processes associ-ated with the short eruptions (days to weeks) thattypify many active basaltic volcanoes (e.g., MaunaLoa, Etna, Piton de la Fournaise, Karthala,Grimsv?tn) and into the mantle. In addition, Pu?u?O?o magmas may partially bypass K?lauea?s sum-mit reservoir (2?6 km depth beneath the summitcaldera) on their way to the East Rift Zone, andmostly avoid its buffering effects [Garcia et al.,2000]. Therefore the Pu?u ?O?o eruption is one ofEarth?s best probes for sampling mantle-derivedmelts almost continuously over nearly threedecades.[4] The study of isotopic and geochemical varia-tion in magmatic events over short time scales(months to years) in oceanic island lavas improvesour temporal and spatial resolution of meltingprocesses and the chemical structure of mantleplumes [Abouchami et al., 2000; Eisele et al.,2003; Hofmann et al., 1984; Vlastelic et al.,2005]. Recent studies of Pb, Sr, and Nd isotoperatios for part of the Pu?u ?O?o eruption [Marskeet al., 2008] and other active basaltic volcanoes[e.g., Piton de la Fournaise; Vlastelic et al., 2005]detected rapid and systematic changes over shortGREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854850time scales (years to decades) resulting fromsource heterogeneity, and variations in crustalprocesses. Here we present new high-precision Pb,Sr, and O isotope ratios, and major- and trace-element abundances for Pu?u ?O?o lavas eruptedbetween January 1983 and June 2010. These data197418231919-20 1971Mauna Ulu   1969-74197719551955196018401790East Rift ZoneSummitMakaopuhi1983-present1790179010 km18681955Pu?u ?O?oNCraterCraterEp. 54Southwest Rift ZoneMauna Loa1983-2010 Pu'u '?'? KupaianahaPACIFIC OCEANNorth Pu'u '?'?N?pauCraterKalapanaEp. 56 February 17, 1992 - Feb. 11, 2010Episodes 50-58July 20, 1986 -  February 17, 1992Episodes 48-49January 3, 1983 -July 20, 1986Episodes 1-48episodicfountaining(mostly centralvent)gentle effusion(lava shield andtube-fed pahoehoe)Kupaianaha Pu?u ?O?oPu?u ?O?oJan. July Feb.Feb.1983 1986 19921997(3.5 years) (5.5 years) (18 years) perched channels,rootless shields,fissure eruption20102007Episode 56(Magma supply ratedoubled)June(uprift)flank  vent eruptions(nearly continuous)(lava shield and tube-fed pahoehoe)2003Episode 54(uprift)010km016miK?lauea Caldera1790-1982kmHalemaumauPu?u ???? East Rift ZoneMauna Ulu1969-1974Makaopuhi N?pau(a)KupaianahaEp. 54Ep. 56(b)123storeddikeJune, 2007?Jan.1997Ep. 58Ep. 58K?lauea Caldera0KohalaMauna Loa K?lauea Mauna KeaL??ihi 50010001500200025003000200015005001000100020003000250050002500 50005500155?156?19?20?Hilo RidgeEast rift zone40 km20Hawai?i Hual?lai(c)(d)Figure 1. Map of flow fields from the Pu?u ?O?o-Kupaianaha eruption on the East Rift Zone of K?lauea Vol-cano from 1983 to 2010 and historical flows, with a timeline summarizing the predominant style of eruptiveactivity. (a) A schematic cross section of summit and East Rift Zone shows the proposed magmatic plumbingsystem for K?lauea Volcano, with locations for episodes 54 and 56 uprift of Pu?u ?O?o. Mantle-derived magmafor this eruption is thought to partially bypass the summit reservoir based on the rapid changes in lava compo-sition [Garcia et al., 1996]. (b) Map of K?lauea East Rift Zone with flow fields from intervals of the Pu?u?O?o eruption. Legend shows episodes in each interval of eruptive activity. Map provided by USGS HawaiianVolcano Observatory. (c) Map of the island of Hawai?i with area of map in Figure 1b indicated with box. (d)Timeline of the Pu?u ?O?o eruption. Episode 54 was a fissure eruption in and downrift of Napau Crater thatoccurred over 23 h in January 1997, following the collapse of the Pu?u ?O?o cone. Episode 56 was a brief (<1day) fissure eruption northeast of Makaopuhi Crater (uprift of Pu?u ?O?o) that occurred in June 2007, coincid-ing with an intrusion and collapse of Pu?u ?O?o crater floor. Dashed lines between 2003 and 2007 indicate pe-riod when magma supply rate nearly doubled [Poland et al., 2012].GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854851are combined with previously published high-precision isotope and trace-element data from1998 to 2005 Pu?u ?O?o lavas [Marske et al.,2008]. The new Pb and Sr isotope and inductivelycoupled plasma mass spectrometry (ICP-MS) dataprovide a record of isotopic and geochemical vari-ation of Pu?u ?O?o lavas at 6 month intervals,whereas X-Ray fluorescence (XRF) data was col-lected at 2 week intervals. This time series anal-ysis of Pu?u ?O?o lavas allows us to distinguish thechanging roles of mantle and crustal processes ingreat detail. The new Pb and Sr isotope ratios areused to assess the short-term expression of mantlesource components throughout the course of theeruption and to evaluate the effects on lava com-position of recent doubling of the magma supply[2003?2007; Poland et al., 2012]. These resultsare compared to the longer-term variations forK?lauea and other Hawaiian shield volcanoes.2. Geologic Background of K?laueaVolcano and the Pu?u ?O?o Eruption(1983?2010)[5] K?lauea Volcano is currently in the middle ofits shield-building stage [DePaolo and Stolper,1996], erupting tholeiitic lava at a rate of 0.13km3/yr [Sutton et al., 2003], one of the highestrates of any volcano on Earth. K?lauea rises 1240m above sea-level on the southern flank of itslarger neighbor, Mauna Loa (4168 m; Figure 1).Geochemical evidence favors a deep mantle plumeorigin for Hawaiian magmas [e.g., Kurz et al.,1982; Weis et al., 2011]. Shield stage magmas arethought to originate from partial melting at mantledepths of 70?120 km within the upper Hawaiianplume [Watson and McKenzie, 1991]. Magmas areextracted from the upwelling mantle within themelting region and transported through chemicallyisolated channels towards the surface [Pietruszkaet al., 2006]. These pooled melts ascend throughthe lithosphere via a primary conduit into a shal-low (2?6 km) magmatic complex within K?lauea[Eaton and Murata, 1960; Ryan, 1987; Tillingand Dvorak, 1993; Wright, 1971]. K?lauea erup-tions occur in and around its summit caldera andEast and Southwest Rift Zones. Approximately90% of the subaerial surface of K?lauea Volcano iscovered with tholeiitic lava less than 1100 yearsold [Holcomb, 1987]. Prior to 1955, historical(post-1820) eruptions on K?lauea occurred mostlyat or near the summit [Macdonald et al., 1983].Subsequently, rift zone eruptions became morecommon, especially along the East Rift Zone,including the 1969?1974 Mauna Ulu eruption, themost voluminous historical eruption prior to Pu?u?O?o [Macdonald et al., 1983].[6] The Pu?u ?O?o-Kupaianaha eruption (referredto as the Pu?u ?O?o eruption throughout this paper)began on 2 January 1983 with the intrusion of adike within K?lauea?s East Rift Zone, although itwas preceded by months of intrusions from thesummit into the rift zone [Wolfe et al., 1987]. Itwas followed 24 h later by eruptive activity alonga discontinuous 7 km long fissure, which localizedto a central vent, Pu?u ?O?o (Figure 1 and Table 1).The eruption can be categorized into three broadphases based on eruptive style and location: (1)1983?1986: brief (mostly less than 24 h), episodiceruptions (24 day average repose between erup-tions) with fountaining up to 400 m, mainly fromthe Pu?u ?O?o vent [Heliker and Mattox, 2003]; (2)1986?1992: nearly continuous effusion from theKupaianaha vent, which was considered to have ashallow (<100 m deep) conduit connection withPu?u ?O?o, 3 km uprift [Garcia et al., 1996]; and(3) 1992?2010: nearly continuous effusion mostlyfrom vents within, and on the southwest and eastflanks of Pu?u ?O?o, and from rootless shields 2km east of Pu?u ?O?o [Poland et al., 2008]. Thispattern was interrupted on 29 January 1997 (epi-sode 54) by the 150 m collapse of the crater floorinside the Pu?u ?O?o cone, and propagation oferuptive fissures 4 km uprift (west) of Pu?u ?O?o,which were active for less than a day [Heliker andMattox, 2003]. This event was followed by a 6week hiatus in effusive activity, although glowreturned to the Pu?u ?O?o vent on 24 February1997 (Table 1). Afterward, and until June 2007,lava erupted nearly continuously from flank ventson Pu?u ?O?o (episode 55). On 19 June 2007, adike intrusion in the upper East Rift Zone resultedin a brief, small (1500 m3) eruption (episode 56)6 km uprift from Pu?u ?O?o [Montgomery-Brownet al., 2010], which was followed by a 2 week hia-tus in effusion [Poland et al., 2008]. Lava produc-tion resumed for 3 weeks in and around Pu?u ?O?ocone (episode 57) until 21 July 2007, when a fis-sure opened on the east flank of Pu?u ?O?o andpropagated eastward towards Kupaianaha (Figure1 and Table 1). This marked the beginning of epi-sode 58 [Poland et al., 2008], which continuedthrough the end of 2010 mostly as tube-fed flowsfrom a vent 2 km east of Pu?u ?O?o. The othernotable K?lauea eruptive activity during the Pu?u?O?o eruption is an ongoing summit eruption thatstarted in March 2008 [Johnson et al., 2010].GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854852Table1.SummaryofthePu?u? O? oEruptionaPrimaryVentEpisodeEpisodeStartDateReposeLength(Days)EpisodeLengthVolume(106xm3)EruptionRate(103m3/day)Vent/LocationBriefDescriptionPu?u? O? o13Jan1983Start20days14?Fissure1;activitylocalizedatPu?uHaluluandPu?uKahauleaeastofPu?u? O? oInitialfissureopenedinNapauCraterafterseismicswarmpropagateddownERZ;fissuresextended8km;fissureslocalizedto1kmnearPu?uKahaulea;fountainsfromPu?uHalulubuilta60m-highconePu?u? O? o2?4710Feb19838?65(betweenepisodes)~3.8years371300MostlyPu?u? O? o;Episodes2?3localizedatPu?uHaluluandPu?uKahauleaeastofPu?u? O? o;Epi-sode4?47PuuOoprimaryventEpisodicfirefountaining;episodesmostly<24hlongsepa-ratedbyanreposelengthaverageof24days;effusionratesincreasedthoughepisode39;maximumlavafoun-tainof470mhigh;firstyearchangedfromlowfountainsandpahoehoeriverstohighfountainsand?a?afans;fountain-fed?a?abyepisode20;conebuilt255mhighand1.4kmindiameter;summitinflatedbetweenfoun-tainingepisodesanddeflatedduringepisodesKupaianaha4818July198624~5.5years500400?0.5Kupaianaha;fissure3kmeastofPu?u? O? oFissuresfirstopenedatthebaseofPu?u? O? oand22hlateropened3kmdownriftataventtobenamedKupaianaha;5.5yearsofnearlycontinuousgentleeffusion;largelavapondformedovervent(140m300m);broadlavashieldformedandtube-fedpahoehoewascommonwaylavaspreadtocoast;homesdestroyedintownofKala-pana;lavatubestoseamid-1987to1989;lavaenteredseaduring68%ofepisode;lavaactiveinPu?u? O? ocraterduringmostofepisodePu?u? O? o498Nov1991None18days110.6Fissure2betweenPu?u? O? oandKupaianahaFissuresopenedonPu?u? O? oandpropagatedtoKupaia-naha;outputwanedduringepisode;gentleeffusion,lavashieldandtube-fedpahoehoe;fissurevents,pahoehoePu?u? O? o5017Feb19921115days3?Pu?u? O? oflank;radialfissureonwestflankofPu?u? O? oconeEruptionreturnedtoPu?u? O? o;radialfissuresonflankofcone;flankventeruptions;lavashieldbankedupagainstthesouthandwestPu?u? O? ocone;spatterconesformedovervents;mostlytube-fedpahoehoe;continuousquieteffusionPu?u? O? o517Mar19924161days32300Pu?u? O? oflankFlankventeruptions;mostlytube-fedpahoehoetothesea;continuousquieteffusion;lavashieldbankedupagainstthesouthandwestPu?u? O? oconePu?u? O? o523Oct1992None15days2300Pu?u? O? oflankFlankventeruptions;mostlytube-fedpahoehoetothesea;continuousquieteffusion;lavashieldbankedupagainstthesouthandwestPu?u? O? oconePu?u? O? o5320Feb1993None~4years535300Pu?u? O? oflankFlankventeruptions;mostlytube-fedpahoehoetothesea;continuousquieteffusion;lavashieldbankedupagainstthesouthandwestPu?u? O? ocone;collapsepitsformedonthesideofPu?u? O? oPu?u? O? o5429Jan1997None1day0.30.3Fissure3;2?4kmupriftofPu?u? O? o(1)LavalakeinsidethePu?u? O? oventdrainedandcraterfloordropped150m;(2)Pu?u? O? owestflankcollapsed;115mgapinwestsideofPu?u? O? o;(3)fissure4kmeruptedupriftfor1day,inanddownriftofNapauCrater,followedbylongesteruptivehiatussince1987(24days);distinctlavachemistryinvolvedmagmamixingwithdifferentiatedmagmastoredinriftzoneGREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854853Table1.(continued)PrimaryVentEpisodeEpisodeStartDateReposeLength(Days)EpisodeLengthVolume(106xm3)EruptionRate(103m3/day)Vent/LocationBriefDescriptionPu?u? O? o5524Feb19972410years265200?500Pu?u? O? oanditsflankLavaspilledfromcratertoformnewpond;lavaspilledfromcrateramonthlater;newflankventeruptionswestandsouthwestofcone;spatterconesonflankscrustedovertoproducemostlytube-fedpahoehoe;byJune1997lavaovertoppedthegapinwallofPu?u? O? oandflowedfromcraterforfirsttimein11years;flankventsunder-minedPu?u? O? oconeinDecember,1997;PukaNuicol-lapsepitformedonsouthwestflankofcone;31pausesoccurredduringepisode55Pu?u? O? o5619June2007None6h0.001450.00036250mlongfissureintheforestnortheastofKaneNuioHamo,approximately6kmwestofPu?u? O? oFather?sDayeruptionnearKaneNuioHamonorthofMakaopuhiCrater;magmasupplytoPu?u? O? owascutoffon17June2007;earthquakeswarmsindicatedmagmamovementintheupperERZ;spattereruptedfromfissureinforestedarea;smalllavaflow(200m50m)accom-paniedintrusioninERZ;craterfloorinPu?u? O? ocol-lapsedanderuptionshutoffPu?u? O? o571July200719daysNone0.82?1.2365Pu?u? O? ocraterAfterabouttwo-weeksofquiet,theeruptionbeganagainon1July.Lavabegantorefillthecrater.On8July,effu-sionwanedasthecraterbegantoupliftinapiston-likefashion.Thecraterthenbegantofillandreachedtowithin30moftheeasternrimofPu?u? O? ocraterbymid-July.Pu?u? O? o5821July2007None~4yearsended7March2011320(asoftheendof2009)FromfissureeastofPu?u? O? ocraterPerchedlavachannel,rootlessshields;forthefirsttimesince7February1992,lavabeginseruptingeastofPu?u? O? ocrater.ThanksgivingEvebreakout,lavabypasses21July2007channelanderuptsonchannelflank;5March2008oceanentryactiveforthefirsttimesinceJune2007;explosioninHalema?uma?uCrateratsummiton19March2008;June2008spatteringventsandasmallpondoflavainPu?u? O? o,lavafountainsgushfromtheTEBtubesystem,channelized?a? aflowsinRoyalGardens,andlargelittoralexplosionsatK ?lauea?soceanentrynearKalapana;Waikupanahaoceanentryactivethroughmuchof2009,andoccasionallyKupapa?uoceanentrytowesta Reposelengthreferstodurationofpausebetweeneruptiveepisodes.Episodeidentifiesoccurrencesoffountainingorlavaflowseparatedbyquiescentperiods.Volumeisdenserockequivalent(DRE)eruptedduringeachepisode.Datasources:Garciaetal.[2000]andreferencestherein,Wolfeetal.[1998],HelikerandMattox[2003].GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.2028548543. Description of Samples andAnalyses Performed in This Study[7] This study presents 52 new high-precision Pband Sr isotope analyses (from 1983 to 1997 and2006 to 2010), 11 new O isotope analyses (fromafter 1997; Table 2), and 13 new ICP-MS trace-element analyses of Pu?u ?O?o lava samples(mostly after 2005; data and analytical methodsare presented in the supporting information).1 NewXRF major- and trace-element analyses for 52Pu?u ?O?o lavas erupted from 2006 to 2010 arealso presented. In addition, new XRF trace-element analyses are given for samples eruptedprior to 1998, when a new, more precise XRFinstrument became available. Almost all of thesamples in this study were collected in a moltenstate and quenched with water to minimize poster-uption crystallization. The sample names are thedate that each lava sample was collected (e.g.,day-month-year), which is generally within a dayof its eruption when lava is flowing in open chan-nels on the surface or in lava tubes [e.g., Garciaet al., 2000] or up to a week or more when it isoozing within slowly advancing pahoehoe flows[K. Ho, personal communication, 2013]. Descrip-tions of the petrography of typical Pu?u ?O?o lavascan be found in Garcia et al. [1989, 1992, 1996,2000] and Marske et al. [2008]. Fourteen high-precision Pb and Sr isotope ratios for Pu?u ?O?olavas erupted from 1998 to 2005 from Marskeet al. [2008] and 15 O isotope analyses from Gar-cia et al. [1998] are listed in Table 2 and areincluded in plots for completeness.4. Temporal Geochemical Variationsin Lavas From 1983 to 2010[9] Early Pu?u ?O?o lavas (1983 to early 1985) re-cord rapid (hours to days) variations in major andcompatible trace-element abundances (Figure 2;Supporting information). These lavas show petro-graphic evidence for both crystal fractionation andmagma mixing [Wolfe et al., 1987; Garcia et al.,1992]. Crystal fractionation of olivine (with minorclinopyroxene and plagioclase, especially for 1983lavas) is the dominant process controlling short-term major-element variation in Pu?u ?O?o lavas[Garcia et al., 1992]. To remove the effects ofcrystal fractionation on parental magma composi-tions, major-element abundances of lavas contain-ing only olivine (MgO >7.0 wt %) werenormalized to 10 wt % MgO by the addition ofequilibrium composition olivine (98.5%) and spi-nel (1.5%) in 0.5 mol % steps, as described byGarcia et al. [2003]. The increases in MgO, CaO/Al2O3, and CaO/TiO2 and decreases in MgO-normalized incompatible element abundances(e.g., TiO2, K2O) between 1983 and early 1985(Figure 2) reflect mixing of new high-MgOmagma with decreasing amounts of a hybridmagma formed at the start of the eruption by mix-ing two differentiated, rift-zone stored magmas[Garcia et al., 1989, 1992]. Lavas erupted afterearly 1985 show no petrographic or geochemicalevidence for mixing [Shamberger and Garcia,2007] until the 1997 uprift eruption, which is dis-cussed below.[10] From 1985 to 1994, Pu?u ?O?o lavas show awide range in MgO reflecting the periodic hiatusesin eruptive activity [Garcia et al., 1992], and grad-ual changes in MgO-normalized concentrations ofmajor elements (TiO2 and K2O), and ratios ofmajor (e.g., CaO/Al2O3; Figure 2) and trace ele-ments (Nb/Y; Figure 3). In 1994, lavas began aperiod of increasing MgO-normalized SiO2 andoverall decreasing MgO-normalized TiO2 that per-sisted until 2001. Other geochemical parameterscontinued their long-term trends (e.g., decreasingCaO/Al2O3, MgO-normalized K2O, and Nb/Y,and increasing Zr/Nb; Figures 2 and 3). Starting inmid- to late 2003, there was an increase in lavaproduction with effusion rates doubling in 2005[Poland et al., 2012]. The lava MgO contentdecreased from 2003 to 2007 and was relativelylow (<7.5 wt %, mostly <7.2 wt %) with limitedvariation (Figure 2). This decrease continued theoverall trend of decreasing MgO that started in1998, as noted by Poland et al. [2012]. There isalso a decrease in MgO-normalized SiO2 and anincrease in MgO-normalized TiO2 and K2O during2003?2007 (Figures 2 and 3). Lava MgO increasedfrom 2008 to 2009 as did CaO/TiO2 and values ofMgO-normalized SiO2, although MgO and SiO2values dropped afterwards for the most recentlyerupted samples that were analyzed in this study(Figure 2). For more on major- and trace-elementvariations in 1983?2005 Pu?u ?O?o lavas, see Gar-cia et al. [1989, 1992, 1996, 2000], Marske et al.[2008], and Thornber [2003].[11] The brief eruptive outbreaks uprift of the Pu?u?O?o vent in 1997 (3 km uprift for episode 54)and 2007 (6 km uprift for episode 56; Figure 1)1Additional supporting information may be found in the onlineversion of this article.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854855Table 2. Pb, Sr and O Isotopic Geochemistry of Pu?u ?O?o Lavas from 1983?2010aSample 206Pb/204Pb 2 207Pb/204Pb 2 208Pb/204Pb 2 87Sr/86Sr 2 18O 123 Jan 1983  18.5247 0.0009 15.4800 0.0008 38.155 0.0020 0.703590 0.000009 4.56 0.029 Apr 1983 18.5309 0.0007 15.4893 0.0007 38.165 0.0017 0.703573 0.0000083 July 1983  18.4780 0.0007 15.4765 0.0008 38.117 0.0017 0.703573 0.000007 4.77 0.0331 Jan 1984 18.4595 0.0009 15.4765 0.0007 38.103 0.0020 0.703587 0.00000812 Sep 1984 18.4417 0.0008 15.4747 0.0007 38.091 0.0019 0.703555 0.0000098 Feb 1985  18.4342 0.0010 15.4743 0.0009 38.087 0.0024 0.703567 0.000008 4.76 0.0521 Apr 1985  4.82 0.1030 Jul 1985 18.4306 0.0007 15.4756 0.0007 38.089 0.0020 0.703571 0.0000082 Jun 1986  18.4138 0.0009 15.4755 0.0008 38.079 0.0020 0.703580 0.000009 4.94 0.3026 Jun 1986  4.77 0.0213 Sep 1986  18.4138 0.0008 15.4726 0.0007 38.074 0.0020 0.703590 0.000008 5.17 0.0116 Mar 1987  18.4108 0.0006 15.4717 0.0005 38.069 0.0014 0.703589 0.000009 5.25 0.0218 Oct 1987 18.3992 0.0008 15.4746 0.0008 38.068 0.0022 0.703597 0.00000919 Jan 1988 18.3952 0.0006 15.4715 0.0006 38.061 0.0016 0.703577 0.00000818 Aug 1988 18.3871 0.0008 15.4703 0.0007 38.052 0.0017 0.703583 0.00000926 Mar 1989  18.3882 0.0007 15.4717 0.0007 38.055 0.0019 0.703584 0.000009 5.11 0.057 Jul 1989 18.3861 0.0010 15.4725 0.0009 38.054 0.0022 0.703581 0.0000097 Jul 1989 ? 18.3851 0.0009 15.4710 0.0008 38.052 0.0021 0.703576 0.0000097 Jan 1990  18.3881 0.0009 15.4745 0.0008 38.056 0.0018 0.703584 0.000008 5.03 0.0127 May 1990 18.3864 0.0008 15.4716 0.0007 38.054 0.0020 0.703603 0.00000721 Oct 1990 18.3856 0.0007 15.4693 0.0006 38.049 0.0017 0.703597 0.00000812 May 1991  18.3992 0.0009 15.4739 0.0007 38.063 0.0019 0.703581 0.000008 5.08 0.041 Aug 1991  5.11 0.046 Jun 1992  18.4048 0.0009 15.4721 0.0009 38.061 0.0019 0.703585 0.000008 5.04 0.1013 Aug 1993  18.4112 0.0007 15.4752 0.0006 38.072 0.0015 0.703611 0.000008 4.98 0.074 Jan 1994 18.4098 0.0009 15.4736 0.0008 38.067 0.0020 0.703607 0.00000825 Apr 1994  18.4100 0.0009 15.4737 0.0009 38.068 0.0027 0.703586 0.000008 5.01 0.029 Oct 1994 18.4059 0.0008 15.4718 0.0007 38.066 0.0019 0.703598 0.00000727 Apr 1995  18.4059 0.0009 15.4721 0.0008 38.066 0.0023 0.703604 0.000009 5.25 0.0514 Oct 1995 18.4068 0.0008 15.4729 0.0008 38.071 0.0021 0.703602 0.00000919 Jan 1996  5.19 0.0715 Mar 1996 18.4064 0.0010 15.4738 0.0008 38.070 0.0023 0.703592 0.00000922 Aug 1996 18.4038 0.0009 15.4722 0.0008 38.065 0.0016 0.703612 0.00000710 Jan 1997  18.4010 0.0012 15.4728 0.0011 38.064 0.0019 0.703606 0.000009 5.2 0.0523 Jul 1997 18.3993 0.0010 15.4729 0.0010 38.068 0.0025 0.703601 0.00000710 Jan 1998 18.3958 0.0007 15.4728 0.0006 38.067 0.0014 0.703591 0.00000810 Jan 1998 ? 18.3940 0.0007 15.4711 0.0006 38.063 0.0016 0.703593 0.00000711 May 1998 18.4005 0.0009 15.4740 0.0008 38.071 0.0020 0.703605 0.0000107 Sep 1998 18.4082 0.0008 15.4775 0.0006 38.083 0.0017 0.703601 0.000006 5.33 0.067 Sep 1998 ? 18.4107 0.0004 15.4727 0.0005 38.075 0.0012 5.29 0.0813 Feb 1999 18.4068 0.0010 15.4783 0.0008 38.085 0.0021 0.703607 0.00000613 Feb 1999 ? 18.4124 0.0004 15.4736 0.0004 38.076 0.001119 Jun 1999 18.3987 0.0010 15.4805 0.0007 38.085 0.0020 0.703620 0.00000927 Oct 1999  18.4018 0.0004 15.4726 0.0004 38.069 0.0011 0.703622 0.000009 5.36 0.0819 Feb 2000  18.4072 0.0004 15.4712 0.0004 38.072 0.0011 0.703624 0.00000721 Jun 2000  18.4067 0.0004 15.4704 0.0004 38.069 0.0011 0.703638 0.000007 5.28 0.088 Jan 2001  18.4116 0.0004 15.4721 0.0004 38.074 0.0011 0.703627 0.000012 5.31 0.087 Jul 2001  18.4137 0.0004 15.4719 0.0004 38.073 0.0013 0.703626 0.0000099 Feb 2002  18.4139 0.0004 15.4707 0.0004 38.069 0.0011 0.703637 0.00000820 Aug 2002  18.4152 0.0004 15.4722 0.0004 38.072 0.0011 0.703639 0.00000512 Apr 2003  18.4161 0.0005 15.4726 0.0005 38.072 0.0013 0.703641 0.000005 5.31 0.0815 Jan 2004  18.4154 0.0005 15.4719 0.0005 38.069 0.0012 0.703632 0.000007 5.21 0.087 Jun 2004  18.4146 0.0003 15.4716 0.0004 38.068 0.0010 0.703624 0.00000731 Jan 2005  18.4170 0.0005 15.4735 0.0006 38.075 0.0012 0.703624 0.000005 4.96 0.138 Aug 2005  18.4119 0.0005 15.4727 0.0005 38.070 0.0013 0.703622 0.00001029 Jan 2006  18.4087 0.0004 15.4720 0.0004 38.065 0.0012 0.703623 0.00000924 Jun 2006 ? 18.4062 0.0028 15.4724 0.0026 38.066 0.0059 0.703612 0.000008 5.23 0.0324 Jun 2006  18.4073 0.0016 15.4714 0.0018 38.063 0.0060 0.703617 0.000013 5.23 0.036 Apr 2007  18.4065 0.0003 15.4715 0.0003 38.063 0.0009 0.703617 0.000007 5.35 0.1317 Jun 2007  18.4019 0.0004 15.4709 0.0004 38.062 0.0012 0.703626 0.000006 5.63 0.1322 Mar 2008  18.4038 0.0003 15.4700 0.0003 38.061 0.0010 0.703607 0.000008 5.45 0.132 May 2008  18.4045 0.0003 15.4721 0.0004 38.066 0.0010 0.703609 0.00000915 Nov 2008 18.3972 0.0009 15.4704 0.0007 38.058 0.0024 0.703600 0.00000829 Jan 2009  18.4003 0.0005 15.4709 0.0006 38.061 0.0012 0.703628 0.0000087 May 2009 18.4005 0.0008 15.4736 0.0007 38.066 0.0020 0.703624 0.0000104 Jun 2009  18.4009 0.0003 15.4720 0.0004 38.064 0.0010 0.703610 0.00000716 Oct 2009  18.3994 0.0005 15.4714 0.0006 38.062 0.0012 0.703622 0.00000722 Jan 2010  18.3987 0.0005 15.4708 0.0005 38.060 0.0013 0.703617 0.000007a indicates analysis at San Diego State University (SDSU), analyses from 7 Sep 1998 to 8 Aug 2005 are from Marske et al. [2008]. Sr isotopeanalyses from 1983?1997 and 2006?2010 were performed at PCIGR. ? Chemistry duplicate.  Published 18O analyses from Garcia et al. [1998].Analytical methods are described in the supporting information. 24 Jun 2006 is an in-house glass standard called Menehune collected from a Pu?u?O?o lava flow (errors are the external 62s of the replicate analyses; average of four analyses for Pb and Sr at PCIGR; 68 for Pb and 26 for Sr atSDSU). US Geological Survey sample numbers for lavas between up to16 Mar 87 are 23 Jan 1983: 1?054, 9 April 1983: 3?117, 3 Jul 1983: 5?139, 31 Jan 1984: 14?232, 12 Sep 1984: KE24?25 310S, 8 Feb 1985: 30?362, 30 Jul 1985: 35?419, 1 Jan 1986:40?484, 2 Jun 1986: 46?536, 13Sep 1986: 48?649, 16 Mar 1987: 48?714F.occurred after major collapses of the Pu?u ?O?ocrater floor (Table 1). The lavas erupted from theseuprift vents were geochemically distinct. Com-pared to coeval Pu?u ?O?o vent lavas, those fromepisode 54 have lower MgO (5.6?6.4 versus 7.5?10.1 wt %), CaO/TiO2 (2.8?3.4 versus 4.4), Sr/Nband Zr/Nb ratios (Figure 3). These geochemicalsignatures and the petrographic evidence of56789100.700.720.740.760.780.800.820.840.8648.849.049.249.449.649.850.050.250.43.03.23.43.63.84.04.24.44.64.82.12.22.32.42.52.60.350.400.450.500.551983198419851986198719881989199 0199119921993199419951996199719981999200020012 00220032004200520062007200820092010198319841985198619871988198919901991199219931994199519961997199 819992000200120022003200420052006200720082009201084 86 88 90 92 94 96 98 00 02 04 06 08 1084 86 88 90 92 94 96 98 00 02 04 06 08 1084 86 88 90 92 94 96 98 00 02 04 06 08 10 84 86 88 90 92 94 96 98 00 02 04 06 08 10Ep. 54TiO2 (wt%)normalized to 10 wt% MgO(Magma supply ratedoubled)2003-071983-19861986-19921992-2010MgO (wt%)CaO/Al2O3 (wt%)SiO2 (wt%)CaO/TiO2 (wt%)K2O (wt%)(b)(d)(c)(a)(f)(e)normalized to 10 wt% MgOnormalized to 10 wt% MgO(Magma supply ratedoubled)2003-07(Magma supply ratedoubled)2003-07(Magma supply ratedoubled)2003-071983-19861986-19921992-2010normalized to 10 wt% MgOnormalized to 10 wt% MgOEp. 56Figure 2. Major-element variation diagrams for Pu?u ?O?o lavas from 1983 to 2010. All major elements andratios except MgO were normalized to 10 wt % MgO [the most primitive lava erupted from Pu?u ?O?o; Gar-cia et al., 2000] by addition of equilibrium composition olivine (98.5%) and spinel (1.5%) in 0.5 mol % steps[Garcia et al., 2003; Rhodes and Vollinger, 2004]. Pu?u ?O?o lavas with <7.2 wt % MgO may have crystal-lized minerals other than olivine (e.g., clinopyroxene and plagioclase) and were not included in the olivinenormalization procedure and are not shown in all the plots, except MgO. Episode 54 (Ep. 54; 29?30 January1997) lavas involved mixing of evolved magmas stored in the rift zone and MgO-rich magma. Three intervalsof eruptive activity in legend and colors correspond with those shown in Figure 1. CaO/TiO2 and CaO/Al2O3ratios also use normalized data although are virtually unaffected by olivine fractionation. Vertical lines indi-cate nearly double magma supply rate between 2003 and 2007 [Poland et al., 2012]. Data are presented in thesupporting information. Uncertainty for analyses is described in the supporting information.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854857disequilibrium in the episode 54 lavas are thoughtto result from mixing Pu?u ?O?o magma withstored, differentiated rift zone magma [Garciaet al., 2000; Thornber et al., 2003]. In contrast,episode 56 lavas have higher MgO (8.5 versus 7.2wt %) and a relatively high 18O value (5.6 versus5.4%) but are otherwise geochemically indistin-guishable from contemporaneous Pu?u ?O?o lavas.[12] The 206Pb/204Pb ratios for Pu?u ?O?o lavasdecreased rapidly through the episodic fountainingperiod (1983?1986) and reached a minimumbetween 1989 and 1991 during the Kupaianahaphase (Figure 4b and Table 2). The rapid decreasein 206Pb/204Pb continues the longer-term trend ofdecreasing Pb isotope ratios for K?lauea lavaserupted following the 1924 collapse of the summitcaldera (Figure 4a). After 1991, the trend of206Pb/204Pb ratios in Pu?u ?O?o lavas shows cyclicvariations with two broad humps, each cycle span-ning approximately 10 years (except for a smalloffset from the overall trend between January1998 and June 1999; Figure 4b). The cyclic varia-tion in Pb isotope ratios is well shown by208Pb/206Pb ratios, which inversely mirror the206Pb/204Pb trend (Figures 4 and 5).[13] The 87Sr/86Sr ratios of Pu?u ?O?o lavas extendthe temporal trend of increasing Sr isotope ratiosfor K?lauea lavas that started following the 1924caldera collapse (Figure 4c). Overall, Pu?u ?O?olavas display an increase in 87Sr/86Sr from 1983 to2003 and a slight decrease after 2004 (Figure4d). Prior to 1999, the 87Sr/86Sr and 206Pb/204Pbratios of the lavas are not well correlated, althoughthere is an overall inverse correlation between the19831984198519861987198819891990199119921993199419951996199719981999200020012002200320042005200620072008200920100.40.50.60.70.819831984198519861987198819891990199119921993199419951996199719981999200020012002200320042005200620072008200920102.02.12.22.32.42.51.99101112161820222426 Sr/NbNb/Y La/Sm(c)(a)(d)1983-19861986-19921992-2010Ep. 54Ep. 54Ep. 5484 86 88 90 92 94 96 98 00 02 04 06 08 10 84 86 88 90 92 94 96 98 00 02 04 06 08 10(Magma supply ratedoubled)2003-07Ep. 54Zr/Nb(b)(Magma supply ratedoubled)2003-07 ?2 SE ?2 SE ?2 SE ?2 SEK?lauea summit(Sep.1982)K?lauea summit(Sep.1982)K?lauea summit(Sep.1982)K?lauea summit(Sep.1982)Figure 3. Trace-element ratios versus time for Pu?u ?O?o lavas from 1983 to 2010. Overall exponential vari-ation in trace-element ratios indicates progressive depletion of the source. In the La/Sm versus time plot, asubset of samples analyzed at PCIGR (April 1983, January 1984, September 1984, and April 2007 to January2010) are normalized to reference material Kil-93 (La/Sm of 2.09, average value from Australia National Uni-versity where most ICP-MS analyses were performed). Vertical lines indicate nearly double magma supplyrate between 2003 and 2007 [Poland et al., 2012]. Trace-element abundances in ppm (data shown in support-ing information). Average 62 bars are shown in a corner of each panel. September 1982 K?lauea summitlava composition from Pietruszka and Garcia [1999].GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854858two ratios for historical summit lavas (Figure 4and Table 2). After 1999, there is a positive corre-lation between 206Pb/204Pb and 87Sr/86Sr (Figure6), which corresponds to the second of the twomajor temporal cycles of Pb isotope ratios (Figure5). The Nd isotope ratios of lavas erupted between1983 and 2005 display no variation outside analyt-ical uncertainty [Marske et al., 2008]. Thus, nonew Nd isotopic data were collected during thisstudy.[14] Oxygen isotopic compositions of Pu?u ?O?olavas erupted over 25 years (1983?2008) show alarger range than historical K?lauea summit lavasspanning 380 years of activity (1.1 versus 0.7% ;Figure 4). Overall, O isotope ratios of Pu?u ?O?olavas have increased over time with early lavas(before 1986) having lower 18O (4.8?4.9%) thansubsequent lavas (5.0?5.6% ; Table 2). The high-est O isotope ratio observed during the Pu?u ?O?oeruption is for an episode 56 lava that erupted 6Uwekahuna BluffUwekahuna BluffHilinasubmarinelavas0.70340.70350.70360.7037198319841985198619871988198919901991199219931994199519961997199819992000200120022003200420052006200720082009201018.3718.3918.4118.4318.4518.4718.4918.5118.530.703550.703600.703651983198419851986198719881989199019911992199319941995199619971998199920002001200220032004200520062007200820092010K?laueasummitK?laueasummitPCIGR analyses (2009)Marske et al. (2008)-SDSUPCIGR analyses (2010)Time (years)Time (years)(summit)(summit)Time (years)K?laueasummitsummit calderacollapse (1924)summit calderacollapse (1924)average ?2 SEMORB mantle rangeSDSU analyses (2007)?18O18.218.418.618.81780 1820 1860 1900 1940 19801000 14001780 1820 1860 1900 1940 19801000 140016001780 1820 1860 1900 1940 198016001600Jun-99Jun-99Jun-86Jun-07(Ep. 56 uprift)(  )(  )(  )4.8 ka 3.5 ka1.7 ka4.8 ka3.5 ka1.7 ka206Pb/204Pb87Sr/ 86Sraverage ?1 SE?18O87Sr/ 86Sr206Pb/204Pb(280-130 ka)tholeiiticalkalictransitionalHilinasubmarinelavas(280-130 ka)tholeiiticalkalictransitional(episodic fountaining)Kupaianaha(gentle effusion)Pu?u ???? Pu?u ???? (mostly flank vent eruptions)(a)(c)(b)(d)Pu?u ???? Pu?u ???? Pu?u ???? (e) (f)data thisstudyMORBmantle summit calderacollapse (1924)PCIGR analyses (2009)Marske et al. (2008)-SDSUPCIGR analyses (2010)SDSU analyses (2007)4.64.85.05.25.45.65.84.64.85.05.25.45.61983198419851986198719881989199019911992199319941995199619971998199920002001200220032004200520062007200820092010no datascale changeK?lauea summit(Sep.1982)K?lauea summit(Sep.1982)Figure 4. Temporal variation of Pb, Sr, and O isotopes for K?lauea Volcano and Pu?u ?O?o lavas. (b)206Pb/204Pb in Pu?u ?O?o lavas shows cyclic variation with two broad humps, each cycle spanning approxi-mately 10 years. (d) 87Sr/86Sr in Pu?u ?O?o lavas increases from 1983 to 2003 and decreases between 2008and 2010, and is correlated with 206Pb/204Pb after 1999. (f) 18O for Pu?u ?O?o lavas shows the same range(0.7%) as historical K?lauea summit lavas and is not well correlated with 206Pb/204Pb or 87Sr/86Sr. Data sour-ces for K?lauea Volcano are Hanyu et al. [2010], Kimura et al. [2006], Marske et al. [2007], Abouchami et al.[2005], Pietruszka and Garcia [1999], Chen et al. [1996], and Garcia et al. [2008]. Data sources for previousanalyses of Pu?u ?O?o lavas areMarske et al. [2008] and Garcia et al. [1998]. For Pu?u ?O?o analyses, average62 for 206Pb/204Pb is smaller than symbol size and uncertainty for 87Sr/86Sr and 18O is shown in the panels.Data for Pu?u ?O?o lavas are presented in Table 2. September 1982 K?lauea summit lava composition fromPietruszka and Garcia [1999]. Colors for symbols in Figures 4b, 4d, and 4f) are the same as Figure 3.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854859km uprift from the Pu?u ?O?o vent in mid-June2007. The episode 56 eruption is related to intru-sion of a dike from the upper East Rift Zone, sothis lava was probably not derived from the shal-low reservoir of magma beneath Pu?u ?O?o [Mont-gomery-Brown et al., 2010]. Thus, its 18O valueis probably representative of the magma feedingthe Pu?u ?O?o eruption. It is identical to the highestvalues observed among historical summit lavas(5.6% ; Figure 4). Variations of 18O in Pu?u ?O?olavas do not correlate with Pb or Sr isotope ratios,or with other geochemical parameters, as wasnoted for previous O isotope work on lavas fromthis eruption [Garcia et al., 1998]. Therefore, Pu?u?O?o lava Pb and Sr isotope ratios were apparentlynot affected by the processes causing variable Oisotope ratios.5. Discussion[15] The high eruption rate and continuous natureof the Pu?u ?O?o eruption provide an exceptionalopportunity to use lava chemistry to evaluate thechanging roles that source, melting, and crustalprocesses play during this single prolonged erup-tion. Previous Pb isotope and trace-element studieson lavas from several multiple-year eruptions ofPiton de la Fournaise Volcano (Reunion Island)have discerned contributions from multiple com-ponents within the Reunion mantle plume and aperiodic role for shallow-level contamination [Pie-truszka et al., 2009; Vlastelic et al., 2005, 2007].Similarly, extreme Pb isotope variability in meltinclusions from Iceland basaltic lavas indicate sig-nificant source heterogeneity, with binary mixingrelationships that may result from combining sol-ids in the mantle and two stages of melt mixing (inporous mantle melt-transport channels and lowercrustal magma chambers) [Maclennan, 2008].Similarly, the geochemistry and petrography ofPu?u ?O?o lavas have been used to interpret theextent of crustal magmatic processes (olivine frac-tionation and accumulation, mixing of higher-MgO and stored rift-zone magmas, and crustalassimilation) and mantle processes (degree of par-tial melting, melt extraction and migration, andsource heterogeneity) during the Pu?u ?O?o erup-tion until 2005 [e.g., Garcia et al., 1998, 2000;Marske et al., 2008]. Here we use new high-precision Pb, Sr, and O isotope ratios, and major-and trace-element data for the entire Pu?u ?O?oeruption (1983?2010) to evaluate the causes ofcyclic and other short-term variability in the proc-esses that operate from the source to the surfacewithin K?lauea Volcano. The effects of crustalprocesses (crystal fractionation, magma mixing,and crustal contamination) on modifying Pu?u?O?o lava compositions are evaluated beforeexamining mantle source and melt transportprocesses.5.1. Magma Mixing and CrystalFractionation During Early EpisodicActivity (1983?1985)[16] The largest compositional changes in Pu?u?O?o lavas occurred from 1983 to 1985. Thesechanges mostly involved two crustal processes:crystal fractionation and magma mixing. Duringsome single eruptive episodes (5?10, 30, and 31),there were relatively large changes in MgO, Ni,and Cr, which are related to minor (3?5%) olivinefractionation in the shallow Pu?u ?O?o reservoirduring eruptive hiatuses [Garcia et al., 1992].These short-term (3?4 weeks) variations are super-imposed on longer term changes that have beenrelated to magma mixing [Garcia et al., 1992;Thornber, 2003]. The longer term variations areevident in plots of MgO-normalized major ele-ments, ratios of incompatible trace elements, andPb isotope ratios (Figures 2?4). Strontium and Oisotopes show less change during this period com-pared to their overall variation during the eruption(Figure 4). The overall progressive compositionalvariation in Pu?u ?O?o lavas from 1983?1985 hasbeen attributed to the mixing of new, relativelyMgO-rich magma (>7.5 wt %) with a decreasingproportion of hybrid, rift-zone stored differen-tiated magma (from 30% of the higher MgOmagma in March 1983 to 100% in September1984) [Garcia et al., 1992; Shamberger andGarcia, 2007].[17] The origin of the higher MgO magma compo-nent from the early phase of the Pu?u ?O?o erup-tion may have been: (1) magma from the summitreservoir, as represented by lavas from the Sep-tember 1982 summit eruption; and/or (2) newmantle-derived magma [Garcia et al., 1992;Shamberger and Garcia, 2007]. Scenario 1involves no change in the composition of thehigher MgO magma from September 1982 to1985, whereas scenario 2 requires it. The 1983?1985 Pu?u ?O?o lavas have both higher and lower206Pb/204Pb ratios than the September 1982 sum-mit lavas (Figures 4b and 7). Therefore, mixing ofa single 1982 summit magma with rift-zone storedmagma (scenario 1) cannot explain the isotopicGREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854860variation of Pu?u ?O?o lavas after 1984, when206Pb/204Pb values are lower than 1982 summitmagma (Figure 7). Ratios of some incompatibletrace elements (Sr/Nb and Zr/Nb) for some lavaswith higher MgO (>7.5 wt %) erupted after mid-1984 also are higher than those for September1982 summit lavas (Figure 3). Thus, if magmafrom the summit reservoir was supplying the Pu?u?O?o eruption, its composition must have changedafter the September 1982 eruption and prior toSeptember 1984 (Figures 4 and 7).[18] The isotopic variations for early Pu?u ?O?olavas are consistent with the eruption being sup-plied by new, compositionally variable, mantle-derived magma in addition to or instead of Sep-tember 1982 summit magma. The rate of206Pb/204Pb variation observed for the period afterthe end of early magma mixing is much faster thanduring the previous 30 years (1952?1982) ofK?lauea summit eruptions (0.016 yr1 versus0.004 yr1). These rapid variations in Pb isotopicratios suggest that magmas supplying Pu?u ?O?opartially bypassed or did not thoroughly mix withthe summit reservoir [Garcia et al., 1996]. Basedon these observations, the composition of the pa-rental magma delivered to Pu?u ?O?o from themantle is interpreted to have continually changedfor the remainder of the eruption (i.e., after 1984).The details and cause of this variation are dis-cussed in section 5.3.5.2. Oxygen Isotope Indications of CrustalContamination and Nature of MantleSource[19] Lavas from oceanic island volcanoes showwide ranges in oxygen isotopic compositions (4.5?7.5%), which have been attributed to composition-ally variable mantle-derived magmas that weremodified by oxygen exchange and/or crustal con-tamination [Harmon and Hoefs, 1995]. Our previ-ous studies revealed that some Pu?u ?O?o andK?lauea summit magmas experienced significantoxygen isotope exchange with metamorphosedK?lauea rocks [Garcia et al., 1998, 2008]. This isindicated by the disequilibrium between matrixand coexisting olivine 18O values, the relativelylow 18O values for these lavas (4.7?5.2%) andthe lack of correlation between 18O values andother geochemical parameters [Garcia et al.,1998, 2008].[20] The highest 18O value observed for any lavaduring the Pu?u ?O?o eruption is for the June 2007208Pb/206PbPCIGR analyses (2009)Marske et al. (2008)-SDSUPCIGR analyses (2010)SDSU analyses (2007)(Magma supply ratedoubled)2003-072.0592.0612.0632.0652.0672.0692.0711983198419851986198719881989199019911992199319941995199619971998199920002001200220032004200520062007200820092010Jun-99(episodic fountaining)Kupaianaha(gentle effusion)Pu?u ???? Pu?u ???? (mostly flank vent eruptions)Jun-861982summitRift-storedmagmaFigure 5. 208Pb/206Pb variation with time for Pu?u ?O?o lavas from 1983 to 2010. Previous analyses of Pu?u?O?o lavas indicated in legend are from Marske et al. [2008]. Average 62 for 208Pb/206Pb is smaller thansymbol size.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854861uprift lava (5.6% ; Figure 4). This value is identi-cal to the highest value reported for historicalK?lauea summit lavas (1820?1982) [Garcia et al.,2008] and lies within the range of normal mid-ocean ridge basalt (MORB) basalt 18O values(5.4?5.8% ; Figure 4) [Eiler, 2001]. These summitlavas (1832, 1866, 1894, and 1917?1921) wereerupted during periods of sustained lava lake ac-tivity, and are thought to be representative of theprimary uncontaminated magma feeding K?lauea[Garcia et al., 2008]. Thus, the 2007 uprift ventlava supports our previous interpretation [Garciaet al., 2008] that the 18O value for the mantlesource of K?lauea?s magma is identical to thesource for MORB.[21] The earliest Pu?u ?O?o lavas (1983?1986)show the strongest signs of disequilibriumbetween coexisting matrix mineral and olivine,and have the lowest O isotope values (<5.0%)[Garcia et al., 1998]. After the shift to continuouseffusion in July 1986, O isotope ratios are higher(5.2%) and the coexisting olivines were in equilib-rium with host matrix for about 1 year [Garciaet al., 1998]. Subsequently, the matrix O isotopevalues decreased somewhat (to 5.0% ; Figure 4)and those for olivine increased, indicating olivine-matrix disequilibrium. This O isotope disequili-brium continued for two more years, and was fol-lowed by a return to olivine-matrix equilibrium in1995?1997 [Garcia et al., 1998]. After 1997, ma-trix O isotope values are relatively low and nearlyconstant (5.36 0.1%) except for a 2005 lava(5.0% ; Figure 4), which was the most evolvedsample (analyzed for O isotopes) since 1984 (6.7wt % MgO). Thus, despite nearly 30 years of vig-orous eruptive activity (producing 4 km3 oflava), oxygen exchange with metamorphosedrocks has probably continued in the Pu?u ?O?omagmatic plumbing system. The magnitude of ox-ygen isotope exchange can be estimated assumingbulk assimilation between a parental magma (asreflected by the 2007 uprift sample with a 18Ovalue of 5.6%) and a hydrothermally alteredK?lauea rift zone lava (1.9%) [Garcia et al., 2008]as a contaminant. Pu?u ?O?o lavas erupted justbefore and after the 2007 uprift event have averageO isotope values of 5.4% (Figure 4), indicating5% bulk contamination, whereas earlier lavas(1986?2006) with average values of 5.2?5.3%,might have experienced 8?11% bulk contamina-tion. This contamination is likely to have occurredin the Pu?u ?O?o reservoir and did not have anyobvious effect on other geochemical parameters[Garcia et al., 1998] (Table 2).5.3. Cyclic Compositional VariationsFrom Mantle Processes (1985?2010)[22] Pu?u ?O?o lavas erupted after the early periodof magma mixing ended in late 1984 show cyclicvariations in several geochemical parameters thatare insensitive to olivine fractionation (e.g., CaO/TiO2, Sr/Nb, Zr/Nb,206Pb/204Pb; Figure 8). Thecyclic variations in CaO/TiO2 and K2O/TiO2ratios for Pu?u ?O?o lavas erupted between 1996and 2001 were reported to be associated with de-formation in the summit magma reservoir[Thornber, 2003]. Although the timing of thehighs and lows in these ratios are not perfectlycoincident with summit tilt changes [see Figure 8,Thornber, 2003], these geochemical cycles wereattributed to mixing of mantle-derived magma ofuniform composition (similar to Pu?u ?O?o lavasaverage ?2 SEaverage ?2 SETime (month-year)18.39018.39518.40018.40518.41018.41518.420Jan-97Jul-97Jan-98Jul-98J an-99Jul- 99Jan-00J ul -00Jan-01Jul-01Jan -02Jul-02Jan-03Jul-03J an-0 4Jul-04Jan-05J ul-05Jan-06Jul-06J an-07J ul-07Jan-08Jul-08Jan-09Jul-09J an-100.703590.703600.703610.703620.703630.703640.70365Jan-97Jul-97Jan-98Jul-98Jan-99Jul-99Jan -00Jul-00Jan-01Jul-01Jan -02Jul-02Jan-03Jul-03Jan -04Jul-04Jan-05Jul-05Jan-06Jul-06Jan-07Jul-07Jan-08Jul-08Jan -09Jul-09Jan -10206Pb/204Pb87Sr/ 86Sr(a)(b)Figure 6. Temporal variation in 206Pb/204Pb and 87Sr/86Srfor Pu?u ?O?o lavas during period of dramatic increase inmagma supply. Dashed lines indicates period of significantincrease in magma supply rate up to 0.25 km3/yr between2003 and 2007 compared to 0.1 km3/yr prior to 2003[Poland et al., 2012]. Analytical uncertainty is shown in thepanels.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854862from 1999 to 2001) with 1982 summit magma[Thornber 2003]. However, Pu?u ?O?o lavaserupted after 1999 have elevated 87Sr/86Sr ratios(at a given 206Pb/204Pb; Figure 7) compared toearlier lavas with low 206Pb/204Pb ratios, and thus,the 1999?2001 lavas cannot serve as a mixingend-members to explain the compositional trendof lavas erupted before 1999. Similar behavior isobserved on a plot of La/Yb versus 206Pb/204Pb(Figure 9), where a relative shift to lower La/Ybratios at a given 206Pb/204Pb occurred after 1999(compared to the trend of pre-1999 lavas). Theserelationships indicate that the temporal variationof Pu?u ?O?o lavas erupted after 1999 cannot beexplained simply by mixing of 1982 summitmagma with a uniform mantle-derived magma(Figure 7) within K?lauea?s shallow magmaticplumbing system. Instead, either a third magma ismixing with the other two or, as we advocatebelow, the composition of the Pu?u ?O?o magmais continually changing due to the melting ofsmall-scale compositional heterogeneities in themantle source.[23] Ratios of Pb isotopes in Pu?u ?O?o lavas showcyclic variations (Figure 5). These variations prob-ably reflect the dynamic process of melt extraction(from a heterogeneous source) over a time scale of0.703500.703550.703600.703650.7037018.36 18.38 18.40 18.42 18.44 18.46 18.48 18.50 18.52 18.54208Pb/204Pb0.703570.703580.703590.703600.703610.703620.703630.703640.7036518.38 18.39 18.40 18.41 18.42208Pb/204Pb38.04538.05538.06538.07518.38 18.39 18.40 18.41 18.4238.0038.0538.1038.1538.2018.36 18.38 18.40 18.42 18.44 18.46 18.48 18.50 18.52 18.5487Sr/ 86Sr206Pb/204PbSep-8687878889 9091909808090700030504020106929394959697 991988-911983-85Jun-9986102000-071992-992008-1096098788Oct-90Jan-90919808090700 0305040201069293949597991988-911983-85Jun-99 8610968788Mar-89Jul-89May-9096979898990005080808Jan-090909862000-071993-992008-1087Sr/ 86Sr206Pb/204Pb19831985198319851986-20101986-2010East Rift Zone(1977)East Rift Zone(1977)K?lauea summit(1982)K?lauea summit(1982)East Rift Zone(1960-69)East Rift Zone(1960-69)1984Sep.Jan.2001 composition1984Sep.Jan.?2?(b)(d)(a)(c)(b)(d)Figure 7. Pb and Sr isotopic compositions for Pu?u ?O?o lavas. Line connects samples in order of increasingeruption date in Figures 7b and 7d. Average 62 for Pb isotope ratios is smaller than symbol size. East RiftZone data is from J. Marske [personal communication, 2013]. September 1982 K?lauea summit lava composi-tion (outline in Figure 7a; orange star in Figure 7c) from Pietruszka and Garcia [1999]. K?lauea summit(1982) field in Figure 7a is new high-precision data from A. Pietruszka [personal communication, 2013]. Bluestar is 2001 composition proposed by Thornber [2003] as mixing end-member with 1982 K?lauea summitcomposition. Pu?u ?O?o lavas erupted after 1999 have elevated 87Sr/86Sr ratios (at a given 206Pb/204Pb). The1999?2001 lavas cannot serve as mixing end-members to explain the compositional trend of lavas eruptedbefore 1999.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854863years to decades rather than movement of small-scale mantle heterogeneities through the meltingzone. This interpretation is based on the hypothe-sis that buoyancy-driven upwelling through themelt-producing region beneath K?lauea occurs onlonger timescales (hundreds to thousands ofyears) than melt extraction (years to decades)[Pietruszka et al., 2006]. The highest estimatesfor solid mantle upwelling in the center of theHawaiian plume are 10 m/yr [Hauri, 1996;Pietruszka and Garcia, 1999], which wouldresult in a maximum of only 270 m of upwell-ing during the first 27 years of the Pu?u ?O?oeruption [cf. 5?10 km maximum thickness forthe zone of melting; Marske et al., 2007]. Forcomparison, estimates for solid mantle upwellingrates beneath Mauna Loa and Lo?ihi based onU-series disequilibria range from 0.4 to 1 m/yr[Sims et al., 1999] and 5?6 cm/yr [Pietruszkaet al., 2011], respectively. Melt extraction rates(or source-to-surface melt velocity) are estimatedto be on the order of 5?17 km/yr [Reiners,2002], which is extremely rapid compared tosolid mantle upwelling rates. Thus, cyclic varia-tion in Pb isotope ratios over short timescales(years) are best explained by variations in theprocess of melting of a heterogeneous source(and the transport of the melt to the surface),rather than upwelling of small-scale mantle het-erogeneities through the melting region.[24] The short-term Pb and Sr isotopic variationsin Pu?u ?O?o lavas may be generated by one ormore processes including: periodic processes ofmelting, melt extraction, or melt aggregation [e.g.,Cordier et al., 2010], changes in melt transportpathways or tapping new source areas [Marskeet al., 2008; Pietruszka et al., 2006], changes inthe volume of the melting region [Pietruszkaet al., 2001], and progressive melt extraction froma source with fine-scale heterogeneities [Garciaet al., 2000]. In the presence of small-scale hetero-geneities, changes in melt pathways over years todecades may lead to tapping compositionally dis-tinct sources and short-term isotopic variation inlavas [Marske et al., 2007]. The scale of composi-tional heterogeneities must be small relative to thesize of the melting region beneath K?lauea Vol-cano to allow for rapid (few years) variation inlava Pb isotope compositions [Pietruszka and Gar-cia, 1999]. Melt pathways within the source regionprobably migrate over years to decades [Pie-truszka et al., 2001, 2006]. Therefore, melt may besupplied from different areas of the melting region(Magma supply ratedoubled)2003-07Time (year)4.04.24.44.64.8CaO/TiO2 1986-19921992-201022232425262790 92 94 96 98 00 02 04 06 08 1090 92 94 96 98 00 02 04 06 08 1090 92 94 96 98 00 02 04 06 08 10206Pb/204PbSr/Nb(a)(d)Zr/Nb(b)(c)18.3518.3618.3718.3818.3918.4018.4118.4218.4318.44199019911992199319941995199619971998199920002001200220032004200520062007200820092010101112Figure 8. Plots of CaO/TiO2, Sr/Nb, Zr/Nb and206Pb/204Pbfor Pu?u ?O?o lavas showing cyclic variation apparent from1990 to 2010. Trace-element abundances in ppm.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854864over a relatively short period of time. These short-term geochemical fluctuations are effectivelytransported from the source to the surface becausePu?u ?O?o magmas are thought to partially bypassthe volcano?s summit magma storage reservoir,and avoid its buffering effects [Garcia et al.,2000].[25] The periodicity and rate of isotopic variationPu?u ?O?o lavas provide clues about the size anddistribution of small-scale heterogeneities in thevolcano?s mantle source. The cycles of Pb isotopicratios at Pu?u ?O?o have a peak to peak duration of10 years (Figure 10). These short-term increasesand decreases of Pb isotope ratios, at similar rateand degree, may represent melt extraction fromsmall-scale heterogeneities with a limited horizon-tal length scale (a few km or less). Modeling byFarnetani and Hofmann [2009] suggests that??filament-like?? structures derived from stretchingof deep-seated mantle heterogeneities may de-velop as the Hawaiian plume rises to the surface.Although their model was developed to explainthe long-term (>100 kyr) geochemical variationobserved for drill core from Mauna Kea [Farne-tani and Hofmann, 2010], we use the filamentmodel to explain the periodic variation in the Pbisotope ratios of Pu?u ?O?o lavas (Figure 10)because it provides a mechanism to link geochem-ical variations with the inferred deep mantle struc-ture. Other geometries have been suggested for thesmall-scale heterogeneities within Hawaiianplume, including a series of vertically stacked,elongated blobs [Blichert-Toft and Albare`de,2009], but we prefer the filament geometry toexplain the Pb isotopic variations of Pu?u ?O?olavas.[26] In this scenario, the periodic variation in Pbisotope ratios of Pu?u ?O?o lavas may reflect meltextraction from a mantle source with verticallyoriented repeating source heterogeneities, or thinstrands, on a small scale (Figure 10). The Pu?u?O?o eruption rate is thought to be greater than therate of melting, so melt must be transferred intochemically isolated channels from successivelyfurther areas within the larger melting region tosustain the eruption [Pietruszka et al., 2006]. Inthe context of the filament model of Farnetani andHofmann [2009], this process might extract meltfrom a succession of strands with different isotopiccompositions, which would potentially create theobserved periodicity in variation of the Pb isotoperatios (Figure 10). The volume of a single compo-sitional strand within the mantle tapped by thePu?u ?O?o eruption can be inferred using estimatesfor lava eruption rate (0.13 km3/yr) [Suttonet al., 2003] and melt zone porosity (1?2%) [Pie-truszka et al., 2001]. This calculation assumes that(1) there have been only two isotopically distinctcomponents since 1986 and (2) the heterogene-ities have the same melt productivity. We do notconsider the effect of melting heterogeneous lith-ologies with different melt productivities (e.g., pe-ridotite versus pyroxenite), despite the potentialsignificance for mixed lithologies in the source forHawaiian lavas [Hauri, 1996; Reiners, 2002;Sobolev et al., 2005]. Indeed, recent modeling ofincompatible trace elements suggests that Pu?u2007-1018.3818.4218.4618.5018.544.5 4.9 5.3 5.7 6.1 6.5 6.9 7.3 7.718.3818.3918.4018.4118.424.5 4.7 4.9 5.1 5.3 5.5 5.7 5.9HistoricalEast RiftZone ?2 ?summit(1982)La/Yb206Pb/204Pb206Pb/204Pb(a)(b)831988-912000-061992-99848586878889908990919394969598989708099900010304050607091092La/Ybmixingsource variationK?laueasummitFigure 9. Plots of 206Pb/204Pb versus La/Yb for Pu?u ?O?olavas with line connecting samples in eruptive order. (a) AllPu?u ?O?o lavas. (b) Smaller variations in lavas eruptedbetween 1988 and 2010 [area indicated by box with dashedline in Figure 9a]. K?lauea summit data is from Pietruszkaand Garcia [1999]. Average 62 for La/Yb shown in Figure9 a. Average 62 for Pb isotope ratios is smaller than symbolsize.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854865?O?o lavas (1986?1998) are derived from a sourcewith 13% recycled oceanic crust in a matrix ofambient depleted Hawaiian mantle [Pietruszkaet al., 2013]. However, this model was unable todistinguish if the recycled oceanic crust was pres-ent as pyroxenite or refertilized peridotite. Itshould also be noted that the compositional rangeof Pu?u ?O?o lavas from 1986 to 2010 is smallcompared to the overall range for Hawaiian volca-noes [e.g., Jackson et al., 2012; Ren et al., 2009;Weis et al., 2011], so the melt productivities of theend-member sources are probably similar.[27] The duration of a single cycle of Pb isotopevariation is approximately 10 years (Figure 10),which suggests melt is extracted from one compo-sitional strand in 5 years (before the trendreverses when melt from a different strand isencountered). Estimates for the height of K?lauea?smelting region range from <5 km [Marske et al.,2008] to 55 km [Watson and McKenzie, 1991].We assume the magma supply rate is roughlyequivalent to the eruption rate given that overallmagma storage in K?lauea has been decreasingslightly since 1983 [Poland et al., 2012]. The meltDepth (km) 13017010 50 90 130 170 210Length(km)Plate motionMelting zoneK?lauea Plate motion9 cm/yr100 0 100 200Length(km)Depth (km)15025050350Melting rate (10-11 kg m-3 s-1)0 2 4 6 8250 225 200 175 150 125?T (?C)50Depth (km)13017020 80 50Depth (km)13017020 80 50Depth (km)13017020 80Depth (km)50Depth (km)13017020 8019902.0652.0672.0692.071208 Pb/206 Pb1987198819891990199119921993199419951996199719981999200020012002200320042005200620072008200920101994     1998 2004(a)(b)(c)(d)Figure 10. Cartoon model for the Hawaiian mantle plume to explain the isotopic variation of Pu?u ?O?olavas, based on assumptions described in the text. (a) 208Pb/206Pb variation with time for Pu?u ?O?o lavas from1990 to 2010 showing cyclic variation. (b) Vertical section of the Hawaiian plume adapted from Farnetaniand Hofmann [2010]. Purple shades indicate the melting rates inside the melting zone, shown in legend.Dashed yellow lines are flow trajectories. Dashed black box is magma capture zone for K?lauea. (c and d)Sketches of the changing melting zone during Pu?u ?O?o eruption. Lithosphere is not shown in Figure 10c.Melting zone from Farnetani and Hofmann [2010].GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854866volume produced in 5 years is 0.5 km3. If meltzone porosity is 2% (estimates based on U-seriesstudies range from 1 to 3%) [Pietruszka et al.,2001; Sims et al., 1999], the total volume of a sin-gle heterogeneity tapped over a 5 year periodwould be 25 km3. If individual filament-shapedheterogeneities extend over the height of the melt-ing zone, then this 25 km3 volume would translateto a diameter of 1?3 km for melting zone heightsof 55?5 km, respectively.5.4. Effects of Doubling of Magma SupplyRate on Lava Composition (2003?2007)[28] One enigmatic question of ocean island volca-nism is whether variations in lava compositions arecorrelated with magma supply rate [e.g., Vlastelicet al., 2005]. Wide variations in magma supplyrate have occurred historically at K?lauea (0.01?0.18 km3/yr between 1840 and 1983) [Dvorak andDzurisin, 1993]. A marked decrease in lava effu-sion rate during the 19th to early 20th century atK?lauea (0.10?0.01 km3/yr) was accompanied byan increase in the ratios of highly over moderatelyincompatible trace-element abundances [Pie-truszka and Garcia, 1999], and in the modal abun-dance of clinopyroxene and plagioclase in thelavas indicating eruption of more fractionated andcooler magma [Garcia et al., 2003]. This change inlava composition is believed to be a direct result ofa decrease in the melt fraction (10?5%) and aswitch to a more depleted source [Pietruszka andGarcia, 1999]. A dramatic short-term increase ineffusion rate was observed for the Pu?u ?O?o erup-tion between 2003 and 2007 [Poland et al., 2012].Here we explore the results of this magma supplysurge on the composition of Pu?u ?O?o lavas.[29] Estimates of lava effusion rate for the Pu?u?O?o eruption prior to 2003 are based on geologicmapping, and measurements of very low fre-quency electromagnetic profiling and gas emis-sions. These techniques indicate an average rate ofmagma supply of 0.13 km3/yr [Sutton et al.,2003]. Lava effusion rate (considered to be aproxy for the magma supply rate by Poland et al.[2012]) was estimated to have increased between2003 and 2007 and to have doubled in 2005 (to0.25 km3/yr), before returning to the previousrate by 2008 [Poland et al., 2012]. An increase inmagma supply is normally expected to result inhigher MgO contents in erupted lavas as magmaundergoes less cooling prior to eruption. This rela-tionship was inferred for Pu?u ?O?o lavas eruptedfrom 1986 to 1992, when changes in tilt were fol-lowed 3 weeks later by changes in MgO [e.g.,Garcia et al., 1996]. However, during the 2003?2007 surge in magma supply, Pu?u ?O?o lavashave consistently lower MgO contents (<7.5 wt%; Figure 2a) than any period since the start ofcontinuous effusion in mid-1986, except duringepisode 54. The lower MgO contents of 2003?2007 lavas was interpreted to have resulted fromthe stirring and flushing of cooler magma withinthe volcano?s shallow magma storage system byan influx of new, hotter more primitive magma[Poland et al., 2012]. Mineralogical evidence (twopopulations of olivine) was noted in support ofthis claim, although no data were presented byPoland et al. [2012]. Our previous study of olivinecompositions in lavas erupted before and duringthe surge found no evidence for two populationsof olivines in any of the lavas, and that olivines inthese weakly phyric rocks are in Fe-Mg equilib-rium with the whole rock [Marske et al., 2008].We re-examined thin sections for these lavas andfound no textural evidence indicating magma mix-ing. In contrast, Pu?u ?O?o lavas from 1983 to1984 and episode 54 display obvious disequili-brium features from magma mixing [Garcia et al.,1989, 2000]. If mixing with a stored, coolermagma was important during the 2003?2007 surgein magma supply, the stored component must nothave differentiated very far beyond olivine control(unlike the situation for Pu?u ?O?o lavas from1983 to 1984 and episode 54).[30] A small increase in lava MgO contentoccurred after the surge, although values werevariable and overlap with those during the surge(7.0?8.1 after versus 6.7?7.4 wt % MgO during;Figure 2). The higher post-surge MgO values wereinterpreted to be a result of the heightened magmasupply from 2003 to 2007 [Poland et al., 2012].By November 2008, MgO dropped to values simi-lar to and lower than during the surge (6.5?7.2 wt%; Figure 2). Since 2000, the MgO content ofPu?u ?O?o lava has been declining with no signifi-cant change of this overall trend during the 2003?2007 surge in magma supply (Figure 2), except forless scatter in MgO content which may simplyreflect fewer interruptions in effusion during thistime. Also, it is noteworthy that the highest sus-tained MgO values were observed for Pu?u ?O?olavas from 1988 to 1993 (Figure 2), a period whenno increase in magma supply was recorded[Poland et al., 2012; Sutton et al., 2003]. Thus,the increase in magma supply from 2003 to 2007appears to have had limited impact on the varia-tion in the MgO content of Pu?u ?O?o lavas.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854867[31] Some of the other compositional features ofthe lavas erupted during the 2003?2007 magmasurge (e.g., slight offsets to lower MgO-normalizedSiO2 contents, Zr/Nb ratios and higher normalizedK2O and TiO2) could be considered to be represen-tative of flushing of differentiated magma similarto the Pu?u ?O?o lavas the late 1990s to early 2000s(Figures 2 and 3). There is also a reversal in87Sr/86Sr and 206Pb/204Pb in 2004 toward values by2007 similar to lavas from 2000 (Figure 6).These trends might be explained by the flushing ofmagma stored since the late 1990s to early 2000sduring the 2003?2007 surge in magma supply.However, based on the close time correlations(weeks to few months) between changes in summittilt and lava composition in earlier Pu?u ?O?o lavas[Garcia et al., 1996; Thornber, 2003], it is hard toimagine why 4? years were required to flushcooler magma from the Pu?u ?O?o system as pro-posed by Poland et al. [2012]. Thus, it is our inter-pretation that the magma erupted during the 2003?2007 surge was new mantle-derived magma andnot stored magma flushed from K?lauea?s shallowcrustal plumbing system. Thus, we interpret theisotopic variations during the Pu?u ?O?o eruption(since 1984) to be generated by the changingcomposition of melts coming from the mantle.5.5. Comparison of Pu?u ?O?o Lavas Withthe Long-Term Isotopic Evolution of OtherHawaiian Volcanoes[32] Studies of the long-term geochemical varia-tion of lavas from individual volcanoes in the Ha-waiian Islands provide understanding of thechemical structure of the Hawaiian mantle plume[e.g., Loa and Kea trends, Abouchami et al., 2005;Ren et al., 2009; Weis et al., 2011], and the vari-ability within single shield volcanoes [e.g., Bryceet al., 2005; Chen and Frey, 1985; Eisele et al.,2003; Marske et al., 2007; Nobre Silva et al.,2013; Rhodes and Hart, 1995; Weis et al., 2011].Here we compare the short-term Pb and Sr iso-topic variation for Pu?u ?O?o lavas to the longerterm variations for lavas from K?lauea and nearby,well-studied shield volcanoes to better understandthe rate and cause of long-term fluctuationsobserved at Hawaiian volcanoes.[33] The Pu?u ?O?o isotopic range covers a rela-tively large part of the long-term variationobserved for Mauna Kea and K?lauea volcanoes(Figures 11 and 12). The isotopic variation ofnearby Mauna Kea volcano was well documentedfor 300 kyr of shield growth using the HSDP2drill core. Lavas from the Pu?u ?O?o eruption afterthe period of magma mixing (early 1985) span30% of the total range of 206Pb/204Pb variationrecorded for HSDP2 (0.07 versus 0.22). Com-pared to K?lauea summit lavas, Pu?u ?O?o lavaserupted since 1985 span 25% of the Pb isotoperange since 950 AD and 47% of the range of Srisotope ratios (Figures 4 and 12). Thus, this singleeruption, which represents <1% of the time cov-ered by the HSDP2 core and 3% of thethousand-year period for K?lauea summit lavas,shows remarkable short-term isotopic variations.However, as seen for K?lauea summit lavas, iso-topic variation in Hawaiian shield lavas is cyclic,with each volcano showing a narrow but com-monly distinctive range (compared to neighboringvolcanoes) as seen by the relatively tight fields forPb and Sr isotope ratios in lavas from K?lauea,Lo?ihi, Mauna Kea and Mauna Loa (Figure 12).Thus, the Pu?u ?O?o eruption may represent thebest known expression of the small-scale compo-sitional heterogeneity of the Hawaiian plume. Asdiscussed above, the rapid rates of isotopic fluctu-ation found in Pu?u ?O?o lavas require a heteroge-neous source (on a scale of less than severalkilometers) that is tapped in only 5 years ofmelt extraction. How does the Pu?u ?O?o sourcecompare with those for nearby, well studied vol-canoes Lo?ihi and Mauna Kea?[34] Compared to other Hawaiian volcanoes, the Srand 206Pb/204Pb isotope ratios for Pu?u ?O?o lavasare most similar to Lo?ihi (Figure 12), althoughLo?ihi lavas have higher 208Pb/204Pb at a given206Pb/204Pb, like other Loa trend volcanoes (Figure11). Pu?u ?O?o lavas overlap with the Kea-mid8 Pbisotope array, the most common lava type inHSDP2 drill core (Figures 11c and 11d) [Eiseleet al., 2003; Nobre Silva et al., 2013]. However, Srisotopic compositions of Pu?u ?O?o lavas eruptedsince 1988 do not overlap those of Mauna Keaand trend orthogonally to the overall inverse arrayfor Hawaiian shield volcanoes and for K?laueasummit lavas (Figure 12). Thus, the Pu?u ?O?osource is isotopically distinct from other Hawaiianvolcanoes and the Pu?u ?O?o data set shows thatindividual eruptions may have trends orthogonal towhat are considered the primary source end-members for Hawaiian shield volcanoes.6. Conclusions[35] The temporal geochemical variation of Pu?u?O?o lavas from 1983 to 2010 provides insights onGREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854868the chemical structure of the Hawaiian mantleplume, and the dynamics of melt transport andmixing within the mantle. Our new results show:[36] 1. The Pu?u ?O?o eruption is being suppliedby new, compositionally variable, mantle-derivedmagma, which is being modified by various crustalprocesses including crystal fractionation (mainlyolivine), magma mixing (during 1983?1984 andepisode 54), and oxygen isotope exchange with orassimilation of altered K?lauea rocks.[37] 2. The episode 56 fissure lava has the highest18O value (5.6%) of any Pu?u ?O?o lava and thehighest MgO of any lava erupted since July 2000.Thus, it may be representative of the magma feed-ing this eruption as of June 2007. The episode 56fissure was fed from an upper East Rift Zone dikerather than the shallow Pu?u ?O?o reservoir. Coe-val lavas erupted from the Pu?u ?O?o vent havelower 18O values (5.36 0.1%) suggesting theywere contaminated, probably in the shallow reser-voir under the vent. Pu?u ?O?o lavas show no cor-relation of 18O values with other geochemicalparameters. For that reason, we suggest that the18O ratios were lowered by oxygen exchangewith or assimilation of altered K?lauea wall rock.[38] 3. Contrary to expectations, the dramaticincrease in magma supply between 2003 and 2007for the Pu?u ?O?o eruption was not accompanied byhigher MgO contents. Instead, lavas erupted duringthe 2003?2007 surge have lower MgO indicative ofgreater cooling of the magma prior to eruption,continuing the long-term trend for the eruption.[39] 4. Rapid and remarkably systematic variationsin Pb and Sr isotopic ratios are present in Pu?u37.737.837.938.038.138.238.338.418.0 18.1 18.2 18.3 18.4 18.5 18.6 18.72.042.052.072.082.092.100.825 0.830 0.835 0.840 0.845 0.850 0.855 0.860L??ihi2.0637.9037.9538.0038.0538.1038.1538.2038.2518.30 18.35 18.40 18.45 18.50 18.55 18.60 18.65 18.702.052.062.072.080.830 0.835 0.840 0.845208Pb/204Pb206Pb/204Pb208Pb/206Pb207Pb/206Pb208Pb/204Pb206Pb/204Pb208Pb/206Pb207Pb/206PbMauna LoaMauna LoaMauna KeaHistorical K?lauea summitPu?u ???? (1983-2010) (b)(d)(c)(a)1986Uwekahuna Bluff (summit)(Pietruszka and Garcia, 1999)Prehistoric K?lauea(Abouchami et al., 2005)L??ihi19831986Hilina BenchHilina Bench(Marske et al., 2007)upper endmemberkea-hi8kea-mid8kea-low8Mauna KeaMauna Kea (HSDP-2)Kea-hi8 arrayKea-mid8 arrayKea-low8 arrayarea of panel c)area of panel d)Historical K?lauea summitLoaKeaLoaKea1983Figure 11. Pb isotopic compositions for Pu?u ?O?o lavas compared to some other Hawaiian shield volca-noes. Top panels are different scales than bottom panels, shown by area of dashed boxes. Colored symbols arelavas from K?lauea Volcano. Fields are for Mauna Loa, Lo?ihi, Mauna Kea, and submarine prehistoricK?lauea (Hilina Bench; >40 ka). References for data sources are listed in the supporting information. Keaarray end-members are from Eisele et al. [2003]. Dashed line with yellow stars in Figure 11c is a best fit linefor Pu?u ?O?o lavas from 1988 to 2010.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854869?O?o lavas. Two cycles of Pb isotopic ratio varia-tion with 10 year periods were found. Thesecycles may be related to extraction of melt from asource with a pattern of vertically oriented sourceheterogeneities, or thin strands. These strands maybe 1?3 km in diameter to explain the scale ofisotopic variation for the Pu?u ?O?o eruption.[40] 5. The Pb isotopic variation of Pu?u ?O?olavas spans 25% of the range observed for the last1000 years of K?lauea summit lavas and 30% for300,000 years of shield volcanism for MaunaKea volcano. There is considerable Pb and Sr iso-topic overlap between Pu?u ?O?o lavas and lavasfrom Mauna Kea and Lo?ihi volcanoes. However,the Pb-Sr isotopic trend for the later Pu?u ?O?olavas (1988?2010) is oblique to the array definedby Hawaiian shield lavas. Thus, each Hawaiianvolcano appears to have an isotopically distinctsource.0.70320.70330.70340.70350.70360.70370.70380.70390.704018.0 18.1 18.2 18.3 18.4 18.5 18.6 18.7 18.80.703450.703500.703550.703600.703650.7037018.2 18.3 18.4 18.5 18.6 18.7 18.8Mauna LoaL??ihiMauna KeaMauna LoaMauna KeaL??ihiSalt Lake Crater xenolithsPu?u ???? (1983-2010) Mauna Kea (HSDP-2)Kea-hi8 arrayKea-mid8 arrayKea-low8 array206Pb/204Pb87Sr/ 86Sr206Pb/204Pb87Sr/ 86Sr198319851986Hilina BenchHilina BenchHistorical K?lauea summitUwekahuna Bluff (summit)(Pietruszka and Garcia, 1999)Prehistoric K?lauea(Abouchami et al., 2005)(Marske et al., 2007)area of panel b)Uwekahuna Bluff (summit)200320100.4-5 kaHistorical and PrehistoricK?lauea summit not shownHistorical K?lauea summit(b)(a)Figure 12. Plots of 206Pb/204Pb and 87Sr/86Sr for Pu?u ?O?o lavas compared to some other Hawaiian shieldvolcanoes. Figure 12b Expanded scale of dashed box in Figure 12a. Colored symbols are lavas from K?laueaVolcano. Fields are for Mauna Loa, Lo?ihi, Mauna Kea and submarine K?lauea (Hilina Bench; >40 ka).References for data sources are listed in the supporting information.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854870Acknowledgments[41] We thank Jane Barling, Bruno Kieffer, and Vivian Laifor their assistance with analyses at PCIGR at UBC, KyleTanguichi and Adonara Murek for sample preparation andcuration at the University of Hawai?i, and J. M. Rhodes forXRF analyses at University of Massachusetts. Claude Maer-schalk assisted with Pb and Sr column chemistry for a subsetof samples. Daniel Heaton provided assistance with severalisotope analyses at San Diego State University. We appreciatereviews by Julie Prytulak, Joel Baker, and Christoph Beier.This research was supported by grants from the National Sci-ence Foundation to M. Garcia (EAR11?18741) and A. Pie-truszka (EAR11?18738). This paper is SOEST ContributionNo. 8939.ReferencesAbouchami, W., S. J. G. Galer, and A. W. Hofmann (2000),High precision lead isotope systematics of lavas from theHawaiian Scientific Drilling Project, Chem. Geol., 169, 187?209, doi:10.1016/S0009?2541(00)00328-4.Abouchami, W., A. W. Hofmann, S. J. G. Galer, F. A. Frey, J.Eisele, and M. Feigenson (2005), Lead isotopes reveal bilat-eral asymmetry and vertical continuity in the Hawaiian man-tle plume, Nature, 434, 851?856, doi:10.1038/nature03402.Albare`de, F., B. Luais, G. Fitton, M. Semet, E. Kaminski, B. G.J. Upton, P. Bache`lery, and J.-L. Cheminee (1997), The geo-chemical regimes of Piton de la Fournaise Volcano (Reunion)during the last 530,000 years, J. Petrol., 38, 171?201.Blichert-Toft, J., and F. Albare`de (2009), Mixing of isotopicheterogeneities in the Mauna Kea plume conduit, EarthPlanet. Sci. Lett., 282, 190?200, doi:10.1016/j.epsl.2009.03.015.Blichert-Toft, J., D. Weis, C. Maerschalk, A. Agranier, and F.Albare`de (2003), Hawaiian hot spot dynamics as inferredfrom the Hf and Pb isotope evolution of Mauna Kea volcano,Geochem. Geophys. Geosyst., 4(2), 8704, doi:10.1029/2002GC000340.Bryce, J. G., D. J. DePaolo, and J. C. Lassiter (2005), Geo-chemical structure of the Hawaiian plume: Sr, Nd, and Osisotopes in the 2.8 km HSDP-2 section of Mauna Kea vol-cano, Geochem. Geophys. Geosyst., 6, Q09G18,doi:10.1029/2004GC000809.Caroff, M., R. C. Maury, P. Vidal, G. Guille, C. Dupuy, J. Cot-ten, H. Guillou, and P.-Y. Gillot (1995), Rapid temporalchanges in ocean island basalt composition: Evidence froman 800 m deep drill hole in Eiao Shield (Marquesas), J. Pet-rol., 36(5), 1333?1365, doi:10.1093/petrology/36.5.1333.Chen, C. Y., and F. A. Frey (1985), Trace element and isotopicgeochemistry of lavas from Haleakala Volcano, east Maui,Hawai?i : Implications for the origin of Hawaiian basalts, J.Geophys. Res., 90, 8743?8768.Chen, C.-Y., F. A. Frey, J. M. Rhodes, and R. M. Easton(1996), Temporal geochemical evolution of K?lauea Vol-cano: Comparison of Hilina and Puna basalt, in Earth Proc-esses, Reading the Isotopic Code, edited by A. Basu and S.R. Hart, pp. 161?181, AGU, Washington, D. C.Cordier, C., M. Benoit, C. Hemond, J. Dyment, B. Le Gall, A.Briais, and M. Kitazawa (2010), Time scales of melt extrac-tion revealed by distribution of lava composition across aridge axis, Geochem. Geophys. Geosyst., 11, Q0AC06,doi:10.1029/2010gc003074.DePaolo, D. J., and E. M. Stolper (1996), Models of Hawaiianvolcano growth and plume structure: Implications from theHawai?i Scientific Drilling Project, J. Geophys. Res., 101,11,643?11,655.Dvorak, J. J., and D. Dzurisin (1993), Variations in magmasupply rate at K?lauea Volcano, Hawai?i, J. Geophys. Res.,98, 22,255?22,268.Eaton, J. P., and K. J. Murata (1960), How volcanoes grow[Hawai?i], Science, 132, 925?938.Eiler, J. M. (2001), Oxygen isotope variations of basaltic lavasand upper mantle rocks, Rev. Mineral. Geochem., 43(1),319?364, doi:10.2138/gsrmg.43.1.319.Eisele, J. R., W. Abouchami, S. J. G. Galer, and A. W. Hof-mann (2003), The 320 kyr Pb isotope evolution of MaunaKea lavas recorded in the HSDP-2 drill core, Geochem. Geo-phys. Geosyst., 4(5), 8710, doi:10.1029/2002GC000339.Farnetani, C. G., and A. W. Hofmann (2009), Dynamics andinternal structure of a lower mantle plume conduit, EarthPlanet. Sci. Lett., 282(1?4), 314?322, doi:10.1016/j.epsl.2009.03.035.Farnetani, C. G., and A. W. Hofmann (2010), Dynamics andinternal structure of the Hawaiian plume, Earth Planet. Sci.Lett., 295, 231?240, doi:10.1016/j.epsl.2010.04.005.Garcia, M. O., R. A. Ho, J. M. Rhodes, and E. W. Wolfe(1989), Petrologic constraints on rift-zone processes:Results from episode 1 of the Pu?u ?O?o eruption of K?laueavolcano, Hawai?i, Bull. Volcanol., 52, 81?96.Garcia, M. O., J. M. Rhodes, E. W. Wolfe, G. E. Ulrich, and R.A. Ho (1992), Petrology of lavas from episodes 2?47 of thePu?u ?O?o eruption of K?lauea Volcano, Hawai?i : Evaluationof magmatic processes, Bull. Volcanol., 55(1), 1?16.Garcia, M. O., J. M. Rhodes, F. A. Trusdell, and A. J. Pie-truszka (1996), Petrology of lavas from the Pu?u ?O?o erup-tion of K?lauea Volcano: III The Kupaianaha episode(1986?1992), Bull. Volcanol., 58, 359?379.Garcia, M. O., E. Ito, J. M. Eiler, and A. J. Pietruszka (1998),Crustal contamination of K?lauea Volcano magmas revealedby oxygen isotope analyses of glass and olivine from Pu?u?O?o eruption lavas, J. Petrol., 39, 803?817.Garcia, M. O., A. J. Pietruszka, J. M. Rhodes, and K. Swanson(2000), Magmatic processes during the prolonged Pu?u ?O?oeruption of K?lauea Volcano, Hawai?i, J. Petrol., 41(7),967?990.Garcia, M. O., A. J. Pietruszka, and J. M. Rhodes (2003), Apetrologic perspective of K?lauea Volcano?s summit magmareservoir, J. Petrol., 44(12), 2313?2339, doi:10.1093/petrol-ogy/egg079.Garcia, M. O., E. Ito, and J. M. Eiler (2008), Oxygen isotopeevidence for chemical interaction of K?lauea historical mag-mas with basement rocks, J. Petrol., 49(4), 757?769,doi :10.1093/petrology/egm034.Hanyu, T., J.-I. Kimura, M. Katakuse, A. T. Calvert, T. W. Sis-son, and S. I. Nakai (2010), Source materials for inceptionstage Hawaiian magmas: Pb-He isotope variations for earlyK?lauea, Geochem. Geophys. Geosyst., 11, Q0AC01,doi:10.1029/2009gc002760.Harmon, R. S., and J. Hoefs (1995), Oxygen isotope heteroge-neity of the mantle deduced from global 18O systematics ofbasalts from different geotectonic settings, Contrib. Mineral.Petrol., 120(1), 95?114.Hauri, E. H. (1996), Major-element variability in the Hawaiianmantle plume, Nature, 382, 415?419.Heliker, C., and T. N. Mattox (2003), The first two decades ofthe Pu?u ?O?o -Kupaianaha eruption; chronology andselected bibliography, in The Pu?u ?O?o -Kupaianaha Erup-tion of K?lauea Volcano, Hawai?i: The First 20 Years, editedGREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854871by C. Heliker et al., pp. 1?28, U.S. Geol. Surv. Prof. Pap.1676.Hofmann, A. W., M. D. Feigenson, and I. Raczek (1984), Casestudies on the orgin of basalt : III Petrogenesis of the MaunaUlu eruption, K?lauea, 1969?1971, Contrib. Mineral. Pet-rol., 88, 24?35.Holcomb, R. T. (1987), Eruptive history and long-term behav-ior of K?lauea Volcano, in Volcanism in Hawai?i, edited byR. W. Decker et al., pp. 261?350, U.S. Geol. Surv. Prof. Pap.1350.Jackson, M. G., D. Weis, and S. Huang (2012), Major elementvariations in Hawaiian shield lavas: Source features and per-spectives from global ocean island basalt (OIB) systematics,Geochem. Geophys. Geosyst., 13, Q09009, doi:10.1029/2012GC004268.Johnson, D. J., A. A. Eggers, and M. Bagnardi (2010), Shallowmagma accumulation at K?lauea Volcano, Hawai?i, revealedby microgravity surveys, Geology 38(12), 1139?1142,doi:10.1130/G31323.1.Kimura, J.-I., T. W. Sisson, N. Nakano, M. L. Coombs, and P.W. Lipman (2006), Isotope geochemistry of early K?laueamagmas from the submarine Hilina bench: The nature of theHilina mantle component, J. Volcanol. Geotherm. Res.,151(1?3), 51?72, doi :10.1016/j.jvolgeores.2005.07.024.Kurz, M. D., W. J. Jenkins, and S. R. Hart (1982), Helium iso-topic systematics of oceanic islands and mantle heterogene-ity, Nature, 297, 43?47.Macdonald, G. A., A. T. Abbott, and F. L. Peterson (1983),Volcanoes in the Sea: The Geology of Hawai?i, 517 pp.,Univ. of Hawaii Press, Honolulu, Hawaii.Maclennan, J. (2008), Lead isotope variability in olivine-hosted melt inclusions from Iceland, Geochim. Cosmochim.Acta, 72(16), 4159?4176, doi:10.1016/j.gca.2008.05.034.Marske, J. P., A. J. Pietruszka, D. Weis, M. O. Garcia, and J.M. Rhodes (2007), Rapid passage of a small-scale mantleheterogeneity through the melting regions of K?lauea andMauna Loa volcanoes, Earth Planet. Sci. Lett., 259, 34?50,doi:10.1016/j.epsl.2007.04.026.Marske, J. P., M. O. Garcia, A. J. Pietruszka, J. M. Rhodes,and M. D. Norman (2008), Geochemical variations duringK?lauea?s Pu?u ?O?o eruption reveal a fine-scale mixture ofmantle heterogeneities within the Hawaiian plume, J. Pet-rol., 49(7), 1297?1318, doi:10.1093/petrology/egn025.Montgomery-Brown, E. K., D. K. Sinnett, P. S. M. Poland, T.Orr, H. Zebker, and A. Miklius (2010), Geodetic evidencefor en echelon dike emplacement and concurrent slow slipduring the June 2007 intrusion and eruption at K?lauea vol-cano, Hawai?i, J. Geophys. Res., 115, B07405, doi:10.1029/2009JB006658.Nobre Silva, I. G., D. Weis, and J. S. Scoates (2013), Isotopicsystematics of the early Mauna Kea shield phase and insightinto the deep mantle beneath the Pacific Ocean, Geochem.Geophys. Geosyst., 14(3), 659?676, doi:10.1029/2012GC004373.Pietruszka, A. J., and M. O. Garcia (1999), A rapid fluctuationin the mantle source and melting history of K?lauea Volcanoinferred from the geochemistry of its historical summit lavas(1790?1982), J. Petrol., 48(8), 1321?1342.Pietruszka, A. J., K. H. Rubin, and M. O. Garcia (2001),226Ra-230Th-238U disequilibria of historical K?lauea lavas(1790?1982) and the dynamics of mantle melting within theHawaiian plume, Earth Planet. Sci. Lett., 186, 15?31.Pietruszka, A. J., E. H. Hauri, R. W. Carlson, and M. O. Garcia(2006), Remelting of recently depleted mantle within theHawaiian plume inferred from the 226Ra-230Th-238U disequi-libria of Pu?u ?O?o eruption lavas, Earth Planet. Sci. Lett.,244(1?2), 155?169, doi:10.1016/j.epsl.2006.01.039.Pietruszka, A. J., E. H. Hauri, and J. Blichert-Toft (2009),Crustal contamination of mantle-derived magmas within Pi-ton de la Fournaise Volcano, Reunion Island, J. Petrol.,50(4), 661?684, doi :10.1093/petrology/egp016.Pietruszka, A. J., M. D. Norman, M. O. Garcia, J. P. Marske,and D. H. Burns (2013), Chemical heterogeneity in the Ha-waiian mantle plume from the alteration and dehydration ofrecycled oceanic crust, Earth Planet. Sci. Lett., 361, 298?309, doi:10.1016/j.epsl.2012.10.030.Poland, M., A. Miklius, T. Orr, A. J. Sutton, C. Thornber, andD. Wilson (2008), New episodes of volcanism at K?laueaVolcano, Hawai?i, Eos Trans. AGU, 89(5), 37?38,doi :10.1029/2008EO050001.Poland, M. P., A. Miklius, A. J. Sutton, and C. R. Thornber(2012), A mantle-driven surge in magma supply to K?laueaVolcano during 2003?2007, Nat. Geosci., 5, 295?300,doi :10.1038/NGEO1426.Reiners, P. W. (2002), Temporal-compositional trends inintraplate basalt eruptions: Implications for mantle hetero-geneity and melting processes, Geochem. Geophys. Geo-syst., 3(2), 1011, doi:10.1029/2001GC000250.Ren, Z.-y., T. Hanyu, T. Miyazaki, Q. Chang, H. Kawabata, T.Takahashi, Y. Hirahara, A. R. L. Nichols, and Y. Tatsumi(2009), Geochemical differences of the Hawaiian shieldlavas: Implications for melting process in the heterogeneousHawaiian plume, J. Petrol., 50(8), 1553?573, doi:10.1093/petrology/egp041.Rhodes, J. M., and S. R. Hart (1995), Episodic trace elementand isotopic variations in Mauna Loa lavas: Implications formagma and plume dynamics, in Mauna Loa Revealed:Structure, Composition, History and Hazards, edited by J.M. Rhodes and J. P. Lockwood, pp. 263?288, AGU, Wash-ington, D. C.Rhodes, J. M., and M. J. Vollinger (2004), Composition of ba-saltic lavas sampled by phase-2 of the Hawai?i ScientificDrilling Project: Geochemical stratigraphy and magmatypes, Geochem. Geophys. Geosyst., 5, Q03G13,doi:10.1029/2002GC000434.Rhodes, J. M., S. Huang, F. A. Frey, M. Pringle, and G. Xu(2012), Compositional diversity of Mauna Kea shield lavasrecovered by the Hawai?i Scientific Drilling Project: Infer-ences on source lithology, magma supply, and the role ofmultiple volcanoes, Geochem. Geophys. Geosyst., 13,Q03014, doi:10.1029/2011GC003812.Ryan, M. P. (1987), Elasticity and contractancy of Hawaiianolivine tholeiite and its role in the stability and structuralevolution of subcaldera magma reservoirs and rift systems,in Volcanism in Hawai?i, edited by R. W. Decker et al., pp.1395?1448, U.S. Geol. Surv. Prof. Pap. 1350.Shamberger, P. J., and M. O. Garcia (2007), Geochemicalmodeling of magma mixing and magma reservoir volumesduring early episodes of K?lauea Volcano?s Pu?u ?O?o erup-tion, Bull. Volcanol., 69, 345?352, doi:10.1007/s00445-006-0074-5.Sims, K. W. W., M. T. Murrell, D. J. DePaolo, W. S. Bal-dridge, S. J. Goldstein, D. Clague, and M. Jull (1999), Poros-ity of the melting zone and variations in the solid mantleupwelling rate beneath Hawai?i : Inferences from238U-230Th-226Ra and 235U-231Pa disequilibria, Geochim.Cosmochim. Acta, 63(23), 4119?4138.Sobolev, A. V., A. W. Hofmann, S. Sobolev, and I. K. Nikogo-sian (2005), An olivine-free mantle source of Hawaiianshield basalts, Nature, 434, 590?595.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854872Sutton, J. A., T. Elias, and J. Kauahikaua (2003), Lava-effu-sion rates for the Pu?u ?O?o -Kupaianaha eruption derivedfrom SO2 emissions and very low frequency (VLF) meas-urements, in The Pu?u ?O?o -Kupaianaha Eruption ofK?lauea Volcano, Hawai?i: The First 20 Years, edited byC. Heliker et al., pp. 137?148, U.S. Geol. Surv. Prof. Pap.1676.Thornber, C. R. (2003), Magma-reservoir processes revealedby geochemistry of the Pu?u ?O?o-Kupaianaha eruption, inThe Pu?u ?O?o -Kupaianaha Eruption of K?lauea Volcano,Hawai?i: The First 20 Years, edited by C. Heliker et al., pp.121?136, U.S. Geol. Surv. Prof. Pap. 1676.Thornber, C. R., C. Heliker, D. R. Sherrod, J. P. Kauahikaua,A. Miklius, P. G. Okubo, F. A. Trusdell, J. R. Budahn, W. I.Ridley, and G. P. Meeker (2003), K?lauea East Rift Zonemagmatism: An episode 54 perspective, J. Petrol., 44(9),1525?1559, doi:10.1093/petrology/egg048.Tilling, R. I., and J. J. Dvorak (1993), Anatomy of a basalticvolcano, Nature, 363, 125?133.Vlastelic, I., T. Staudacher, and M. Semet (2005), Rapidchange of lava composition from 1998 to 2002 at Piton de laFournaise (Reunion) inferred from Pb isotopes and trace ele-ments: Evidence for variable crustal contamination, J. Pet-rol., 46(1), 79?107, doi :10.1093/petrology/egh062.Vlastelic, I., A. Peltier, and T. Staudacher (2007), Short-term(1998?2006) fluctuations of Pb isotopes at Piton de la Four-naise volcano (Reunion Island): Origins and constraints onthe size and shape of the magma reservoir, Chem. Geol.,244, 202?220, doi:10.1016/j.chemgeo.2007.06.015.Watson, S., and D. McKenzie (1991), Melt generation byplumes: A study of Hawaiian volcanism, J. Petrol., 32(3),501?537.Weis, D., M. O. Garcia, J. M. Rhodes, A. M. Jellinek, and J. S.Scoates (2011), Role of the deep mantle in generating thecompositional asymmetry of the Hawaiian mantle plume,Nat. Geosci., 4, 831?838, doi:10.1038/ngeo1328.Wolfe, E. W., M. O. Garcia, D. B. Jackson, R. Y. Koyanagi, C.A. Neal, and A. T. Okamura (1987), The Pu?u ?O?o eruptionof K?lauea Volcano, episodes 1?20, January 3, 1983 to June8, 1984, in Volcanism in Hawai?i, edited by R. W. Deckeret al., pp. 471?508, U.S. Geol. Surv. Prof. Pap. 1350.Wright, T. L. (1971), Chemistry of K?lauea and Mauna Loalava in space and time, U.S. Geol. Surv. Prof. Pap., 735,1?45.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854873 Temporal geochemical variations in lavas fromK?lauea?s Pu?u ?O?o eruption (1983?2010): Cyclicvariations from melting of source heterogeneitiesAndrew R. GreeneDepartment of Natural Sciences, Hawai?i Pacific University, 45-045 Kamehameha Hwy, Kane?ohe, Hawaii,96744, USA (agreene@hpu.edu)Michael O. GarciaDepartment of Geology and Geophysics, University of Hawai?i, Honolulu, Hawaii, USAAaron J. PietruszkaDepartment of Geological Sciences, San Diego State University, San Diego, California, USANow at U. S. Geological Survey, Denver Federal Center, Denver, Colorado, USADominique WeisPacific Centre for Isotopic and Geochemical Research, Department of Earth and Ocean Sciences, University ofBritish Columbia, Vancouver, British Columbia, CanadaJared P. MarskeDepartment of Terrestrial Magnetism, Carnegie Institution, Washington, D.C., USAMichael J. VollingerRonald B. Gilmore XRF Lab, Department of Geosciences, University of Massachusetts, Amherst, Massachusetts, USAJohn EilerPlanetary and Geological Sciences Institute, California Institute of Technology, Pasadena, California, USA[1] Geochemical time series analysis of lavas from K?lauea?s ongoing Pu?u ?O?o eruption chroniclemantle and crustal processes during a single, prolonged (1983 to present) magmatic event, which hasshown nearly two-fold variation in lava effusion rates. Here we present an update of our ongoingmonitoring of the geochemical variations of Pu?u ?O?o lavas for the entire eruption through 2010. Oxygenisotope measurements on Pu?u ?O?o lavas show a remarkable range (18O values of 4.6?5.6%), which areinterpreted to reflect moderate levels of oxygen isotope exchange with or crustal contamination byhydrothermally altered K?lauea lavas, probably in the shallow reservoir under the Pu?u ?O?o vent. Thisprocess has not measurably affected ratios of radiogenic isotope or incompatible trace elements, whichare thought to vary due to mantle-derived changes in the composition of the parental magma delivered tothe volcano. High-precision Pb and Sr isotopic measurements were performed on lavas erupted at 6month intervals since 1983 to provide insights about melting dynamics and the compositional structure ofthe Hawaiian plume. The new results show systematic variations of Pb and Sr isotope ratios thatcontinued the long-term compositional trend for K?lauea until 1990. Afterward, Pb isotope ratios showtwo cycles with 10 year periods, whereas the Sr isotope ratios continued to increase until 2003 andthen shifted toward slightly less radiogenic values. The short-term periodicity of Pb isotope ratios mayreflect melt extraction from mantle with a fine-scale pattern of repeating source heterogeneities or strands,which are about 1?3 km in diameter. Over the last 30 years, Pu?u ?O?o lavas show 15% and 25% of the? 2013. American Geophysical Union. All Rights Reserved. 4849ArticleVolume 14, Number 1115 November 2013doi: 10.1002/ggge.20285ISSN: 1525-2027known isotopic variation for K?lauea and Mauna Kea, respectively. This observation illustrates that thedominant time scale of mantle-derived compositional variation for Hawaiian lavas is years to decades.Components: 13,235 words, 12 figures, 2 tables.Keywords: Hawaiian plume; tholeiitic volcanism; melt extraction; oceanic island.Index Terms: 3619 Magma genesis and partial melting: Mineralogy and Petrology; 3621 Mantle processes: Mineralogyand Petrology; 1025 Composition of the mantle: Geochemistry; 1037 Magma genesis and partial melting: Geochemistry;1038 Mantle processes: Geochemistry.Received 5 March 2013; Revised 9 September 2013; Accepted 7 October 2013; Published 15 November 2013.Greene, A. R., M. O. Garcia, A. J. Pietruszka, D. Weis, J. P. Marske, M. J. Vollinger, and J. Eiler (2013), Temporal geo-chemical variations in lavas from K?lauea?s Pu?u ?O?o eruption (1983?2010): Cyclic variations from melting of source het-erogeneities, Geochem. Geophys. Geosyst., 14, 4849?4873, doi:10.1002/ggge.20285.1. Introduction[2] K?lauea, on the Island of Hawai?i (Figure 1), isone of the most active and best-monitored volca-noes in the world [Heliker and Mattox, 2003;Wolfe et al., 1987]. The ongoing Pu?u ?O?o erup-tion on K?lauea?s East Rift Zone (Figure 1) hasbeen active nearly continuously for 30 years and isHawai?i?s longest and most voluminous (4 km3)historical eruption [Poland et al., 2012]. The con-tinuous petrologic and geochemical monitoring ofthe Pu?u ?O?o eruption [e.g., Garcia et al., 2000;Marske et al., 2008; Thornber, 2003] has allowedus to witness the dynamic changes in the meltingprocess and mantle source composition during asingle, long-lasting magmatic event. Extractionand transport of melt through open channels dur-ing the Pu?u ?O?o eruption has efficiently transmit-ted variations of melting in the heterogeneoussource to lavas erupted at the surface without sig-nificant pooling and homogenization, preservingshort-term isotopic and geochemical variations[Pietruszka et al., 2006].[3] The long-term geochemical variations (manythousands of years) of Hawaiian and other oceanisland basalts has been well documented due todetailed geochemical work on 3? km deep drillcore [e.g., Albare`de et al., 1997; Blichert-Toftet al., 2003; Bryce et al., 2005; Caroff et al.,1995; Rhodes et al., 2012]. These studieschronicle processes on millennium time scales butmiss potential short-term variations (<100 years),which may provide better insights into meltingand crustal processes. K?lauea?s historical (1823?1982) and prehistoric (AD 900?1400) summitlavas reveal rapid and systematic changes in Pb,Sr, Nd, O, and U-series isotope ratios on a timescale of decades to centuries [Garcia et al., 2003,2008; Marske et al., 2007; Pietruszka and Garcia,1999; Pietruszka et al., 2001]. The Pu?u ?O?oeruption (sampled from hourly to monthly) showscompositional change over hours (in rare cases)for major elements to a few years for isotope ratios[Garcia et al., 2000; Marske et al., 2008]. Thelong duration and vigorous activity (0.35  106m3 of lava erupted daily) of Pu?u ?O?o [e.g., Suttonet al., 2003] provides a rare opportunity to lookbeyond the shallow-level crustal processes associ-ated with the short eruptions (days to weeks) thattypify many active basaltic volcanoes (e.g., MaunaLoa, Etna, Piton de la Fournaise, Karthala,Grimsv?tn) and into the mantle. In addition, Pu?u?O?o magmas may partially bypass K?lauea?s sum-mit reservoir (2?6 km depth beneath the summitcaldera) on their way to the East Rift Zone, andmostly avoid its buffering effects [Garcia et al.,2000]. Therefore the Pu?u ?O?o eruption is one ofEarth?s best probes for sampling mantle-derivedmelts almost continuously over nearly threedecades.[4] The study of isotopic and geochemical varia-tion in magmatic events over short time scales(months to years) in oceanic island lavas improvesour temporal and spatial resolution of meltingprocesses and the chemical structure of mantleplumes [Abouchami et al., 2000; Eisele et al.,2003; Hofmann et al., 1984; Vlastelic et al.,2005]. Recent studies of Pb, Sr, and Nd isotoperatios for part of the Pu?u ?O?o eruption [Marskeet al., 2008] and other active basaltic volcanoes[e.g., Piton de la Fournaise; Vlastelic et al., 2005]detected rapid and systematic changes over shortGREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854850time scales (years to decades) resulting fromsource heterogeneity, and variations in crustalprocesses. Here we present new high-precision Pb,Sr, and O isotope ratios, and major- and trace-element abundances for Pu?u ?O?o lavas eruptedbetween January 1983 and June 2010. These data197418231919-20 1971Mauna Ulu   1969-74197719551955196018401790East Rift ZoneSummitMakaopuhi1983-present1790179010 km18681955Pu?u ?O?oNCraterCraterEp. 54Southwest Rift ZoneMauna Loa1983-2010 Pu'u '?'? KupaianahaPACIFIC OCEANNorth Pu'u '?'?N?pauCraterKalapanaEp. 56 February 17, 1992 - Feb. 11, 2010Episodes 50-58July 20, 1986 -  February 17, 1992Episodes 48-49January 3, 1983 -July 20, 1986Episodes 1-48episodicfountaining(mostly centralvent)gentle effusion(lava shield andtube-fed pahoehoe)Kupaianaha Pu?u ?O?oPu?u ?O?oJan. July Feb.Feb.1983 1986 19921997(3.5 years) (5.5 years) (18 years) perched channels,rootless shields,fissure eruption20102007Episode 56(Magma supply ratedoubled)June(uprift)flank  vent eruptions(nearly continuous)(lava shield and tube-fed pahoehoe)2003Episode 54(uprift)010km016miK?lauea Caldera1790-1982kmHalemaumauPu?u ???? East Rift ZoneMauna Ulu1969-1974Makaopuhi N?pau(a)KupaianahaEp. 54Ep. 56(b)123storeddikeJune, 2007?Jan.1997Ep. 58Ep. 58K?lauea Caldera0KohalaMauna Loa K?lauea Mauna KeaL??ihi 50010001500200025003000200015005001000100020003000250050002500 50005500155?156?19?20?Hilo RidgeEast rift zone40 km20Hawai?i Hual?lai(c)(d)Figure 1. Map of flow fields from the Pu?u ?O?o-Kupaianaha eruption on the East Rift Zone of K?lauea Vol-cano from 1983 to 2010 and historical flows, with a timeline summarizing the predominant style of eruptiveactivity. (a) A schematic cross section of summit and East Rift Zone shows the proposed magmatic plumbingsystem for K?lauea Volcano, with locations for episodes 54 and 56 uprift of Pu?u ?O?o. Mantle-derived magmafor this eruption is thought to partially bypass the summit reservoir based on the rapid changes in lava compo-sition [Garcia et al., 1996]. (b) Map of K?lauea East Rift Zone with flow fields from intervals of the Pu?u?O?o eruption. Legend shows episodes in each interval of eruptive activity. Map provided by USGS HawaiianVolcano Observatory. (c) Map of the island of Hawai?i with area of map in Figure 1b indicated with box. (d)Timeline of the Pu?u ?O?o eruption. Episode 54 was a fissure eruption in and downrift of Napau Crater thatoccurred over 23 h in January 1997, following the collapse of the Pu?u ?O?o cone. Episode 56 was a brief (<1day) fissure eruption northeast of Makaopuhi Crater (uprift of Pu?u ?O?o) that occurred in June 2007, coincid-ing with an intrusion and collapse of Pu?u ?O?o crater floor. Dashed lines between 2003 and 2007 indicate pe-riod when magma supply rate nearly doubled [Poland et al., 2012].GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854851are combined with previously published high-precision isotope and trace-element data from1998 to 2005 Pu?u ?O?o lavas [Marske et al.,2008]. The new Pb and Sr isotope and inductivelycoupled plasma mass spectrometry (ICP-MS) dataprovide a record of isotopic and geochemical vari-ation of Pu?u ?O?o lavas at 6 month intervals,whereas X-Ray fluorescence (XRF) data was col-lected at 2 week intervals. This time series anal-ysis of Pu?u ?O?o lavas allows us to distinguish thechanging roles of mantle and crustal processes ingreat detail. The new Pb and Sr isotope ratios areused to assess the short-term expression of mantlesource components throughout the course of theeruption and to evaluate the effects on lava com-position of recent doubling of the magma supply[2003?2007; Poland et al., 2012]. These resultsare compared to the longer-term variations forK?lauea and other Hawaiian shield volcanoes.2. Geologic Background of K?laueaVolcano and the Pu?u ?O?o Eruption(1983?2010)[5] K?lauea Volcano is currently in the middle ofits shield-building stage [DePaolo and Stolper,1996], erupting tholeiitic lava at a rate of 0.13km3/yr [Sutton et al., 2003], one of the highestrates of any volcano on Earth. K?lauea rises 1240m above sea-level on the southern flank of itslarger neighbor, Mauna Loa (4168 m; Figure 1).Geochemical evidence favors a deep mantle plumeorigin for Hawaiian magmas [e.g., Kurz et al.,1982; Weis et al., 2011]. Shield stage magmas arethought to originate from partial melting at mantledepths of 70?120 km within the upper Hawaiianplume [Watson and McKenzie, 1991]. Magmas areextracted from the upwelling mantle within themelting region and transported through chemicallyisolated channels towards the surface [Pietruszkaet al., 2006]. These pooled melts ascend throughthe lithosphere via a primary conduit into a shal-low (2?6 km) magmatic complex within K?lauea[Eaton and Murata, 1960; Ryan, 1987; Tillingand Dvorak, 1993; Wright, 1971]. K?lauea erup-tions occur in and around its summit caldera andEast and Southwest Rift Zones. Approximately90% of the subaerial surface of K?lauea Volcano iscovered with tholeiitic lava less than 1100 yearsold [Holcomb, 1987]. Prior to 1955, historical(post-1820) eruptions on K?lauea occurred mostlyat or near the summit [Macdonald et al., 1983].Subsequently, rift zone eruptions became morecommon, especially along the East Rift Zone,including the 1969?1974 Mauna Ulu eruption, themost voluminous historical eruption prior to Pu?u?O?o [Macdonald et al., 1983].[6] The Pu?u ?O?o-Kupaianaha eruption (referredto as the Pu?u ?O?o eruption throughout this paper)began on 2 January 1983 with the intrusion of adike within K?lauea?s East Rift Zone, although itwas preceded by months of intrusions from thesummit into the rift zone [Wolfe et al., 1987]. Itwas followed 24 h later by eruptive activity alonga discontinuous 7 km long fissure, which localizedto a central vent, Pu?u ?O?o (Figure 1 and Table 1).The eruption can be categorized into three broadphases based on eruptive style and location: (1)1983?1986: brief (mostly less than 24 h), episodiceruptions (24 day average repose between erup-tions) with fountaining up to 400 m, mainly fromthe Pu?u ?O?o vent [Heliker and Mattox, 2003]; (2)1986?1992: nearly continuous effusion from theKupaianaha vent, which was considered to have ashallow (<100 m deep) conduit connection withPu?u ?O?o, 3 km uprift [Garcia et al., 1996]; and(3) 1992?2010: nearly continuous effusion mostlyfrom vents within, and on the southwest and eastflanks of Pu?u ?O?o, and from rootless shields 2km east of Pu?u ?O?o [Poland et al., 2008]. Thispattern was interrupted on 29 January 1997 (epi-sode 54) by the 150 m collapse of the crater floorinside the Pu?u ?O?o cone, and propagation oferuptive fissures 4 km uprift (west) of Pu?u ?O?o,which were active for less than a day [Heliker andMattox, 2003]. This event was followed by a 6week hiatus in effusive activity, although glowreturned to the Pu?u ?O?o vent on 24 February1997 (Table 1). Afterward, and until June 2007,lava erupted nearly continuously from flank ventson Pu?u ?O?o (episode 55). On 19 June 2007, adike intrusion in the upper East Rift Zone resultedin a brief, small (1500 m3) eruption (episode 56)6 km uprift from Pu?u ?O?o [Montgomery-Brownet al., 2010], which was followed by a 2 week hia-tus in effusion [Poland et al., 2008]. Lava produc-tion resumed for 3 weeks in and around Pu?u ?O?ocone (episode 57) until 21 July 2007, when a fis-sure opened on the east flank of Pu?u ?O?o andpropagated eastward towards Kupaianaha (Figure1 and Table 1). This marked the beginning of epi-sode 58 [Poland et al., 2008], which continuedthrough the end of 2010 mostly as tube-fed flowsfrom a vent 2 km east of Pu?u ?O?o. The othernotable K?lauea eruptive activity during the Pu?u?O?o eruption is an ongoing summit eruption thatstarted in March 2008 [Johnson et al., 2010].GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854852Table1.SummaryofthePu?u? O? oEruptionaPrimaryVentEpisodeEpisodeStartDateReposeLength(Days)EpisodeLengthVolume(106xm3)EruptionRate(103m3/day)Vent/LocationBriefDescriptionPu?u? O? o13Jan1983Start20days14?Fissure1;activitylocalizedatPu?uHaluluandPu?uKahauleaeastofPu?u? O? oInitialfissureopenedinNapauCraterafterseismicswarmpropagateddownERZ;fissuresextended8km;fissureslocalizedto1kmnearPu?uKahaulea;fountainsfromPu?uHalulubuilta60m-highconePu?u? O? o2?4710Feb19838?65(betweenepisodes)~3.8years371300MostlyPu?u? O? o;Episodes2?3localizedatPu?uHaluluandPu?uKahauleaeastofPu?u? O? o;Epi-sode4?47PuuOoprimaryventEpisodicfirefountaining;episodesmostly<24hlongsepa-ratedbyanreposelengthaverageof24days;effusionratesincreasedthoughepisode39;maximumlavafoun-tainof470mhigh;firstyearchangedfromlowfountainsandpahoehoeriverstohighfountainsand?a?afans;fountain-fed?a?abyepisode20;conebuilt255mhighand1.4kmindiameter;summitinflatedbetweenfoun-tainingepisodesanddeflatedduringepisodesKupaianaha4818July198624~5.5years500400?0.5Kupaianaha;fissure3kmeastofPu?u? O? oFissuresfirstopenedatthebaseofPu?u? O? oand22hlateropened3kmdownriftataventtobenamedKupaianaha;5.5yearsofnearlycontinuousgentleeffusion;largelavapondformedovervent(140m300m);broadlavashieldformedandtube-fedpahoehoewascommonwaylavaspreadtocoast;homesdestroyedintownofKala-pana;lavatubestoseamid-1987to1989;lavaenteredseaduring68%ofepisode;lavaactiveinPu?u? O? ocraterduringmostofepisodePu?u? O? o498Nov1991None18days110.6Fissure2betweenPu?u? O? oandKupaianahaFissuresopenedonPu?u? O? oandpropagatedtoKupaia-naha;outputwanedduringepisode;gentleeffusion,lavashieldandtube-fedpahoehoe;fissurevents,pahoehoePu?u? O? o5017Feb19921115days3?Pu?u? O? oflank;radialfissureonwestflankofPu?u? O? oconeEruptionreturnedtoPu?u? O? o;radialfissuresonflankofcone;flankventeruptions;lavashieldbankedupagainstthesouthandwestPu?u? O? ocone;spatterconesformedovervents;mostlytube-fedpahoehoe;continuousquieteffusionPu?u? O? o517Mar19924161days32300Pu?u? O? oflankFlankventeruptions;mostlytube-fedpahoehoetothesea;continuousquieteffusion;lavashieldbankedupagainstthesouthandwestPu?u? O? oconePu?u? O? o523Oct1992None15days2300Pu?u? O? oflankFlankventeruptions;mostlytube-fedpahoehoetothesea;continuousquieteffusion;lavashieldbankedupagainstthesouthandwestPu?u? O? oconePu?u? O? o5320Feb1993None~4years535300Pu?u? O? oflankFlankventeruptions;mostlytube-fedpahoehoetothesea;continuousquieteffusion;lavashieldbankedupagainstthesouthandwestPu?u? O? ocone;collapsepitsformedonthesideofPu?u? O? oPu?u? O? o5429Jan1997None1day0.30.3Fissure3;2?4kmupriftofPu?u? O? o(1)LavalakeinsidethePu?u? O? oventdrainedandcraterfloordropped150m;(2)Pu?u? O? owestflankcollapsed;115mgapinwestsideofPu?u? O? o;(3)fissure4kmeruptedupriftfor1day,inanddownriftofNapauCrater,followedbylongesteruptivehiatussince1987(24days);distinctlavachemistryinvolvedmagmamixingwithdifferentiatedmagmastoredinriftzoneGREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854853Table1.(continued)PrimaryVentEpisodeEpisodeStartDateReposeLength(Days)EpisodeLengthVolume(106xm3)EruptionRate(103m3/day)Vent/LocationBriefDescriptionPu?u? O? o5524Feb19972410years265200?500Pu?u? O? oanditsflankLavaspilledfromcratertoformnewpond;lavaspilledfromcrateramonthlater;newflankventeruptionswestandsouthwestofcone;spatterconesonflankscrustedovertoproducemostlytube-fedpahoehoe;byJune1997lavaovertoppedthegapinwallofPu?u? O? oandflowedfromcraterforfirsttimein11years;flankventsunder-minedPu?u? O? oconeinDecember,1997;PukaNuicol-lapsepitformedonsouthwestflankofcone;31pausesoccurredduringepisode55Pu?u? O? o5619June2007None6h0.001450.00036250mlongfissureintheforestnortheastofKaneNuioHamo,approximately6kmwestofPu?u? O? oFather?sDayeruptionnearKaneNuioHamonorthofMakaopuhiCrater;magmasupplytoPu?u? O? owascutoffon17June2007;earthquakeswarmsindicatedmagmamovementintheupperERZ;spattereruptedfromfissureinforestedarea;smalllavaflow(200m50m)accom-paniedintrusioninERZ;craterfloorinPu?u? O? ocol-lapsedanderuptionshutoffPu?u? O? o571July200719daysNone0.82?1.2365Pu?u? O? ocraterAfterabouttwo-weeksofquiet,theeruptionbeganagainon1July.Lavabegantorefillthecrater.On8July,effu-sionwanedasthecraterbegantoupliftinapiston-likefashion.Thecraterthenbegantofillandreachedtowithin30moftheeasternrimofPu?u? O? ocraterbymid-July.Pu?u? O? o5821July2007None~4yearsended7March2011320(asoftheendof2009)FromfissureeastofPu?u? O? ocraterPerchedlavachannel,rootlessshields;forthefirsttimesince7February1992,lavabeginseruptingeastofPu?u? O? ocrater.ThanksgivingEvebreakout,lavabypasses21July2007channelanderuptsonchannelflank;5March2008oceanentryactiveforthefirsttimesinceJune2007;explosioninHalema?uma?uCrateratsummiton19March2008;June2008spatteringventsandasmallpondoflavainPu?u? O? o,lavafountainsgushfromtheTEBtubesystem,channelized?a? aflowsinRoyalGardens,andlargelittoralexplosionsatK ?lauea?soceanentrynearKalapana;Waikupanahaoceanentryactivethroughmuchof2009,andoccasionallyKupapa?uoceanentrytowesta Reposelengthreferstodurationofpausebetweeneruptiveepisodes.Episodeidentifiesoccurrencesoffountainingorlavaflowseparatedbyquiescentperiods.Volumeisdenserockequivalent(DRE)eruptedduringeachepisode.Datasources:Garciaetal.[2000]andreferencestherein,Wolfeetal.[1998],HelikerandMattox[2003].GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.2028548543. Description of Samples andAnalyses Performed in This Study[7] This study presents 52 new high-precision Pband Sr isotope analyses (from 1983 to 1997 and2006 to 2010), 11 new O isotope analyses (fromafter 1997; Table 2), and 13 new ICP-MS trace-element analyses of Pu?u ?O?o lava samples(mostly after 2005; data and analytical methodsare presented in the supporting information).1 NewXRF major- and trace-element analyses for 52Pu?u ?O?o lavas erupted from 2006 to 2010 arealso presented. In addition, new XRF trace-element analyses are given for samples eruptedprior to 1998, when a new, more precise XRFinstrument became available. Almost all of thesamples in this study were collected in a moltenstate and quenched with water to minimize poster-uption crystallization. The sample names are thedate that each lava sample was collected (e.g.,day-month-year), which is generally within a dayof its eruption when lava is flowing in open chan-nels on the surface or in lava tubes [e.g., Garciaet al., 2000] or up to a week or more when it isoozing within slowly advancing pahoehoe flows[K. Ho, personal communication, 2013]. Descrip-tions of the petrography of typical Pu?u ?O?o lavascan be found in Garcia et al. [1989, 1992, 1996,2000] and Marske et al. [2008]. Fourteen high-precision Pb and Sr isotope ratios for Pu?u ?O?olavas erupted from 1998 to 2005 from Marskeet al. [2008] and 15 O isotope analyses from Gar-cia et al. [1998] are listed in Table 2 and areincluded in plots for completeness.4. Temporal Geochemical Variationsin Lavas From 1983 to 2010[9] Early Pu?u ?O?o lavas (1983 to early 1985) re-cord rapid (hours to days) variations in major andcompatible trace-element abundances (Figure 2;Supporting information). These lavas show petro-graphic evidence for both crystal fractionation andmagma mixing [Wolfe et al., 1987; Garcia et al.,1992]. Crystal fractionation of olivine (with minorclinopyroxene and plagioclase, especially for 1983lavas) is the dominant process controlling short-term major-element variation in Pu?u ?O?o lavas[Garcia et al., 1992]. To remove the effects ofcrystal fractionation on parental magma composi-tions, major-element abundances of lavas contain-ing only olivine (MgO >7.0 wt %) werenormalized to 10 wt % MgO by the addition ofequilibrium composition olivine (98.5%) and spi-nel (1.5%) in 0.5 mol % steps, as described byGarcia et al. [2003]. The increases in MgO, CaO/Al2O3, and CaO/TiO2 and decreases in MgO-normalized incompatible element abundances(e.g., TiO2, K2O) between 1983 and early 1985(Figure 2) reflect mixing of new high-MgOmagma with decreasing amounts of a hybridmagma formed at the start of the eruption by mix-ing two differentiated, rift-zone stored magmas[Garcia et al., 1989, 1992]. Lavas erupted afterearly 1985 show no petrographic or geochemicalevidence for mixing [Shamberger and Garcia,2007] until the 1997 uprift eruption, which is dis-cussed below.[10] From 1985 to 1994, Pu?u ?O?o lavas show awide range in MgO reflecting the periodic hiatusesin eruptive activity [Garcia et al., 1992], and grad-ual changes in MgO-normalized concentrations ofmajor elements (TiO2 and K2O), and ratios ofmajor (e.g., CaO/Al2O3; Figure 2) and trace ele-ments (Nb/Y; Figure 3). In 1994, lavas began aperiod of increasing MgO-normalized SiO2 andoverall decreasing MgO-normalized TiO2 that per-sisted until 2001. Other geochemical parameterscontinued their long-term trends (e.g., decreasingCaO/Al2O3, MgO-normalized K2O, and Nb/Y,and increasing Zr/Nb; Figures 2 and 3). Starting inmid- to late 2003, there was an increase in lavaproduction with effusion rates doubling in 2005[Poland et al., 2012]. The lava MgO contentdecreased from 2003 to 2007 and was relativelylow (<7.5 wt %, mostly <7.2 wt %) with limitedvariation (Figure 2). This decrease continued theoverall trend of decreasing MgO that started in1998, as noted by Poland et al. [2012]. There isalso a decrease in MgO-normalized SiO2 and anincrease in MgO-normalized TiO2 and K2O during2003?2007 (Figures 2 and 3). Lava MgO increasedfrom 2008 to 2009 as did CaO/TiO2 and values ofMgO-normalized SiO2, although MgO and SiO2values dropped afterwards for the most recentlyerupted samples that were analyzed in this study(Figure 2). For more on major- and trace-elementvariations in 1983?2005 Pu?u ?O?o lavas, see Gar-cia et al. [1989, 1992, 1996, 2000], Marske et al.[2008], and Thornber [2003].[11] The brief eruptive outbreaks uprift of the Pu?u?O?o vent in 1997 (3 km uprift for episode 54)and 2007 (6 km uprift for episode 56; Figure 1)1Additional supporting information may be found in the onlineversion of this article.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854855Table 2. Pb, Sr and O Isotopic Geochemistry of Pu?u ?O?o Lavas from 1983?2010aSample 206Pb/204Pb 2 207Pb/204Pb 2 208Pb/204Pb 2 87Sr/86Sr 2 18O 123 Jan 1983  18.5247 0.0009 15.4800 0.0008 38.155 0.0020 0.703590 0.000009 4.56 0.029 Apr 1983 18.5309 0.0007 15.4893 0.0007 38.165 0.0017 0.703573 0.0000083 July 1983  18.4780 0.0007 15.4765 0.0008 38.117 0.0017 0.703573 0.000007 4.77 0.0331 Jan 1984 18.4595 0.0009 15.4765 0.0007 38.103 0.0020 0.703587 0.00000812 Sep 1984 18.4417 0.0008 15.4747 0.0007 38.091 0.0019 0.703555 0.0000098 Feb 1985  18.4342 0.0010 15.4743 0.0009 38.087 0.0024 0.703567 0.000008 4.76 0.0521 Apr 1985  4.82 0.1030 Jul 1985 18.4306 0.0007 15.4756 0.0007 38.089 0.0020 0.703571 0.0000082 Jun 1986  18.4138 0.0009 15.4755 0.0008 38.079 0.0020 0.703580 0.000009 4.94 0.3026 Jun 1986  4.77 0.0213 Sep 1986  18.4138 0.0008 15.4726 0.0007 38.074 0.0020 0.703590 0.000008 5.17 0.0116 Mar 1987  18.4108 0.0006 15.4717 0.0005 38.069 0.0014 0.703589 0.000009 5.25 0.0218 Oct 1987 18.3992 0.0008 15.4746 0.0008 38.068 0.0022 0.703597 0.00000919 Jan 1988 18.3952 0.0006 15.4715 0.0006 38.061 0.0016 0.703577 0.00000818 Aug 1988 18.3871 0.0008 15.4703 0.0007 38.052 0.0017 0.703583 0.00000926 Mar 1989  18.3882 0.0007 15.4717 0.0007 38.055 0.0019 0.703584 0.000009 5.11 0.057 Jul 1989 18.3861 0.0010 15.4725 0.0009 38.054 0.0022 0.703581 0.0000097 Jul 1989 ? 18.3851 0.0009 15.4710 0.0008 38.052 0.0021 0.703576 0.0000097 Jan 1990  18.3881 0.0009 15.4745 0.0008 38.056 0.0018 0.703584 0.000008 5.03 0.0127 May 1990 18.3864 0.0008 15.4716 0.0007 38.054 0.0020 0.703603 0.00000721 Oct 1990 18.3856 0.0007 15.4693 0.0006 38.049 0.0017 0.703597 0.00000812 May 1991  18.3992 0.0009 15.4739 0.0007 38.063 0.0019 0.703581 0.000008 5.08 0.041 Aug 1991  5.11 0.046 Jun 1992  18.4048 0.0009 15.4721 0.0009 38.061 0.0019 0.703585 0.000008 5.04 0.1013 Aug 1993  18.4112 0.0007 15.4752 0.0006 38.072 0.0015 0.703611 0.000008 4.98 0.074 Jan 1994 18.4098 0.0009 15.4736 0.0008 38.067 0.0020 0.703607 0.00000825 Apr 1994  18.4100 0.0009 15.4737 0.0009 38.068 0.0027 0.703586 0.000008 5.01 0.029 Oct 1994 18.4059 0.0008 15.4718 0.0007 38.066 0.0019 0.703598 0.00000727 Apr 1995  18.4059 0.0009 15.4721 0.0008 38.066 0.0023 0.703604 0.000009 5.25 0.0514 Oct 1995 18.4068 0.0008 15.4729 0.0008 38.071 0.0021 0.703602 0.00000919 Jan 1996  5.19 0.0715 Mar 1996 18.4064 0.0010 15.4738 0.0008 38.070 0.0023 0.703592 0.00000922 Aug 1996 18.4038 0.0009 15.4722 0.0008 38.065 0.0016 0.703612 0.00000710 Jan 1997  18.4010 0.0012 15.4728 0.0011 38.064 0.0019 0.703606 0.000009 5.2 0.0523 Jul 1997 18.3993 0.0010 15.4729 0.0010 38.068 0.0025 0.703601 0.00000710 Jan 1998 18.3958 0.0007 15.4728 0.0006 38.067 0.0014 0.703591 0.00000810 Jan 1998 ? 18.3940 0.0007 15.4711 0.0006 38.063 0.0016 0.703593 0.00000711 May 1998 18.4005 0.0009 15.4740 0.0008 38.071 0.0020 0.703605 0.0000107 Sep 1998 18.4082 0.0008 15.4775 0.0006 38.083 0.0017 0.703601 0.000006 5.33 0.067 Sep 1998 ? 18.4107 0.0004 15.4727 0.0005 38.075 0.0012 5.29 0.0813 Feb 1999 18.4068 0.0010 15.4783 0.0008 38.085 0.0021 0.703607 0.00000613 Feb 1999 ? 18.4124 0.0004 15.4736 0.0004 38.076 0.001119 Jun 1999 18.3987 0.0010 15.4805 0.0007 38.085 0.0020 0.703620 0.00000927 Oct 1999  18.4018 0.0004 15.4726 0.0004 38.069 0.0011 0.703622 0.000009 5.36 0.0819 Feb 2000  18.4072 0.0004 15.4712 0.0004 38.072 0.0011 0.703624 0.00000721 Jun 2000  18.4067 0.0004 15.4704 0.0004 38.069 0.0011 0.703638 0.000007 5.28 0.088 Jan 2001  18.4116 0.0004 15.4721 0.0004 38.074 0.0011 0.703627 0.000012 5.31 0.087 Jul 2001  18.4137 0.0004 15.4719 0.0004 38.073 0.0013 0.703626 0.0000099 Feb 2002  18.4139 0.0004 15.4707 0.0004 38.069 0.0011 0.703637 0.00000820 Aug 2002  18.4152 0.0004 15.4722 0.0004 38.072 0.0011 0.703639 0.00000512 Apr 2003  18.4161 0.0005 15.4726 0.0005 38.072 0.0013 0.703641 0.000005 5.31 0.0815 Jan 2004  18.4154 0.0005 15.4719 0.0005 38.069 0.0012 0.703632 0.000007 5.21 0.087 Jun 2004  18.4146 0.0003 15.4716 0.0004 38.068 0.0010 0.703624 0.00000731 Jan 2005  18.4170 0.0005 15.4735 0.0006 38.075 0.0012 0.703624 0.000005 4.96 0.138 Aug 2005  18.4119 0.0005 15.4727 0.0005 38.070 0.0013 0.703622 0.00001029 Jan 2006  18.4087 0.0004 15.4720 0.0004 38.065 0.0012 0.703623 0.00000924 Jun 2006 ? 18.4062 0.0028 15.4724 0.0026 38.066 0.0059 0.703612 0.000008 5.23 0.0324 Jun 2006  18.4073 0.0016 15.4714 0.0018 38.063 0.0060 0.703617 0.000013 5.23 0.036 Apr 2007  18.4065 0.0003 15.4715 0.0003 38.063 0.0009 0.703617 0.000007 5.35 0.1317 Jun 2007  18.4019 0.0004 15.4709 0.0004 38.062 0.0012 0.703626 0.000006 5.63 0.1322 Mar 2008  18.4038 0.0003 15.4700 0.0003 38.061 0.0010 0.703607 0.000008 5.45 0.132 May 2008  18.4045 0.0003 15.4721 0.0004 38.066 0.0010 0.703609 0.00000915 Nov 2008 18.3972 0.0009 15.4704 0.0007 38.058 0.0024 0.703600 0.00000829 Jan 2009  18.4003 0.0005 15.4709 0.0006 38.061 0.0012 0.703628 0.0000087 May 2009 18.4005 0.0008 15.4736 0.0007 38.066 0.0020 0.703624 0.0000104 Jun 2009  18.4009 0.0003 15.4720 0.0004 38.064 0.0010 0.703610 0.00000716 Oct 2009  18.3994 0.0005 15.4714 0.0006 38.062 0.0012 0.703622 0.00000722 Jan 2010  18.3987 0.0005 15.4708 0.0005 38.060 0.0013 0.703617 0.000007a indicates analysis at San Diego State University (SDSU), analyses from 7 Sep 1998 to 8 Aug 2005 are from Marske et al. [2008]. Sr isotopeanalyses from 1983?1997 and 2006?2010 were performed at PCIGR. ? Chemistry duplicate.  Published 18O analyses from Garcia et al. [1998].Analytical methods are described in the supporting information. 24 Jun 2006 is an in-house glass standard called Menehune collected from a Pu?u?O?o lava flow (errors are the external 62s of the replicate analyses; average of four analyses for Pb and Sr at PCIGR; 68 for Pb and 26 for Sr atSDSU). US Geological Survey sample numbers for lavas between up to16 Mar 87 are 23 Jan 1983: 1?054, 9 April 1983: 3?117, 3 Jul 1983: 5?139, 31 Jan 1984: 14?232, 12 Sep 1984: KE24?25 310S, 8 Feb 1985: 30?362, 30 Jul 1985: 35?419, 1 Jan 1986:40?484, 2 Jun 1986: 46?536, 13Sep 1986: 48?649, 16 Mar 1987: 48?714F.occurred after major collapses of the Pu?u ?O?ocrater floor (Table 1). The lavas erupted from theseuprift vents were geochemically distinct. Com-pared to coeval Pu?u ?O?o vent lavas, those fromepisode 54 have lower MgO (5.6?6.4 versus 7.5?10.1 wt %), CaO/TiO2 (2.8?3.4 versus 4.4), Sr/Nband Zr/Nb ratios (Figure 3). These geochemicalsignatures and the petrographic evidence of56789100.700.720.740.760.780.800.820.840.8648.849.049.249.449.649.850.050.250.43.03.23.43.63.84.04.24.44.64.82.12.22.32.42.52.60.350.400.450.500.551983198419851986198719881989199 0199119921993199419951996199719981999200020012 00220032004200520062007200820092010198319841985198619871988198919901991199219931994199519961997199 819992000200120022003200420052006200720082009201084 86 88 90 92 94 96 98 00 02 04 06 08 1084 86 88 90 92 94 96 98 00 02 04 06 08 1084 86 88 90 92 94 96 98 00 02 04 06 08 10 84 86 88 90 92 94 96 98 00 02 04 06 08 10Ep. 54TiO2 (wt%)normalized to 10 wt% MgO(Magma supply ratedoubled)2003-071983-19861986-19921992-2010MgO (wt%)CaO/Al2O3 (wt%)SiO2 (wt%)CaO/TiO2 (wt%)K2O (wt%)(b)(d)(c)(a)(f)(e)normalized to 10 wt% MgOnormalized to 10 wt% MgO(Magma supply ratedoubled)2003-07(Magma supply ratedoubled)2003-07(Magma supply ratedoubled)2003-071983-19861986-19921992-2010normalized to 10 wt% MgOnormalized to 10 wt% MgOEp. 56Figure 2. Major-element variation diagrams for Pu?u ?O?o lavas from 1983 to 2010. All major elements andratios except MgO were normalized to 10 wt % MgO [the most primitive lava erupted from Pu?u ?O?o; Gar-cia et al., 2000] by addition of equilibrium composition olivine (98.5%) and spinel (1.5%) in 0.5 mol % steps[Garcia et al., 2003; Rhodes and Vollinger, 2004]. Pu?u ?O?o lavas with <7.2 wt % MgO may have crystal-lized minerals other than olivine (e.g., clinopyroxene and plagioclase) and were not included in the olivinenormalization procedure and are not shown in all the plots, except MgO. Episode 54 (Ep. 54; 29?30 January1997) lavas involved mixing of evolved magmas stored in the rift zone and MgO-rich magma. Three intervalsof eruptive activity in legend and colors correspond with those shown in Figure 1. CaO/TiO2 and CaO/Al2O3ratios also use normalized data although are virtually unaffected by olivine fractionation. Vertical lines indi-cate nearly double magma supply rate between 2003 and 2007 [Poland et al., 2012]. Data are presented in thesupporting information. Uncertainty for analyses is described in the supporting information.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854857disequilibrium in the episode 54 lavas are thoughtto result from mixing Pu?u ?O?o magma withstored, differentiated rift zone magma [Garciaet al., 2000; Thornber et al., 2003]. In contrast,episode 56 lavas have higher MgO (8.5 versus 7.2wt %) and a relatively high 18O value (5.6 versus5.4%) but are otherwise geochemically indistin-guishable from contemporaneous Pu?u ?O?o lavas.[12] The 206Pb/204Pb ratios for Pu?u ?O?o lavasdecreased rapidly through the episodic fountainingperiod (1983?1986) and reached a minimumbetween 1989 and 1991 during the Kupaianahaphase (Figure 4b and Table 2). The rapid decreasein 206Pb/204Pb continues the longer-term trend ofdecreasing Pb isotope ratios for K?lauea lavaserupted following the 1924 collapse of the summitcaldera (Figure 4a). After 1991, the trend of206Pb/204Pb ratios in Pu?u ?O?o lavas shows cyclicvariations with two broad humps, each cycle span-ning approximately 10 years (except for a smalloffset from the overall trend between January1998 and June 1999; Figure 4b). The cyclic varia-tion in Pb isotope ratios is well shown by208Pb/206Pb ratios, which inversely mirror the206Pb/204Pb trend (Figures 4 and 5).[13] The 87Sr/86Sr ratios of Pu?u ?O?o lavas extendthe temporal trend of increasing Sr isotope ratiosfor K?lauea lavas that started following the 1924caldera collapse (Figure 4c). Overall, Pu?u ?O?olavas display an increase in 87Sr/86Sr from 1983 to2003 and a slight decrease after 2004 (Figure4d). Prior to 1999, the 87Sr/86Sr and 206Pb/204Pbratios of the lavas are not well correlated, althoughthere is an overall inverse correlation between the19831984198519861987198819891990199119921993199419951996199719981999200020012002200320042005200620072008200920100.40.50.60.70.819831984198519861987198819891990199119921993199419951996199719981999200020012002200320042005200620072008200920102.02.12.22.32.42.51.99101112161820222426 Sr/NbNb/Y La/Sm(c)(a)(d)1983-19861986-19921992-2010Ep. 54Ep. 54Ep. 5484 86 88 90 92 94 96 98 00 02 04 06 08 10 84 86 88 90 92 94 96 98 00 02 04 06 08 10(Magma supply ratedoubled)2003-07Ep. 54Zr/Nb(b)(Magma supply ratedoubled)2003-07 ?2 SE ?2 SE ?2 SE ?2 SEK?lauea summit(Sep.1982)K?lauea summit(Sep.1982)K?lauea summit(Sep.1982)K?lauea summit(Sep.1982)Figure 3. Trace-element ratios versus time for Pu?u ?O?o lavas from 1983 to 2010. Overall exponential vari-ation in trace-element ratios indicates progressive depletion of the source. In the La/Sm versus time plot, asubset of samples analyzed at PCIGR (April 1983, January 1984, September 1984, and April 2007 to January2010) are normalized to reference material Kil-93 (La/Sm of 2.09, average value from Australia National Uni-versity where most ICP-MS analyses were performed). Vertical lines indicate nearly double magma supplyrate between 2003 and 2007 [Poland et al., 2012]. Trace-element abundances in ppm (data shown in support-ing information). Average 62 bars are shown in a corner of each panel. September 1982 K?lauea summitlava composition from Pietruszka and Garcia [1999].GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854858two ratios for historical summit lavas (Figure 4and Table 2). After 1999, there is a positive corre-lation between 206Pb/204Pb and 87Sr/86Sr (Figure6), which corresponds to the second of the twomajor temporal cycles of Pb isotope ratios (Figure5). The Nd isotope ratios of lavas erupted between1983 and 2005 display no variation outside analyt-ical uncertainty [Marske et al., 2008]. Thus, nonew Nd isotopic data were collected during thisstudy.[14] Oxygen isotopic compositions of Pu?u ?O?olavas erupted over 25 years (1983?2008) show alarger range than historical K?lauea summit lavasspanning 380 years of activity (1.1 versus 0.7% ;Figure 4). Overall, O isotope ratios of Pu?u ?O?olavas have increased over time with early lavas(before 1986) having lower 18O (4.8?4.9%) thansubsequent lavas (5.0?5.6% ; Table 2). The high-est O isotope ratio observed during the Pu?u ?O?oeruption is for an episode 56 lava that erupted 6Uwekahuna BluffUwekahuna BluffHilinasubmarinelavas0.70340.70350.70360.7037198319841985198619871988198919901991199219931994199519961997199819992000200120022003200420052006200720082009201018.3718.3918.4118.4318.4518.4718.4918.5118.530.703550.703600.703651983198419851986198719881989199019911992199319941995199619971998199920002001200220032004200520062007200820092010K?laueasummitK?laueasummitPCIGR analyses (2009)Marske et al. (2008)-SDSUPCIGR analyses (2010)Time (years)Time (years)(summit)(summit)Time (years)K?laueasummitsummit calderacollapse (1924)summit calderacollapse (1924)average ?2 SEMORB mantle rangeSDSU analyses (2007)?18O18.218.418.618.81780 1820 1860 1900 1940 19801000 14001780 1820 1860 1900 1940 19801000 140016001780 1820 1860 1900 1940 198016001600Jun-99Jun-99Jun-86Jun-07(Ep. 56 uprift)(  )(  )(  )4.8 ka 3.5 ka1.7 ka4.8 ka3.5 ka1.7 ka206Pb/204Pb87Sr/ 86Sraverage ?1 SE?18O87Sr/ 86Sr206Pb/204Pb(280-130 ka)tholeiiticalkalictransitionalHilinasubmarinelavas(280-130 ka)tholeiiticalkalictransitional(episodic fountaining)Kupaianaha(gentle effusion)Pu?u ???? Pu?u ???? (mostly flank vent eruptions)(a)(c)(b)(d)Pu?u ???? Pu?u ???? Pu?u ???? (e) (f)data thisstudyMORBmantle summit calderacollapse (1924)PCIGR analyses (2009)Marske et al. (2008)-SDSUPCIGR analyses (2010)SDSU analyses (2007)4.64.85.05.25.45.65.84.64.85.05.25.45.61983198419851986198719881989199019911992199319941995199619971998199920002001200220032004200520062007200820092010no datascale changeK?lauea summit(Sep.1982)K?lauea summit(Sep.1982)Figure 4. Temporal variation of Pb, Sr, and O isotopes for K?lauea Volcano and Pu?u ?O?o lavas. (b)206Pb/204Pb in Pu?u ?O?o lavas shows cyclic variation with two broad humps, each cycle spanning approxi-mately 10 years. (d) 87Sr/86Sr in Pu?u ?O?o lavas increases from 1983 to 2003 and decreases between 2008and 2010, and is correlated with 206Pb/204Pb after 1999. (f) 18O for Pu?u ?O?o lavas shows the same range(0.7%) as historical K?lauea summit lavas and is not well correlated with 206Pb/204Pb or 87Sr/86Sr. Data sour-ces for K?lauea Volcano are Hanyu et al. [2010], Kimura et al. [2006], Marske et al. [2007], Abouchami et al.[2005], Pietruszka and Garcia [1999], Chen et al. [1996], and Garcia et al. [2008]. Data sources for previousanalyses of Pu?u ?O?o lavas areMarske et al. [2008] and Garcia et al. [1998]. For Pu?u ?O?o analyses, average62 for 206Pb/204Pb is smaller than symbol size and uncertainty for 87Sr/86Sr and 18O is shown in the panels.Data for Pu?u ?O?o lavas are presented in Table 2. September 1982 K?lauea summit lava composition fromPietruszka and Garcia [1999]. Colors for symbols in Figures 4b, 4d, and 4f) are the same as Figure 3.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854859km uprift from the Pu?u ?O?o vent in mid-June2007. The episode 56 eruption is related to intru-sion of a dike from the upper East Rift Zone, sothis lava was probably not derived from the shal-low reservoir of magma beneath Pu?u ?O?o [Mont-gomery-Brown et al., 2010]. Thus, its 18O valueis probably representative of the magma feedingthe Pu?u ?O?o eruption. It is identical to the highestvalues observed among historical summit lavas(5.6% ; Figure 4). Variations of 18O in Pu?u ?O?olavas do not correlate with Pb or Sr isotope ratios,or with other geochemical parameters, as wasnoted for previous O isotope work on lavas fromthis eruption [Garcia et al., 1998]. Therefore, Pu?u?O?o lava Pb and Sr isotope ratios were apparentlynot affected by the processes causing variable Oisotope ratios.5. Discussion[15] The high eruption rate and continuous natureof the Pu?u ?O?o eruption provide an exceptionalopportunity to use lava chemistry to evaluate thechanging roles that source, melting, and crustalprocesses play during this single prolonged erup-tion. Previous Pb isotope and trace-element studieson lavas from several multiple-year eruptions ofPiton de la Fournaise Volcano (Reunion Island)have discerned contributions from multiple com-ponents within the Reunion mantle plume and aperiodic role for shallow-level contamination [Pie-truszka et al., 2009; Vlastelic et al., 2005, 2007].Similarly, extreme Pb isotope variability in meltinclusions from Iceland basaltic lavas indicate sig-nificant source heterogeneity, with binary mixingrelationships that may result from combining sol-ids in the mantle and two stages of melt mixing (inporous mantle melt-transport channels and lowercrustal magma chambers) [Maclennan, 2008].Similarly, the geochemistry and petrography ofPu?u ?O?o lavas have been used to interpret theextent of crustal magmatic processes (olivine frac-tionation and accumulation, mixing of higher-MgO and stored rift-zone magmas, and crustalassimilation) and mantle processes (degree of par-tial melting, melt extraction and migration, andsource heterogeneity) during the Pu?u ?O?o erup-tion until 2005 [e.g., Garcia et al., 1998, 2000;Marske et al., 2008]. Here we use new high-precision Pb, Sr, and O isotope ratios, and major-and trace-element data for the entire Pu?u ?O?oeruption (1983?2010) to evaluate the causes ofcyclic and other short-term variability in the proc-esses that operate from the source to the surfacewithin K?lauea Volcano. The effects of crustalprocesses (crystal fractionation, magma mixing,and crustal contamination) on modifying Pu?u?O?o lava compositions are evaluated beforeexamining mantle source and melt transportprocesses.5.1. Magma Mixing and CrystalFractionation During Early EpisodicActivity (1983?1985)[16] The largest compositional changes in Pu?u?O?o lavas occurred from 1983 to 1985. Thesechanges mostly involved two crustal processes:crystal fractionation and magma mixing. Duringsome single eruptive episodes (5?10, 30, and 31),there were relatively large changes in MgO, Ni,and Cr, which are related to minor (3?5%) olivinefractionation in the shallow Pu?u ?O?o reservoirduring eruptive hiatuses [Garcia et al., 1992].These short-term (3?4 weeks) variations are super-imposed on longer term changes that have beenrelated to magma mixing [Garcia et al., 1992;Thornber, 2003]. The longer term variations areevident in plots of MgO-normalized major ele-ments, ratios of incompatible trace elements, andPb isotope ratios (Figures 2?4). Strontium and Oisotopes show less change during this period com-pared to their overall variation during the eruption(Figure 4). The overall progressive compositionalvariation in Pu?u ?O?o lavas from 1983?1985 hasbeen attributed to the mixing of new, relativelyMgO-rich magma (>7.5 wt %) with a decreasingproportion of hybrid, rift-zone stored differen-tiated magma (from 30% of the higher MgOmagma in March 1983 to 100% in September1984) [Garcia et al., 1992; Shamberger andGarcia, 2007].[17] The origin of the higher MgO magma compo-nent from the early phase of the Pu?u ?O?o erup-tion may have been: (1) magma from the summitreservoir, as represented by lavas from the Sep-tember 1982 summit eruption; and/or (2) newmantle-derived magma [Garcia et al., 1992;Shamberger and Garcia, 2007]. Scenario 1involves no change in the composition of thehigher MgO magma from September 1982 to1985, whereas scenario 2 requires it. The 1983?1985 Pu?u ?O?o lavas have both higher and lower206Pb/204Pb ratios than the September 1982 sum-mit lavas (Figures 4b and 7). Therefore, mixing ofa single 1982 summit magma with rift-zone storedmagma (scenario 1) cannot explain the isotopicGREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854860variation of Pu?u ?O?o lavas after 1984, when206Pb/204Pb values are lower than 1982 summitmagma (Figure 7). Ratios of some incompatibletrace elements (Sr/Nb and Zr/Nb) for some lavaswith higher MgO (>7.5 wt %) erupted after mid-1984 also are higher than those for September1982 summit lavas (Figure 3). Thus, if magmafrom the summit reservoir was supplying the Pu?u?O?o eruption, its composition must have changedafter the September 1982 eruption and prior toSeptember 1984 (Figures 4 and 7).[18] The isotopic variations for early Pu?u ?O?olavas are consistent with the eruption being sup-plied by new, compositionally variable, mantle-derived magma in addition to or instead of Sep-tember 1982 summit magma. The rate of206Pb/204Pb variation observed for the period afterthe end of early magma mixing is much faster thanduring the previous 30 years (1952?1982) ofK?lauea summit eruptions (0.016 yr1 versus0.004 yr1). These rapid variations in Pb isotopicratios suggest that magmas supplying Pu?u ?O?opartially bypassed or did not thoroughly mix withthe summit reservoir [Garcia et al., 1996]. Basedon these observations, the composition of the pa-rental magma delivered to Pu?u ?O?o from themantle is interpreted to have continually changedfor the remainder of the eruption (i.e., after 1984).The details and cause of this variation are dis-cussed in section 5.3.5.2. Oxygen Isotope Indications of CrustalContamination and Nature of MantleSource[19] Lavas from oceanic island volcanoes showwide ranges in oxygen isotopic compositions (4.5?7.5%), which have been attributed to composition-ally variable mantle-derived magmas that weremodified by oxygen exchange and/or crustal con-tamination [Harmon and Hoefs, 1995]. Our previ-ous studies revealed that some Pu?u ?O?o andK?lauea summit magmas experienced significantoxygen isotope exchange with metamorphosedK?lauea rocks [Garcia et al., 1998, 2008]. This isindicated by the disequilibrium between matrixand coexisting olivine 18O values, the relativelylow 18O values for these lavas (4.7?5.2%) andthe lack of correlation between 18O values andother geochemical parameters [Garcia et al.,1998, 2008].[20] The highest 18O value observed for any lavaduring the Pu?u ?O?o eruption is for the June 2007208Pb/206PbPCIGR analyses (2009)Marske et al. (2008)-SDSUPCIGR analyses (2010)SDSU analyses (2007)(Magma supply ratedoubled)2003-072.0592.0612.0632.0652.0672.0692.0711983198419851986198719881989199019911992199319941995199619971998199920002001200220032004200520062007200820092010Jun-99(episodic fountaining)Kupaianaha(gentle effusion)Pu?u ???? Pu?u ???? (mostly flank vent eruptions)Jun-861982summitRift-storedmagmaFigure 5. 208Pb/206Pb variation with time for Pu?u ?O?o lavas from 1983 to 2010. Previous analyses of Pu?u?O?o lavas indicated in legend are from Marske et al. [2008]. Average 62 for 208Pb/206Pb is smaller thansymbol size.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854861uprift lava (5.6% ; Figure 4). This value is identi-cal to the highest value reported for historicalK?lauea summit lavas (1820?1982) [Garcia et al.,2008] and lies within the range of normal mid-ocean ridge basalt (MORB) basalt 18O values(5.4?5.8% ; Figure 4) [Eiler, 2001]. These summitlavas (1832, 1866, 1894, and 1917?1921) wereerupted during periods of sustained lava lake ac-tivity, and are thought to be representative of theprimary uncontaminated magma feeding K?lauea[Garcia et al., 2008]. Thus, the 2007 uprift ventlava supports our previous interpretation [Garciaet al., 2008] that the 18O value for the mantlesource of K?lauea?s magma is identical to thesource for MORB.[21] The earliest Pu?u ?O?o lavas (1983?1986)show the strongest signs of disequilibriumbetween coexisting matrix mineral and olivine,and have the lowest O isotope values (<5.0%)[Garcia et al., 1998]. After the shift to continuouseffusion in July 1986, O isotope ratios are higher(5.2%) and the coexisting olivines were in equilib-rium with host matrix for about 1 year [Garciaet al., 1998]. Subsequently, the matrix O isotopevalues decreased somewhat (to 5.0% ; Figure 4)and those for olivine increased, indicating olivine-matrix disequilibrium. This O isotope disequili-brium continued for two more years, and was fol-lowed by a return to olivine-matrix equilibrium in1995?1997 [Garcia et al., 1998]. After 1997, ma-trix O isotope values are relatively low and nearlyconstant (5.36 0.1%) except for a 2005 lava(5.0% ; Figure 4), which was the most evolvedsample (analyzed for O isotopes) since 1984 (6.7wt % MgO). Thus, despite nearly 30 years of vig-orous eruptive activity (producing 4 km3 oflava), oxygen exchange with metamorphosedrocks has probably continued in the Pu?u ?O?omagmatic plumbing system. The magnitude of ox-ygen isotope exchange can be estimated assumingbulk assimilation between a parental magma (asreflected by the 2007 uprift sample with a 18Ovalue of 5.6%) and a hydrothermally alteredK?lauea rift zone lava (1.9%) [Garcia et al., 2008]as a contaminant. Pu?u ?O?o lavas erupted justbefore and after the 2007 uprift event have averageO isotope values of 5.4% (Figure 4), indicating5% bulk contamination, whereas earlier lavas(1986?2006) with average values of 5.2?5.3%,might have experienced 8?11% bulk contamina-tion. This contamination is likely to have occurredin the Pu?u ?O?o reservoir and did not have anyobvious effect on other geochemical parameters[Garcia et al., 1998] (Table 2).5.3. Cyclic Compositional VariationsFrom Mantle Processes (1985?2010)[22] Pu?u ?O?o lavas erupted after the early periodof magma mixing ended in late 1984 show cyclicvariations in several geochemical parameters thatare insensitive to olivine fractionation (e.g., CaO/TiO2, Sr/Nb, Zr/Nb,206Pb/204Pb; Figure 8). Thecyclic variations in CaO/TiO2 and K2O/TiO2ratios for Pu?u ?O?o lavas erupted between 1996and 2001 were reported to be associated with de-formation in the summit magma reservoir[Thornber, 2003]. Although the timing of thehighs and lows in these ratios are not perfectlycoincident with summit tilt changes [see Figure 8,Thornber, 2003], these geochemical cycles wereattributed to mixing of mantle-derived magma ofuniform composition (similar to Pu?u ?O?o lavasaverage ?2 SEaverage ?2 SETime (month-year)18.39018.39518.40018.40518.41018.41518.420Jan-97Jul-97Jan-98Jul-98J an-99Jul- 99Jan-00J ul -00Jan-01Jul-01Jan -02Jul-02Jan-03Jul-03J an-0 4Jul-04Jan-05J ul-05Jan-06Jul-06J an-07J ul-07Jan-08Jul-08Jan-09Jul-09J an-100.703590.703600.703610.703620.703630.703640.70365Jan-97Jul-97Jan-98Jul-98Jan-99Jul-99Jan -00Jul-00Jan-01Jul-01Jan -02Jul-02Jan-03Jul-03Jan -04Jul-04Jan-05Jul-05Jan-06Jul-06Jan-07Jul-07Jan-08Jul-08Jan -09Jul-09Jan -10206Pb/204Pb87Sr/ 86Sr(a)(b)Figure 6. Temporal variation in 206Pb/204Pb and 87Sr/86Srfor Pu?u ?O?o lavas during period of dramatic increase inmagma supply. Dashed lines indicates period of significantincrease in magma supply rate up to 0.25 km3/yr between2003 and 2007 compared to 0.1 km3/yr prior to 2003[Poland et al., 2012]. Analytical uncertainty is shown in thepanels.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854862from 1999 to 2001) with 1982 summit magma[Thornber 2003]. However, Pu?u ?O?o lavaserupted after 1999 have elevated 87Sr/86Sr ratios(at a given 206Pb/204Pb; Figure 7) compared toearlier lavas with low 206Pb/204Pb ratios, and thus,the 1999?2001 lavas cannot serve as a mixingend-members to explain the compositional trendof lavas erupted before 1999. Similar behavior isobserved on a plot of La/Yb versus 206Pb/204Pb(Figure 9), where a relative shift to lower La/Ybratios at a given 206Pb/204Pb occurred after 1999(compared to the trend of pre-1999 lavas). Theserelationships indicate that the temporal variationof Pu?u ?O?o lavas erupted after 1999 cannot beexplained simply by mixing of 1982 summitmagma with a uniform mantle-derived magma(Figure 7) within K?lauea?s shallow magmaticplumbing system. Instead, either a third magma ismixing with the other two or, as we advocatebelow, the composition of the Pu?u ?O?o magmais continually changing due to the melting ofsmall-scale compositional heterogeneities in themantle source.[23] Ratios of Pb isotopes in Pu?u ?O?o lavas showcyclic variations (Figure 5). These variations prob-ably reflect the dynamic process of melt extraction(from a heterogeneous source) over a time scale of0.703500.703550.703600.703650.7037018.36 18.38 18.40 18.42 18.44 18.46 18.48 18.50 18.52 18.54208Pb/204Pb0.703570.703580.703590.703600.703610.703620.703630.703640.7036518.38 18.39 18.40 18.41 18.42208Pb/204Pb38.04538.05538.06538.07518.38 18.39 18.40 18.41 18.4238.0038.0538.1038.1538.2018.36 18.38 18.40 18.42 18.44 18.46 18.48 18.50 18.52 18.5487Sr/ 86Sr206Pb/204PbSep-8687878889 9091909808090700030504020106929394959697 991988-911983-85Jun-9986102000-071992-992008-1096098788Oct-90Jan-90919808090700 0305040201069293949597991988-911983-85Jun-99 8610968788Mar-89Jul-89May-9096979898990005080808Jan-090909862000-071993-992008-1087Sr/ 86Sr206Pb/204Pb19831985198319851986-20101986-2010East Rift Zone(1977)East Rift Zone(1977)K?lauea summit(1982)K?lauea summit(1982)East Rift Zone(1960-69)East Rift Zone(1960-69)1984Sep.Jan.2001 composition1984Sep.Jan.?2?(b)(d)(a)(c)(b)(d)Figure 7. Pb and Sr isotopic compositions for Pu?u ?O?o lavas. Line connects samples in order of increasingeruption date in Figures 7b and 7d. Average 62 for Pb isotope ratios is smaller than symbol size. East RiftZone data is from J. Marske [personal communication, 2013]. September 1982 K?lauea summit lava composi-tion (outline in Figure 7a; orange star in Figure 7c) from Pietruszka and Garcia [1999]. K?lauea summit(1982) field in Figure 7a is new high-precision data from A. Pietruszka [personal communication, 2013]. Bluestar is 2001 composition proposed by Thornber [2003] as mixing end-member with 1982 K?lauea summitcomposition. Pu?u ?O?o lavas erupted after 1999 have elevated 87Sr/86Sr ratios (at a given 206Pb/204Pb). The1999?2001 lavas cannot serve as mixing end-members to explain the compositional trend of lavas eruptedbefore 1999.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854863years to decades rather than movement of small-scale mantle heterogeneities through the meltingzone. This interpretation is based on the hypothe-sis that buoyancy-driven upwelling through themelt-producing region beneath K?lauea occurs onlonger timescales (hundreds to thousands ofyears) than melt extraction (years to decades)[Pietruszka et al., 2006]. The highest estimatesfor solid mantle upwelling in the center of theHawaiian plume are 10 m/yr [Hauri, 1996;Pietruszka and Garcia, 1999], which wouldresult in a maximum of only 270 m of upwell-ing during the first 27 years of the Pu?u ?O?oeruption [cf. 5?10 km maximum thickness forthe zone of melting; Marske et al., 2007]. Forcomparison, estimates for solid mantle upwellingrates beneath Mauna Loa and Lo?ihi based onU-series disequilibria range from 0.4 to 1 m/yr[Sims et al., 1999] and 5?6 cm/yr [Pietruszkaet al., 2011], respectively. Melt extraction rates(or source-to-surface melt velocity) are estimatedto be on the order of 5?17 km/yr [Reiners,2002], which is extremely rapid compared tosolid mantle upwelling rates. Thus, cyclic varia-tion in Pb isotope ratios over short timescales(years) are best explained by variations in theprocess of melting of a heterogeneous source(and the transport of the melt to the surface),rather than upwelling of small-scale mantle het-erogeneities through the melting region.[24] The short-term Pb and Sr isotopic variationsin Pu?u ?O?o lavas may be generated by one ormore processes including: periodic processes ofmelting, melt extraction, or melt aggregation [e.g.,Cordier et al., 2010], changes in melt transportpathways or tapping new source areas [Marskeet al., 2008; Pietruszka et al., 2006], changes inthe volume of the melting region [Pietruszkaet al., 2001], and progressive melt extraction froma source with fine-scale heterogeneities [Garciaet al., 2000]. In the presence of small-scale hetero-geneities, changes in melt pathways over years todecades may lead to tapping compositionally dis-tinct sources and short-term isotopic variation inlavas [Marske et al., 2007]. The scale of composi-tional heterogeneities must be small relative to thesize of the melting region beneath K?lauea Vol-cano to allow for rapid (few years) variation inlava Pb isotope compositions [Pietruszka and Gar-cia, 1999]. Melt pathways within the source regionprobably migrate over years to decades [Pie-truszka et al., 2001, 2006]. Therefore, melt may besupplied from different areas of the melting region(Magma supply ratedoubled)2003-07Time (year)4.04.24.44.64.8CaO/TiO2 1986-19921992-201022232425262790 92 94 96 98 00 02 04 06 08 1090 92 94 96 98 00 02 04 06 08 1090 92 94 96 98 00 02 04 06 08 10206Pb/204PbSr/Nb(a)(d)Zr/Nb(b)(c)18.3518.3618.3718.3818.3918.4018.4118.4218.4318.44199019911992199319941995199619971998199920002001200220032004200520062007200820092010101112Figure 8. Plots of CaO/TiO2, Sr/Nb, Zr/Nb and206Pb/204Pbfor Pu?u ?O?o lavas showing cyclic variation apparent from1990 to 2010. Trace-element abundances in ppm.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854864over a relatively short period of time. These short-term geochemical fluctuations are effectivelytransported from the source to the surface becausePu?u ?O?o magmas are thought to partially bypassthe volcano?s summit magma storage reservoir,and avoid its buffering effects [Garcia et al.,2000].[25] The periodicity and rate of isotopic variationPu?u ?O?o lavas provide clues about the size anddistribution of small-scale heterogeneities in thevolcano?s mantle source. The cycles of Pb isotopicratios at Pu?u ?O?o have a peak to peak duration of10 years (Figure 10). These short-term increasesand decreases of Pb isotope ratios, at similar rateand degree, may represent melt extraction fromsmall-scale heterogeneities with a limited horizon-tal length scale (a few km or less). Modeling byFarnetani and Hofmann [2009] suggests that??filament-like?? structures derived from stretchingof deep-seated mantle heterogeneities may de-velop as the Hawaiian plume rises to the surface.Although their model was developed to explainthe long-term (>100 kyr) geochemical variationobserved for drill core from Mauna Kea [Farne-tani and Hofmann, 2010], we use the filamentmodel to explain the periodic variation in the Pbisotope ratios of Pu?u ?O?o lavas (Figure 10)because it provides a mechanism to link geochem-ical variations with the inferred deep mantle struc-ture. Other geometries have been suggested for thesmall-scale heterogeneities within Hawaiianplume, including a series of vertically stacked,elongated blobs [Blichert-Toft and Albare`de,2009], but we prefer the filament geometry toexplain the Pb isotopic variations of Pu?u ?O?olavas.[26] In this scenario, the periodic variation in Pbisotope ratios of Pu?u ?O?o lavas may reflect meltextraction from a mantle source with verticallyoriented repeating source heterogeneities, or thinstrands, on a small scale (Figure 10). The Pu?u?O?o eruption rate is thought to be greater than therate of melting, so melt must be transferred intochemically isolated channels from successivelyfurther areas within the larger melting region tosustain the eruption [Pietruszka et al., 2006]. Inthe context of the filament model of Farnetani andHofmann [2009], this process might extract meltfrom a succession of strands with different isotopiccompositions, which would potentially create theobserved periodicity in variation of the Pb isotoperatios (Figure 10). The volume of a single compo-sitional strand within the mantle tapped by thePu?u ?O?o eruption can be inferred using estimatesfor lava eruption rate (0.13 km3/yr) [Suttonet al., 2003] and melt zone porosity (1?2%) [Pie-truszka et al., 2001]. This calculation assumes that(1) there have been only two isotopically distinctcomponents since 1986 and (2) the heterogene-ities have the same melt productivity. We do notconsider the effect of melting heterogeneous lith-ologies with different melt productivities (e.g., pe-ridotite versus pyroxenite), despite the potentialsignificance for mixed lithologies in the source forHawaiian lavas [Hauri, 1996; Reiners, 2002;Sobolev et al., 2005]. Indeed, recent modeling ofincompatible trace elements suggests that Pu?u2007-1018.3818.4218.4618.5018.544.5 4.9 5.3 5.7 6.1 6.5 6.9 7.3 7.718.3818.3918.4018.4118.424.5 4.7 4.9 5.1 5.3 5.5 5.7 5.9HistoricalEast RiftZone ?2 ?summit(1982)La/Yb206Pb/204Pb206Pb/204Pb(a)(b)831988-912000-061992-99848586878889908990919394969598989708099900010304050607091092La/Ybmixingsource variationK?laueasummitFigure 9. Plots of 206Pb/204Pb versus La/Yb for Pu?u ?O?olavas with line connecting samples in eruptive order. (a) AllPu?u ?O?o lavas. (b) Smaller variations in lavas eruptedbetween 1988 and 2010 [area indicated by box with dashedline in Figure 9a]. K?lauea summit data is from Pietruszkaand Garcia [1999]. Average 62 for La/Yb shown in Figure9 a. Average 62 for Pb isotope ratios is smaller than symbolsize.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854865?O?o lavas (1986?1998) are derived from a sourcewith 13% recycled oceanic crust in a matrix ofambient depleted Hawaiian mantle [Pietruszkaet al., 2013]. However, this model was unable todistinguish if the recycled oceanic crust was pres-ent as pyroxenite or refertilized peridotite. Itshould also be noted that the compositional rangeof Pu?u ?O?o lavas from 1986 to 2010 is smallcompared to the overall range for Hawaiian volca-noes [e.g., Jackson et al., 2012; Ren et al., 2009;Weis et al., 2011], so the melt productivities of theend-member sources are probably similar.[27] The duration of a single cycle of Pb isotopevariation is approximately 10 years (Figure 10),which suggests melt is extracted from one compo-sitional strand in 5 years (before the trendreverses when melt from a different strand isencountered). Estimates for the height of K?lauea?smelting region range from <5 km [Marske et al.,2008] to 55 km [Watson and McKenzie, 1991].We assume the magma supply rate is roughlyequivalent to the eruption rate given that overallmagma storage in K?lauea has been decreasingslightly since 1983 [Poland et al., 2012]. The meltDepth (km) 13017010 50 90 130 170 210Length(km)Plate motionMelting zoneK?lauea Plate motion9 cm/yr100 0 100 200Length(km)Depth (km)15025050350Melting rate (10-11 kg m-3 s-1)0 2 4 6 8250 225 200 175 150 125?T (?C)50Depth (km)13017020 80 50Depth (km)13017020 80 50Depth (km)13017020 80Depth (km)50Depth (km)13017020 8019902.0652.0672.0692.071208 Pb/206 Pb1987198819891990199119921993199419951996199719981999200020012002200320042005200620072008200920101994     1998 2004(a)(b)(c)(d)Figure 10. Cartoon model for the Hawaiian mantle plume to explain the isotopic variation of Pu?u ?O?olavas, based on assumptions described in the text. (a) 208Pb/206Pb variation with time for Pu?u ?O?o lavas from1990 to 2010 showing cyclic variation. (b) Vertical section of the Hawaiian plume adapted from Farnetaniand Hofmann [2010]. Purple shades indicate the melting rates inside the melting zone, shown in legend.Dashed yellow lines are flow trajectories. Dashed black box is magma capture zone for K?lauea. (c and d)Sketches of the changing melting zone during Pu?u ?O?o eruption. Lithosphere is not shown in Figure 10c.Melting zone from Farnetani and Hofmann [2010].GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854866volume produced in 5 years is 0.5 km3. If meltzone porosity is 2% (estimates based on U-seriesstudies range from 1 to 3%) [Pietruszka et al.,2001; Sims et al., 1999], the total volume of a sin-gle heterogeneity tapped over a 5 year periodwould be 25 km3. If individual filament-shapedheterogeneities extend over the height of the melt-ing zone, then this 25 km3 volume would translateto a diameter of 1?3 km for melting zone heightsof 55?5 km, respectively.5.4. Effects of Doubling of Magma SupplyRate on Lava Composition (2003?2007)[28] One enigmatic question of ocean island volca-nism is whether variations in lava compositions arecorrelated with magma supply rate [e.g., Vlastelicet al., 2005]. Wide variations in magma supplyrate have occurred historically at K?lauea (0.01?0.18 km3/yr between 1840 and 1983) [Dvorak andDzurisin, 1993]. A marked decrease in lava effu-sion rate during the 19th to early 20th century atK?lauea (0.10?0.01 km3/yr) was accompanied byan increase in the ratios of highly over moderatelyincompatible trace-element abundances [Pie-truszka and Garcia, 1999], and in the modal abun-dance of clinopyroxene and plagioclase in thelavas indicating eruption of more fractionated andcooler magma [Garcia et al., 2003]. This change inlava composition is believed to be a direct result ofa decrease in the melt fraction (10?5%) and aswitch to a more depleted source [Pietruszka andGarcia, 1999]. A dramatic short-term increase ineffusion rate was observed for the Pu?u ?O?o erup-tion between 2003 and 2007 [Poland et al., 2012].Here we explore the results of this magma supplysurge on the composition of Pu?u ?O?o lavas.[29] Estimates of lava effusion rate for the Pu?u?O?o eruption prior to 2003 are based on geologicmapping, and measurements of very low fre-quency electromagnetic profiling and gas emis-sions. These techniques indicate an average rate ofmagma supply of 0.13 km3/yr [Sutton et al.,2003]. Lava effusion rate (considered to be aproxy for the magma supply rate by Poland et al.[2012]) was estimated to have increased between2003 and 2007 and to have doubled in 2005 (to0.25 km3/yr), before returning to the previousrate by 2008 [Poland et al., 2012]. An increase inmagma supply is normally expected to result inhigher MgO contents in erupted lavas as magmaundergoes less cooling prior to eruption. This rela-tionship was inferred for Pu?u ?O?o lavas eruptedfrom 1986 to 1992, when changes in tilt were fol-lowed 3 weeks later by changes in MgO [e.g.,Garcia et al., 1996]. However, during the 2003?2007 surge in magma supply, Pu?u ?O?o lavashave consistently lower MgO contents (<7.5 wt%; Figure 2a) than any period since the start ofcontinuous effusion in mid-1986, except duringepisode 54. The lower MgO contents of 2003?2007 lavas was interpreted to have resulted fromthe stirring and flushing of cooler magma withinthe volcano?s shallow magma storage system byan influx of new, hotter more primitive magma[Poland et al., 2012]. Mineralogical evidence (twopopulations of olivine) was noted in support ofthis claim, although no data were presented byPoland et al. [2012]. Our previous study of olivinecompositions in lavas erupted before and duringthe surge found no evidence for two populationsof olivines in any of the lavas, and that olivines inthese weakly phyric rocks are in Fe-Mg equilib-rium with the whole rock [Marske et al., 2008].We re-examined thin sections for these lavas andfound no textural evidence indicating magma mix-ing. In contrast, Pu?u ?O?o lavas from 1983 to1984 and episode 54 display obvious disequili-brium features from magma mixing [Garcia et al.,1989, 2000]. If mixing with a stored, coolermagma was important during the 2003?2007 surgein magma supply, the stored component must nothave differentiated very far beyond olivine control(unlike the situation for Pu?u ?O?o lavas from1983 to 1984 and episode 54).[30] A small increase in lava MgO contentoccurred after the surge, although values werevariable and overlap with those during the surge(7.0?8.1 after versus 6.7?7.4 wt % MgO during;Figure 2). The higher post-surge MgO values wereinterpreted to be a result of the heightened magmasupply from 2003 to 2007 [Poland et al., 2012].By November 2008, MgO dropped to values simi-lar to and lower than during the surge (6.5?7.2 wt%; Figure 2). Since 2000, the MgO content ofPu?u ?O?o lava has been declining with no signifi-cant change of this overall trend during the 2003?2007 surge in magma supply (Figure 2), except forless scatter in MgO content which may simplyreflect fewer interruptions in effusion during thistime. Also, it is noteworthy that the highest sus-tained MgO values were observed for Pu?u ?O?olavas from 1988 to 1993 (Figure 2), a period whenno increase in magma supply was recorded[Poland et al., 2012; Sutton et al., 2003]. Thus,the increase in magma supply from 2003 to 2007appears to have had limited impact on the varia-tion in the MgO content of Pu?u ?O?o lavas.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854867[31] Some of the other compositional features ofthe lavas erupted during the 2003?2007 magmasurge (e.g., slight offsets to lower MgO-normalizedSiO2 contents, Zr/Nb ratios and higher normalizedK2O and TiO2) could be considered to be represen-tative of flushing of differentiated magma similarto the Pu?u ?O?o lavas the late 1990s to early 2000s(Figures 2 and 3). There is also a reversal in87Sr/86Sr and 206Pb/204Pb in 2004 toward values by2007 similar to lavas from 2000 (Figure 6).These trends might be explained by the flushing ofmagma stored since the late 1990s to early 2000sduring the 2003?2007 surge in magma supply.However, based on the close time correlations(weeks to few months) between changes in summittilt and lava composition in earlier Pu?u ?O?o lavas[Garcia et al., 1996; Thornber, 2003], it is hard toimagine why 4? years were required to flushcooler magma from the Pu?u ?O?o system as pro-posed by Poland et al. [2012]. Thus, it is our inter-pretation that the magma erupted during the 2003?2007 surge was new mantle-derived magma andnot stored magma flushed from K?lauea?s shallowcrustal plumbing system. Thus, we interpret theisotopic variations during the Pu?u ?O?o eruption(since 1984) to be generated by the changingcomposition of melts coming from the mantle.5.5. Comparison of Pu?u ?O?o Lavas Withthe Long-Term Isotopic Evolution of OtherHawaiian Volcanoes[32] Studies of the long-term geochemical varia-tion of lavas from individual volcanoes in the Ha-waiian Islands provide understanding of thechemical structure of the Hawaiian mantle plume[e.g., Loa and Kea trends, Abouchami et al., 2005;Ren et al., 2009; Weis et al., 2011], and the vari-ability within single shield volcanoes [e.g., Bryceet al., 2005; Chen and Frey, 1985; Eisele et al.,2003; Marske et al., 2007; Nobre Silva et al.,2013; Rhodes and Hart, 1995; Weis et al., 2011].Here we compare the short-term Pb and Sr iso-topic variation for Pu?u ?O?o lavas to the longerterm variations for lavas from K?lauea and nearby,well-studied shield volcanoes to better understandthe rate and cause of long-term fluctuationsobserved at Hawaiian volcanoes.[33] The Pu?u ?O?o isotopic range covers a rela-tively large part of the long-term variationobserved for Mauna Kea and K?lauea volcanoes(Figures 11 and 12). The isotopic variation ofnearby Mauna Kea volcano was well documentedfor 300 kyr of shield growth using the HSDP2drill core. Lavas from the Pu?u ?O?o eruption afterthe period of magma mixing (early 1985) span30% of the total range of 206Pb/204Pb variationrecorded for HSDP2 (0.07 versus 0.22). Com-pared to K?lauea summit lavas, Pu?u ?O?o lavaserupted since 1985 span 25% of the Pb isotoperange since 950 AD and 47% of the range of Srisotope ratios (Figures 4 and 12). Thus, this singleeruption, which represents <1% of the time cov-ered by the HSDP2 core and 3% of thethousand-year period for K?lauea summit lavas,shows remarkable short-term isotopic variations.However, as seen for K?lauea summit lavas, iso-topic variation in Hawaiian shield lavas is cyclic,with each volcano showing a narrow but com-monly distinctive range (compared to neighboringvolcanoes) as seen by the relatively tight fields forPb and Sr isotope ratios in lavas from K?lauea,Lo?ihi, Mauna Kea and Mauna Loa (Figure 12).Thus, the Pu?u ?O?o eruption may represent thebest known expression of the small-scale compo-sitional heterogeneity of the Hawaiian plume. Asdiscussed above, the rapid rates of isotopic fluctu-ation found in Pu?u ?O?o lavas require a heteroge-neous source (on a scale of less than severalkilometers) that is tapped in only 5 years ofmelt extraction. How does the Pu?u ?O?o sourcecompare with those for nearby, well studied vol-canoes Lo?ihi and Mauna Kea?[34] Compared to other Hawaiian volcanoes, the Srand 206Pb/204Pb isotope ratios for Pu?u ?O?o lavasare most similar to Lo?ihi (Figure 12), althoughLo?ihi lavas have higher 208Pb/204Pb at a given206Pb/204Pb, like other Loa trend volcanoes (Figure11). Pu?u ?O?o lavas overlap with the Kea-mid8 Pbisotope array, the most common lava type inHSDP2 drill core (Figures 11c and 11d) [Eiseleet al., 2003; Nobre Silva et al., 2013]. However, Srisotopic compositions of Pu?u ?O?o lavas eruptedsince 1988 do not overlap those of Mauna Keaand trend orthogonally to the overall inverse arrayfor Hawaiian shield volcanoes and for K?laueasummit lavas (Figure 12). Thus, the Pu?u ?O?osource is isotopically distinct from other Hawaiianvolcanoes and the Pu?u ?O?o data set shows thatindividual eruptions may have trends orthogonal towhat are considered the primary source end-members for Hawaiian shield volcanoes.6. Conclusions[35] The temporal geochemical variation of Pu?u?O?o lavas from 1983 to 2010 provides insights onGREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854868the chemical structure of the Hawaiian mantleplume, and the dynamics of melt transport andmixing within the mantle. Our new results show:[36] 1. The Pu?u ?O?o eruption is being suppliedby new, compositionally variable, mantle-derivedmagma, which is being modified by various crustalprocesses including crystal fractionation (mainlyolivine), magma mixing (during 1983?1984 andepisode 54), and oxygen isotope exchange with orassimilation of altered K?lauea rocks.[37] 2. The episode 56 fissure lava has the highest18O value (5.6%) of any Pu?u ?O?o lava and thehighest MgO of any lava erupted since July 2000.Thus, it may be representative of the magma feed-ing this eruption as of June 2007. The episode 56fissure was fed from an upper East Rift Zone dikerather than the shallow Pu?u ?O?o reservoir. Coe-val lavas erupted from the Pu?u ?O?o vent havelower 18O values (5.36 0.1%) suggesting theywere contaminated, probably in the shallow reser-voir under the vent. Pu?u ?O?o lavas show no cor-relation of 18O values with other geochemicalparameters. For that reason, we suggest that the18O ratios were lowered by oxygen exchangewith or assimilation of altered K?lauea wall rock.[38] 3. Contrary to expectations, the dramaticincrease in magma supply between 2003 and 2007for the Pu?u ?O?o eruption was not accompanied byhigher MgO contents. Instead, lavas erupted duringthe 2003?2007 surge have lower MgO indicative ofgreater cooling of the magma prior to eruption,continuing the long-term trend for the eruption.[39] 4. Rapid and remarkably systematic variationsin Pb and Sr isotopic ratios are present in Pu?u37.737.837.938.038.138.238.338.418.0 18.1 18.2 18.3 18.4 18.5 18.6 18.72.042.052.072.082.092.100.825 0.830 0.835 0.840 0.845 0.850 0.855 0.860L??ihi2.0637.9037.9538.0038.0538.1038.1538.2038.2518.30 18.35 18.40 18.45 18.50 18.55 18.60 18.65 18.702.052.062.072.080.830 0.835 0.840 0.845208Pb/204Pb206Pb/204Pb208Pb/206Pb207Pb/206Pb208Pb/204Pb206Pb/204Pb208Pb/206Pb207Pb/206PbMauna LoaMauna LoaMauna KeaHistorical K?lauea summitPu?u ???? (1983-2010) (b)(d)(c)(a)1986Uwekahuna Bluff (summit)(Pietruszka and Garcia, 1999)Prehistoric K?lauea(Abouchami et al., 2005)L??ihi19831986Hilina BenchHilina Bench(Marske et al., 2007)upper endmemberkea-hi8kea-mid8kea-low8Mauna KeaMauna Kea (HSDP-2)Kea-hi8 arrayKea-mid8 arrayKea-low8 arrayarea of panel c)area of panel d)Historical K?lauea summitLoaKeaLoaKea1983Figure 11. Pb isotopic compositions for Pu?u ?O?o lavas compared to some other Hawaiian shield volca-noes. Top panels are different scales than bottom panels, shown by area of dashed boxes. Colored symbols arelavas from K?lauea Volcano. Fields are for Mauna Loa, Lo?ihi, Mauna Kea, and submarine prehistoricK?lauea (Hilina Bench; >40 ka). References for data sources are listed in the supporting information. Keaarray end-members are from Eisele et al. [2003]. Dashed line with yellow stars in Figure 11c is a best fit linefor Pu?u ?O?o lavas from 1988 to 2010.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854869?O?o lavas. Two cycles of Pb isotopic ratio varia-tion with 10 year periods were found. Thesecycles may be related to extraction of melt from asource with a pattern of vertically oriented sourceheterogeneities, or thin strands. These strands maybe 1?3 km in diameter to explain the scale ofisotopic variation for the Pu?u ?O?o eruption.[40] 5. The Pb isotopic variation of Pu?u ?O?olavas spans 25% of the range observed for the last1000 years of K?lauea summit lavas and 30% for300,000 years of shield volcanism for MaunaKea volcano. There is considerable Pb and Sr iso-topic overlap between Pu?u ?O?o lavas and lavasfrom Mauna Kea and Lo?ihi volcanoes. However,the Pb-Sr isotopic trend for the later Pu?u ?O?olavas (1988?2010) is oblique to the array definedby Hawaiian shield lavas. Thus, each Hawaiianvolcano appears to have an isotopically distinctsource.0.70320.70330.70340.70350.70360.70370.70380.70390.704018.0 18.1 18.2 18.3 18.4 18.5 18.6 18.7 18.80.703450.703500.703550.703600.703650.7037018.2 18.3 18.4 18.5 18.6 18.7 18.8Mauna LoaL??ihiMauna KeaMauna LoaMauna KeaL??ihiSalt Lake Crater xenolithsPu?u ???? (1983-2010) Mauna Kea (HSDP-2)Kea-hi8 arrayKea-mid8 arrayKea-low8 array206Pb/204Pb87Sr/ 86Sr206Pb/204Pb87Sr/ 86Sr198319851986Hilina BenchHilina BenchHistorical K?lauea summitUwekahuna Bluff (summit)(Pietruszka and Garcia, 1999)Prehistoric K?lauea(Abouchami et al., 2005)(Marske et al., 2007)area of panel b)Uwekahuna Bluff (summit)200320100.4-5 kaHistorical and PrehistoricK?lauea summit not shownHistorical K?lauea summit(b)(a)Figure 12. Plots of 206Pb/204Pb and 87Sr/86Sr for Pu?u ?O?o lavas compared to some other Hawaiian shieldvolcanoes. Figure 12b Expanded scale of dashed box in Figure 12a. Colored symbols are lavas from K?laueaVolcano. Fields are for Mauna Loa, Lo?ihi, Mauna Kea and submarine K?lauea (Hilina Bench; >40 ka).References for data sources are listed in the supporting information.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854870Acknowledgments[41] We thank Jane Barling, Bruno Kieffer, and Vivian Laifor their assistance with analyses at PCIGR at UBC, KyleTanguichi and Adonara Murek for sample preparation andcuration at the University of Hawai?i, and J. M. Rhodes forXRF analyses at University of Massachusetts. Claude Maer-schalk assisted with Pb and Sr column chemistry for a subsetof samples. Daniel Heaton provided assistance with severalisotope analyses at San Diego State University. We appreciatereviews by Julie Prytulak, Joel Baker, and Christoph Beier.This research was supported by grants from the National Sci-ence Foundation to M. Garcia (EAR11?18741) and A. Pie-truszka (EAR11?18738). This paper is SOEST ContributionNo. 8939.ReferencesAbouchami, W., S. J. G. Galer, and A. W. Hofmann (2000),High precision lead isotope systematics of lavas from theHawaiian Scientific Drilling Project, Chem. Geol., 169, 187?209, doi:10.1016/S0009?2541(00)00328-4.Abouchami, W., A. W. Hofmann, S. J. G. Galer, F. A. Frey, J.Eisele, and M. Feigenson (2005), Lead isotopes reveal bilat-eral asymmetry and vertical continuity in the Hawaiian man-tle plume, Nature, 434, 851?856, doi:10.1038/nature03402.Albare`de, F., B. Luais, G. Fitton, M. Semet, E. Kaminski, B. G.J. Upton, P. Bache`lery, and J.-L. Cheminee (1997), The geo-chemical regimes of Piton de la Fournaise Volcano (Reunion)during the last 530,000 years, J. Petrol., 38, 171?201.Blichert-Toft, J., and F. Albare`de (2009), Mixing of isotopicheterogeneities in the Mauna Kea plume conduit, EarthPlanet. Sci. Lett., 282, 190?200, doi:10.1016/j.epsl.2009.03.015.Blichert-Toft, J., D. Weis, C. Maerschalk, A. Agranier, and F.Albare`de (2003), Hawaiian hot spot dynamics as inferredfrom the Hf and Pb isotope evolution of Mauna Kea volcano,Geochem. Geophys. Geosyst., 4(2), 8704, doi:10.1029/2002GC000340.Bryce, J. G., D. J. DePaolo, and J. C. Lassiter (2005), Geo-chemical structure of the Hawaiian plume: Sr, Nd, and Osisotopes in the 2.8 km HSDP-2 section of Mauna Kea vol-cano, Geochem. Geophys. Geosyst., 6, Q09G18,doi:10.1029/2004GC000809.Caroff, M., R. C. Maury, P. Vidal, G. Guille, C. Dupuy, J. Cot-ten, H. Guillou, and P.-Y. Gillot (1995), Rapid temporalchanges in ocean island basalt composition: Evidence froman 800 m deep drill hole in Eiao Shield (Marquesas), J. Pet-rol., 36(5), 1333?1365, doi:10.1093/petrology/36.5.1333.Chen, C. Y., and F. A. Frey (1985), Trace element and isotopicgeochemistry of lavas from Haleakala Volcano, east Maui,Hawai?i : Implications for the origin of Hawaiian basalts, J.Geophys. Res., 90, 8743?8768.Chen, C.-Y., F. A. Frey, J. M. Rhodes, and R. M. Easton(1996), Temporal geochemical evolution of K?lauea Vol-cano: Comparison of Hilina and Puna basalt, in Earth Proc-esses, Reading the Isotopic Code, edited by A. Basu and S.R. Hart, pp. 161?181, AGU, Washington, D. C.Cordier, C., M. Benoit, C. Hemond, J. Dyment, B. Le Gall, A.Briais, and M. Kitazawa (2010), Time scales of melt extrac-tion revealed by distribution of lava composition across aridge axis, Geochem. Geophys. Geosyst., 11, Q0AC06,doi:10.1029/2010gc003074.DePaolo, D. J., and E. M. Stolper (1996), Models of Hawaiianvolcano growth and plume structure: Implications from theHawai?i Scientific Drilling Project, J. Geophys. Res., 101,11,643?11,655.Dvorak, J. J., and D. Dzurisin (1993), Variations in magmasupply rate at K?lauea Volcano, Hawai?i, J. Geophys. Res.,98, 22,255?22,268.Eaton, J. P., and K. J. Murata (1960), How volcanoes grow[Hawai?i], Science, 132, 925?938.Eiler, J. M. (2001), Oxygen isotope variations of basaltic lavasand upper mantle rocks, Rev. Mineral. Geochem., 43(1),319?364, doi:10.2138/gsrmg.43.1.319.Eisele, J. R., W. Abouchami, S. J. G. Galer, and A. W. Hof-mann (2003), The 320 kyr Pb isotope evolution of MaunaKea lavas recorded in the HSDP-2 drill core, Geochem. Geo-phys. Geosyst., 4(5), 8710, doi:10.1029/2002GC000339.Farnetani, C. G., and A. W. Hofmann (2009), Dynamics andinternal structure of a lower mantle plume conduit, EarthPlanet. Sci. Lett., 282(1?4), 314?322, doi:10.1016/j.epsl.2009.03.035.Farnetani, C. G., and A. W. Hofmann (2010), Dynamics andinternal structure of the Hawaiian plume, Earth Planet. Sci.Lett., 295, 231?240, doi:10.1016/j.epsl.2010.04.005.Garcia, M. O., R. A. Ho, J. M. Rhodes, and E. W. Wolfe(1989), Petrologic constraints on rift-zone processes:Results from episode 1 of the Pu?u ?O?o eruption of K?laueavolcano, Hawai?i, Bull. Volcanol., 52, 81?96.Garcia, M. O., J. M. Rhodes, E. W. Wolfe, G. E. Ulrich, and R.A. Ho (1992), Petrology of lavas from episodes 2?47 of thePu?u ?O?o eruption of K?lauea Volcano, Hawai?i : Evaluationof magmatic processes, Bull. Volcanol., 55(1), 1?16.Garcia, M. O., J. M. Rhodes, F. A. Trusdell, and A. J. Pie-truszka (1996), Petrology of lavas from the Pu?u ?O?o erup-tion of K?lauea Volcano: III The Kupaianaha episode(1986?1992), Bull. Volcanol., 58, 359?379.Garcia, M. O., E. Ito, J. M. Eiler, and A. J. Pietruszka (1998),Crustal contamination of K?lauea Volcano magmas revealedby oxygen isotope analyses of glass and olivine from Pu?u?O?o eruption lavas, J. Petrol., 39, 803?817.Garcia, M. O., A. J. Pietruszka, J. M. Rhodes, and K. Swanson(2000), Magmatic processes during the prolonged Pu?u ?O?oeruption of K?lauea Volcano, Hawai?i, J. Petrol., 41(7),967?990.Garcia, M. O., A. J. Pietruszka, and J. M. Rhodes (2003), Apetrologic perspective of K?lauea Volcano?s summit magmareservoir, J. Petrol., 44(12), 2313?2339, doi:10.1093/petrol-ogy/egg079.Garcia, M. O., E. Ito, and J. M. Eiler (2008), Oxygen isotopeevidence for chemical interaction of K?lauea historical mag-mas with basement rocks, J. Petrol., 49(4), 757?769,doi :10.1093/petrology/egm034.Hanyu, T., J.-I. Kimura, M. Katakuse, A. T. Calvert, T. W. Sis-son, and S. I. Nakai (2010), Source materials for inceptionstage Hawaiian magmas: Pb-He isotope variations for earlyK?lauea, Geochem. Geophys. Geosyst., 11, Q0AC01,doi:10.1029/2009gc002760.Harmon, R. S., and J. Hoefs (1995), Oxygen isotope heteroge-neity of the mantle deduced from global 18O systematics ofbasalts from different geotectonic settings, Contrib. Mineral.Petrol., 120(1), 95?114.Hauri, E. H. (1996), Major-element variability in the Hawaiianmantle plume, Nature, 382, 415?419.Heliker, C., and T. N. Mattox (2003), The first two decades ofthe Pu?u ?O?o -Kupaianaha eruption; chronology andselected bibliography, in The Pu?u ?O?o -Kupaianaha Erup-tion of K?lauea Volcano, Hawai?i: The First 20 Years, editedGREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854871by C. Heliker et al., pp. 1?28, U.S. Geol. Surv. Prof. Pap.1676.Hofmann, A. W., M. D. Feigenson, and I. Raczek (1984), Casestudies on the orgin of basalt : III Petrogenesis of the MaunaUlu eruption, K?lauea, 1969?1971, Contrib. Mineral. Pet-rol., 88, 24?35.Holcomb, R. T. (1987), Eruptive history and long-term behav-ior of K?lauea Volcano, in Volcanism in Hawai?i, edited byR. W. Decker et al., pp. 261?350, U.S. Geol. Surv. Prof. Pap.1350.Jackson, M. G., D. Weis, and S. Huang (2012), Major elementvariations in Hawaiian shield lavas: Source features and per-spectives from global ocean island basalt (OIB) systematics,Geochem. Geophys. Geosyst., 13, Q09009, doi:10.1029/2012GC004268.Johnson, D. J., A. A. Eggers, and M. Bagnardi (2010), Shallowmagma accumulation at K?lauea Volcano, Hawai?i, revealedby microgravity surveys, Geology 38(12), 1139?1142,doi:10.1130/G31323.1.Kimura, J.-I., T. W. Sisson, N. Nakano, M. L. Coombs, and P.W. Lipman (2006), Isotope geochemistry of early K?laueamagmas from the submarine Hilina bench: The nature of theHilina mantle component, J. Volcanol. Geotherm. Res.,151(1?3), 51?72, doi :10.1016/j.jvolgeores.2005.07.024.Kurz, M. D., W. J. Jenkins, and S. R. Hart (1982), Helium iso-topic systematics of oceanic islands and mantle heterogene-ity, Nature, 297, 43?47.Macdonald, G. A., A. T. Abbott, and F. L. Peterson (1983),Volcanoes in the Sea: The Geology of Hawai?i, 517 pp.,Univ. of Hawaii Press, Honolulu, Hawaii.Maclennan, J. (2008), Lead isotope variability in olivine-hosted melt inclusions from Iceland, Geochim. Cosmochim.Acta, 72(16), 4159?4176, doi:10.1016/j.gca.2008.05.034.Marske, J. P., A. J. Pietruszka, D. Weis, M. O. Garcia, and J.M. Rhodes (2007), Rapid passage of a small-scale mantleheterogeneity through the melting regions of K?lauea andMauna Loa volcanoes, Earth Planet. Sci. Lett., 259, 34?50,doi:10.1016/j.epsl.2007.04.026.Marske, J. P., M. O. Garcia, A. J. Pietruszka, J. M. Rhodes,and M. D. Norman (2008), Geochemical variations duringK?lauea?s Pu?u ?O?o eruption reveal a fine-scale mixture ofmantle heterogeneities within the Hawaiian plume, J. Pet-rol., 49(7), 1297?1318, doi:10.1093/petrology/egn025.Montgomery-Brown, E. K., D. K. Sinnett, P. S. M. Poland, T.Orr, H. Zebker, and A. Miklius (2010), Geodetic evidencefor en echelon dike emplacement and concurrent slow slipduring the June 2007 intrusion and eruption at K?lauea vol-cano, Hawai?i, J. Geophys. Res., 115, B07405, doi:10.1029/2009JB006658.Nobre Silva, I. G., D. Weis, and J. S. Scoates (2013), Isotopicsystematics of the early Mauna Kea shield phase and insightinto the deep mantle beneath the Pacific Ocean, Geochem.Geophys. Geosyst., 14(3), 659?676, doi:10.1029/2012GC004373.Pietruszka, A. J., and M. O. Garcia (1999), A rapid fluctuationin the mantle source and melting history of K?lauea Volcanoinferred from the geochemistry of its historical summit lavas(1790?1982), J. Petrol., 48(8), 1321?1342.Pietruszka, A. J., K. H. Rubin, and M. O. Garcia (2001),226Ra-230Th-238U disequilibria of historical K?lauea lavas(1790?1982) and the dynamics of mantle melting within theHawaiian plume, Earth Planet. Sci. Lett., 186, 15?31.Pietruszka, A. J., E. H. Hauri, R. W. Carlson, and M. O. Garcia(2006), Remelting of recently depleted mantle within theHawaiian plume inferred from the 226Ra-230Th-238U disequi-libria of Pu?u ?O?o eruption lavas, Earth Planet. Sci. Lett.,244(1?2), 155?169, doi:10.1016/j.epsl.2006.01.039.Pietruszka, A. J., E. H. Hauri, and J. Blichert-Toft (2009),Crustal contamination of mantle-derived magmas within Pi-ton de la Fournaise Volcano, Reunion Island, J. Petrol.,50(4), 661?684, doi :10.1093/petrology/egp016.Pietruszka, A. J., M. D. Norman, M. O. Garcia, J. P. Marske,and D. H. Burns (2013), Chemical heterogeneity in the Ha-waiian mantle plume from the alteration and dehydration ofrecycled oceanic crust, Earth Planet. Sci. Lett., 361, 298?309, doi:10.1016/j.epsl.2012.10.030.Poland, M., A. Miklius, T. Orr, A. J. Sutton, C. Thornber, andD. Wilson (2008), New episodes of volcanism at K?laueaVolcano, Hawai?i, Eos Trans. AGU, 89(5), 37?38,doi :10.1029/2008EO050001.Poland, M. P., A. Miklius, A. J. Sutton, and C. R. Thornber(2012), A mantle-driven surge in magma supply to K?laueaVolcano during 2003?2007, Nat. Geosci., 5, 295?300,doi :10.1038/NGEO1426.Reiners, P. W. (2002), Temporal-compositional trends inintraplate basalt eruptions: Implications for mantle hetero-geneity and melting processes, Geochem. Geophys. Geo-syst., 3(2), 1011, doi:10.1029/2001GC000250.Ren, Z.-y., T. Hanyu, T. Miyazaki, Q. Chang, H. Kawabata, T.Takahashi, Y. Hirahara, A. R. L. Nichols, and Y. Tatsumi(2009), Geochemical differences of the Hawaiian shieldlavas: Implications for melting process in the heterogeneousHawaiian plume, J. Petrol., 50(8), 1553?573, doi:10.1093/petrology/egp041.Rhodes, J. M., and S. R. Hart (1995), Episodic trace elementand isotopic variations in Mauna Loa lavas: Implications formagma and plume dynamics, in Mauna Loa Revealed:Structure, Composition, History and Hazards, edited by J.M. Rhodes and J. P. Lockwood, pp. 263?288, AGU, Wash-ington, D. C.Rhodes, J. M., and M. J. Vollinger (2004), Composition of ba-saltic lavas sampled by phase-2 of the Hawai?i ScientificDrilling Project: Geochemical stratigraphy and magmatypes, Geochem. Geophys. Geosyst., 5, Q03G13,doi:10.1029/2002GC000434.Rhodes, J. M., S. Huang, F. A. Frey, M. Pringle, and G. Xu(2012), Compositional diversity of Mauna Kea shield lavasrecovered by the Hawai?i Scientific Drilling Project: Infer-ences on source lithology, magma supply, and the role ofmultiple volcanoes, Geochem. Geophys. Geosyst., 13,Q03014, doi:10.1029/2011GC003812.Ryan, M. P. (1987), Elasticity and contractancy of Hawaiianolivine tholeiite and its role in the stability and structuralevolution of subcaldera magma reservoirs and rift systems,in Volcanism in Hawai?i, edited by R. W. Decker et al., pp.1395?1448, U.S. Geol. Surv. Prof. Pap. 1350.Shamberger, P. J., and M. O. Garcia (2007), Geochemicalmodeling of magma mixing and magma reservoir volumesduring early episodes of K?lauea Volcano?s Pu?u ?O?o erup-tion, Bull. Volcanol., 69, 345?352, doi:10.1007/s00445-006-0074-5.Sims, K. W. W., M. T. Murrell, D. J. DePaolo, W. S. Bal-dridge, S. J. Goldstein, D. Clague, and M. Jull (1999), Poros-ity of the melting zone and variations in the solid mantleupwelling rate beneath Hawai?i : Inferences from238U-230Th-226Ra and 235U-231Pa disequilibria, Geochim.Cosmochim. Acta, 63(23), 4119?4138.Sobolev, A. V., A. W. Hofmann, S. Sobolev, and I. K. Nikogo-sian (2005), An olivine-free mantle source of Hawaiianshield basalts, Nature, 434, 590?595.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854872Sutton, J. A., T. Elias, and J. Kauahikaua (2003), Lava-effu-sion rates for the Pu?u ?O?o -Kupaianaha eruption derivedfrom SO2 emissions and very low frequency (VLF) meas-urements, in The Pu?u ?O?o -Kupaianaha Eruption ofK?lauea Volcano, Hawai?i: The First 20 Years, edited byC. Heliker et al., pp. 137?148, U.S. Geol. Surv. Prof. Pap.1676.Thornber, C. R. (2003), Magma-reservoir processes revealedby geochemistry of the Pu?u ?O?o-Kupaianaha eruption, inThe Pu?u ?O?o -Kupaianaha Eruption of K?lauea Volcano,Hawai?i: The First 20 Years, edited by C. Heliker et al., pp.121?136, U.S. Geol. Surv. Prof. Pap. 1676.Thornber, C. R., C. Heliker, D. R. Sherrod, J. P. Kauahikaua,A. Miklius, P. G. Okubo, F. A. Trusdell, J. R. Budahn, W. I.Ridley, and G. P. Meeker (2003), K?lauea East Rift Zonemagmatism: An episode 54 perspective, J. Petrol., 44(9),1525?1559, doi:10.1093/petrology/egg048.Tilling, R. I., and J. J. Dvorak (1993), Anatomy of a basalticvolcano, Nature, 363, 125?133.Vlastelic, I., T. Staudacher, and M. Semet (2005), Rapidchange of lava composition from 1998 to 2002 at Piton de laFournaise (Reunion) inferred from Pb isotopes and trace ele-ments: Evidence for variable crustal contamination, J. Pet-rol., 46(1), 79?107, doi :10.1093/petrology/egh062.Vlastelic, I., A. Peltier, and T. Staudacher (2007), Short-term(1998?2006) fluctuations of Pb isotopes at Piton de la Four-naise volcano (Reunion Island): Origins and constraints onthe size and shape of the magma reservoir, Chem. Geol.,244, 202?220, doi:10.1016/j.chemgeo.2007.06.015.Watson, S., and D. McKenzie (1991), Melt generation byplumes: A study of Hawaiian volcanism, J. Petrol., 32(3),501?537.Weis, D., M. O. Garcia, J. M. Rhodes, A. M. Jellinek, and J. S.Scoates (2011), Role of the deep mantle in generating thecompositional asymmetry of the Hawaiian mantle plume,Nat. Geosci., 4, 831?838, doi:10.1038/ngeo1328.Wolfe, E. W., M. O. Garcia, D. B. Jackson, R. Y. Koyanagi, C.A. Neal, and A. T. Okamura (1987), The Pu?u ?O?o eruptionof K?lauea Volcano, episodes 1?20, January 3, 1983 to June8, 1984, in Volcanism in Hawai?i, edited by R. W. Deckeret al., pp. 471?508, U.S. Geol. Surv. Prof. Pap. 1350.Wright, T. L. (1971), Chemistry of K?lauea and Mauna Loalava in space and time, U.S. Geol. Surv. Prof. Pap., 735,1?45.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854873 1  Sample Preparation and Analytical Methods  A1. University of Massachusetts XRF Analytical Methods New XRF analyses are given in the Auxiliary Materials (ts1) with previous major-element analyses. New XRF trace element analyses are also given (and indicated in table) for samples erupted prior to 1998, when a new, more precise XRF instrument became available. Fresh pieces of quenched lava were crushed in a tungsten carbide jaw crusher to ~2 mm diameter and cleaned in a beaker with deionized water and dried in an oven at 70?C for 24 hours to drive off excess water. Lava fragments were then powdered using a Rocklabs WC shatterbox for 1 ? 2 minutes. All XRF analyses were performed at the University of Massachusetts XRF Laboratory where whole-rock, major-element abundances were measured using the procedures of Rhodes and Vollinger [2004]. One sigma accuracy and precision estimates for the XRF data are ~0.5% for major elements [Rhodes, 1988]. Five grams of a powdered sample were heated in a muffle furnace at 1020?C for 10 minutes to limit the amount of ferrous iron formation [Rhodes and Vollinger, 2004]. The weight loss is LOI. Major elements were measured on a fused La-bearing lithium borate glass disc using a Siemens MRS-400 spectrometer with a Rh X-ray tube operating at 2700 W (the voltage was 45 kV and 60mA and the count times were 200 seconds). Trace element concentrations (Rb, Sr, Ba, La, Ce, Nb, Zr, Y, Zn, Ga, Ni, Cr, V) were measured on a separate powder pellet using a Philips PW2400 sequential spectrometer with a Rh X-ray tube operating at 3000 W (the voltage was 60kV and 50mA and 2.5 hour total count time). Precision and accuracy estimates for the trace element data are described by Rhodes [1996] and Rhodes and Vollinger [2004]. Results for each sample are the average of two separate analyses (2 disks for one sample) for major elements, but one analysis (one pellet run once) for trace elements. A summary of XRF trace element values of USGS standards analyzed with lavas from K?lauea?s Pu?u ???? eruption since 2002 are given in the Auxiliary Materials (ts3).  A2. PCIGR and SDSU Trace Element and Isotopic Analytical Methods Forty-one Pu?u ???? lava samples from 1983 to 1999 were selected for Pb and Sr isotopic analysis at the Pacific Centre for Isotopic and Geochemical Research (PCIGR) at the University of British Columbia (UBC; Table 2). The analytical methods for thirteen additional Pb isotopic analyses of Pu?u ???? lavas collected between 2006 and 2010, and fourteen analyses from Marske et al. [2008] that are used in this study, are described in Marske et al. [2007] (samples 2  are indicated in Table 2). Thirteen samples without trace-element data were selected for high-precision trace-element analysis and results are shown in Auxiliary Materials (ts2) with previous trace-element analyses for Pu?u ???? lavas (corrected data is also shown, as described in table caption). Isotopic ratios of Pb for these samples were measured with a Nu Plasma (Nu Instruments) 1700 Multiple Collector-Inductively Coupled Plasma-Mass Spectrometer (MC-ICP-MS) at San Diego State University (SDSU). Sr isotopic ratios for 1998-2005 lava samples from Marske et al. [2008] were measured using a VG Sector 54 thermal ionization mass spectrometer (TIMS) at SDSU. Lava samples from 2006 to 2010 were analyzed for Sr isotopic ratios at the PCIGR, according the the methods described below.  Samples were prepared for trace-element analysis at the PCIGR by the technique described by Pretorius et al. [2006] and Carpentier et al. [2013] on unleached lava fragments. Lava fragments <2 mm in diameter (~100 mg) were weighed in 7 mL screw-top Savillex? beakers and dissolved in 1 mL ~14N HNO3 and 5 mL 48% HF on a hotplate for 48 hours at 130?C with periodic ultrasonication. Samples were dried and redissolved in 6 mL 6N HCl on a hotplate for 24 hours and then dried and redissolved in 1 mL concentrated HNO3 for 24 hours before final drying. Trace element abundances were measured with a Thermo Finnigan Element2 High Resolution-Inductively Coupled Plasma-Mass Spectrometer (HR-ICP-MS) within 24 hours of redissolution, following the procedures described by Pretorius et al. [2006] and Carpentier et al. [2013]. High field strength elements (HFSE) and large ion lithophile elements (LILE) were measured in medium resolution mode at 2000x dilution using a PFA teflon spray chamber washed with Aqua Regia for 3 minutes between samples. Rare earth elements (REE) were measured in high resolution mode, and U and Pb in low resolution mode, at 2000x dilution using a glass spray chamber washed with 2% HNO3 between samples. Total procedural blanks and reference material (Kil93) was analyzed with the batch of samples. Indium was used as an internal standard in all samples and standard solutions. Background and standard solutions were analyzed after every 5 samples to detect memory effects and mass drift.  For isotope chemistry at PCIGR, fresh pieces of crushed lava (<2mm diameter; ~0.6-0.8 g total) were carefully picked using a binocular microscope to avoid signs of post-eruptive alteration. 3  Each sample was ultra-sonicated (15 minutes) and rinsed three times in ultra-pure water in a clean Savillex? beaker prior to rinsing with room temperature 0.5 M HBr for 15 seconds to remove any surface contamination. The sample was washed again in ultra-pure water, dried, and weighed. Sample digestion for purification of Pb and Sr by column chemistry involved dissolving ~100-250 mg of the cleaned, crushed lava in 1 mL ~14N HNO3 and 10 mL 48% HF on a hotplate for 48 hours at 130?C with periodic ultrasonication. Samples were dried and redissolved in 6 mL 6N HCl on a hotplate for 24 hours and then dried. Pb was separated and purified in two passes through anion exchange columns and the discard was used for Sr separation. Detailed procedures for column chemistry for separating Pb and Sr at the PCIGR are described in Weis et al. [2006] and Nobre Silva et al. [2009]. Sr isotope ratios were measured on a Thermo Finnigan Triton Thermal Ionization Mass Spectrometer (TIMS) in static mode with relay matrix rotation on a single Ta and double Re-Ta filament, respectively. Four to 5 filaments per barrel of 21 were occupied by standard NIST SRM 987 for each barrel where samples were run. Sample Sr isotopic compositions were corrected for mass fractionation using 86Sr/88Sr = 0.1194. Each sample was then normalized using the barrel average of the reference material relative to the values of 87Sr/86Sr = 0.710248 [Weis et al., 2006]. During the period when the samples were analyzed, the NIST SRM 987 standard gave an average of 0.710243 ? 13 (n = 18; 2? error is reported as times 106). PCIGR internal reference material from the K?lauea summit eruption in 1919 (Kil1919) was processed with the samples and yielded Sr isotopic ratios of 0.703468 ? 8. Twenty-six previous analyses of Kil1919 at SDSU average 0.703471 ? 13. Four chemistry duplicates of Menehune standard (an in-house glass standard collected from a Pu?u ???? lava flow of K?lauea Volcano that was quenched on June 24, 2006), processed with the other lava samples, had Sr isotopic compositions of  0.703608 ? 7, 0.703605 ? 8, 0.703611 ? 8, and 0.703624 ? 9. Twenty-six previous analyses of Menehune standard performed at SDSU averaged 0.703617 ? 13.  Pb isotopic compositions at PCIGR were analyzed by static multi-collection on a Nu Plasma MC-ICP-MS. The detailed analytical procedure for Pb isotopic analyses at the PCIGR is described in Weis et al. [2005]. The configuration for Pb analyses allows for collection of Pb, Tl, and Hg together. Tl and Hg are used to monitor instrumental mass discrimination and isobaric overlap, repectively. All sample solutions were analyzed with approximately the same Pb/Tl 4  ratio (~4) as the reference material NIST SRM 981. To accomplish this, a small aliquot of each sample solution from the Pb columns was analyzed on the Nu Plasma to determine the amount of Pb available. The SRM 981 standard was run after every two samples on the Nu Plasma. During the time samples were run, analyses of the SRM 981 Pb reference material gave values of 206Pb/204Pb = 16.9430 ? 19, 207Pb/204Pb = 15.4997 ? 18, and 208Pb/204Pb = 36.7221 ? 58 (n = 38; 2? error is reported as times 104); these values are in excellent agreement with reported TIMS triple-spike values of Galer and Abouchami [1998]. Results were further corrected by the sample-standard bracketing method described by White et al. [2000]. Reference material Kil1919 analyzed at PCIGR yielded Pb isotopic ratios of 206Pb/204Pb = 18.6556 ? 8, 207Pb/204Pb = 15.4924 ? 8, and 208Pb/204Pb = 38.2148 ? 21. Thirty-three analyses of Kil1919 performed at SDSU over the course of this study averaged 206Pb/204Pb = 18.6552 ? 27, 207Pb/204Pb = 15.4897 ? 20, and 208Pb/204Pb = 38.2068 ? 57. Four chemistry duplicates of Menehune standard analyzed at PCIGR averaged 206Pb/204Pb = 18.4062 ? 28, 207Pb/204Pb = 15.4724 ? 26, and 208Pb/204Pb = 38.0662 ? 59. Sixty-eight previous analyses of Menehune standard performed at SDSU over the course of this study averaged 206Pb/204Pb = 18.4073 ? 16, 207Pb/204Pb = 15.4714 ? 18, and 208Pb/204Pb = 38.0627 ? 60.  A3. Glass and matrix oxygen isotopes Glass and holocrystalline matrix material were separated from Pu?u ???? lavas by hand picking from coarsely crushed but otherwise untreated samples. Oxygen isotope compositions of c. 2 mg aliquots of these separates were determined by laser fluorination, using a 50W CO2 laser and BrF5 as reagent [Sharp, 1990; Valley et al., 1995]. Product O2 was converted to CO2 by reaction with hot graphite; CO2 was then analyzed for its isotopic composition by dual-inlet gas source mass spectrometry on a Thermo Finnegan Delta XL gas source isotope ratio mass spectrometer at California Institute of Technology. Data are reported in units of per mil versus the VSMOW standard. Analyses were standardized by comparison with measurements of Gore Mountain garnet standard [Valley et al., 1995]. This standard was analyzed between two and five times each day of analyses, and the data for unknowns analyzed on that day were corrected by the average difference between measured and accepted values for that standard. The external precision of repeat measurements of separate splits of unknown samples averaged 0.05? (1?). Seven analyses of Gore Mtn. Garnet were run concurrently with these samples, with a standard 5  deviation of ?0.06 ? (1?; averaged over all days).  This is comparable with the typical external precision for replicate measurements of silicate standards for this laboratory and technique [e.g., Bindeman et al., 2006; Eiler et al., 1995; Eiler et al., 1996], suggesting that the glass separates analyzed in this study are homogeneous in ?18O at the scale of c. 2 mg aliquots. The results reported here are relative to ?18OVSMOW [Coplen, 1988]. Four analyses of Menehune standard performed with the other Pu?u ???? lavas yielded 5.22 ? 0.03 ? (1?).  References  Bindeman, I. N., O. Sigmarsson, and J. M. Eiler (2006), Time constraints on the origin of large volume basalts derived from O-isotope and trace element mineral zoning and U-series disequilibria in the Laki and Grimsvotn volcanic system, Earth Planet. Sci. Lett., 245, 245-259. Carpentier, M., D. Weis, and C. Chauvel (2013), Large U loss during weathering of upper continental crust: The sedimentary record. , Chemical Geology (Isotope Geosciences Section), 340, 91-104. Coplen, T. B. (1988), Normalization of oxygen and hydrogen isotope data, Chemical Geology (Isotope Geosciences Section)  72, 293-297. Eiler, J. M., K. A. Farley, J. W. Valley, E. M. Stolper, E. H. Hauri, and H. Craig (1995), Oxygen isotope evidence against bulk recycled sediment in the mantle sources of Pitcairn Island lavas, Nature, 377, 138-141. Eiler, J. M., J. W. Valley, and E. M. Stolper (1996), Oxygen isotope ratios in olivine from the Hawaii Scientific Drilling Project, J. Geophys. Res., 101(B5), 11,807-11,813. Galer, S. J. G., and W. Abouchami (1998), Practical application of lead triple spiking for correction of instrumental mass discrimination, Mineral. Mag., 62A, 491-492. Marske, J. P., A. J. Pietruszka, D. Weis, M. O. Garcia, and J. M. Rhodes (2007), Rapid passage of a small-scale mantle heterogeneity through the melting regions of Kilauea and Mauna Loa volcanoes, Earth Planet. Sci. Lett., 259, 34-50. Marske, J. P., M. O. Garcia, A. J. Pietruszka, J. M. Rhodes, and M. D. Norman (2008), Geochemical variations during Kilauea's Pu'u 'O'o eruption reveal a fine-scale mixture of mantle heterogeneities within the Hawaiian plume, J. Petrol., 49(7), 1297-1318. Nobre Silva, I. G., D. Weis, J. Barling, and J. S. Scoates (2009), Leaching systematics and matrix elimination for the determination of high-precision Pb isotope compositions of ocean island basalts, Geochem. Geophys. Geosyst., 10(8), Q08012. Pretorius, W., D. Weis, G. Williams, D. Hanano, B. Kieffer, and J. S. Scoates (2006), Complete trace elemental characterization of granitoid (USGSG-2,GSP-2) reference materials by high resolution inductively coupled plasma-mass spectrometry, Geost. and Geoanaly. Res., 30(1), 39-54. Rhodes, J. M. (1988), Geochemistry of the 1984 Mauna Loa eruption: Implications for magma storage and supply, J. Geophys. Res., 93, 4,453-454,466. Rhodes, J. M., and M. J. Vollinger (2004), Composition of basaltic lavas sampled by phase-2 of the Hawaii Scientific Drilling Project: Geochemical stratigraphy and magma types, Geochem. Geophys. Geosyst., 5(3). 6  Sharp, Z. D. (1990), A laser-based microanalytical method for the in situ determination of oxygen isotope ratios of silicates and oxides, Geochim. Cosmochim. Acta, 54(5), 1353-1357. Valley, J. W., N. Kitchen, and M. J. Kohn (1995), Strategies for high-precision oxygen isotope analysis by laser fluorination, Geochimica Cosmoschimica Acta, 59, 5223-5231. Weis, D., B. Kieffer, C. Maerschalk, W. Pretorius, and J. Barling (2005), High-precision Pb-Sr-Nd-Hf isotopic characterization of USGS BHVO-1 and BHVO-2 reference materials, Geochem. Geophys. Geosyst., 6, (Q02002). Weis, D., B. Kieffer, C. Maerschalk, J. Barling, J. de Jong, G. A. Williams, D. Hanano, N. Mattielli, J. S. Scoates, A. Goolaerts, R. A. Friedman, and J. B. Mahoney (2006), High-precision isotopic characterization of USGS reference materials by TIMS and MC-ICP-MS, Geochem. Geophys. Geosyst., 7(Q08006). White, W. M., F. Albar?de, and P. T?louk (2000), High-precision analysis of Pb isotope ratios by multi-collector ICP-MS, Chem. Geol., 167, 257-270.    Temporal geochemical variations in lavas fromK?lauea?s Pu?u ?O?o eruption (1983?2010): Cyclicvariations from melting of source heterogeneitiesAndrew R. GreeneDepartment of Natural Sciences, Hawai?i Pacific University, 45-045 Kamehameha Hwy, Kane?ohe, Hawaii,96744, USA (agreene@hpu.edu)Michael O. GarciaDepartment of Geology and Geophysics, University of Hawai?i, Honolulu, Hawaii, USAAaron J. PietruszkaDepartment of Geological Sciences, San Diego State University, San Diego, California, USANow at U. S. Geological Survey, Denver Federal Center, Denver, Colorado, USADominique WeisPacific Centre for Isotopic and Geochemical Research, Department of Earth and Ocean Sciences, University ofBritish Columbia, Vancouver, British Columbia, CanadaJared P. MarskeDepartment of Terrestrial Magnetism, Carnegie Institution, Washington, D.C., USAMichael J. VollingerRonald B. Gilmore XRF Lab, Department of Geosciences, University of Massachusetts, Amherst, Massachusetts, USAJohn EilerPlanetary and Geological Sciences Institute, California Institute of Technology, Pasadena, California, USA[1] Geochemical time series analysis of lavas from K?lauea?s ongoing Pu?u ?O?o eruption chroniclemantle and crustal processes during a single, prolonged (1983 to present) magmatic event, which hasshown nearly two-fold variation in lava effusion rates. Here we present an update of our ongoingmonitoring of the geochemical variations of Pu?u ?O?o lavas for the entire eruption through 2010. Oxygenisotope measurements on Pu?u ?O?o lavas show a remarkable range (18O values of 4.6?5.6%), which areinterpreted to reflect moderate levels of oxygen isotope exchange with or crustal contamination byhydrothermally altered K?lauea lavas, probably in the shallow reservoir under the Pu?u ?O?o vent. Thisprocess has not measurably affected ratios of radiogenic isotope or incompatible trace elements, whichare thought to vary due to mantle-derived changes in the composition of the parental magma delivered tothe volcano. High-precision Pb and Sr isotopic measurements were performed on lavas erupted at 6month intervals since 1983 to provide insights about melting dynamics and the compositional structure ofthe Hawaiian plume. The new results show systematic variations of Pb and Sr isotope ratios thatcontinued the long-term compositional trend for K?lauea until 1990. Afterward, Pb isotope ratios showtwo cycles with 10 year periods, whereas the Sr isotope ratios continued to increase until 2003 andthen shifted toward slightly less radiogenic values. The short-term periodicity of Pb isotope ratios mayreflect melt extraction from mantle with a fine-scale pattern of repeating source heterogeneities or strands,which are about 1?3 km in diameter. Over the last 30 years, Pu?u ?O?o lavas show 15% and 25% of the? 2013. American Geophysical Union. All Rights Reserved. 4849ArticleVolume 14, Number 1115 November 2013doi: 10.1002/ggge.20285ISSN: 1525-2027known isotopic variation for K?lauea and Mauna Kea, respectively. This observation illustrates that thedominant time scale of mantle-derived compositional variation for Hawaiian lavas is years to decades.Components: 13,235 words, 12 figures, 2 tables.Keywords: Hawaiian plume; tholeiitic volcanism; melt extraction; oceanic island.Index Terms: 3619 Magma genesis and partial melting: Mineralogy and Petrology; 3621 Mantle processes: Mineralogyand Petrology; 1025 Composition of the mantle: Geochemistry; 1037 Magma genesis and partial melting: Geochemistry;1038 Mantle processes: Geochemistry.Received 5 March 2013; Revised 9 September 2013; Accepted 7 October 2013; Published 15 November 2013.Greene, A. R., M. O. Garcia, A. J. Pietruszka, D. Weis, J. P. Marske, M. J. Vollinger, and J. Eiler (2013), Temporal geo-chemical variations in lavas from K?lauea?s Pu?u ?O?o eruption (1983?2010): Cyclic variations from melting of source het-erogeneities, Geochem. Geophys. Geosyst., 14, 4849?4873, doi:10.1002/ggge.20285.1. Introduction[2] K?lauea, on the Island of Hawai?i (Figure 1), isone of the most active and best-monitored volca-noes in the world [Heliker and Mattox, 2003;Wolfe et al., 1987]. The ongoing Pu?u ?O?o erup-tion on K?lauea?s East Rift Zone (Figure 1) hasbeen active nearly continuously for 30 years and isHawai?i?s longest and most voluminous (4 km3)historical eruption [Poland et al., 2012]. The con-tinuous petrologic and geochemical monitoring ofthe Pu?u ?O?o eruption [e.g., Garcia et al., 2000;Marske et al., 2008; Thornber, 2003] has allowedus to witness the dynamic changes in the meltingprocess and mantle source composition during asingle, long-lasting magmatic event. Extractionand transport of melt through open channels dur-ing the Pu?u ?O?o eruption has efficiently transmit-ted variations of melting in the heterogeneoussource to lavas erupted at the surface without sig-nificant pooling and homogenization, preservingshort-term isotopic and geochemical variations[Pietruszka et al., 2006].[3] The long-term geochemical variations (manythousands of years) of Hawaiian and other oceanisland basalts has been well documented due todetailed geochemical work on 3? km deep drillcore [e.g., Albare`de et al., 1997; Blichert-Toftet al., 2003; Bryce et al., 2005; Caroff et al.,1995; Rhodes et al., 2012]. These studieschronicle processes on millennium time scales butmiss potential short-term variations (<100 years),which may provide better insights into meltingand crustal processes. K?lauea?s historical (1823?1982) and prehistoric (AD 900?1400) summitlavas reveal rapid and systematic changes in Pb,Sr, Nd, O, and U-series isotope ratios on a timescale of decades to centuries [Garcia et al., 2003,2008; Marske et al., 2007; Pietruszka and Garcia,1999; Pietruszka et al., 2001]. The Pu?u ?O?oeruption (sampled from hourly to monthly) showscompositional change over hours (in rare cases)for major elements to a few years for isotope ratios[Garcia et al., 2000; Marske et al., 2008]. Thelong duration and vigorous activity (0.35  106m3 of lava erupted daily) of Pu?u ?O?o [e.g., Suttonet al., 2003] provides a rare opportunity to lookbeyond the shallow-level crustal processes associ-ated with the short eruptions (days to weeks) thattypify many active basaltic volcanoes (e.g., MaunaLoa, Etna, Piton de la Fournaise, Karthala,Grimsv?tn) and into the mantle. In addition, Pu?u?O?o magmas may partially bypass K?lauea?s sum-mit reservoir (2?6 km depth beneath the summitcaldera) on their way to the East Rift Zone, andmostly avoid its buffering effects [Garcia et al.,2000]. Therefore the Pu?u ?O?o eruption is one ofEarth?s best probes for sampling mantle-derivedmelts almost continuously over nearly threedecades.[4] The study of isotopic and geochemical varia-tion in magmatic events over short time scales(months to years) in oceanic island lavas improvesour temporal and spatial resolution of meltingprocesses and the chemical structure of mantleplumes [Abouchami et al., 2000; Eisele et al.,2003; Hofmann et al., 1984; Vlastelic et al.,2005]. Recent studies of Pb, Sr, and Nd isotoperatios for part of the Pu?u ?O?o eruption [Marskeet al., 2008] and other active basaltic volcanoes[e.g., Piton de la Fournaise; Vlastelic et al., 2005]detected rapid and systematic changes over shortGREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854850time scales (years to decades) resulting fromsource heterogeneity, and variations in crustalprocesses. Here we present new high-precision Pb,Sr, and O isotope ratios, and major- and trace-element abundances for Pu?u ?O?o lavas eruptedbetween January 1983 and June 2010. These data197418231919-20 1971Mauna Ulu   1969-74197719551955196018401790East Rift ZoneSummitMakaopuhi1983-present1790179010 km18681955Pu?u ?O?oNCraterCraterEp. 54Southwest Rift ZoneMauna Loa1983-2010 Pu'u '?'? KupaianahaPACIFIC OCEANNorth Pu'u '?'?N?pauCraterKalapanaEp. 56 February 17, 1992 - Feb. 11, 2010Episodes 50-58July 20, 1986 -  February 17, 1992Episodes 48-49January 3, 1983 -July 20, 1986Episodes 1-48episodicfountaining(mostly centralvent)gentle effusion(lava shield andtube-fed pahoehoe)Kupaianaha Pu?u ?O?oPu?u ?O?oJan. July Feb.Feb.1983 1986 19921997(3.5 years) (5.5 years) (18 years) perched channels,rootless shields,fissure eruption20102007Episode 56(Magma supply ratedoubled)June(uprift)flank  vent eruptions(nearly continuous)(lava shield and tube-fed pahoehoe)2003Episode 54(uprift)010km016miK?lauea Caldera1790-1982kmHalemaumauPu?u ???? East Rift ZoneMauna Ulu1969-1974Makaopuhi N?pau(a)KupaianahaEp. 54Ep. 56(b)123storeddikeJune, 2007?Jan.1997Ep. 58Ep. 58K?lauea Caldera0KohalaMauna Loa K?lauea Mauna KeaL??ihi 50010001500200025003000200015005001000100020003000250050002500 50005500155?156?19?20?Hilo RidgeEast rift zone40 km20Hawai?i Hual?lai(c)(d)Figure 1. Map of flow fields from the Pu?u ?O?o-Kupaianaha eruption on the East Rift Zone of K?lauea Vol-cano from 1983 to 2010 and historical flows, with a timeline summarizing the predominant style of eruptiveactivity. (a) A schematic cross section of summit and East Rift Zone shows the proposed magmatic plumbingsystem for K?lauea Volcano, with locations for episodes 54 and 56 uprift of Pu?u ?O?o. Mantle-derived magmafor this eruption is thought to partially bypass the summit reservoir based on the rapid changes in lava compo-sition [Garcia et al., 1996]. (b) Map of K?lauea East Rift Zone with flow fields from intervals of the Pu?u?O?o eruption. Legend shows episodes in each interval of eruptive activity. Map provided by USGS HawaiianVolcano Observatory. (c) Map of the island of Hawai?i with area of map in Figure 1b indicated with box. (d)Timeline of the Pu?u ?O?o eruption. Episode 54 was a fissure eruption in and downrift of Napau Crater thatoccurred over 23 h in January 1997, following the collapse of the Pu?u ?O?o cone. Episode 56 was a brief (<1day) fissure eruption northeast of Makaopuhi Crater (uprift of Pu?u ?O?o) that occurred in June 2007, coincid-ing with an intrusion and collapse of Pu?u ?O?o crater floor. Dashed lines between 2003 and 2007 indicate pe-riod when magma supply rate nearly doubled [Poland et al., 2012].GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854851are combined with previously published high-precision isotope and trace-element data from1998 to 2005 Pu?u ?O?o lavas [Marske et al.,2008]. The new Pb and Sr isotope and inductivelycoupled plasma mass spectrometry (ICP-MS) dataprovide a record of isotopic and geochemical vari-ation of Pu?u ?O?o lavas at 6 month intervals,whereas X-Ray fluorescence (XRF) data was col-lected at 2 week intervals. This time series anal-ysis of Pu?u ?O?o lavas allows us to distinguish thechanging roles of mantle and crustal processes ingreat detail. The new Pb and Sr isotope ratios areused to assess the short-term expression of mantlesource components throughout the course of theeruption and to evaluate the effects on lava com-position of recent doubling of the magma supply[2003?2007; Poland et al., 2012]. These resultsare compared to the longer-term variations forK?lauea and other Hawaiian shield volcanoes.2. Geologic Background of K?laueaVolcano and the Pu?u ?O?o Eruption(1983?2010)[5] K?lauea Volcano is currently in the middle ofits shield-building stage [DePaolo and Stolper,1996], erupting tholeiitic lava at a rate of 0.13km3/yr [Sutton et al., 2003], one of the highestrates of any volcano on Earth. K?lauea rises 1240m above sea-level on the southern flank of itslarger neighbor, Mauna Loa (4168 m; Figure 1).Geochemical evidence favors a deep mantle plumeorigin for Hawaiian magmas [e.g., Kurz et al.,1982; Weis et al., 2011]. Shield stage magmas arethought to originate from partial melting at mantledepths of 70?120 km within the upper Hawaiianplume [Watson and McKenzie, 1991]. Magmas areextracted from the upwelling mantle within themelting region and transported through chemicallyisolated channels towards the surface [Pietruszkaet al., 2006]. These pooled melts ascend throughthe lithosphere via a primary conduit into a shal-low (2?6 km) magmatic complex within K?lauea[Eaton and Murata, 1960; Ryan, 1987; Tillingand Dvorak, 1993; Wright, 1971]. K?lauea erup-tions occur in and around its summit caldera andEast and Southwest Rift Zones. Approximately90% of the subaerial surface of K?lauea Volcano iscovered with tholeiitic lava less than 1100 yearsold [Holcomb, 1987]. Prior to 1955, historical(post-1820) eruptions on K?lauea occurred mostlyat or near the summit [Macdonald et al., 1983].Subsequently, rift zone eruptions became morecommon, especially along the East Rift Zone,including the 1969?1974 Mauna Ulu eruption, themost voluminous historical eruption prior to Pu?u?O?o [Macdonald et al., 1983].[6] The Pu?u ?O?o-Kupaianaha eruption (referredto as the Pu?u ?O?o eruption throughout this paper)began on 2 January 1983 with the intrusion of adike within K?lauea?s East Rift Zone, although itwas preceded by months of intrusions from thesummit into the rift zone [Wolfe et al., 1987]. Itwas followed 24 h later by eruptive activity alonga discontinuous 7 km long fissure, which localizedto a central vent, Pu?u ?O?o (Figure 1 and Table 1).The eruption can be categorized into three broadphases based on eruptive style and location: (1)1983?1986: brief (mostly less than 24 h), episodiceruptions (24 day average repose between erup-tions) with fountaining up to 400 m, mainly fromthe Pu?u ?O?o vent [Heliker and Mattox, 2003]; (2)1986?1992: nearly continuous effusion from theKupaianaha vent, which was considered to have ashallow (<100 m deep) conduit connection withPu?u ?O?o, 3 km uprift [Garcia et al., 1996]; and(3) 1992?2010: nearly continuous effusion mostlyfrom vents within, and on the southwest and eastflanks of Pu?u ?O?o, and from rootless shields 2km east of Pu?u ?O?o [Poland et al., 2008]. Thispattern was interrupted on 29 January 1997 (epi-sode 54) by the 150 m collapse of the crater floorinside the Pu?u ?O?o cone, and propagation oferuptive fissures 4 km uprift (west) of Pu?u ?O?o,which were active for less than a day [Heliker andMattox, 2003]. This event was followed by a 6week hiatus in effusive activity, although glowreturned to the Pu?u ?O?o vent on 24 February1997 (Table 1). Afterward, and until June 2007,lava erupted nearly continuously from flank ventson Pu?u ?O?o (episode 55). On 19 June 2007, adike intrusion in the upper East Rift Zone resultedin a brief, small (1500 m3) eruption (episode 56)6 km uprift from Pu?u ?O?o [Montgomery-Brownet al., 2010], which was followed by a 2 week hia-tus in effusion [Poland et al., 2008]. Lava produc-tion resumed for 3 weeks in and around Pu?u ?O?ocone (episode 57) until 21 July 2007, when a fis-sure opened on the east flank of Pu?u ?O?o andpropagated eastward towards Kupaianaha (Figure1 and Table 1). This marked the beginning of epi-sode 58 [Poland et al., 2008], which continuedthrough the end of 2010 mostly as tube-fed flowsfrom a vent 2 km east of Pu?u ?O?o. The othernotable K?lauea eruptive activity during the Pu?u?O?o eruption is an ongoing summit eruption thatstarted in March 2008 [Johnson et al., 2010].GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854852Table1.SummaryofthePu?u? O? oEruptionaPrimaryVentEpisodeEpisodeStartDateReposeLength(Days)EpisodeLengthVolume(106xm3)EruptionRate(103m3/day)Vent/LocationBriefDescriptionPu?u? O? o13Jan1983Start20days14?Fissure1;activitylocalizedatPu?uHaluluandPu?uKahauleaeastofPu?u? O? oInitialfissureopenedinNapauCraterafterseismicswarmpropagateddownERZ;fissuresextended8km;fissureslocalizedto1kmnearPu?uKahaulea;fountainsfromPu?uHalulubuilta60m-highconePu?u? O? o2?4710Feb19838?65(betweenepisodes)~3.8years371300MostlyPu?u? O? o;Episodes2?3localizedatPu?uHaluluandPu?uKahauleaeastofPu?u? O? o;Epi-sode4?47PuuOoprimaryventEpisodicfirefountaining;episodesmostly<24hlongsepa-ratedbyanreposelengthaverageof24days;effusionratesincreasedthoughepisode39;maximumlavafoun-tainof470mhigh;firstyearchangedfromlowfountainsandpahoehoeriverstohighfountainsand?a?afans;fountain-fed?a?abyepisode20;conebuilt255mhighand1.4kmindiameter;summitinflatedbetweenfoun-tainingepisodesanddeflatedduringepisodesKupaianaha4818July198624~5.5years500400?0.5Kupaianaha;fissure3kmeastofPu?u? O? oFissuresfirstopenedatthebaseofPu?u? O? oand22hlateropened3kmdownriftataventtobenamedKupaianaha;5.5yearsofnearlycontinuousgentleeffusion;largelavapondformedovervent(140m300m);broadlavashieldformedandtube-fedpahoehoewascommonwaylavaspreadtocoast;homesdestroyedintownofKala-pana;lavatubestoseamid-1987to1989;lavaenteredseaduring68%ofepisode;lavaactiveinPu?u? O? ocraterduringmostofepisodePu?u? O? o498Nov1991None18days110.6Fissure2betweenPu?u? O? oandKupaianahaFissuresopenedonPu?u? O? oandpropagatedtoKupaia-naha;outputwanedduringepisode;gentleeffusion,lavashieldandtube-fedpahoehoe;fissurevents,pahoehoePu?u? O? o5017Feb19921115days3?Pu?u? O? oflank;radialfissureonwestflankofPu?u? O? oconeEruptionreturnedtoPu?u? O? o;radialfissuresonflankofcone;flankventeruptions;lavashieldbankedupagainstthesouthandwestPu?u? O? ocone;spatterconesformedovervents;mostlytube-fedpahoehoe;continuousquieteffusionPu?u? O? o517Mar19924161days32300Pu?u? O? oflankFlankventeruptions;mostlytube-fedpahoehoetothesea;continuousquieteffusion;lavashieldbankedupagainstthesouthandwestPu?u? O? oconePu?u? O? o523Oct1992None15days2300Pu?u? O? oflankFlankventeruptions;mostlytube-fedpahoehoetothesea;continuousquieteffusion;lavashieldbankedupagainstthesouthandwestPu?u? O? oconePu?u? O? o5320Feb1993None~4years535300Pu?u? O? oflankFlankventeruptions;mostlytube-fedpahoehoetothesea;continuousquieteffusion;lavashieldbankedupagainstthesouthandwestPu?u? O? ocone;collapsepitsformedonthesideofPu?u? O? oPu?u? O? o5429Jan1997None1day0.30.3Fissure3;2?4kmupriftofPu?u? O? o(1)LavalakeinsidethePu?u? O? oventdrainedandcraterfloordropped150m;(2)Pu?u? O? owestflankcollapsed;115mgapinwestsideofPu?u? O? o;(3)fissure4kmeruptedupriftfor1day,inanddownriftofNapauCrater,followedbylongesteruptivehiatussince1987(24days);distinctlavachemistryinvolvedmagmamixingwithdifferentiatedmagmastoredinriftzoneGREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854853Table1.(continued)PrimaryVentEpisodeEpisodeStartDateReposeLength(Days)EpisodeLengthVolume(106xm3)EruptionRate(103m3/day)Vent/LocationBriefDescriptionPu?u? O? o5524Feb19972410years265200?500Pu?u? O? oanditsflankLavaspilledfromcratertoformnewpond;lavaspilledfromcrateramonthlater;newflankventeruptionswestandsouthwestofcone;spatterconesonflankscrustedovertoproducemostlytube-fedpahoehoe;byJune1997lavaovertoppedthegapinwallofPu?u? O? oandflowedfromcraterforfirsttimein11years;flankventsunder-minedPu?u? O? oconeinDecember,1997;PukaNuicol-lapsepitformedonsouthwestflankofcone;31pausesoccurredduringepisode55Pu?u? O? o5619June2007None6h0.001450.00036250mlongfissureintheforestnortheastofKaneNuioHamo,approximately6kmwestofPu?u? O? oFather?sDayeruptionnearKaneNuioHamonorthofMakaopuhiCrater;magmasupplytoPu?u? O? owascutoffon17June2007;earthquakeswarmsindicatedmagmamovementintheupperERZ;spattereruptedfromfissureinforestedarea;smalllavaflow(200m50m)accom-paniedintrusioninERZ;craterfloorinPu?u? O? ocol-lapsedanderuptionshutoffPu?u? O? o571July200719daysNone0.82?1.2365Pu?u? O? ocraterAfterabouttwo-weeksofquiet,theeruptionbeganagainon1July.Lavabegantorefillthecrater.On8July,effu-sionwanedasthecraterbegantoupliftinapiston-likefashion.Thecraterthenbegantofillandreachedtowithin30moftheeasternrimofPu?u? O? ocraterbymid-July.Pu?u? O? o5821July2007None~4yearsended7March2011320(asoftheendof2009)FromfissureeastofPu?u? O? ocraterPerchedlavachannel,rootlessshields;forthefirsttimesince7February1992,lavabeginseruptingeastofPu?u? O? ocrater.ThanksgivingEvebreakout,lavabypasses21July2007channelanderuptsonchannelflank;5March2008oceanentryactiveforthefirsttimesinceJune2007;explosioninHalema?uma?uCrateratsummiton19March2008;June2008spatteringventsandasmallpondoflavainPu?u? O? o,lavafountainsgushfromtheTEBtubesystem,channelized?a? aflowsinRoyalGardens,andlargelittoralexplosionsatK ?lauea?soceanentrynearKalapana;Waikupanahaoceanentryactivethroughmuchof2009,andoccasionallyKupapa?uoceanentrytowesta Reposelengthreferstodurationofpausebetweeneruptiveepisodes.Episodeidentifiesoccurrencesoffountainingorlavaflowseparatedbyquiescentperiods.Volumeisdenserockequivalent(DRE)eruptedduringeachepisode.Datasources:Garciaetal.[2000]andreferencestherein,Wolfeetal.[1998],HelikerandMattox[2003].GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.2028548543. Description of Samples andAnalyses Performed in This Study[7] This study presents 52 new high-precision Pband Sr isotope analyses (from 1983 to 1997 and2006 to 2010), 11 new O isotope analyses (fromafter 1997; Table 2), and 13 new ICP-MS trace-element analyses of Pu?u ?O?o lava samples(mostly after 2005; data and analytical methodsare presented in the supporting information).1 NewXRF major- and trace-element analyses for 52Pu?u ?O?o lavas erupted from 2006 to 2010 arealso presented. In addition, new XRF trace-element analyses are given for samples eruptedprior to 1998, when a new, more precise XRFinstrument became available. Almost all of thesamples in this study were collected in a moltenstate and quenched with water to minimize poster-uption crystallization. The sample names are thedate that each lava sample was collected (e.g.,day-month-year), which is generally within a dayof its eruption when lava is flowing in open chan-nels on the surface or in lava tubes [e.g., Garciaet al., 2000] or up to a week or more when it isoozing within slowly advancing pahoehoe flows[K. Ho, personal communication, 2013]. Descrip-tions of the petrography of typical Pu?u ?O?o lavascan be found in Garcia et al. [1989, 1992, 1996,2000] and Marske et al. [2008]. Fourteen high-precision Pb and Sr isotope ratios for Pu?u ?O?olavas erupted from 1998 to 2005 from Marskeet al. [2008] and 15 O isotope analyses from Gar-cia et al. [1998] are listed in Table 2 and areincluded in plots for completeness.4. Temporal Geochemical Variationsin Lavas From 1983 to 2010[9] Early Pu?u ?O?o lavas (1983 to early 1985) re-cord rapid (hours to days) variations in major andcompatible trace-element abundances (Figure 2;Supporting information). These lavas show petro-graphic evidence for both crystal fractionation andmagma mixing [Wolfe et al., 1987; Garcia et al.,1992]. Crystal fractionation of olivine (with minorclinopyroxene and plagioclase, especially for 1983lavas) is the dominant process controlling short-term major-element variation in Pu?u ?O?o lavas[Garcia et al., 1992]. To remove the effects ofcrystal fractionation on parental magma composi-tions, major-element abundances of lavas contain-ing only olivine (MgO >7.0 wt %) werenormalized to 10 wt % MgO by the addition ofequilibrium composition olivine (98.5%) and spi-nel (1.5%) in 0.5 mol % steps, as described byGarcia et al. [2003]. The increases in MgO, CaO/Al2O3, and CaO/TiO2 and decreases in MgO-normalized incompatible element abundances(e.g., TiO2, K2O) between 1983 and early 1985(Figure 2) reflect mixing of new high-MgOmagma with decreasing amounts of a hybridmagma formed at the start of the eruption by mix-ing two differentiated, rift-zone stored magmas[Garcia et al., 1989, 1992]. Lavas erupted afterearly 1985 show no petrographic or geochemicalevidence for mixing [Shamberger and Garcia,2007] until the 1997 uprift eruption, which is dis-cussed below.[10] From 1985 to 1994, Pu?u ?O?o lavas show awide range in MgO reflecting the periodic hiatusesin eruptive activity [Garcia et al., 1992], and grad-ual changes in MgO-normalized concentrations ofmajor elements (TiO2 and K2O), and ratios ofmajor (e.g., CaO/Al2O3; Figure 2) and trace ele-ments (Nb/Y; Figure 3). In 1994, lavas began aperiod of increasing MgO-normalized SiO2 andoverall decreasing MgO-normalized TiO2 that per-sisted until 2001. Other geochemical parameterscontinued their long-term trends (e.g., decreasingCaO/Al2O3, MgO-normalized K2O, and Nb/Y,and increasing Zr/Nb; Figures 2 and 3). Starting inmid- to late 2003, there was an increase in lavaproduction with effusion rates doubling in 2005[Poland et al., 2012]. The lava MgO contentdecreased from 2003 to 2007 and was relativelylow (<7.5 wt %, mostly <7.2 wt %) with limitedvariation (Figure 2). This decrease continued theoverall trend of decreasing MgO that started in1998, as noted by Poland et al. [2012]. There isalso a decrease in MgO-normalized SiO2 and anincrease in MgO-normalized TiO2 and K2O during2003?2007 (Figures 2 and 3). Lava MgO increasedfrom 2008 to 2009 as did CaO/TiO2 and values ofMgO-normalized SiO2, although MgO and SiO2values dropped afterwards for the most recentlyerupted samples that were analyzed in this study(Figure 2). For more on major- and trace-elementvariations in 1983?2005 Pu?u ?O?o lavas, see Gar-cia et al. [1989, 1992, 1996, 2000], Marske et al.[2008], and Thornber [2003].[11] The brief eruptive outbreaks uprift of the Pu?u?O?o vent in 1997 (3 km uprift for episode 54)and 2007 (6 km uprift for episode 56; Figure 1)1Additional supporting information may be found in the onlineversion of this article.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854855Table 2. Pb, Sr and O Isotopic Geochemistry of Pu?u ?O?o Lavas from 1983?2010aSample 206Pb/204Pb 2 207Pb/204Pb 2 208Pb/204Pb 2 87Sr/86Sr 2 18O 123 Jan 1983  18.5247 0.0009 15.4800 0.0008 38.155 0.0020 0.703590 0.000009 4.56 0.029 Apr 1983 18.5309 0.0007 15.4893 0.0007 38.165 0.0017 0.703573 0.0000083 July 1983  18.4780 0.0007 15.4765 0.0008 38.117 0.0017 0.703573 0.000007 4.77 0.0331 Jan 1984 18.4595 0.0009 15.4765 0.0007 38.103 0.0020 0.703587 0.00000812 Sep 1984 18.4417 0.0008 15.4747 0.0007 38.091 0.0019 0.703555 0.0000098 Feb 1985  18.4342 0.0010 15.4743 0.0009 38.087 0.0024 0.703567 0.000008 4.76 0.0521 Apr 1985  4.82 0.1030 Jul 1985 18.4306 0.0007 15.4756 0.0007 38.089 0.0020 0.703571 0.0000082 Jun 1986  18.4138 0.0009 15.4755 0.0008 38.079 0.0020 0.703580 0.000009 4.94 0.3026 Jun 1986  4.77 0.0213 Sep 1986  18.4138 0.0008 15.4726 0.0007 38.074 0.0020 0.703590 0.000008 5.17 0.0116 Mar 1987  18.4108 0.0006 15.4717 0.0005 38.069 0.0014 0.703589 0.000009 5.25 0.0218 Oct 1987 18.3992 0.0008 15.4746 0.0008 38.068 0.0022 0.703597 0.00000919 Jan 1988 18.3952 0.0006 15.4715 0.0006 38.061 0.0016 0.703577 0.00000818 Aug 1988 18.3871 0.0008 15.4703 0.0007 38.052 0.0017 0.703583 0.00000926 Mar 1989  18.3882 0.0007 15.4717 0.0007 38.055 0.0019 0.703584 0.000009 5.11 0.057 Jul 1989 18.3861 0.0010 15.4725 0.0009 38.054 0.0022 0.703581 0.0000097 Jul 1989 ? 18.3851 0.0009 15.4710 0.0008 38.052 0.0021 0.703576 0.0000097 Jan 1990  18.3881 0.0009 15.4745 0.0008 38.056 0.0018 0.703584 0.000008 5.03 0.0127 May 1990 18.3864 0.0008 15.4716 0.0007 38.054 0.0020 0.703603 0.00000721 Oct 1990 18.3856 0.0007 15.4693 0.0006 38.049 0.0017 0.703597 0.00000812 May 1991  18.3992 0.0009 15.4739 0.0007 38.063 0.0019 0.703581 0.000008 5.08 0.041 Aug 1991  5.11 0.046 Jun 1992  18.4048 0.0009 15.4721 0.0009 38.061 0.0019 0.703585 0.000008 5.04 0.1013 Aug 1993  18.4112 0.0007 15.4752 0.0006 38.072 0.0015 0.703611 0.000008 4.98 0.074 Jan 1994 18.4098 0.0009 15.4736 0.0008 38.067 0.0020 0.703607 0.00000825 Apr 1994  18.4100 0.0009 15.4737 0.0009 38.068 0.0027 0.703586 0.000008 5.01 0.029 Oct 1994 18.4059 0.0008 15.4718 0.0007 38.066 0.0019 0.703598 0.00000727 Apr 1995  18.4059 0.0009 15.4721 0.0008 38.066 0.0023 0.703604 0.000009 5.25 0.0514 Oct 1995 18.4068 0.0008 15.4729 0.0008 38.071 0.0021 0.703602 0.00000919 Jan 1996  5.19 0.0715 Mar 1996 18.4064 0.0010 15.4738 0.0008 38.070 0.0023 0.703592 0.00000922 Aug 1996 18.4038 0.0009 15.4722 0.0008 38.065 0.0016 0.703612 0.00000710 Jan 1997  18.4010 0.0012 15.4728 0.0011 38.064 0.0019 0.703606 0.000009 5.2 0.0523 Jul 1997 18.3993 0.0010 15.4729 0.0010 38.068 0.0025 0.703601 0.00000710 Jan 1998 18.3958 0.0007 15.4728 0.0006 38.067 0.0014 0.703591 0.00000810 Jan 1998 ? 18.3940 0.0007 15.4711 0.0006 38.063 0.0016 0.703593 0.00000711 May 1998 18.4005 0.0009 15.4740 0.0008 38.071 0.0020 0.703605 0.0000107 Sep 1998 18.4082 0.0008 15.4775 0.0006 38.083 0.0017 0.703601 0.000006 5.33 0.067 Sep 1998 ? 18.4107 0.0004 15.4727 0.0005 38.075 0.0012 5.29 0.0813 Feb 1999 18.4068 0.0010 15.4783 0.0008 38.085 0.0021 0.703607 0.00000613 Feb 1999 ? 18.4124 0.0004 15.4736 0.0004 38.076 0.001119 Jun 1999 18.3987 0.0010 15.4805 0.0007 38.085 0.0020 0.703620 0.00000927 Oct 1999  18.4018 0.0004 15.4726 0.0004 38.069 0.0011 0.703622 0.000009 5.36 0.0819 Feb 2000  18.4072 0.0004 15.4712 0.0004 38.072 0.0011 0.703624 0.00000721 Jun 2000  18.4067 0.0004 15.4704 0.0004 38.069 0.0011 0.703638 0.000007 5.28 0.088 Jan 2001  18.4116 0.0004 15.4721 0.0004 38.074 0.0011 0.703627 0.000012 5.31 0.087 Jul 2001  18.4137 0.0004 15.4719 0.0004 38.073 0.0013 0.703626 0.0000099 Feb 2002  18.4139 0.0004 15.4707 0.0004 38.069 0.0011 0.703637 0.00000820 Aug 2002  18.4152 0.0004 15.4722 0.0004 38.072 0.0011 0.703639 0.00000512 Apr 2003  18.4161 0.0005 15.4726 0.0005 38.072 0.0013 0.703641 0.000005 5.31 0.0815 Jan 2004  18.4154 0.0005 15.4719 0.0005 38.069 0.0012 0.703632 0.000007 5.21 0.087 Jun 2004  18.4146 0.0003 15.4716 0.0004 38.068 0.0010 0.703624 0.00000731 Jan 2005  18.4170 0.0005 15.4735 0.0006 38.075 0.0012 0.703624 0.000005 4.96 0.138 Aug 2005  18.4119 0.0005 15.4727 0.0005 38.070 0.0013 0.703622 0.00001029 Jan 2006  18.4087 0.0004 15.4720 0.0004 38.065 0.0012 0.703623 0.00000924 Jun 2006 ? 18.4062 0.0028 15.4724 0.0026 38.066 0.0059 0.703612 0.000008 5.23 0.0324 Jun 2006  18.4073 0.0016 15.4714 0.0018 38.063 0.0060 0.703617 0.000013 5.23 0.036 Apr 2007  18.4065 0.0003 15.4715 0.0003 38.063 0.0009 0.703617 0.000007 5.35 0.1317 Jun 2007  18.4019 0.0004 15.4709 0.0004 38.062 0.0012 0.703626 0.000006 5.63 0.1322 Mar 2008  18.4038 0.0003 15.4700 0.0003 38.061 0.0010 0.703607 0.000008 5.45 0.132 May 2008  18.4045 0.0003 15.4721 0.0004 38.066 0.0010 0.703609 0.00000915 Nov 2008 18.3972 0.0009 15.4704 0.0007 38.058 0.0024 0.703600 0.00000829 Jan 2009  18.4003 0.0005 15.4709 0.0006 38.061 0.0012 0.703628 0.0000087 May 2009 18.4005 0.0008 15.4736 0.0007 38.066 0.0020 0.703624 0.0000104 Jun 2009  18.4009 0.0003 15.4720 0.0004 38.064 0.0010 0.703610 0.00000716 Oct 2009  18.3994 0.0005 15.4714 0.0006 38.062 0.0012 0.703622 0.00000722 Jan 2010  18.3987 0.0005 15.4708 0.0005 38.060 0.0013 0.703617 0.000007a indicates analysis at San Diego State University (SDSU), analyses from 7 Sep 1998 to 8 Aug 2005 are from Marske et al. [2008]. Sr isotopeanalyses from 1983?1997 and 2006?2010 were performed at PCIGR. ? Chemistry duplicate.  Published 18O analyses from Garcia et al. [1998].Analytical methods are described in the supporting information. 24 Jun 2006 is an in-house glass standard called Menehune collected from a Pu?u?O?o lava flow (errors are the external 62s of the replicate analyses; average of four analyses for Pb and Sr at PCIGR; 68 for Pb and 26 for Sr atSDSU). US Geological Survey sample numbers for lavas between up to16 Mar 87 are 23 Jan 1983: 1?054, 9 April 1983: 3?117, 3 Jul 1983: 5?139, 31 Jan 1984: 14?232, 12 Sep 1984: KE24?25 310S, 8 Feb 1985: 30?362, 30 Jul 1985: 35?419, 1 Jan 1986:40?484, 2 Jun 1986: 46?536, 13Sep 1986: 48?649, 16 Mar 1987: 48?714F.occurred after major collapses of the Pu?u ?O?ocrater floor (Table 1). The lavas erupted from theseuprift vents were geochemically distinct. Com-pared to coeval Pu?u ?O?o vent lavas, those fromepisode 54 have lower MgO (5.6?6.4 versus 7.5?10.1 wt %), CaO/TiO2 (2.8?3.4 versus 4.4), Sr/Nband Zr/Nb ratios (Figure 3). These geochemicalsignatures and the petrographic evidence of56789100.700.720.740.760.780.800.820.840.8648.849.049.249.449.649.850.050.250.43.03.23.43.63.84.04.24.44.64.82.12.22.32.42.52.60.350.400.450.500.551983198419851986198719881989199 0199119921993199419951996199719981999200020012 00220032004200520062007200820092010198319841985198619871988198919901991199219931994199519961997199 819992000200120022003200420052006200720082009201084 86 88 90 92 94 96 98 00 02 04 06 08 1084 86 88 90 92 94 96 98 00 02 04 06 08 1084 86 88 90 92 94 96 98 00 02 04 06 08 10 84 86 88 90 92 94 96 98 00 02 04 06 08 10Ep. 54TiO2 (wt%)normalized to 10 wt% MgO(Magma supply ratedoubled)2003-071983-19861986-19921992-2010MgO (wt%)CaO/Al2O3 (wt%)SiO2 (wt%)CaO/TiO2 (wt%)K2O (wt%)(b)(d)(c)(a)(f)(e)normalized to 10 wt% MgOnormalized to 10 wt% MgO(Magma supply ratedoubled)2003-07(Magma supply ratedoubled)2003-07(Magma supply ratedoubled)2003-071983-19861986-19921992-2010normalized to 10 wt% MgOnormalized to 10 wt% MgOEp. 56Figure 2. Major-element variation diagrams for Pu?u ?O?o lavas from 1983 to 2010. All major elements andratios except MgO were normalized to 10 wt % MgO [the most primitive lava erupted from Pu?u ?O?o; Gar-cia et al., 2000] by addition of equilibrium composition olivine (98.5%) and spinel (1.5%) in 0.5 mol % steps[Garcia et al., 2003; Rhodes and Vollinger, 2004]. Pu?u ?O?o lavas with <7.2 wt % MgO may have crystal-lized minerals other than olivine (e.g., clinopyroxene and plagioclase) and were not included in the olivinenormalization procedure and are not shown in all the plots, except MgO. Episode 54 (Ep. 54; 29?30 January1997) lavas involved mixing of evolved magmas stored in the rift zone and MgO-rich magma. Three intervalsof eruptive activity in legend and colors correspond with those shown in Figure 1. CaO/TiO2 and CaO/Al2O3ratios also use normalized data although are virtually unaffected by olivine fractionation. Vertical lines indi-cate nearly double magma supply rate between 2003 and 2007 [Poland et al., 2012]. Data are presented in thesupporting information. Uncertainty for analyses is described in the supporting information.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854857disequilibrium in the episode 54 lavas are thoughtto result from mixing Pu?u ?O?o magma withstored, differentiated rift zone magma [Garciaet al., 2000; Thornber et al., 2003]. In contrast,episode 56 lavas have higher MgO (8.5 versus 7.2wt %) and a relatively high 18O value (5.6 versus5.4%) but are otherwise geochemically indistin-guishable from contemporaneous Pu?u ?O?o lavas.[12] The 206Pb/204Pb ratios for Pu?u ?O?o lavasdecreased rapidly through the episodic fountainingperiod (1983?1986) and reached a minimumbetween 1989 and 1991 during the Kupaianahaphase (Figure 4b and Table 2). The rapid decreasein 206Pb/204Pb continues the longer-term trend ofdecreasing Pb isotope ratios for K?lauea lavaserupted following the 1924 collapse of the summitcaldera (Figure 4a). After 1991, the trend of206Pb/204Pb ratios in Pu?u ?O?o lavas shows cyclicvariations with two broad humps, each cycle span-ning approximately 10 years (except for a smalloffset from the overall trend between January1998 and June 1999; Figure 4b). The cyclic varia-tion in Pb isotope ratios is well shown by208Pb/206Pb ratios, which inversely mirror the206Pb/204Pb trend (Figures 4 and 5).[13] The 87Sr/86Sr ratios of Pu?u ?O?o lavas extendthe temporal trend of increasing Sr isotope ratiosfor K?lauea lavas that started following the 1924caldera collapse (Figure 4c). Overall, Pu?u ?O?olavas display an increase in 87Sr/86Sr from 1983 to2003 and a slight decrease after 2004 (Figure4d). Prior to 1999, the 87Sr/86Sr and 206Pb/204Pbratios of the lavas are not well correlated, althoughthere is an overall inverse correlation between the19831984198519861987198819891990199119921993199419951996199719981999200020012002200320042005200620072008200920100.40.50.60.70.819831984198519861987198819891990199119921993199419951996199719981999200020012002200320042005200620072008200920102.02.12.22.32.42.51.99101112161820222426 Sr/NbNb/Y La/Sm(c)(a)(d)1983-19861986-19921992-2010Ep. 54Ep. 54Ep. 5484 86 88 90 92 94 96 98 00 02 04 06 08 10 84 86 88 90 92 94 96 98 00 02 04 06 08 10(Magma supply ratedoubled)2003-07Ep. 54Zr/Nb(b)(Magma supply ratedoubled)2003-07 ?2 SE ?2 SE ?2 SE ?2 SEK?lauea summit(Sep.1982)K?lauea summit(Sep.1982)K?lauea summit(Sep.1982)K?lauea summit(Sep.1982)Figure 3. Trace-element ratios versus time for Pu?u ?O?o lavas from 1983 to 2010. Overall exponential vari-ation in trace-element ratios indicates progressive depletion of the source. In the La/Sm versus time plot, asubset of samples analyzed at PCIGR (April 1983, January 1984, September 1984, and April 2007 to January2010) are normalized to reference material Kil-93 (La/Sm of 2.09, average value from Australia National Uni-versity where most ICP-MS analyses were performed). Vertical lines indicate nearly double magma supplyrate between 2003 and 2007 [Poland et al., 2012]. Trace-element abundances in ppm (data shown in support-ing information). Average 62 bars are shown in a corner of each panel. September 1982 K?lauea summitlava composition from Pietruszka and Garcia [1999].GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854858two ratios for historical summit lavas (Figure 4and Table 2). After 1999, there is a positive corre-lation between 206Pb/204Pb and 87Sr/86Sr (Figure6), which corresponds to the second of the twomajor temporal cycles of Pb isotope ratios (Figure5). The Nd isotope ratios of lavas erupted between1983 and 2005 display no variation outside analyt-ical uncertainty [Marske et al., 2008]. Thus, nonew Nd isotopic data were collected during thisstudy.[14] Oxygen isotopic compositions of Pu?u ?O?olavas erupted over 25 years (1983?2008) show alarger range than historical K?lauea summit lavasspanning 380 years of activity (1.1 versus 0.7% ;Figure 4). Overall, O isotope ratios of Pu?u ?O?olavas have increased over time with early lavas(before 1986) having lower 18O (4.8?4.9%) thansubsequent lavas (5.0?5.6% ; Table 2). The high-est O isotope ratio observed during the Pu?u ?O?oeruption is for an episode 56 lava that erupted 6Uwekahuna BluffUwekahuna BluffHilinasubmarinelavas0.70340.70350.70360.7037198319841985198619871988198919901991199219931994199519961997199819992000200120022003200420052006200720082009201018.3718.3918.4118.4318.4518.4718.4918.5118.530.703550.703600.703651983198419851986198719881989199019911992199319941995199619971998199920002001200220032004200520062007200820092010K?laueasummitK?laueasummitPCIGR analyses (2009)Marske et al. (2008)-SDSUPCIGR analyses (2010)Time (years)Time (years)(summit)(summit)Time (years)K?laueasummitsummit calderacollapse (1924)summit calderacollapse (1924)average ?2 SEMORB mantle rangeSDSU analyses (2007)?18O18.218.418.618.81780 1820 1860 1900 1940 19801000 14001780 1820 1860 1900 1940 19801000 140016001780 1820 1860 1900 1940 198016001600Jun-99Jun-99Jun-86Jun-07(Ep. 56 uprift)(  )(  )(  )4.8 ka 3.5 ka1.7 ka4.8 ka3.5 ka1.7 ka206Pb/204Pb87Sr/ 86Sraverage ?1 SE?18O87Sr/ 86Sr206Pb/204Pb(280-130 ka)tholeiiticalkalictransitionalHilinasubmarinelavas(280-130 ka)tholeiiticalkalictransitional(episodic fountaining)Kupaianaha(gentle effusion)Pu?u ???? Pu?u ???? (mostly flank vent eruptions)(a)(c)(b)(d)Pu?u ???? Pu?u ???? Pu?u ???? (e) (f)data thisstudyMORBmantle summit calderacollapse (1924)PCIGR analyses (2009)Marske et al. (2008)-SDSUPCIGR analyses (2010)SDSU analyses (2007)4.64.85.05.25.45.65.84.64.85.05.25.45.61983198419851986198719881989199019911992199319941995199619971998199920002001200220032004200520062007200820092010no datascale changeK?lauea summit(Sep.1982)K?lauea summit(Sep.1982)Figure 4. Temporal variation of Pb, Sr, and O isotopes for K?lauea Volcano and Pu?u ?O?o lavas. (b)206Pb/204Pb in Pu?u ?O?o lavas shows cyclic variation with two broad humps, each cycle spanning approxi-mately 10 years. (d) 87Sr/86Sr in Pu?u ?O?o lavas increases from 1983 to 2003 and decreases between 2008and 2010, and is correlated with 206Pb/204Pb after 1999. (f) 18O for Pu?u ?O?o lavas shows the same range(0.7%) as historical K?lauea summit lavas and is not well correlated with 206Pb/204Pb or 87Sr/86Sr. Data sour-ces for K?lauea Volcano are Hanyu et al. [2010], Kimura et al. [2006], Marske et al. [2007], Abouchami et al.[2005], Pietruszka and Garcia [1999], Chen et al. [1996], and Garcia et al. [2008]. Data sources for previousanalyses of Pu?u ?O?o lavas areMarske et al. [2008] and Garcia et al. [1998]. For Pu?u ?O?o analyses, average62 for 206Pb/204Pb is smaller than symbol size and uncertainty for 87Sr/86Sr and 18O is shown in the panels.Data for Pu?u ?O?o lavas are presented in Table 2. September 1982 K?lauea summit lava composition fromPietruszka and Garcia [1999]. Colors for symbols in Figures 4b, 4d, and 4f) are the same as Figure 3.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854859km uprift from the Pu?u ?O?o vent in mid-June2007. The episode 56 eruption is related to intru-sion of a dike from the upper East Rift Zone, sothis lava was probably not derived from the shal-low reservoir of magma beneath Pu?u ?O?o [Mont-gomery-Brown et al., 2010]. Thus, its 18O valueis probably representative of the magma feedingthe Pu?u ?O?o eruption. It is identical to the highestvalues observed among historical summit lavas(5.6% ; Figure 4). Variations of 18O in Pu?u ?O?olavas do not correlate with Pb or Sr isotope ratios,or with other geochemical parameters, as wasnoted for previous O isotope work on lavas fromthis eruption [Garcia et al., 1998]. Therefore, Pu?u?O?o lava Pb and Sr isotope ratios were apparentlynot affected by the processes causing variable Oisotope ratios.5. Discussion[15] The high eruption rate and continuous natureof the Pu?u ?O?o eruption provide an exceptionalopportunity to use lava chemistry to evaluate thechanging roles that source, melting, and crustalprocesses play during this single prolonged erup-tion. Previous Pb isotope and trace-element studieson lavas from several multiple-year eruptions ofPiton de la Fournaise Volcano (Reunion Island)have discerned contributions from multiple com-ponents within the Reunion mantle plume and aperiodic role for shallow-level contamination [Pie-truszka et al., 2009; Vlastelic et al., 2005, 2007].Similarly, extreme Pb isotope variability in meltinclusions from Iceland basaltic lavas indicate sig-nificant source heterogeneity, with binary mixingrelationships that may result from combining sol-ids in the mantle and two stages of melt mixing (inporous mantle melt-transport channels and lowercrustal magma chambers) [Maclennan, 2008].Similarly, the geochemistry and petrography ofPu?u ?O?o lavas have been used to interpret theextent of crustal magmatic processes (olivine frac-tionation and accumulation, mixing of higher-MgO and stored rift-zone magmas, and crustalassimilation) and mantle processes (degree of par-tial melting, melt extraction and migration, andsource heterogeneity) during the Pu?u ?O?o erup-tion until 2005 [e.g., Garcia et al., 1998, 2000;Marske et al., 2008]. Here we use new high-precision Pb, Sr, and O isotope ratios, and major-and trace-element data for the entire Pu?u ?O?oeruption (1983?2010) to evaluate the causes ofcyclic and other short-term variability in the proc-esses that operate from the source to the surfacewithin K?lauea Volcano. The effects of crustalprocesses (crystal fractionation, magma mixing,and crustal contamination) on modifying Pu?u?O?o lava compositions are evaluated beforeexamining mantle source and melt transportprocesses.5.1. Magma Mixing and CrystalFractionation During Early EpisodicActivity (1983?1985)[16] The largest compositional changes in Pu?u?O?o lavas occurred from 1983 to 1985. Thesechanges mostly involved two crustal processes:crystal fractionation and magma mixing. Duringsome single eruptive episodes (5?10, 30, and 31),there were relatively large changes in MgO, Ni,and Cr, which are related to minor (3?5%) olivinefractionation in the shallow Pu?u ?O?o reservoirduring eruptive hiatuses [Garcia et al., 1992].These short-term (3?4 weeks) variations are super-imposed on longer term changes that have beenrelated to magma mixing [Garcia et al., 1992;Thornber, 2003]. The longer term variations areevident in plots of MgO-normalized major ele-ments, ratios of incompatible trace elements, andPb isotope ratios (Figures 2?4). Strontium and Oisotopes show less change during this period com-pared to their overall variation during the eruption(Figure 4). The overall progressive compositionalvariation in Pu?u ?O?o lavas from 1983?1985 hasbeen attributed to the mixing of new, relativelyMgO-rich magma (>7.5 wt %) with a decreasingproportion of hybrid, rift-zone stored differen-tiated magma (from 30% of the higher MgOmagma in March 1983 to 100% in September1984) [Garcia et al., 1992; Shamberger andGarcia, 2007].[17] The origin of the higher MgO magma compo-nent from the early phase of the Pu?u ?O?o erup-tion may have been: (1) magma from the summitreservoir, as represented by lavas from the Sep-tember 1982 summit eruption; and/or (2) newmantle-derived magma [Garcia et al., 1992;Shamberger and Garcia, 2007]. Scenario 1involves no change in the composition of thehigher MgO magma from September 1982 to1985, whereas scenario 2 requires it. The 1983?1985 Pu?u ?O?o lavas have both higher and lower206Pb/204Pb ratios than the September 1982 sum-mit lavas (Figures 4b and 7). Therefore, mixing ofa single 1982 summit magma with rift-zone storedmagma (scenario 1) cannot explain the isotopicGREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854860variation of Pu?u ?O?o lavas after 1984, when206Pb/204Pb values are lower than 1982 summitmagma (Figure 7). Ratios of some incompatibletrace elements (Sr/Nb and Zr/Nb) for some lavaswith higher MgO (>7.5 wt %) erupted after mid-1984 also are higher than those for September1982 summit lavas (Figure 3). Thus, if magmafrom the summit reservoir was supplying the Pu?u?O?o eruption, its composition must have changedafter the September 1982 eruption and prior toSeptember 1984 (Figures 4 and 7).[18] The isotopic variations for early Pu?u ?O?olavas are consistent with the eruption being sup-plied by new, compositionally variable, mantle-derived magma in addition to or instead of Sep-tember 1982 summit magma. The rate of206Pb/204Pb variation observed for the period afterthe end of early magma mixing is much faster thanduring the previous 30 years (1952?1982) ofK?lauea summit eruptions (0.016 yr1 versus0.004 yr1). These rapid variations in Pb isotopicratios suggest that magmas supplying Pu?u ?O?opartially bypassed or did not thoroughly mix withthe summit reservoir [Garcia et al., 1996]. Basedon these observations, the composition of the pa-rental magma delivered to Pu?u ?O?o from themantle is interpreted to have continually changedfor the remainder of the eruption (i.e., after 1984).The details and cause of this variation are dis-cussed in section 5.3.5.2. Oxygen Isotope Indications of CrustalContamination and Nature of MantleSource[19] Lavas from oceanic island volcanoes showwide ranges in oxygen isotopic compositions (4.5?7.5%), which have been attributed to composition-ally variable mantle-derived magmas that weremodified by oxygen exchange and/or crustal con-tamination [Harmon and Hoefs, 1995]. Our previ-ous studies revealed that some Pu?u ?O?o andK?lauea summit magmas experienced significantoxygen isotope exchange with metamorphosedK?lauea rocks [Garcia et al., 1998, 2008]. This isindicated by the disequilibrium between matrixand coexisting olivine 18O values, the relativelylow 18O values for these lavas (4.7?5.2%) andthe lack of correlation between 18O values andother geochemical parameters [Garcia et al.,1998, 2008].[20] The highest 18O value observed for any lavaduring the Pu?u ?O?o eruption is for the June 2007208Pb/206PbPCIGR analyses (2009)Marske et al. (2008)-SDSUPCIGR analyses (2010)SDSU analyses (2007)(Magma supply ratedoubled)2003-072.0592.0612.0632.0652.0672.0692.0711983198419851986198719881989199019911992199319941995199619971998199920002001200220032004200520062007200820092010Jun-99(episodic fountaining)Kupaianaha(gentle effusion)Pu?u ???? Pu?u ???? (mostly flank vent eruptions)Jun-861982summitRift-storedmagmaFigure 5. 208Pb/206Pb variation with time for Pu?u ?O?o lavas from 1983 to 2010. Previous analyses of Pu?u?O?o lavas indicated in legend are from Marske et al. [2008]. Average 62 for 208Pb/206Pb is smaller thansymbol size.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854861uprift lava (5.6% ; Figure 4). This value is identi-cal to the highest value reported for historicalK?lauea summit lavas (1820?1982) [Garcia et al.,2008] and lies within the range of normal mid-ocean ridge basalt (MORB) basalt 18O values(5.4?5.8% ; Figure 4) [Eiler, 2001]. These summitlavas (1832, 1866, 1894, and 1917?1921) wereerupted during periods of sustained lava lake ac-tivity, and are thought to be representative of theprimary uncontaminated magma feeding K?lauea[Garcia et al., 2008]. Thus, the 2007 uprift ventlava supports our previous interpretation [Garciaet al., 2008] that the 18O value for the mantlesource of K?lauea?s magma is identical to thesource for MORB.[21] The earliest Pu?u ?O?o lavas (1983?1986)show the strongest signs of disequilibriumbetween coexisting matrix mineral and olivine,and have the lowest O isotope values (<5.0%)[Garcia et al., 1998]. After the shift to continuouseffusion in July 1986, O isotope ratios are higher(5.2%) and the coexisting olivines were in equilib-rium with host matrix for about 1 year [Garciaet al., 1998]. Subsequently, the matrix O isotopevalues decreased somewhat (to 5.0% ; Figure 4)and those for olivine increased, indicating olivine-matrix disequilibrium. This O isotope disequili-brium continued for two more years, and was fol-lowed by a return to olivine-matrix equilibrium in1995?1997 [Garcia et al., 1998]. After 1997, ma-trix O isotope values are relatively low and nearlyconstant (5.36 0.1%) except for a 2005 lava(5.0% ; Figure 4), which was the most evolvedsample (analyzed for O isotopes) since 1984 (6.7wt % MgO). Thus, despite nearly 30 years of vig-orous eruptive activity (producing 4 km3 oflava), oxygen exchange with metamorphosedrocks has probably continued in the Pu?u ?O?omagmatic plumbing system. The magnitude of ox-ygen isotope exchange can be estimated assumingbulk assimilation between a parental magma (asreflected by the 2007 uprift sample with a 18Ovalue of 5.6%) and a hydrothermally alteredK?lauea rift zone lava (1.9%) [Garcia et al., 2008]as a contaminant. Pu?u ?O?o lavas erupted justbefore and after the 2007 uprift event have averageO isotope values of 5.4% (Figure 4), indicating5% bulk contamination, whereas earlier lavas(1986?2006) with average values of 5.2?5.3%,might have experienced 8?11% bulk contamina-tion. This contamination is likely to have occurredin the Pu?u ?O?o reservoir and did not have anyobvious effect on other geochemical parameters[Garcia et al., 1998] (Table 2).5.3. Cyclic Compositional VariationsFrom Mantle Processes (1985?2010)[22] Pu?u ?O?o lavas erupted after the early periodof magma mixing ended in late 1984 show cyclicvariations in several geochemical parameters thatare insensitive to olivine fractionation (e.g., CaO/TiO2, Sr/Nb, Zr/Nb,206Pb/204Pb; Figure 8). Thecyclic variations in CaO/TiO2 and K2O/TiO2ratios for Pu?u ?O?o lavas erupted between 1996and 2001 were reported to be associated with de-formation in the summit magma reservoir[Thornber, 2003]. Although the timing of thehighs and lows in these ratios are not perfectlycoincident with summit tilt changes [see Figure 8,Thornber, 2003], these geochemical cycles wereattributed to mixing of mantle-derived magma ofuniform composition (similar to Pu?u ?O?o lavasaverage ?2 SEaverage ?2 SETime (month-year)18.39018.39518.40018.40518.41018.41518.420Jan-97Jul-97Jan-98Jul-98J an-99Jul- 99Jan-00J ul -00Jan-01Jul-01Jan -02Jul-02Jan-03Jul-03J an-0 4Jul-04Jan-05J ul-05Jan-06Jul-06J an-07J ul-07Jan-08Jul-08Jan-09Jul-09J an-100.703590.703600.703610.703620.703630.703640.70365Jan-97Jul-97Jan-98Jul-98Jan-99Jul-99Jan -00Jul-00Jan-01Jul-01Jan -02Jul-02Jan-03Jul-03Jan -04Jul-04Jan-05Jul-05Jan-06Jul-06Jan-07Jul-07Jan-08Jul-08Jan -09Jul-09Jan -10206Pb/204Pb87Sr/ 86Sr(a)(b)Figure 6. Temporal variation in 206Pb/204Pb and 87Sr/86Srfor Pu?u ?O?o lavas during period of dramatic increase inmagma supply. Dashed lines indicates period of significantincrease in magma supply rate up to 0.25 km3/yr between2003 and 2007 compared to 0.1 km3/yr prior to 2003[Poland et al., 2012]. Analytical uncertainty is shown in thepanels.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854862from 1999 to 2001) with 1982 summit magma[Thornber 2003]. However, Pu?u ?O?o lavaserupted after 1999 have elevated 87Sr/86Sr ratios(at a given 206Pb/204Pb; Figure 7) compared toearlier lavas with low 206Pb/204Pb ratios, and thus,the 1999?2001 lavas cannot serve as a mixingend-members to explain the compositional trendof lavas erupted before 1999. Similar behavior isobserved on a plot of La/Yb versus 206Pb/204Pb(Figure 9), where a relative shift to lower La/Ybratios at a given 206Pb/204Pb occurred after 1999(compared to the trend of pre-1999 lavas). Theserelationships indicate that the temporal variationof Pu?u ?O?o lavas erupted after 1999 cannot beexplained simply by mixing of 1982 summitmagma with a uniform mantle-derived magma(Figure 7) within K?lauea?s shallow magmaticplumbing system. Instead, either a third magma ismixing with the other two or, as we advocatebelow, the composition of the Pu?u ?O?o magmais continually changing due to the melting ofsmall-scale compositional heterogeneities in themantle source.[23] Ratios of Pb isotopes in Pu?u ?O?o lavas showcyclic variations (Figure 5). These variations prob-ably reflect the dynamic process of melt extraction(from a heterogeneous source) over a time scale of0.703500.703550.703600.703650.7037018.36 18.38 18.40 18.42 18.44 18.46 18.48 18.50 18.52 18.54208Pb/204Pb0.703570.703580.703590.703600.703610.703620.703630.703640.7036518.38 18.39 18.40 18.41 18.42208Pb/204Pb38.04538.05538.06538.07518.38 18.39 18.40 18.41 18.4238.0038.0538.1038.1538.2018.36 18.38 18.40 18.42 18.44 18.46 18.48 18.50 18.52 18.5487Sr/ 86Sr206Pb/204PbSep-8687878889 9091909808090700030504020106929394959697 991988-911983-85Jun-9986102000-071992-992008-1096098788Oct-90Jan-90919808090700 0305040201069293949597991988-911983-85Jun-99 8610968788Mar-89Jul-89May-9096979898990005080808Jan-090909862000-071993-992008-1087Sr/ 86Sr206Pb/204Pb19831985198319851986-20101986-2010East Rift Zone(1977)East Rift Zone(1977)K?lauea summit(1982)K?lauea summit(1982)East Rift Zone(1960-69)East Rift Zone(1960-69)1984Sep.Jan.2001 composition1984Sep.Jan.?2?(b)(d)(a)(c)(b)(d)Figure 7. Pb and Sr isotopic compositions for Pu?u ?O?o lavas. Line connects samples in order of increasingeruption date in Figures 7b and 7d. Average 62 for Pb isotope ratios is smaller than symbol size. East RiftZone data is from J. Marske [personal communication, 2013]. September 1982 K?lauea summit lava composi-tion (outline in Figure 7a; orange star in Figure 7c) from Pietruszka and Garcia [1999]. K?lauea summit(1982) field in Figure 7a is new high-precision data from A. Pietruszka [personal communication, 2013]. Bluestar is 2001 composition proposed by Thornber [2003] as mixing end-member with 1982 K?lauea summitcomposition. Pu?u ?O?o lavas erupted after 1999 have elevated 87Sr/86Sr ratios (at a given 206Pb/204Pb). The1999?2001 lavas cannot serve as mixing end-members to explain the compositional trend of lavas eruptedbefore 1999.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854863years to decades rather than movement of small-scale mantle heterogeneities through the meltingzone. This interpretation is based on the hypothe-sis that buoyancy-driven upwelling through themelt-producing region beneath K?lauea occurs onlonger timescales (hundreds to thousands ofyears) than melt extraction (years to decades)[Pietruszka et al., 2006]. The highest estimatesfor solid mantle upwelling in the center of theHawaiian plume are 10 m/yr [Hauri, 1996;Pietruszka and Garcia, 1999], which wouldresult in a maximum of only 270 m of upwell-ing during the first 27 years of the Pu?u ?O?oeruption [cf. 5?10 km maximum thickness forthe zone of melting; Marske et al., 2007]. Forcomparison, estimates for solid mantle upwellingrates beneath Mauna Loa and Lo?ihi based onU-series disequilibria range from 0.4 to 1 m/yr[Sims et al., 1999] and 5?6 cm/yr [Pietruszkaet al., 2011], respectively. Melt extraction rates(or source-to-surface melt velocity) are estimatedto be on the order of 5?17 km/yr [Reiners,2002], which is extremely rapid compared tosolid mantle upwelling rates. Thus, cyclic varia-tion in Pb isotope ratios over short timescales(years) are best explained by variations in theprocess of melting of a heterogeneous source(and the transport of the melt to the surface),rather than upwelling of small-scale mantle het-erogeneities through the melting region.[24] The short-term Pb and Sr isotopic variationsin Pu?u ?O?o lavas may be generated by one ormore processes including: periodic processes ofmelting, melt extraction, or melt aggregation [e.g.,Cordier et al., 2010], changes in melt transportpathways or tapping new source areas [Marskeet al., 2008; Pietruszka et al., 2006], changes inthe volume of the melting region [Pietruszkaet al., 2001], and progressive melt extraction froma source with fine-scale heterogeneities [Garciaet al., 2000]. In the presence of small-scale hetero-geneities, changes in melt pathways over years todecades may lead to tapping compositionally dis-tinct sources and short-term isotopic variation inlavas [Marske et al., 2007]. The scale of composi-tional heterogeneities must be small relative to thesize of the melting region beneath K?lauea Vol-cano to allow for rapid (few years) variation inlava Pb isotope compositions [Pietruszka and Gar-cia, 1999]. Melt pathways within the source regionprobably migrate over years to decades [Pie-truszka et al., 2001, 2006]. Therefore, melt may besupplied from different areas of the melting region(Magma supply ratedoubled)2003-07Time (year)4.04.24.44.64.8CaO/TiO2 1986-19921992-201022232425262790 92 94 96 98 00 02 04 06 08 1090 92 94 96 98 00 02 04 06 08 1090 92 94 96 98 00 02 04 06 08 10206Pb/204PbSr/Nb(a)(d)Zr/Nb(b)(c)18.3518.3618.3718.3818.3918.4018.4118.4218.4318.44199019911992199319941995199619971998199920002001200220032004200520062007200820092010101112Figure 8. Plots of CaO/TiO2, Sr/Nb, Zr/Nb and206Pb/204Pbfor Pu?u ?O?o lavas showing cyclic variation apparent from1990 to 2010. Trace-element abundances in ppm.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854864over a relatively short period of time. These short-term geochemical fluctuations are effectivelytransported from the source to the surface becausePu?u ?O?o magmas are thought to partially bypassthe volcano?s summit magma storage reservoir,and avoid its buffering effects [Garcia et al.,2000].[25] The periodicity and rate of isotopic variationPu?u ?O?o lavas provide clues about the size anddistribution of small-scale heterogeneities in thevolcano?s mantle source. The cycles of Pb isotopicratios at Pu?u ?O?o have a peak to peak duration of10 years (Figure 10). These short-term increasesand decreases of Pb isotope ratios, at similar rateand degree, may represent melt extraction fromsmall-scale heterogeneities with a limited horizon-tal length scale (a few km or less). Modeling byFarnetani and Hofmann [2009] suggests that??filament-like?? structures derived from stretchingof deep-seated mantle heterogeneities may de-velop as the Hawaiian plume rises to the surface.Although their model was developed to explainthe long-term (>100 kyr) geochemical variationobserved for drill core from Mauna Kea [Farne-tani and Hofmann, 2010], we use the filamentmodel to explain the periodic variation in the Pbisotope ratios of Pu?u ?O?o lavas (Figure 10)because it provides a mechanism to link geochem-ical variations with the inferred deep mantle struc-ture. Other geometries have been suggested for thesmall-scale heterogeneities within Hawaiianplume, including a series of vertically stacked,elongated blobs [Blichert-Toft and Albare`de,2009], but we prefer the filament geometry toexplain the Pb isotopic variations of Pu?u ?O?olavas.[26] In this scenario, the periodic variation in Pbisotope ratios of Pu?u ?O?o lavas may reflect meltextraction from a mantle source with verticallyoriented repeating source heterogeneities, or thinstrands, on a small scale (Figure 10). The Pu?u?O?o eruption rate is thought to be greater than therate of melting, so melt must be transferred intochemically isolated channels from successivelyfurther areas within the larger melting region tosustain the eruption [Pietruszka et al., 2006]. Inthe context of the filament model of Farnetani andHofmann [2009], this process might extract meltfrom a succession of strands with different isotopiccompositions, which would potentially create theobserved periodicity in variation of the Pb isotoperatios (Figure 10). The volume of a single compo-sitional strand within the mantle tapped by thePu?u ?O?o eruption can be inferred using estimatesfor lava eruption rate (0.13 km3/yr) [Suttonet al., 2003] and melt zone porosity (1?2%) [Pie-truszka et al., 2001]. This calculation assumes that(1) there have been only two isotopically distinctcomponents since 1986 and (2) the heterogene-ities have the same melt productivity. We do notconsider the effect of melting heterogeneous lith-ologies with different melt productivities (e.g., pe-ridotite versus pyroxenite), despite the potentialsignificance for mixed lithologies in the source forHawaiian lavas [Hauri, 1996; Reiners, 2002;Sobolev et al., 2005]. Indeed, recent modeling ofincompatible trace elements suggests that Pu?u2007-1018.3818.4218.4618.5018.544.5 4.9 5.3 5.7 6.1 6.5 6.9 7.3 7.718.3818.3918.4018.4118.424.5 4.7 4.9 5.1 5.3 5.5 5.7 5.9HistoricalEast RiftZone ?2 ?summit(1982)La/Yb206Pb/204Pb206Pb/204Pb(a)(b)831988-912000-061992-99848586878889908990919394969598989708099900010304050607091092La/Ybmixingsource variationK?laueasummitFigure 9. Plots of 206Pb/204Pb versus La/Yb for Pu?u ?O?olavas with line connecting samples in eruptive order. (a) AllPu?u ?O?o lavas. (b) Smaller variations in lavas eruptedbetween 1988 and 2010 [area indicated by box with dashedline in Figure 9a]. K?lauea summit data is from Pietruszkaand Garcia [1999]. Average 62 for La/Yb shown in Figure9 a. Average 62 for Pb isotope ratios is smaller than symbolsize.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854865?O?o lavas (1986?1998) are derived from a sourcewith 13% recycled oceanic crust in a matrix ofambient depleted Hawaiian mantle [Pietruszkaet al., 2013]. However, this model was unable todistinguish if the recycled oceanic crust was pres-ent as pyroxenite or refertilized peridotite. Itshould also be noted that the compositional rangeof Pu?u ?O?o lavas from 1986 to 2010 is smallcompared to the overall range for Hawaiian volca-noes [e.g., Jackson et al., 2012; Ren et al., 2009;Weis et al., 2011], so the melt productivities of theend-member sources are probably similar.[27] The duration of a single cycle of Pb isotopevariation is approximately 10 years (Figure 10),which suggests melt is extracted from one compo-sitional strand in 5 years (before the trendreverses when melt from a different strand isencountered). Estimates for the height of K?lauea?smelting region range from <5 km [Marske et al.,2008] to 55 km [Watson and McKenzie, 1991].We assume the magma supply rate is roughlyequivalent to the eruption rate given that overallmagma storage in K?lauea has been decreasingslightly since 1983 [Poland et al., 2012]. The meltDepth (km) 13017010 50 90 130 170 210Length(km)Plate motionMelting zoneK?lauea Plate motion9 cm/yr100 0 100 200Length(km)Depth (km)15025050350Melting rate (10-11 kg m-3 s-1)0 2 4 6 8250 225 200 175 150 125?T (?C)50Depth (km)13017020 80 50Depth (km)13017020 80 50Depth (km)13017020 80Depth (km)50Depth (km)13017020 8019902.0652.0672.0692.071208 Pb/206 Pb1987198819891990199119921993199419951996199719981999200020012002200320042005200620072008200920101994     1998 2004(a)(b)(c)(d)Figure 10. Cartoon model for the Hawaiian mantle plume to explain the isotopic variation of Pu?u ?O?olavas, based on assumptions described in the text. (a) 208Pb/206Pb variation with time for Pu?u ?O?o lavas from1990 to 2010 showing cyclic variation. (b) Vertical section of the Hawaiian plume adapted from Farnetaniand Hofmann [2010]. Purple shades indicate the melting rates inside the melting zone, shown in legend.Dashed yellow lines are flow trajectories. Dashed black box is magma capture zone for K?lauea. (c and d)Sketches of the changing melting zone during Pu?u ?O?o eruption. Lithosphere is not shown in Figure 10c.Melting zone from Farnetani and Hofmann [2010].GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854866volume produced in 5 years is 0.5 km3. If meltzone porosity is 2% (estimates based on U-seriesstudies range from 1 to 3%) [Pietruszka et al.,2001; Sims et al., 1999], the total volume of a sin-gle heterogeneity tapped over a 5 year periodwould be 25 km3. If individual filament-shapedheterogeneities extend over the height of the melt-ing zone, then this 25 km3 volume would translateto a diameter of 1?3 km for melting zone heightsof 55?5 km, respectively.5.4. Effects of Doubling of Magma SupplyRate on Lava Composition (2003?2007)[28] One enigmatic question of ocean island volca-nism is whether variations in lava compositions arecorrelated with magma supply rate [e.g., Vlastelicet al., 2005]. Wide variations in magma supplyrate have occurred historically at K?lauea (0.01?0.18 km3/yr between 1840 and 1983) [Dvorak andDzurisin, 1993]. A marked decrease in lava effu-sion rate during the 19th to early 20th century atK?lauea (0.10?0.01 km3/yr) was accompanied byan increase in the ratios of highly over moderatelyincompatible trace-element abundances [Pie-truszka and Garcia, 1999], and in the modal abun-dance of clinopyroxene and plagioclase in thelavas indicating eruption of more fractionated andcooler magma [Garcia et al., 2003]. This change inlava composition is believed to be a direct result ofa decrease in the melt fraction (10?5%) and aswitch to a more depleted source [Pietruszka andGarcia, 1999]. A dramatic short-term increase ineffusion rate was observed for the Pu?u ?O?o erup-tion between 2003 and 2007 [Poland et al., 2012].Here we explore the results of this magma supplysurge on the composition of Pu?u ?O?o lavas.[29] Estimates of lava effusion rate for the Pu?u?O?o eruption prior to 2003 are based on geologicmapping, and measurements of very low fre-quency electromagnetic profiling and gas emis-sions. These techniques indicate an average rate ofmagma supply of 0.13 km3/yr [Sutton et al.,2003]. Lava effusion rate (considered to be aproxy for the magma supply rate by Poland et al.[2012]) was estimated to have increased between2003 and 2007 and to have doubled in 2005 (to0.25 km3/yr), before returning to the previousrate by 2008 [Poland et al., 2012]. An increase inmagma supply is normally expected to result inhigher MgO contents in erupted lavas as magmaundergoes less cooling prior to eruption. This rela-tionship was inferred for Pu?u ?O?o lavas eruptedfrom 1986 to 1992, when changes in tilt were fol-lowed 3 weeks later by changes in MgO [e.g.,Garcia et al., 1996]. However, during the 2003?2007 surge in magma supply, Pu?u ?O?o lavashave consistently lower MgO contents (<7.5 wt%; Figure 2a) than any period since the start ofcontinuous effusion in mid-1986, except duringepisode 54. The lower MgO contents of 2003?2007 lavas was interpreted to have resulted fromthe stirring and flushing of cooler magma withinthe volcano?s shallow magma storage system byan influx of new, hotter more primitive magma[Poland et al., 2012]. Mineralogical evidence (twopopulations of olivine) was noted in support ofthis claim, although no data were presented byPoland et al. [2012]. Our previous study of olivinecompositions in lavas erupted before and duringthe surge found no evidence for two populationsof olivines in any of the lavas, and that olivines inthese weakly phyric rocks are in Fe-Mg equilib-rium with the whole rock [Marske et al., 2008].We re-examined thin sections for these lavas andfound no textural evidence indicating magma mix-ing. In contrast, Pu?u ?O?o lavas from 1983 to1984 and episode 54 display obvious disequili-brium features from magma mixing [Garcia et al.,1989, 2000]. If mixing with a stored, coolermagma was important during the 2003?2007 surgein magma supply, the stored component must nothave differentiated very far beyond olivine control(unlike the situation for Pu?u ?O?o lavas from1983 to 1984 and episode 54).[30] A small increase in lava MgO contentoccurred after the surge, although values werevariable and overlap with those during the surge(7.0?8.1 after versus 6.7?7.4 wt % MgO during;Figure 2). The higher post-surge MgO values wereinterpreted to be a result of the heightened magmasupply from 2003 to 2007 [Poland et al., 2012].By November 2008, MgO dropped to values simi-lar to and lower than during the surge (6.5?7.2 wt%; Figure 2). Since 2000, the MgO content ofPu?u ?O?o lava has been declining with no signifi-cant change of this overall trend during the 2003?2007 surge in magma supply (Figure 2), except forless scatter in MgO content which may simplyreflect fewer interruptions in effusion during thistime. Also, it is noteworthy that the highest sus-tained MgO values were observed for Pu?u ?O?olavas from 1988 to 1993 (Figure 2), a period whenno increase in magma supply was recorded[Poland et al., 2012; Sutton et al., 2003]. Thus,the increase in magma supply from 2003 to 2007appears to have had limited impact on the varia-tion in the MgO content of Pu?u ?O?o lavas.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854867[31] Some of the other compositional features ofthe lavas erupted during the 2003?2007 magmasurge (e.g., slight offsets to lower MgO-normalizedSiO2 contents, Zr/Nb ratios and higher normalizedK2O and TiO2) could be considered to be represen-tative of flushing of differentiated magma similarto the Pu?u ?O?o lavas the late 1990s to early 2000s(Figures 2 and 3). There is also a reversal in87Sr/86Sr and 206Pb/204Pb in 2004 toward values by2007 similar to lavas from 2000 (Figure 6).These trends might be explained by the flushing ofmagma stored since the late 1990s to early 2000sduring the 2003?2007 surge in magma supply.However, based on the close time correlations(weeks to few months) between changes in summittilt and lava composition in earlier Pu?u ?O?o lavas[Garcia et al., 1996; Thornber, 2003], it is hard toimagine why 4? years were required to flushcooler magma from the Pu?u ?O?o system as pro-posed by Poland et al. [2012]. Thus, it is our inter-pretation that the magma erupted during the 2003?2007 surge was new mantle-derived magma andnot stored magma flushed from K?lauea?s shallowcrustal plumbing system. Thus, we interpret theisotopic variations during the Pu?u ?O?o eruption(since 1984) to be generated by the changingcomposition of melts coming from the mantle.5.5. Comparison of Pu?u ?O?o Lavas Withthe Long-Term Isotopic Evolution of OtherHawaiian Volcanoes[32] Studies of the long-term geochemical varia-tion of lavas from individual volcanoes in the Ha-waiian Islands provide understanding of thechemical structure of the Hawaiian mantle plume[e.g., Loa and Kea trends, Abouchami et al., 2005;Ren et al., 2009; Weis et al., 2011], and the vari-ability within single shield volcanoes [e.g., Bryceet al., 2005; Chen and Frey, 1985; Eisele et al.,2003; Marske et al., 2007; Nobre Silva et al.,2013; Rhodes and Hart, 1995; Weis et al., 2011].Here we compare the short-term Pb and Sr iso-topic variation for Pu?u ?O?o lavas to the longerterm variations for lavas from K?lauea and nearby,well-studied shield volcanoes to better understandthe rate and cause of long-term fluctuationsobserved at Hawaiian volcanoes.[33] The Pu?u ?O?o isotopic range covers a rela-tively large part of the long-term variationobserved for Mauna Kea and K?lauea volcanoes(Figures 11 and 12). The isotopic variation ofnearby Mauna Kea volcano was well documentedfor 300 kyr of shield growth using the HSDP2drill core. Lavas from the Pu?u ?O?o eruption afterthe period of magma mixing (early 1985) span30% of the total range of 206Pb/204Pb variationrecorded for HSDP2 (0.07 versus 0.22). Com-pared to K?lauea summit lavas, Pu?u ?O?o lavaserupted since 1985 span 25% of the Pb isotoperange since 950 AD and 47% of the range of Srisotope ratios (Figures 4 and 12). Thus, this singleeruption, which represents <1% of the time cov-ered by the HSDP2 core and 3% of thethousand-year period for K?lauea summit lavas,shows remarkable short-term isotopic variations.However, as seen for K?lauea summit lavas, iso-topic variation in Hawaiian shield lavas is cyclic,with each volcano showing a narrow but com-monly distinctive range (compared to neighboringvolcanoes) as seen by the relatively tight fields forPb and Sr isotope ratios in lavas from K?lauea,Lo?ihi, Mauna Kea and Mauna Loa (Figure 12).Thus, the Pu?u ?O?o eruption may represent thebest known expression of the small-scale compo-sitional heterogeneity of the Hawaiian plume. Asdiscussed above, the rapid rates of isotopic fluctu-ation found in Pu?u ?O?o lavas require a heteroge-neous source (on a scale of less than severalkilometers) that is tapped in only 5 years ofmelt extraction. How does the Pu?u ?O?o sourcecompare with those for nearby, well studied vol-canoes Lo?ihi and Mauna Kea?[34] Compared to other Hawaiian volcanoes, the Srand 206Pb/204Pb isotope ratios for Pu?u ?O?o lavasare most similar to Lo?ihi (Figure 12), althoughLo?ihi lavas have higher 208Pb/204Pb at a given206Pb/204Pb, like other Loa trend volcanoes (Figure11). Pu?u ?O?o lavas overlap with the Kea-mid8 Pbisotope array, the most common lava type inHSDP2 drill core (Figures 11c and 11d) [Eiseleet al., 2003; Nobre Silva et al., 2013]. However, Srisotopic compositions of Pu?u ?O?o lavas eruptedsince 1988 do not overlap those of Mauna Keaand trend orthogonally to the overall inverse arrayfor Hawaiian shield volcanoes and for K?laueasummit lavas (Figure 12). Thus, the Pu?u ?O?osource is isotopically distinct from other Hawaiianvolcanoes and the Pu?u ?O?o data set shows thatindividual eruptions may have trends orthogonal towhat are considered the primary source end-members for Hawaiian shield volcanoes.6. Conclusions[35] The temporal geochemical variation of Pu?u?O?o lavas from 1983 to 2010 provides insights onGREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854868the chemical structure of the Hawaiian mantleplume, and the dynamics of melt transport andmixing within the mantle. Our new results show:[36] 1. The Pu?u ?O?o eruption is being suppliedby new, compositionally variable, mantle-derivedmagma, which is being modified by various crustalprocesses including crystal fractionation (mainlyolivine), magma mixing (during 1983?1984 andepisode 54), and oxygen isotope exchange with orassimilation of altered K?lauea rocks.[37] 2. The episode 56 fissure lava has the highest18O value (5.6%) of any Pu?u ?O?o lava and thehighest MgO of any lava erupted since July 2000.Thus, it may be representative of the magma feed-ing this eruption as of June 2007. The episode 56fissure was fed from an upper East Rift Zone dikerather than the shallow Pu?u ?O?o reservoir. Coe-val lavas erupted from the Pu?u ?O?o vent havelower 18O values (5.36 0.1%) suggesting theywere contaminated, probably in the shallow reser-voir under the vent. Pu?u ?O?o lavas show no cor-relation of 18O values with other geochemicalparameters. For that reason, we suggest that the18O ratios were lowered by oxygen exchangewith or assimilation of altered K?lauea wall rock.[38] 3. Contrary to expectations, the dramaticincrease in magma supply between 2003 and 2007for the Pu?u ?O?o eruption was not accompanied byhigher MgO contents. Instead, lavas erupted duringthe 2003?2007 surge have lower MgO indicative ofgreater cooling of the magma prior to eruption,continuing the long-term trend for the eruption.[39] 4. Rapid and remarkably systematic variationsin Pb and Sr isotopic ratios are present in Pu?u37.737.837.938.038.138.238.338.418.0 18.1 18.2 18.3 18.4 18.5 18.6 18.72.042.052.072.082.092.100.825 0.830 0.835 0.840 0.845 0.850 0.855 0.860L??ihi2.0637.9037.9538.0038.0538.1038.1538.2038.2518.30 18.35 18.40 18.45 18.50 18.55 18.60 18.65 18.702.052.062.072.080.830 0.835 0.840 0.845208Pb/204Pb206Pb/204Pb208Pb/206Pb207Pb/206Pb208Pb/204Pb206Pb/204Pb208Pb/206Pb207Pb/206PbMauna LoaMauna LoaMauna KeaHistorical K?lauea summitPu?u ???? (1983-2010) (b)(d)(c)(a)1986Uwekahuna Bluff (summit)(Pietruszka and Garcia, 1999)Prehistoric K?lauea(Abouchami et al., 2005)L??ihi19831986Hilina BenchHilina Bench(Marske et al., 2007)upper endmemberkea-hi8kea-mid8kea-low8Mauna KeaMauna Kea (HSDP-2)Kea-hi8 arrayKea-mid8 arrayKea-low8 arrayarea of panel c)area of panel d)Historical K?lauea summitLoaKeaLoaKea1983Figure 11. Pb isotopic compositions for Pu?u ?O?o lavas compared to some other Hawaiian shield volca-noes. Top panels are different scales than bottom panels, shown by area of dashed boxes. Colored symbols arelavas from K?lauea Volcano. Fields are for Mauna Loa, Lo?ihi, Mauna Kea, and submarine prehistoricK?lauea (Hilina Bench; >40 ka). References for data sources are listed in the supporting information. Keaarray end-members are from Eisele et al. [2003]. Dashed line with yellow stars in Figure 11c is a best fit linefor Pu?u ?O?o lavas from 1988 to 2010.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854869?O?o lavas. Two cycles of Pb isotopic ratio varia-tion with 10 year periods were found. Thesecycles may be related to extraction of melt from asource with a pattern of vertically oriented sourceheterogeneities, or thin strands. These strands maybe 1?3 km in diameter to explain the scale ofisotopic variation for the Pu?u ?O?o eruption.[40] 5. The Pb isotopic variation of Pu?u ?O?olavas spans 25% of the range observed for the last1000 years of K?lauea summit lavas and 30% for300,000 years of shield volcanism for MaunaKea volcano. There is considerable Pb and Sr iso-topic overlap between Pu?u ?O?o lavas and lavasfrom Mauna Kea and Lo?ihi volcanoes. However,the Pb-Sr isotopic trend for the later Pu?u ?O?olavas (1988?2010) is oblique to the array definedby Hawaiian shield lavas. Thus, each Hawaiianvolcano appears to have an isotopically distinctsource.0.70320.70330.70340.70350.70360.70370.70380.70390.704018.0 18.1 18.2 18.3 18.4 18.5 18.6 18.7 18.80.703450.703500.703550.703600.703650.7037018.2 18.3 18.4 18.5 18.6 18.7 18.8Mauna LoaL??ihiMauna KeaMauna LoaMauna KeaL??ihiSalt Lake Crater xenolithsPu?u ???? (1983-2010) Mauna Kea (HSDP-2)Kea-hi8 arrayKea-mid8 arrayKea-low8 array206Pb/204Pb87Sr/ 86Sr206Pb/204Pb87Sr/ 86Sr198319851986Hilina BenchHilina BenchHistorical K?lauea summitUwekahuna Bluff (summit)(Pietruszka and Garcia, 1999)Prehistoric K?lauea(Abouchami et al., 2005)(Marske et al., 2007)area of panel b)Uwekahuna Bluff (summit)200320100.4-5 kaHistorical and PrehistoricK?lauea summit not shownHistorical K?lauea summit(b)(a)Figure 12. Plots of 206Pb/204Pb and 87Sr/86Sr for Pu?u ?O?o lavas compared to some other Hawaiian shieldvolcanoes. Figure 12b Expanded scale of dashed box in Figure 12a. Colored symbols are lavas from K?laueaVolcano. Fields are for Mauna Loa, Lo?ihi, Mauna Kea and submarine K?lauea (Hilina Bench; >40 ka).References for data sources are listed in the supporting information.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854870Acknowledgments[41] We thank Jane Barling, Bruno Kieffer, and Vivian Laifor their assistance with analyses at PCIGR at UBC, KyleTanguichi and Adonara Murek for sample preparation andcuration at the University of Hawai?i, and J. M. Rhodes forXRF analyses at University of Massachusetts. Claude Maer-schalk assisted with Pb and Sr column chemistry for a subsetof samples. Daniel Heaton provided assistance with severalisotope analyses at San Diego State University. We appreciatereviews by Julie Prytulak, Joel Baker, and Christoph Beier.This research was supported by grants from the National Sci-ence Foundation to M. Garcia (EAR11?18741) and A. Pie-truszka (EAR11?18738). This paper is SOEST ContributionNo. 8939.ReferencesAbouchami, W., S. J. G. Galer, and A. W. Hofmann (2000),High precision lead isotope systematics of lavas from theHawaiian Scientific Drilling Project, Chem. Geol., 169, 187?209, doi:10.1016/S0009?2541(00)00328-4.Abouchami, W., A. W. Hofmann, S. J. G. Galer, F. A. Frey, J.Eisele, and M. Feigenson (2005), Lead isotopes reveal bilat-eral asymmetry and vertical continuity in the Hawaiian man-tle plume, Nature, 434, 851?856, doi:10.1038/nature03402.Albare`de, F., B. Luais, G. Fitton, M. Semet, E. Kaminski, B. G.J. Upton, P. Bache`lery, and J.-L. Cheminee (1997), The geo-chemical regimes of Piton de la Fournaise Volcano (Reunion)during the last 530,000 years, J. Petrol., 38, 171?201.Blichert-Toft, J., and F. Albare`de (2009), Mixing of isotopicheterogeneities in the Mauna Kea plume conduit, EarthPlanet. Sci. Lett., 282, 190?200, doi:10.1016/j.epsl.2009.03.015.Blichert-Toft, J., D. Weis, C. Maerschalk, A. Agranier, and F.Albare`de (2003), Hawaiian hot spot dynamics as inferredfrom the Hf and Pb isotope evolution of Mauna Kea volcano,Geochem. Geophys. Geosyst., 4(2), 8704, doi:10.1029/2002GC000340.Bryce, J. G., D. J. DePaolo, and J. C. Lassiter (2005), Geo-chemical structure of the Hawaiian plume: Sr, Nd, and Osisotopes in the 2.8 km HSDP-2 section of Mauna Kea vol-cano, Geochem. Geophys. Geosyst., 6, Q09G18,doi:10.1029/2004GC000809.Caroff, M., R. C. Maury, P. Vidal, G. Guille, C. Dupuy, J. Cot-ten, H. Guillou, and P.-Y. Gillot (1995), Rapid temporalchanges in ocean island basalt composition: Evidence froman 800 m deep drill hole in Eiao Shield (Marquesas), J. Pet-rol., 36(5), 1333?1365, doi:10.1093/petrology/36.5.1333.Chen, C. Y., and F. A. Frey (1985), Trace element and isotopicgeochemistry of lavas from Haleakala Volcano, east Maui,Hawai?i : Implications for the origin of Hawaiian basalts, J.Geophys. Res., 90, 8743?8768.Chen, C.-Y., F. A. Frey, J. M. Rhodes, and R. M. Easton(1996), Temporal geochemical evolution of K?lauea Vol-cano: Comparison of Hilina and Puna basalt, in Earth Proc-esses, Reading the Isotopic Code, edited by A. Basu and S.R. Hart, pp. 161?181, AGU, Washington, D. C.Cordier, C., M. Benoit, C. Hemond, J. Dyment, B. Le Gall, A.Briais, and M. Kitazawa (2010), Time scales of melt extrac-tion revealed by distribution of lava composition across aridge axis, Geochem. Geophys. Geosyst., 11, Q0AC06,doi:10.1029/2010gc003074.DePaolo, D. J., and E. M. Stolper (1996), Models of Hawaiianvolcano growth and plume structure: Implications from theHawai?i Scientific Drilling Project, J. Geophys. Res., 101,11,643?11,655.Dvorak, J. J., and D. Dzurisin (1993), Variations in magmasupply rate at K?lauea Volcano, Hawai?i, J. Geophys. Res.,98, 22,255?22,268.Eaton, J. P., and K. J. Murata (1960), How volcanoes grow[Hawai?i], Science, 132, 925?938.Eiler, J. M. (2001), Oxygen isotope variations of basaltic lavasand upper mantle rocks, Rev. Mineral. Geochem., 43(1),319?364, doi:10.2138/gsrmg.43.1.319.Eisele, J. R., W. Abouchami, S. J. G. Galer, and A. W. Hof-mann (2003), The 320 kyr Pb isotope evolution of MaunaKea lavas recorded in the HSDP-2 drill core, Geochem. Geo-phys. Geosyst., 4(5), 8710, doi:10.1029/2002GC000339.Farnetani, C. G., and A. W. Hofmann (2009), Dynamics andinternal structure of a lower mantle plume conduit, EarthPlanet. Sci. Lett., 282(1?4), 314?322, doi:10.1016/j.epsl.2009.03.035.Farnetani, C. G., and A. W. Hofmann (2010), Dynamics andinternal structure of the Hawaiian plume, Earth Planet. Sci.Lett., 295, 231?240, doi:10.1016/j.epsl.2010.04.005.Garcia, M. O., R. A. Ho, J. M. Rhodes, and E. W. Wolfe(1989), Petrologic constraints on rift-zone processes:Results from episode 1 of the Pu?u ?O?o eruption of K?laueavolcano, Hawai?i, Bull. Volcanol., 52, 81?96.Garcia, M. O., J. M. Rhodes, E. W. Wolfe, G. E. Ulrich, and R.A. Ho (1992), Petrology of lavas from episodes 2?47 of thePu?u ?O?o eruption of K?lauea Volcano, Hawai?i : Evaluationof magmatic processes, Bull. Volcanol., 55(1), 1?16.Garcia, M. O., J. M. Rhodes, F. A. Trusdell, and A. J. Pie-truszka (1996), Petrology of lavas from the Pu?u ?O?o erup-tion of K?lauea Volcano: III The Kupaianaha episode(1986?1992), Bull. Volcanol., 58, 359?379.Garcia, M. O., E. Ito, J. M. Eiler, and A. J. Pietruszka (1998),Crustal contamination of K?lauea Volcano magmas revealedby oxygen isotope analyses of glass and olivine from Pu?u?O?o eruption lavas, J. Petrol., 39, 803?817.Garcia, M. O., A. J. Pietruszka, J. M. Rhodes, and K. Swanson(2000), Magmatic processes during the prolonged Pu?u ?O?oeruption of K?lauea Volcano, Hawai?i, J. Petrol., 41(7),967?990.Garcia, M. O., A. J. Pietruszka, and J. M. Rhodes (2003), Apetrologic perspective of K?lauea Volcano?s summit magmareservoir, J. Petrol., 44(12), 2313?2339, doi:10.1093/petrol-ogy/egg079.Garcia, M. O., E. Ito, and J. M. Eiler (2008), Oxygen isotopeevidence for chemical interaction of K?lauea historical mag-mas with basement rocks, J. Petrol., 49(4), 757?769,doi :10.1093/petrology/egm034.Hanyu, T., J.-I. Kimura, M. Katakuse, A. T. Calvert, T. W. Sis-son, and S. I. Nakai (2010), Source materials for inceptionstage Hawaiian magmas: Pb-He isotope variations for earlyK?lauea, Geochem. Geophys. Geosyst., 11, Q0AC01,doi:10.1029/2009gc002760.Harmon, R. S., and J. Hoefs (1995), Oxygen isotope heteroge-neity of the mantle deduced from global 18O systematics ofbasalts from different geotectonic settings, Contrib. Mineral.Petrol., 120(1), 95?114.Hauri, E. H. (1996), Major-element variability in the Hawaiianmantle plume, Nature, 382, 415?419.Heliker, C., and T. N. Mattox (2003), The first two decades ofthe Pu?u ?O?o -Kupaianaha eruption; chronology andselected bibliography, in The Pu?u ?O?o -Kupaianaha Erup-tion of K?lauea Volcano, Hawai?i: The First 20 Years, editedGREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854871by C. Heliker et al., pp. 1?28, U.S. Geol. Surv. Prof. Pap.1676.Hofmann, A. W., M. D. Feigenson, and I. Raczek (1984), Casestudies on the orgin of basalt : III Petrogenesis of the MaunaUlu eruption, K?lauea, 1969?1971, Contrib. Mineral. Pet-rol., 88, 24?35.Holcomb, R. T. (1987), Eruptive history and long-term behav-ior of K?lauea Volcano, in Volcanism in Hawai?i, edited byR. W. Decker et al., pp. 261?350, U.S. Geol. Surv. Prof. Pap.1350.Jackson, M. G., D. Weis, and S. Huang (2012), Major elementvariations in Hawaiian shield lavas: Source features and per-spectives from global ocean island basalt (OIB) systematics,Geochem. Geophys. Geosyst., 13, Q09009, doi:10.1029/2012GC004268.Johnson, D. J., A. A. Eggers, and M. Bagnardi (2010), Shallowmagma accumulation at K?lauea Volcano, Hawai?i, revealedby microgravity surveys, Geology 38(12), 1139?1142,doi:10.1130/G31323.1.Kimura, J.-I., T. W. Sisson, N. Nakano, M. L. Coombs, and P.W. Lipman (2006), Isotope geochemistry of early K?laueamagmas from the submarine Hilina bench: The nature of theHilina mantle component, J. Volcanol. Geotherm. Res.,151(1?3), 51?72, doi :10.1016/j.jvolgeores.2005.07.024.Kurz, M. D., W. J. Jenkins, and S. R. Hart (1982), Helium iso-topic systematics of oceanic islands and mantle heterogene-ity, Nature, 297, 43?47.Macdonald, G. A., A. T. Abbott, and F. L. Peterson (1983),Volcanoes in the Sea: The Geology of Hawai?i, 517 pp.,Univ. of Hawaii Press, Honolulu, Hawaii.Maclennan, J. (2008), Lead isotope variability in olivine-hosted melt inclusions from Iceland, Geochim. Cosmochim.Acta, 72(16), 4159?4176, doi:10.1016/j.gca.2008.05.034.Marske, J. P., A. J. Pietruszka, D. Weis, M. O. Garcia, and J.M. Rhodes (2007), Rapid passage of a small-scale mantleheterogeneity through the melting regions of K?lauea andMauna Loa volcanoes, Earth Planet. Sci. Lett., 259, 34?50,doi:10.1016/j.epsl.2007.04.026.Marske, J. P., M. O. Garcia, A. J. Pietruszka, J. M. Rhodes,and M. D. Norman (2008), Geochemical variations duringK?lauea?s Pu?u ?O?o eruption reveal a fine-scale mixture ofmantle heterogeneities within the Hawaiian plume, J. Pet-rol., 49(7), 1297?1318, doi:10.1093/petrology/egn025.Montgomery-Brown, E. K., D. K. Sinnett, P. S. M. Poland, T.Orr, H. Zebker, and A. Miklius (2010), Geodetic evidencefor en echelon dike emplacement and concurrent slow slipduring the June 2007 intrusion and eruption at K?lauea vol-cano, Hawai?i, J. Geophys. Res., 115, B07405, doi:10.1029/2009JB006658.Nobre Silva, I. G., D. Weis, and J. S. Scoates (2013), Isotopicsystematics of the early Mauna Kea shield phase and insightinto the deep mantle beneath the Pacific Ocean, Geochem.Geophys. Geosyst., 14(3), 659?676, doi:10.1029/2012GC004373.Pietruszka, A. J., and M. O. Garcia (1999), A rapid fluctuationin the mantle source and melting history of K?lauea Volcanoinferred from the geochemistry of its historical summit lavas(1790?1982), J. Petrol., 48(8), 1321?1342.Pietruszka, A. J., K. H. Rubin, and M. O. Garcia (2001),226Ra-230Th-238U disequilibria of historical K?lauea lavas(1790?1982) and the dynamics of mantle melting within theHawaiian plume, Earth Planet. Sci. Lett., 186, 15?31.Pietruszka, A. J., E. H. Hauri, R. W. Carlson, and M. O. Garcia(2006), Remelting of recently depleted mantle within theHawaiian plume inferred from the 226Ra-230Th-238U disequi-libria of Pu?u ?O?o eruption lavas, Earth Planet. Sci. Lett.,244(1?2), 155?169, doi:10.1016/j.epsl.2006.01.039.Pietruszka, A. J., E. H. Hauri, and J. Blichert-Toft (2009),Crustal contamination of mantle-derived magmas within Pi-ton de la Fournaise Volcano, Reunion Island, J. Petrol.,50(4), 661?684, doi :10.1093/petrology/egp016.Pietruszka, A. J., M. D. Norman, M. O. Garcia, J. P. Marske,and D. H. Burns (2013), Chemical heterogeneity in the Ha-waiian mantle plume from the alteration and dehydration ofrecycled oceanic crust, Earth Planet. Sci. Lett., 361, 298?309, doi:10.1016/j.epsl.2012.10.030.Poland, M., A. Miklius, T. Orr, A. J. Sutton, C. Thornber, andD. Wilson (2008), New episodes of volcanism at K?laueaVolcano, Hawai?i, Eos Trans. AGU, 89(5), 37?38,doi :10.1029/2008EO050001.Poland, M. P., A. Miklius, A. J. Sutton, and C. R. Thornber(2012), A mantle-driven surge in magma supply to K?laueaVolcano during 2003?2007, Nat. Geosci., 5, 295?300,doi :10.1038/NGEO1426.Reiners, P. W. (2002), Temporal-compositional trends inintraplate basalt eruptions: Implications for mantle hetero-geneity and melting processes, Geochem. Geophys. Geo-syst., 3(2), 1011, doi:10.1029/2001GC000250.Ren, Z.-y., T. Hanyu, T. Miyazaki, Q. Chang, H. Kawabata, T.Takahashi, Y. Hirahara, A. R. L. Nichols, and Y. Tatsumi(2009), Geochemical differences of the Hawaiian shieldlavas: Implications for melting process in the heterogeneousHawaiian plume, J. Petrol., 50(8), 1553?573, doi:10.1093/petrology/egp041.Rhodes, J. M., and S. R. Hart (1995), Episodic trace elementand isotopic variations in Mauna Loa lavas: Implications formagma and plume dynamics, in Mauna Loa Revealed:Structure, Composition, History and Hazards, edited by J.M. Rhodes and J. P. Lockwood, pp. 263?288, AGU, Wash-ington, D. C.Rhodes, J. M., and M. J. Vollinger (2004), Composition of ba-saltic lavas sampled by phase-2 of the Hawai?i ScientificDrilling Project: Geochemical stratigraphy and magmatypes, Geochem. Geophys. Geosyst., 5, Q03G13,doi:10.1029/2002GC000434.Rhodes, J. M., S. Huang, F. A. Frey, M. Pringle, and G. Xu(2012), Compositional diversity of Mauna Kea shield lavasrecovered by the Hawai?i Scientific Drilling Project: Infer-ences on source lithology, magma supply, and the role ofmultiple volcanoes, Geochem. Geophys. Geosyst., 13,Q03014, doi:10.1029/2011GC003812.Ryan, M. P. (1987), Elasticity and contractancy of Hawaiianolivine tholeiite and its role in the stability and structuralevolution of subcaldera magma reservoirs and rift systems,in Volcanism in Hawai?i, edited by R. W. Decker et al., pp.1395?1448, U.S. Geol. Surv. Prof. Pap. 1350.Shamberger, P. J., and M. O. Garcia (2007), Geochemicalmodeling of magma mixing and magma reservoir volumesduring early episodes of K?lauea Volcano?s Pu?u ?O?o erup-tion, Bull. Volcanol., 69, 345?352, doi:10.1007/s00445-006-0074-5.Sims, K. W. W., M. T. Murrell, D. J. DePaolo, W. S. Bal-dridge, S. J. Goldstein, D. Clague, and M. Jull (1999), Poros-ity of the melting zone and variations in the solid mantleupwelling rate beneath Hawai?i : Inferences from238U-230Th-226Ra and 235U-231Pa disequilibria, Geochim.Cosmochim. Acta, 63(23), 4119?4138.Sobolev, A. V., A. W. Hofmann, S. Sobolev, and I. K. Nikogo-sian (2005), An olivine-free mantle source of Hawaiianshield basalts, Nature, 434, 590?595.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854872Sutton, J. A., T. Elias, and J. Kauahikaua (2003), Lava-effu-sion rates for the Pu?u ?O?o -Kupaianaha eruption derivedfrom SO2 emissions and very low frequency (VLF) meas-urements, in The Pu?u ?O?o -Kupaianaha Eruption ofK?lauea Volcano, Hawai?i: The First 20 Years, edited byC. Heliker et al., pp. 137?148, U.S. Geol. Surv. Prof. Pap.1676.Thornber, C. R. (2003), Magma-reservoir processes revealedby geochemistry of the Pu?u ?O?o-Kupaianaha eruption, inThe Pu?u ?O?o -Kupaianaha Eruption of K?lauea Volcano,Hawai?i: The First 20 Years, edited by C. Heliker et al., pp.121?136, U.S. Geol. Surv. Prof. Pap. 1676.Thornber, C. R., C. Heliker, D. R. Sherrod, J. P. Kauahikaua,A. Miklius, P. G. Okubo, F. A. Trusdell, J. R. Budahn, W. I.Ridley, and G. P. Meeker (2003), K?lauea East Rift Zonemagmatism: An episode 54 perspective, J. Petrol., 44(9),1525?1559, doi:10.1093/petrology/egg048.Tilling, R. I., and J. J. Dvorak (1993), Anatomy of a basalticvolcano, Nature, 363, 125?133.Vlastelic, I., T. Staudacher, and M. Semet (2005), Rapidchange of lava composition from 1998 to 2002 at Piton de laFournaise (Reunion) inferred from Pb isotopes and trace ele-ments: Evidence for variable crustal contamination, J. Pet-rol., 46(1), 79?107, doi :10.1093/petrology/egh062.Vlastelic, I., A. Peltier, and T. Staudacher (2007), Short-term(1998?2006) fluctuations of Pb isotopes at Piton de la Four-naise volcano (Reunion Island): Origins and constraints onthe size and shape of the magma reservoir, Chem. Geol.,244, 202?220, doi:10.1016/j.chemgeo.2007.06.015.Watson, S., and D. McKenzie (1991), Melt generation byplumes: A study of Hawaiian volcanism, J. Petrol., 32(3),501?537.Weis, D., M. O. Garcia, J. M. Rhodes, A. M. Jellinek, and J. S.Scoates (2011), Role of the deep mantle in generating thecompositional asymmetry of the Hawaiian mantle plume,Nat. Geosci., 4, 831?838, doi:10.1038/ngeo1328.Wolfe, E. W., M. O. Garcia, D. B. Jackson, R. Y. Koyanagi, C.A. Neal, and A. T. Okamura (1987), The Pu?u ?O?o eruptionof K?lauea Volcano, episodes 1?20, January 3, 1983 to June8, 1984, in Volcanism in Hawai?i, edited by R. W. Deckeret al., pp. 471?508, U.S. Geol. Surv. Prof. Pap. 1350.Wright, T. L. (1971), Chemistry of K?lauea and Mauna Loalava in space and time, U.S. Geol. Surv. Prof. Pap., 735,1?45.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854873 1  Sample Preparation and Analytical Methods  A1. University of Massachusetts XRF Analytical Methods New XRF analyses are given in the Auxiliary Materials (ts1) with previous major-element analyses. New XRF trace element analyses are also given (and indicated in table) for samples erupted prior to 1998, when a new, more precise XRF instrument became available. Fresh pieces of quenched lava were crushed in a tungsten carbide jaw crusher to ~2 mm diameter and cleaned in a beaker with deionized water and dried in an oven at 70?C for 24 hours to drive off excess water. Lava fragments were then powdered using a Rocklabs WC shatterbox for 1 ? 2 minutes. All XRF analyses were performed at the University of Massachusetts XRF Laboratory where whole-rock, major-element abundances were measured using the procedures of Rhodes and Vollinger [2004]. One sigma accuracy and precision estimates for the XRF data are ~0.5% for major elements [Rhodes, 1988]. Five grams of a powdered sample were heated in a muffle furnace at 1020?C for 10 minutes to limit the amount of ferrous iron formation [Rhodes and Vollinger, 2004]. The weight loss is LOI. Major elements were measured on a fused La-bearing lithium borate glass disc using a Siemens MRS-400 spectrometer with a Rh X-ray tube operating at 2700 W (the voltage was 45 kV and 60mA and the count times were 200 seconds). Trace element concentrations (Rb, Sr, Ba, La, Ce, Nb, Zr, Y, Zn, Ga, Ni, Cr, V) were measured on a separate powder pellet using a Philips PW2400 sequential spectrometer with a Rh X-ray tube operating at 3000 W (the voltage was 60kV and 50mA and 2.5 hour total count time). Precision and accuracy estimates for the trace element data are described by Rhodes [1996] and Rhodes and Vollinger [2004]. Results for each sample are the average of two separate analyses (2 disks for one sample) for major elements, but one analysis (one pellet run once) for trace elements. A summary of XRF trace element values of USGS standards analyzed with lavas from K?lauea?s Pu?u ???? eruption since 2002 are given in the Auxiliary Materials (ts3).  A2. PCIGR and SDSU Trace Element and Isotopic Analytical Methods Forty-one Pu?u ???? lava samples from 1983 to 1999 were selected for Pb and Sr isotopic analysis at the Pacific Centre for Isotopic and Geochemical Research (PCIGR) at the University of British Columbia (UBC; Table 2). The analytical methods for thirteen additional Pb isotopic analyses of Pu?u ???? lavas collected between 2006 and 2010, and fourteen analyses from Marske et al. [2008] that are used in this study, are described in Marske et al. [2007] (samples 2  are indicated in Table 2). Thirteen samples without trace-element data were selected for high-precision trace-element analysis and results are shown in Auxiliary Materials (ts2) with previous trace-element analyses for Pu?u ???? lavas (corrected data is also shown, as described in table caption). Isotopic ratios of Pb for these samples were measured with a Nu Plasma (Nu Instruments) 1700 Multiple Collector-Inductively Coupled Plasma-Mass Spectrometer (MC-ICP-MS) at San Diego State University (SDSU). Sr isotopic ratios for 1998-2005 lava samples from Marske et al. [2008] were measured using a VG Sector 54 thermal ionization mass spectrometer (TIMS) at SDSU. Lava samples from 2006 to 2010 were analyzed for Sr isotopic ratios at the PCIGR, according the the methods described below.  Samples were prepared for trace-element analysis at the PCIGR by the technique described by Pretorius et al. [2006] and Carpentier et al. [2013] on unleached lava fragments. Lava fragments <2 mm in diameter (~100 mg) were weighed in 7 mL screw-top Savillex? beakers and dissolved in 1 mL ~14N HNO3 and 5 mL 48% HF on a hotplate for 48 hours at 130?C with periodic ultrasonication. Samples were dried and redissolved in 6 mL 6N HCl on a hotplate for 24 hours and then dried and redissolved in 1 mL concentrated HNO3 for 24 hours before final drying. Trace element abundances were measured with a Thermo Finnigan Element2 High Resolution-Inductively Coupled Plasma-Mass Spectrometer (HR-ICP-MS) within 24 hours of redissolution, following the procedures described by Pretorius et al. [2006] and Carpentier et al. [2013]. High field strength elements (HFSE) and large ion lithophile elements (LILE) were measured in medium resolution mode at 2000x dilution using a PFA teflon spray chamber washed with Aqua Regia for 3 minutes between samples. Rare earth elements (REE) were measured in high resolution mode, and U and Pb in low resolution mode, at 2000x dilution using a glass spray chamber washed with 2% HNO3 between samples. Total procedural blanks and reference material (Kil93) was analyzed with the batch of samples. Indium was used as an internal standard in all samples and standard solutions. Background and standard solutions were analyzed after every 5 samples to detect memory effects and mass drift.  For isotope chemistry at PCIGR, fresh pieces of crushed lava (<2mm diameter; ~0.6-0.8 g total) were carefully picked using a binocular microscope to avoid signs of post-eruptive alteration. 3  Each sample was ultra-sonicated (15 minutes) and rinsed three times in ultra-pure water in a clean Savillex? beaker prior to rinsing with room temperature 0.5 M HBr for 15 seconds to remove any surface contamination. The sample was washed again in ultra-pure water, dried, and weighed. Sample digestion for purification of Pb and Sr by column chemistry involved dissolving ~100-250 mg of the cleaned, crushed lava in 1 mL ~14N HNO3 and 10 mL 48% HF on a hotplate for 48 hours at 130?C with periodic ultrasonication. Samples were dried and redissolved in 6 mL 6N HCl on a hotplate for 24 hours and then dried. Pb was separated and purified in two passes through anion exchange columns and the discard was used for Sr separation. Detailed procedures for column chemistry for separating Pb and Sr at the PCIGR are described in Weis et al. [2006] and Nobre Silva et al. [2009]. Sr isotope ratios were measured on a Thermo Finnigan Triton Thermal Ionization Mass Spectrometer (TIMS) in static mode with relay matrix rotation on a single Ta and double Re-Ta filament, respectively. Four to 5 filaments per barrel of 21 were occupied by standard NIST SRM 987 for each barrel where samples were run. Sample Sr isotopic compositions were corrected for mass fractionation using 86Sr/88Sr = 0.1194. Each sample was then normalized using the barrel average of the reference material relative to the values of 87Sr/86Sr = 0.710248 [Weis et al., 2006]. During the period when the samples were analyzed, the NIST SRM 987 standard gave an average of 0.710243 ? 13 (n = 18; 2? error is reported as times 106). PCIGR internal reference material from the K?lauea summit eruption in 1919 (Kil1919) was processed with the samples and yielded Sr isotopic ratios of 0.703468 ? 8. Twenty-six previous analyses of Kil1919 at SDSU average 0.703471 ? 13. Four chemistry duplicates of Menehune standard (an in-house glass standard collected from a Pu?u ???? lava flow of K?lauea Volcano that was quenched on June 24, 2006), processed with the other lava samples, had Sr isotopic compositions of  0.703608 ? 7, 0.703605 ? 8, 0.703611 ? 8, and 0.703624 ? 9. Twenty-six previous analyses of Menehune standard performed at SDSU averaged 0.703617 ? 13.  Pb isotopic compositions at PCIGR were analyzed by static multi-collection on a Nu Plasma MC-ICP-MS. The detailed analytical procedure for Pb isotopic analyses at the PCIGR is described in Weis et al. [2005]. The configuration for Pb analyses allows for collection of Pb, Tl, and Hg together. Tl and Hg are used to monitor instrumental mass discrimination and isobaric overlap, repectively. All sample solutions were analyzed with approximately the same Pb/Tl 4  ratio (~4) as the reference material NIST SRM 981. To accomplish this, a small aliquot of each sample solution from the Pb columns was analyzed on the Nu Plasma to determine the amount of Pb available. The SRM 981 standard was run after every two samples on the Nu Plasma. During the time samples were run, analyses of the SRM 981 Pb reference material gave values of 206Pb/204Pb = 16.9430 ? 19, 207Pb/204Pb = 15.4997 ? 18, and 208Pb/204Pb = 36.7221 ? 58 (n = 38; 2? error is reported as times 104); these values are in excellent agreement with reported TIMS triple-spike values of Galer and Abouchami [1998]. Results were further corrected by the sample-standard bracketing method described by White et al. [2000]. Reference material Kil1919 analyzed at PCIGR yielded Pb isotopic ratios of 206Pb/204Pb = 18.6556 ? 8, 207Pb/204Pb = 15.4924 ? 8, and 208Pb/204Pb = 38.2148 ? 21. Thirty-three analyses of Kil1919 performed at SDSU over the course of this study averaged 206Pb/204Pb = 18.6552 ? 27, 207Pb/204Pb = 15.4897 ? 20, and 208Pb/204Pb = 38.2068 ? 57. Four chemistry duplicates of Menehune standard analyzed at PCIGR averaged 206Pb/204Pb = 18.4062 ? 28, 207Pb/204Pb = 15.4724 ? 26, and 208Pb/204Pb = 38.0662 ? 59. Sixty-eight previous analyses of Menehune standard performed at SDSU over the course of this study averaged 206Pb/204Pb = 18.4073 ? 16, 207Pb/204Pb = 15.4714 ? 18, and 208Pb/204Pb = 38.0627 ? 60.  A3. Glass and matrix oxygen isotopes Glass and holocrystalline matrix material were separated from Pu?u ???? lavas by hand picking from coarsely crushed but otherwise untreated samples. Oxygen isotope compositions of c. 2 mg aliquots of these separates were determined by laser fluorination, using a 50W CO2 laser and BrF5 as reagent [Sharp, 1990; Valley et al., 1995]. Product O2 was converted to CO2 by reaction with hot graphite; CO2 was then analyzed for its isotopic composition by dual-inlet gas source mass spectrometry on a Thermo Finnegan Delta XL gas source isotope ratio mass spectrometer at California Institute of Technology. Data are reported in units of per mil versus the VSMOW standard. Analyses were standardized by comparison with measurements of Gore Mountain garnet standard [Valley et al., 1995]. This standard was analyzed between two and five times each day of analyses, and the data for unknowns analyzed on that day were corrected by the average difference between measured and accepted values for that standard. The external precision of repeat measurements of separate splits of unknown samples averaged 0.05? (1?). Seven analyses of Gore Mtn. Garnet were run concurrently with these samples, with a standard 5  deviation of ?0.06 ? (1?; averaged over all days).  This is comparable with the typical external precision for replicate measurements of silicate standards for this laboratory and technique [e.g., Bindeman et al., 2006; Eiler et al., 1995; Eiler et al., 1996], suggesting that the glass separates analyzed in this study are homogeneous in ?18O at the scale of c. 2 mg aliquots. The results reported here are relative to ?18OVSMOW [Coplen, 1988]. Four analyses of Menehune standard performed with the other Pu?u ???? lavas yielded 5.22 ? 0.03 ? (1?).  References  Bindeman, I. N., O. Sigmarsson, and J. M. Eiler (2006), Time constraints on the origin of large volume basalts derived from O-isotope and trace element mineral zoning and U-series disequilibria in the Laki and Grimsvotn volcanic system, Earth Planet. Sci. Lett., 245, 245-259. Carpentier, M., D. Weis, and C. Chauvel (2013), Large U loss during weathering of upper continental crust: The sedimentary record. , Chemical Geology (Isotope Geosciences Section), 340, 91-104. Coplen, T. B. (1988), Normalization of oxygen and hydrogen isotope data, Chemical Geology (Isotope Geosciences Section)  72, 293-297. Eiler, J. M., K. A. Farley, J. W. Valley, E. M. Stolper, E. H. Hauri, and H. Craig (1995), Oxygen isotope evidence against bulk recycled sediment in the mantle sources of Pitcairn Island lavas, Nature, 377, 138-141. Eiler, J. M., J. W. Valley, and E. M. Stolper (1996), Oxygen isotope ratios in olivine from the Hawaii Scientific Drilling Project, J. Geophys. Res., 101(B5), 11,807-11,813. Galer, S. J. G., and W. Abouchami (1998), Practical application of lead triple spiking for correction of instrumental mass discrimination, Mineral. Mag., 62A, 491-492. Marske, J. P., A. J. Pietruszka, D. Weis, M. O. Garcia, and J. M. Rhodes (2007), Rapid passage of a small-scale mantle heterogeneity through the melting regions of Kilauea and Mauna Loa volcanoes, Earth Planet. Sci. Lett., 259, 34-50. Marske, J. P., M. O. Garcia, A. J. Pietruszka, J. M. Rhodes, and M. D. Norman (2008), Geochemical variations during Kilauea's Pu'u 'O'o eruption reveal a fine-scale mixture of mantle heterogeneities within the Hawaiian plume, J. Petrol., 49(7), 1297-1318. Nobre Silva, I. G., D. Weis, J. Barling, and J. S. Scoates (2009), Leaching systematics and matrix elimination for the determination of high-precision Pb isotope compositions of ocean island basalts, Geochem. Geophys. Geosyst., 10(8), Q08012. Pretorius, W., D. Weis, G. Williams, D. Hanano, B. Kieffer, and J. S. Scoates (2006), Complete trace elemental characterization of granitoid (USGSG-2,GSP-2) reference materials by high resolution inductively coupled plasma-mass spectrometry, Geost. and Geoanaly. Res., 30(1), 39-54. Rhodes, J. M. (1988), Geochemistry of the 1984 Mauna Loa eruption: Implications for magma storage and supply, J. Geophys. Res., 93, 4,453-454,466. Rhodes, J. M., and M. J. Vollinger (2004), Composition of basaltic lavas sampled by phase-2 of the Hawaii Scientific Drilling Project: Geochemical stratigraphy and magma types, Geochem. Geophys. Geosyst., 5(3). 6  Sharp, Z. D. (1990), A laser-based microanalytical method for the in situ determination of oxygen isotope ratios of silicates and oxides, Geochim. Cosmochim. Acta, 54(5), 1353-1357. Valley, J. W., N. Kitchen, and M. J. Kohn (1995), Strategies for high-precision oxygen isotope analysis by laser fluorination, Geochimica Cosmoschimica Acta, 59, 5223-5231. Weis, D., B. Kieffer, C. Maerschalk, W. Pretorius, and J. Barling (2005), High-precision Pb-Sr-Nd-Hf isotopic characterization of USGS BHVO-1 and BHVO-2 reference materials, Geochem. Geophys. Geosyst., 6, (Q02002). Weis, D., B. Kieffer, C. Maerschalk, J. Barling, J. de Jong, G. A. Williams, D. Hanano, N. Mattielli, J. S. Scoates, A. Goolaerts, R. A. Friedman, and J. B. Mahoney (2006), High-precision isotopic characterization of USGS reference materials by TIMS and MC-ICP-MS, Geochem. Geophys. Geosyst., 7(Q08006). White, W. M., F. Albar?de, and P. T?louk (2000), High-precision analysis of Pb isotope ratios by multi-collector ICP-MS, Chem. Geol., 167, 257-270.     Temporal Geochemical Variations in Lavas from K?lauea?s Pu?u ???? Eruption (1983-2010): Cyclic Variations from Melting of Source Heterogeneities  Andrew R. Greene Dept. of Natural Sciences, Hawai?i Pacific University, K?ne`ohe, HI, 96744 USA  Michael O. Garcia  Dept. of Geology and Geophysics, University of Hawai?i, Honolulu, HI, 96822 USA  Aaron J. Pietruszka  Dept. of Geological Sciences, San Diego State University, San Diego, CA, 92182 USA   Dominique Weis Pacific Centre for Isotopic and Geochemical Research, Dept. of Earth and Ocean Sciences, University of British Columbia, Vancouver, BC, V6T 1Z4 Canada  Jared P. Marske Dept. of Terrestrial Magnetism, Carnegie Institution, Washington, D.C., 20015 USA   Michael J. Vollinger Ronald B. Gilmore XRF Lab, Dept. of Geosciences, University of Massachusetts, Amherst, MA, 01003 USA  John Eiler Planetary and Geological Sciences Institute, California Institute of Technology, Pasadena, CA, 91125 USA    Five supplementary files include descriptions of analytical methods, a full table of XRF major and trace element data, a full table of ICP-MS trace element data, a summary of XRF trace element values of USGS standards analyzed with Pu?u ???? lavas since 2002, and a list of references for data used in figures 11 and 12. The tables are Microsoft Excel data files. The name and contents of individual files are listed below.   2013G3Greene-methods  Sample Preparation and Analytical Methods   2013G3Greene-ts1-XRF_major_trace_elements  XRF Major Element (wt% oxide) and Trace Element (ppm) Abundances in Lavas from K?lauea?s Pu?u ???? Eruption (1983-2010)  2013G3Greene-ts2-ICP-MS_trace_elements  ICP-MS Trace Element (ppm) Abundances in Lavas from K?lauea?s Pu?u ???? Eruption (1983-2010)  2013G3Greene-ts3-XRF_trace_USGS-standards  Summary of XRF Trace Element Values of USGS Standards Analyzed with Lavas from K?lauea?s Pu?u ???? Eruption Since 2002  2013G3Greene-references-figs11-12  List of References for Data Used in Figures 11 and 12    Temporal geochemical variations in lavas fromK?lauea?s Pu?u ?O?o eruption (1983?2010): Cyclicvariations from melting of source heterogeneitiesAndrew R. GreeneDepartment of Natural Sciences, Hawai?i Pacific University, 45-045 Kamehameha Hwy, Kane?ohe, Hawaii,96744, USA (agreene@hpu.edu)Michael O. GarciaDepartment of Geology and Geophysics, University of Hawai?i, Honolulu, Hawaii, USAAaron J. PietruszkaDepartment of Geological Sciences, San Diego State University, San Diego, California, USANow at U. S. Geological Survey, Denver Federal Center, Denver, Colorado, USADominique WeisPacific Centre for Isotopic and Geochemical Research, Department of Earth and Ocean Sciences, University ofBritish Columbia, Vancouver, British Columbia, CanadaJared P. MarskeDepartment of Terrestrial Magnetism, Carnegie Institution, Washington, D.C., USAMichael J. VollingerRonald B. Gilmore XRF Lab, Department of Geosciences, University of Massachusetts, Amherst, Massachusetts, USAJohn EilerPlanetary and Geological Sciences Institute, California Institute of Technology, Pasadena, California, USA[1] Geochemical time series analysis of lavas from K?lauea?s ongoing Pu?u ?O?o eruption chroniclemantle and crustal processes during a single, prolonged (1983 to present) magmatic event, which hasshown nearly two-fold variation in lava effusion rates. Here we present an update of our ongoingmonitoring of the geochemical variations of Pu?u ?O?o lavas for the entire eruption through 2010. Oxygenisotope measurements on Pu?u ?O?o lavas show a remarkable range (18O values of 4.6?5.6%), which areinterpreted to reflect moderate levels of oxygen isotope exchange with or crustal contamination byhydrothermally altered K?lauea lavas, probably in the shallow reservoir under the Pu?u ?O?o vent. Thisprocess has not measurably affected ratios of radiogenic isotope or incompatible trace elements, whichare thought to vary due to mantle-derived changes in the composition of the parental magma delivered tothe volcano. High-precision Pb and Sr isotopic measurements were performed on lavas erupted at 6month intervals since 1983 to provide insights about melting dynamics and the compositional structure ofthe Hawaiian plume. The new results show systematic variations of Pb and Sr isotope ratios thatcontinued the long-term compositional trend for K?lauea until 1990. Afterward, Pb isotope ratios showtwo cycles with 10 year periods, whereas the Sr isotope ratios continued to increase until 2003 andthen shifted toward slightly less radiogenic values. The short-term periodicity of Pb isotope ratios mayreflect melt extraction from mantle with a fine-scale pattern of repeating source heterogeneities or strands,which are about 1?3 km in diameter. Over the last 30 years, Pu?u ?O?o lavas show 15% and 25% of the? 2013. American Geophysical Union. All Rights Reserved. 4849ArticleVolume 14, Number 1115 November 2013doi: 10.1002/ggge.20285ISSN: 1525-2027known isotopic variation for K?lauea and Mauna Kea, respectively. This observation illustrates that thedominant time scale of mantle-derived compositional variation for Hawaiian lavas is years to decades.Components: 13,235 words, 12 figures, 2 tables.Keywords: Hawaiian plume; tholeiitic volcanism; melt extraction; oceanic island.Index Terms: 3619 Magma genesis and partial melting: Mineralogy and Petrology; 3621 Mantle processes: Mineralogyand Petrology; 1025 Composition of the mantle: Geochemistry; 1037 Magma genesis and partial melting: Geochemistry;1038 Mantle processes: Geochemistry.Received 5 March 2013; Revised 9 September 2013; Accepted 7 October 2013; Published 15 November 2013.Greene, A. R., M. O. Garcia, A. J. Pietruszka, D. Weis, J. P. Marske, M. J. Vollinger, and J. Eiler (2013), Temporal geo-chemical variations in lavas from K?lauea?s Pu?u ?O?o eruption (1983?2010): Cyclic variations from melting of source het-erogeneities, Geochem. Geophys. Geosyst., 14, 4849?4873, doi:10.1002/ggge.20285.1. Introduction[2] K?lauea, on the Island of Hawai?i (Figure 1), isone of the most active and best-monitored volca-noes in the world [Heliker and Mattox, 2003;Wolfe et al., 1987]. The ongoing Pu?u ?O?o erup-tion on K?lauea?s East Rift Zone (Figure 1) hasbeen active nearly continuously for 30 years and isHawai?i?s longest and most voluminous (4 km3)historical eruption [Poland et al., 2012]. The con-tinuous petrologic and geochemical monitoring ofthe Pu?u ?O?o eruption [e.g., Garcia et al., 2000;Marske et al., 2008; Thornber, 2003] has allowedus to witness the dynamic changes in the meltingprocess and mantle source composition during asingle, long-lasting magmatic event. Extractionand transport of melt through open channels dur-ing the Pu?u ?O?o eruption has efficiently transmit-ted variations of melting in the heterogeneoussource to lavas erupted at the surface without sig-nificant pooling and homogenization, preservingshort-term isotopic and geochemical variations[Pietruszka et al., 2006].[3] The long-term geochemical variations (manythousands of years) of Hawaiian and other oceanisland basalts has been well documented due todetailed geochemical work on 3? km deep drillcore [e.g., Albare`de et al., 1997; Blichert-Toftet al., 2003; Bryce et al., 2005; Caroff et al.,1995; Rhodes et al., 2012]. These studieschronicle processes on millennium time scales butmiss potential short-term variations (<100 years),which may provide better insights into meltingand crustal processes. K?lauea?s historical (1823?1982) and prehistoric (AD 900?1400) summitlavas reveal rapid and systematic changes in Pb,Sr, Nd, O, and U-series isotope ratios on a timescale of decades to centuries [Garcia et al., 2003,2008; Marske et al., 2007; Pietruszka and Garcia,1999; Pietruszka et al., 2001]. The Pu?u ?O?oeruption (sampled from hourly to monthly) showscompositional change over hours (in rare cases)for major elements to a few years for isotope ratios[Garcia et al., 2000; Marske et al., 2008]. Thelong duration and vigorous activity (0.35  106m3 of lava erupted daily) of Pu?u ?O?o [e.g., Suttonet al., 2003] provides a rare opportunity to lookbeyond the shallow-level crustal processes associ-ated with the short eruptions (days to weeks) thattypify many active basaltic volcanoes (e.g., MaunaLoa, Etna, Piton de la Fournaise, Karthala,Grimsv?tn) and into the mantle. In addition, Pu?u?O?o magmas may partially bypass K?lauea?s sum-mit reservoir (2?6 km depth beneath the summitcaldera) on their way to the East Rift Zone, andmostly avoid its buffering effects [Garcia et al.,2000]. Therefore the Pu?u ?O?o eruption is one ofEarth?s best probes for sampling mantle-derivedmelts almost continuously over nearly threedecades.[4] The study of isotopic and geochemical varia-tion in magmatic events over short time scales(months to years) in oceanic island lavas improvesour temporal and spatial resolution of meltingprocesses and the chemical structure of mantleplumes [Abouchami et al., 2000; Eisele et al.,2003; Hofmann et al., 1984; Vlastelic et al.,2005]. Recent studies of Pb, Sr, and Nd isotoperatios for part of the Pu?u ?O?o eruption [Marskeet al., 2008] and other active basaltic volcanoes[e.g., Piton de la Fournaise; Vlastelic et al., 2005]detected rapid and systematic changes over shortGREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854850time scales (years to decades) resulting fromsource heterogeneity, and variations in crustalprocesses. Here we present new high-precision Pb,Sr, and O isotope ratios, and major- and trace-element abundances for Pu?u ?O?o lavas eruptedbetween January 1983 and June 2010. These data197418231919-20 1971Mauna Ulu   1969-74197719551955196018401790East Rift ZoneSummitMakaopuhi1983-present1790179010 km18681955Pu?u ?O?oNCraterCraterEp. 54Southwest Rift ZoneMauna Loa1983-2010 Pu'u '?'? KupaianahaPACIFIC OCEANNorth Pu'u '?'?N?pauCraterKalapanaEp. 56 February 17, 1992 - Feb. 11, 2010Episodes 50-58July 20, 1986 -  February 17, 1992Episodes 48-49January 3, 1983 -July 20, 1986Episodes 1-48episodicfountaining(mostly centralvent)gentle effusion(lava shield andtube-fed pahoehoe)Kupaianaha Pu?u ?O?oPu?u ?O?oJan. July Feb.Feb.1983 1986 19921997(3.5 years) (5.5 years) (18 years) perched channels,rootless shields,fissure eruption20102007Episode 56(Magma supply ratedoubled)June(uprift)flank  vent eruptions(nearly continuous)(lava shield and tube-fed pahoehoe)2003Episode 54(uprift)010km016miK?lauea Caldera1790-1982kmHalemaumauPu?u ???? East Rift ZoneMauna Ulu1969-1974Makaopuhi N?pau(a)KupaianahaEp. 54Ep. 56(b)123storeddikeJune, 2007?Jan.1997Ep. 58Ep. 58K?lauea Caldera0KohalaMauna Loa K?lauea Mauna KeaL??ihi 50010001500200025003000200015005001000100020003000250050002500 50005500155?156?19?20?Hilo RidgeEast rift zone40 km20Hawai?i Hual?lai(c)(d)Figure 1. Map of flow fields from the Pu?u ?O?o-Kupaianaha eruption on the East Rift Zone of K?lauea Vol-cano from 1983 to 2010 and historical flows, with a timeline summarizing the predominant style of eruptiveactivity. (a) A schematic cross section of summit and East Rift Zone shows the proposed magmatic plumbingsystem for K?lauea Volcano, with locations for episodes 54 and 56 uprift of Pu?u ?O?o. Mantle-derived magmafor this eruption is thought to partially bypass the summit reservoir based on the rapid changes in lava compo-sition [Garcia et al., 1996]. (b) Map of K?lauea East Rift Zone with flow fields from intervals of the Pu?u?O?o eruption. Legend shows episodes in each interval of eruptive activity. Map provided by USGS HawaiianVolcano Observatory. (c) Map of the island of Hawai?i with area of map in Figure 1b indicated with box. (d)Timeline of the Pu?u ?O?o eruption. Episode 54 was a fissure eruption in and downrift of Napau Crater thatoccurred over 23 h in January 1997, following the collapse of the Pu?u ?O?o cone. Episode 56 was a brief (<1day) fissure eruption northeast of Makaopuhi Crater (uprift of Pu?u ?O?o) that occurred in June 2007, coincid-ing with an intrusion and collapse of Pu?u ?O?o crater floor. Dashed lines between 2003 and 2007 indicate pe-riod when magma supply rate nearly doubled [Poland et al., 2012].GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854851are combined with previously published high-precision isotope and trace-element data from1998 to 2005 Pu?u ?O?o lavas [Marske et al.,2008]. The new Pb and Sr isotope and inductivelycoupled plasma mass spectrometry (ICP-MS) dataprovide a record of isotopic and geochemical vari-ation of Pu?u ?O?o lavas at 6 month intervals,whereas X-Ray fluorescence (XRF) data was col-lected at 2 week intervals. This time series anal-ysis of Pu?u ?O?o lavas allows us to distinguish thechanging roles of mantle and crustal processes ingreat detail. The new Pb and Sr isotope ratios areused to assess the short-term expression of mantlesource components throughout the course of theeruption and to evaluate the effects on lava com-position of recent doubling of the magma supply[2003?2007; Poland et al., 2012]. These resultsare compared to the longer-term variations forK?lauea and other Hawaiian shield volcanoes.2. Geologic Background of K?laueaVolcano and the Pu?u ?O?o Eruption(1983?2010)[5] K?lauea Volcano is currently in the middle ofits shield-building stage [DePaolo and Stolper,1996], erupting tholeiitic lava at a rate of 0.13km3/yr [Sutton et al., 2003], one of the highestrates of any volcano on Earth. K?lauea rises 1240m above sea-level on the southern flank of itslarger neighbor, Mauna Loa (4168 m; Figure 1).Geochemical evidence favors a deep mantle plumeorigin for Hawaiian magmas [e.g., Kurz et al.,1982; Weis et al., 2011]. Shield stage magmas arethought to originate from partial melting at mantledepths of 70?120 km within the upper Hawaiianplume [Watson and McKenzie, 1991]. Magmas areextracted from the upwelling mantle within themelting region and transported through chemicallyisolated channels towards the surface [Pietruszkaet al., 2006]. These pooled melts ascend throughthe lithosphere via a primary conduit into a shal-low (2?6 km) magmatic complex within K?lauea[Eaton and Murata, 1960; Ryan, 1987; Tillingand Dvorak, 1993; Wright, 1971]. K?lauea erup-tions occur in and around its summit caldera andEast and Southwest Rift Zones. Approximately90% of the subaerial surface of K?lauea Volcano iscovered with tholeiitic lava less than 1100 yearsold [Holcomb, 1987]. Prior to 1955, historical(post-1820) eruptions on K?lauea occurred mostlyat or near the summit [Macdonald et al., 1983].Subsequently, rift zone eruptions became morecommon, especially along the East Rift Zone,including the 1969?1974 Mauna Ulu eruption, themost voluminous historical eruption prior to Pu?u?O?o [Macdonald et al., 1983].[6] The Pu?u ?O?o-Kupaianaha eruption (referredto as the Pu?u ?O?o eruption throughout this paper)began on 2 January 1983 with the intrusion of adike within K?lauea?s East Rift Zone, although itwas preceded by months of intrusions from thesummit into the rift zone [Wolfe et al., 1987]. Itwas followed 24 h later by eruptive activity alonga discontinuous 7 km long fissure, which localizedto a central vent, Pu?u ?O?o (Figure 1 and Table 1).The eruption can be categorized into three broadphases based on eruptive style and location: (1)1983?1986: brief (mostly less than 24 h), episodiceruptions (24 day average repose between erup-tions) with fountaining up to 400 m, mainly fromthe Pu?u ?O?o vent [Heliker and Mattox, 2003]; (2)1986?1992: nearly continuous effusion from theKupaianaha vent, which was considered to have ashallow (<100 m deep) conduit connection withPu?u ?O?o, 3 km uprift [Garcia et al., 1996]; and(3) 1992?2010: nearly continuous effusion mostlyfrom vents within, and on the southwest and eastflanks of Pu?u ?O?o, and from rootless shields 2km east of Pu?u ?O?o [Poland et al., 2008]. Thispattern was interrupted on 29 January 1997 (epi-sode 54) by the 150 m collapse of the crater floorinside the Pu?u ?O?o cone, and propagation oferuptive fissures 4 km uprift (west) of Pu?u ?O?o,which were active for less than a day [Heliker andMattox, 2003]. This event was followed by a 6week hiatus in effusive activity, although glowreturned to the Pu?u ?O?o vent on 24 February1997 (Table 1). Afterward, and until June 2007,lava erupted nearly continuously from flank ventson Pu?u ?O?o (episode 55). On 19 June 2007, adike intrusion in the upper East Rift Zone resultedin a brief, small (1500 m3) eruption (episode 56)6 km uprift from Pu?u ?O?o [Montgomery-Brownet al., 2010], which was followed by a 2 week hia-tus in effusion [Poland et al., 2008]. Lava produc-tion resumed for 3 weeks in and around Pu?u ?O?ocone (episode 57) until 21 July 2007, when a fis-sure opened on the east flank of Pu?u ?O?o andpropagated eastward towards Kupaianaha (Figure1 and Table 1). This marked the beginning of epi-sode 58 [Poland et al., 2008], which continuedthrough the end of 2010 mostly as tube-fed flowsfrom a vent 2 km east of Pu?u ?O?o. The othernotable K?lauea eruptive activity during the Pu?u?O?o eruption is an ongoing summit eruption thatstarted in March 2008 [Johnson et al., 2010].GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854852Table1.SummaryofthePu?u? O? oEruptionaPrimaryVentEpisodeEpisodeStartDateReposeLength(Days)EpisodeLengthVolume(106xm3)EruptionRate(103m3/day)Vent/LocationBriefDescriptionPu?u? O? o13Jan1983Start20days14?Fissure1;activitylocalizedatPu?uHaluluandPu?uKahauleaeastofPu?u? O? oInitialfissureopenedinNapauCraterafterseismicswarmpropagateddownERZ;fissuresextended8km;fissureslocalizedto1kmnearPu?uKahaulea;fountainsfromPu?uHalulubuilta60m-highconePu?u? O? o2?4710Feb19838?65(betweenepisodes)~3.8years371300MostlyPu?u? O? o;Episodes2?3localizedatPu?uHaluluandPu?uKahauleaeastofPu?u? O? o;Epi-sode4?47PuuOoprimaryventEpisodicfirefountaining;episodesmostly<24hlongsepa-ratedbyanreposelengthaverageof24days;effusionratesincreasedthoughepisode39;maximumlavafoun-tainof470mhigh;firstyearchangedfromlowfountainsandpahoehoeriverstohighfountainsand?a?afans;fountain-fed?a?abyepisode20;conebuilt255mhighand1.4kmindiameter;summitinflatedbetweenfoun-tainingepisodesanddeflatedduringepisodesKupaianaha4818July198624~5.5years500400?0.5Kupaianaha;fissure3kmeastofPu?u? O? oFissuresfirstopenedatthebaseofPu?u? O? oand22hlateropened3kmdownriftataventtobenamedKupaianaha;5.5yearsofnearlycontinuousgentleeffusion;largelavapondformedovervent(140m300m);broadlavashieldformedandtube-fedpahoehoewascommonwaylavaspreadtocoast;homesdestroyedintownofKala-pana;lavatubestoseamid-1987to1989;lavaenteredseaduring68%ofepisode;lavaactiveinPu?u? O? ocraterduringmostofepisodePu?u? O? o498Nov1991None18days110.6Fissure2betweenPu?u? O? oandKupaianahaFissuresopenedonPu?u? O? oandpropagatedtoKupaia-naha;outputwanedduringepisode;gentleeffusion,lavashieldandtube-fedpahoehoe;fissurevents,pahoehoePu?u? O? o5017Feb19921115days3?Pu?u? O? oflank;radialfissureonwestflankofPu?u? O? oconeEruptionreturnedtoPu?u? O? o;radialfissuresonflankofcone;flankventeruptions;lavashieldbankedupagainstthesouthandwestPu?u? O? ocone;spatterconesformedovervents;mostlytube-fedpahoehoe;continuousquieteffusionPu?u? O? o517Mar19924161days32300Pu?u? O? oflankFlankventeruptions;mostlytube-fedpahoehoetothesea;continuousquieteffusion;lavashieldbankedupagainstthesouthandwestPu?u? O? oconePu?u? O? o523Oct1992None15days2300Pu?u? O? oflankFlankventeruptions;mostlytube-fedpahoehoetothesea;continuousquieteffusion;lavashieldbankedupagainstthesouthandwestPu?u? O? oconePu?u? O? o5320Feb1993None~4years535300Pu?u? O? oflankFlankventeruptions;mostlytube-fedpahoehoetothesea;continuousquieteffusion;lavashieldbankedupagainstthesouthandwestPu?u? O? ocone;collapsepitsformedonthesideofPu?u? O? oPu?u? O? o5429Jan1997None1day0.30.3Fissure3;2?4kmupriftofPu?u? O? o(1)LavalakeinsidethePu?u? O? oventdrainedandcraterfloordropped150m;(2)Pu?u? O? owestflankcollapsed;115mgapinwestsideofPu?u? O? o;(3)fissure4kmeruptedupriftfor1day,inanddownriftofNapauCrater,followedbylongesteruptivehiatussince1987(24days);distinctlavachemistryinvolvedmagmamixingwithdifferentiatedmagmastoredinriftzoneGREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854853Table1.(continued)PrimaryVentEpisodeEpisodeStartDateReposeLength(Days)EpisodeLengthVolume(106xm3)EruptionRate(103m3/day)Vent/LocationBriefDescriptionPu?u? O? o5524Feb19972410years265200?500Pu?u? O? oanditsflankLavaspilledfromcratertoformnewpond;lavaspilledfromcrateramonthlater;newflankventeruptionswestandsouthwestofcone;spatterconesonflankscrustedovertoproducemostlytube-fedpahoehoe;byJune1997lavaovertoppedthegapinwallofPu?u? O? oandflowedfromcraterforfirsttimein11years;flankventsunder-minedPu?u? O? oconeinDecember,1997;PukaNuicol-lapsepitformedonsouthwestflankofcone;31pausesoccurredduringepisode55Pu?u? O? o5619June2007None6h0.001450.00036250mlongfissureintheforestnortheastofKaneNuioHamo,approximately6kmwestofPu?u? O? oFather?sDayeruptionnearKaneNuioHamonorthofMakaopuhiCrater;magmasupplytoPu?u? O? owascutoffon17June2007;earthquakeswarmsindicatedmagmamovementintheupperERZ;spattereruptedfromfissureinforestedarea;smalllavaflow(200m50m)accom-paniedintrusioninERZ;craterfloorinPu?u? O? ocol-lapsedanderuptionshutoffPu?u? O? o571July200719daysNone0.82?1.2365Pu?u? O? ocraterAfterabouttwo-weeksofquiet,theeruptionbeganagainon1July.Lavabegantorefillthecrater.On8July,effu-sionwanedasthecraterbegantoupliftinapiston-likefashion.Thecraterthenbegantofillandreachedtowithin30moftheeasternrimofPu?u? O? ocraterbymid-July.Pu?u? O? o5821July2007None~4yearsended7March2011320(asoftheendof2009)FromfissureeastofPu?u? O? ocraterPerchedlavachannel,rootlessshields;forthefirsttimesince7February1992,lavabeginseruptingeastofPu?u? O? ocrater.ThanksgivingEvebreakout,lavabypasses21July2007channelanderuptsonchannelflank;5March2008oceanentryactiveforthefirsttimesinceJune2007;explosioninHalema?uma?uCrateratsummiton19March2008;June2008spatteringventsandasmallpondoflavainPu?u? O? o,lavafountainsgushfromtheTEBtubesystem,channelized?a? aflowsinRoyalGardens,andlargelittoralexplosionsatK ?lauea?soceanentrynearKalapana;Waikupanahaoceanentryactivethroughmuchof2009,andoccasionallyKupapa?uoceanentrytowesta Reposelengthreferstodurationofpausebetweeneruptiveepisodes.Episodeidentifiesoccurrencesoffountainingorlavaflowseparatedbyquiescentperiods.Volumeisdenserockequivalent(DRE)eruptedduringeachepisode.Datasources:Garciaetal.[2000]andreferencestherein,Wolfeetal.[1998],HelikerandMattox[2003].GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.2028548543. Description of Samples andAnalyses Performed in This Study[7] This study presents 52 new high-precision Pband Sr isotope analyses (from 1983 to 1997 and2006 to 2010), 11 new O isotope analyses (fromafter 1997; Table 2), and 13 new ICP-MS trace-element analyses of Pu?u ?O?o lava samples(mostly after 2005; data and analytical methodsare presented in the supporting information).1 NewXRF major- and trace-element analyses for 52Pu?u ?O?o lavas erupted from 2006 to 2010 arealso presented. In addition, new XRF trace-element analyses are given for samples eruptedprior to 1998, when a new, more precise XRFinstrument became available. Almost all of thesamples in this study were collected in a moltenstate and quenched with water to minimize poster-uption crystallization. The sample names are thedate that each lava sample was collected (e.g.,day-month-year), which is generally within a dayof its eruption when lava is flowing in open chan-nels on the surface or in lava tubes [e.g., Garciaet al., 2000] or up to a week or more when it isoozing within slowly advancing pahoehoe flows[K. Ho, personal communication, 2013]. Descrip-tions of the petrography of typical Pu?u ?O?o lavascan be found in Garcia et al. [1989, 1992, 1996,2000] and Marske et al. [2008]. Fourteen high-precision Pb and Sr isotope ratios for Pu?u ?O?olavas erupted from 1998 to 2005 from Marskeet al. [2008] and 15 O isotope analyses from Gar-cia et al. [1998] are listed in Table 2 and areincluded in plots for completeness.4. Temporal Geochemical Variationsin Lavas From 1983 to 2010[9] Early Pu?u ?O?o lavas (1983 to early 1985) re-cord rapid (hours to days) variations in major andcompatible trace-element abundances (Figure 2;Supporting information). These lavas show petro-graphic evidence for both crystal fractionation andmagma mixing [Wolfe et al., 1987; Garcia et al.,1992]. Crystal fractionation of olivine (with minorclinopyroxene and plagioclase, especially for 1983lavas) is the dominant process controlling short-term major-element variation in Pu?u ?O?o lavas[Garcia et al., 1992]. To remove the effects ofcrystal fractionation on parental magma composi-tions, major-element abundances of lavas contain-ing only olivine (MgO >7.0 wt %) werenormalized to 10 wt % MgO by the addition ofequilibrium composition olivine (98.5%) and spi-nel (1.5%) in 0.5 mol % steps, as described byGarcia et al. [2003]. The increases in MgO, CaO/Al2O3, and CaO/TiO2 and decreases in MgO-normalized incompatible element abundances(e.g., TiO2, K2O) between 1983 and early 1985(Figure 2) reflect mixing of new high-MgOmagma with decreasing amounts of a hybridmagma formed at the start of the eruption by mix-ing two differentiated, rift-zone stored magmas[Garcia et al., 1989, 1992]. Lavas erupted afterearly 1985 show no petrographic or geochemicalevidence for mixing [Shamberger and Garcia,2007] until the 1997 uprift eruption, which is dis-cussed below.[10] From 1985 to 1994, Pu?u ?O?o lavas show awide range in MgO reflecting the periodic hiatusesin eruptive activity [Garcia et al., 1992], and grad-ual changes in MgO-normalized concentrations ofmajor elements (TiO2 and K2O), and ratios ofmajor (e.g., CaO/Al2O3; Figure 2) and trace ele-ments (Nb/Y; Figure 3). In 1994, lavas began aperiod of increasing MgO-normalized SiO2 andoverall decreasing MgO-normalized TiO2 that per-sisted until 2001. Other geochemical parameterscontinued their long-term trends (e.g., decreasingCaO/Al2O3, MgO-normalized K2O, and Nb/Y,and increasing Zr/Nb; Figures 2 and 3). Starting inmid- to late 2003, there was an increase in lavaproduction with effusion rates doubling in 2005[Poland et al., 2012]. The lava MgO contentdecreased from 2003 to 2007 and was relativelylow (<7.5 wt %, mostly <7.2 wt %) with limitedvariation (Figure 2). This decrease continued theoverall trend of decreasing MgO that started in1998, as noted by Poland et al. [2012]. There isalso a decrease in MgO-normalized SiO2 and anincrease in MgO-normalized TiO2 and K2O during2003?2007 (Figures 2 and 3). Lava MgO increasedfrom 2008 to 2009 as did CaO/TiO2 and values ofMgO-normalized SiO2, although MgO and SiO2values dropped afterwards for the most recentlyerupted samples that were analyzed in this study(Figure 2). For more on major- and trace-elementvariations in 1983?2005 Pu?u ?O?o lavas, see Gar-cia et al. [1989, 1992, 1996, 2000], Marske et al.[2008], and Thornber [2003].[11] The brief eruptive outbreaks uprift of the Pu?u?O?o vent in 1997 (3 km uprift for episode 54)and 2007 (6 km uprift for episode 56; Figure 1)1Additional supporting information may be found in the onlineversion of this article.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854855Table 2. Pb, Sr and O Isotopic Geochemistry of Pu?u ?O?o Lavas from 1983?2010aSample 206Pb/204Pb 2 207Pb/204Pb 2 208Pb/204Pb 2 87Sr/86Sr 2 18O 123 Jan 1983  18.5247 0.0009 15.4800 0.0008 38.155 0.0020 0.703590 0.000009 4.56 0.029 Apr 1983 18.5309 0.0007 15.4893 0.0007 38.165 0.0017 0.703573 0.0000083 July 1983  18.4780 0.0007 15.4765 0.0008 38.117 0.0017 0.703573 0.000007 4.77 0.0331 Jan 1984 18.4595 0.0009 15.4765 0.0007 38.103 0.0020 0.703587 0.00000812 Sep 1984 18.4417 0.0008 15.4747 0.0007 38.091 0.0019 0.703555 0.0000098 Feb 1985  18.4342 0.0010 15.4743 0.0009 38.087 0.0024 0.703567 0.000008 4.76 0.0521 Apr 1985  4.82 0.1030 Jul 1985 18.4306 0.0007 15.4756 0.0007 38.089 0.0020 0.703571 0.0000082 Jun 1986  18.4138 0.0009 15.4755 0.0008 38.079 0.0020 0.703580 0.000009 4.94 0.3026 Jun 1986  4.77 0.0213 Sep 1986  18.4138 0.0008 15.4726 0.0007 38.074 0.0020 0.703590 0.000008 5.17 0.0116 Mar 1987  18.4108 0.0006 15.4717 0.0005 38.069 0.0014 0.703589 0.000009 5.25 0.0218 Oct 1987 18.3992 0.0008 15.4746 0.0008 38.068 0.0022 0.703597 0.00000919 Jan 1988 18.3952 0.0006 15.4715 0.0006 38.061 0.0016 0.703577 0.00000818 Aug 1988 18.3871 0.0008 15.4703 0.0007 38.052 0.0017 0.703583 0.00000926 Mar 1989  18.3882 0.0007 15.4717 0.0007 38.055 0.0019 0.703584 0.000009 5.11 0.057 Jul 1989 18.3861 0.0010 15.4725 0.0009 38.054 0.0022 0.703581 0.0000097 Jul 1989 ? 18.3851 0.0009 15.4710 0.0008 38.052 0.0021 0.703576 0.0000097 Jan 1990  18.3881 0.0009 15.4745 0.0008 38.056 0.0018 0.703584 0.000008 5.03 0.0127 May 1990 18.3864 0.0008 15.4716 0.0007 38.054 0.0020 0.703603 0.00000721 Oct 1990 18.3856 0.0007 15.4693 0.0006 38.049 0.0017 0.703597 0.00000812 May 1991  18.3992 0.0009 15.4739 0.0007 38.063 0.0019 0.703581 0.000008 5.08 0.041 Aug 1991  5.11 0.046 Jun 1992  18.4048 0.0009 15.4721 0.0009 38.061 0.0019 0.703585 0.000008 5.04 0.1013 Aug 1993  18.4112 0.0007 15.4752 0.0006 38.072 0.0015 0.703611 0.000008 4.98 0.074 Jan 1994 18.4098 0.0009 15.4736 0.0008 38.067 0.0020 0.703607 0.00000825 Apr 1994  18.4100 0.0009 15.4737 0.0009 38.068 0.0027 0.703586 0.000008 5.01 0.029 Oct 1994 18.4059 0.0008 15.4718 0.0007 38.066 0.0019 0.703598 0.00000727 Apr 1995  18.4059 0.0009 15.4721 0.0008 38.066 0.0023 0.703604 0.000009 5.25 0.0514 Oct 1995 18.4068 0.0008 15.4729 0.0008 38.071 0.0021 0.703602 0.00000919 Jan 1996  5.19 0.0715 Mar 1996 18.4064 0.0010 15.4738 0.0008 38.070 0.0023 0.703592 0.00000922 Aug 1996 18.4038 0.0009 15.4722 0.0008 38.065 0.0016 0.703612 0.00000710 Jan 1997  18.4010 0.0012 15.4728 0.0011 38.064 0.0019 0.703606 0.000009 5.2 0.0523 Jul 1997 18.3993 0.0010 15.4729 0.0010 38.068 0.0025 0.703601 0.00000710 Jan 1998 18.3958 0.0007 15.4728 0.0006 38.067 0.0014 0.703591 0.00000810 Jan 1998 ? 18.3940 0.0007 15.4711 0.0006 38.063 0.0016 0.703593 0.00000711 May 1998 18.4005 0.0009 15.4740 0.0008 38.071 0.0020 0.703605 0.0000107 Sep 1998 18.4082 0.0008 15.4775 0.0006 38.083 0.0017 0.703601 0.000006 5.33 0.067 Sep 1998 ? 18.4107 0.0004 15.4727 0.0005 38.075 0.0012 5.29 0.0813 Feb 1999 18.4068 0.0010 15.4783 0.0008 38.085 0.0021 0.703607 0.00000613 Feb 1999 ? 18.4124 0.0004 15.4736 0.0004 38.076 0.001119 Jun 1999 18.3987 0.0010 15.4805 0.0007 38.085 0.0020 0.703620 0.00000927 Oct 1999  18.4018 0.0004 15.4726 0.0004 38.069 0.0011 0.703622 0.000009 5.36 0.0819 Feb 2000  18.4072 0.0004 15.4712 0.0004 38.072 0.0011 0.703624 0.00000721 Jun 2000  18.4067 0.0004 15.4704 0.0004 38.069 0.0011 0.703638 0.000007 5.28 0.088 Jan 2001  18.4116 0.0004 15.4721 0.0004 38.074 0.0011 0.703627 0.000012 5.31 0.087 Jul 2001  18.4137 0.0004 15.4719 0.0004 38.073 0.0013 0.703626 0.0000099 Feb 2002  18.4139 0.0004 15.4707 0.0004 38.069 0.0011 0.703637 0.00000820 Aug 2002  18.4152 0.0004 15.4722 0.0004 38.072 0.0011 0.703639 0.00000512 Apr 2003  18.4161 0.0005 15.4726 0.0005 38.072 0.0013 0.703641 0.000005 5.31 0.0815 Jan 2004  18.4154 0.0005 15.4719 0.0005 38.069 0.0012 0.703632 0.000007 5.21 0.087 Jun 2004  18.4146 0.0003 15.4716 0.0004 38.068 0.0010 0.703624 0.00000731 Jan 2005  18.4170 0.0005 15.4735 0.0006 38.075 0.0012 0.703624 0.000005 4.96 0.138 Aug 2005  18.4119 0.0005 15.4727 0.0005 38.070 0.0013 0.703622 0.00001029 Jan 2006  18.4087 0.0004 15.4720 0.0004 38.065 0.0012 0.703623 0.00000924 Jun 2006 ? 18.4062 0.0028 15.4724 0.0026 38.066 0.0059 0.703612 0.000008 5.23 0.0324 Jun 2006  18.4073 0.0016 15.4714 0.0018 38.063 0.0060 0.703617 0.000013 5.23 0.036 Apr 2007  18.4065 0.0003 15.4715 0.0003 38.063 0.0009 0.703617 0.000007 5.35 0.1317 Jun 2007  18.4019 0.0004 15.4709 0.0004 38.062 0.0012 0.703626 0.000006 5.63 0.1322 Mar 2008  18.4038 0.0003 15.4700 0.0003 38.061 0.0010 0.703607 0.000008 5.45 0.132 May 2008  18.4045 0.0003 15.4721 0.0004 38.066 0.0010 0.703609 0.00000915 Nov 2008 18.3972 0.0009 15.4704 0.0007 38.058 0.0024 0.703600 0.00000829 Jan 2009  18.4003 0.0005 15.4709 0.0006 38.061 0.0012 0.703628 0.0000087 May 2009 18.4005 0.0008 15.4736 0.0007 38.066 0.0020 0.703624 0.0000104 Jun 2009  18.4009 0.0003 15.4720 0.0004 38.064 0.0010 0.703610 0.00000716 Oct 2009  18.3994 0.0005 15.4714 0.0006 38.062 0.0012 0.703622 0.00000722 Jan 2010  18.3987 0.0005 15.4708 0.0005 38.060 0.0013 0.703617 0.000007a indicates analysis at San Diego State University (SDSU), analyses from 7 Sep 1998 to 8 Aug 2005 are from Marske et al. [2008]. Sr isotopeanalyses from 1983?1997 and 2006?2010 were performed at PCIGR. ? Chemistry duplicate.  Published 18O analyses from Garcia et al. [1998].Analytical methods are described in the supporting information. 24 Jun 2006 is an in-house glass standard called Menehune collected from a Pu?u?O?o lava flow (errors are the external 62s of the replicate analyses; average of four analyses for Pb and Sr at PCIGR; 68 for Pb and 26 for Sr atSDSU). US Geological Survey sample numbers for lavas between up to16 Mar 87 are 23 Jan 1983: 1?054, 9 April 1983: 3?117, 3 Jul 1983: 5?139, 31 Jan 1984: 14?232, 12 Sep 1984: KE24?25 310S, 8 Feb 1985: 30?362, 30 Jul 1985: 35?419, 1 Jan 1986:40?484, 2 Jun 1986: 46?536, 13Sep 1986: 48?649, 16 Mar 1987: 48?714F.occurred after major collapses of the Pu?u ?O?ocrater floor (Table 1). The lavas erupted from theseuprift vents were geochemically distinct. Com-pared to coeval Pu?u ?O?o vent lavas, those fromepisode 54 have lower MgO (5.6?6.4 versus 7.5?10.1 wt %), CaO/TiO2 (2.8?3.4 versus 4.4), Sr/Nband Zr/Nb ratios (Figure 3). These geochemicalsignatures and the petrographic evidence of56789100.700.720.740.760.780.800.820.840.8648.849.049.249.449.649.850.050.250.43.03.23.43.63.84.04.24.44.64.82.12.22.32.42.52.60.350.400.450.500.551983198419851986198719881989199 0199119921993199419951996199719981999200020012 00220032004200520062007200820092010198319841985198619871988198919901991199219931994199519961997199 819992000200120022003200420052006200720082009201084 86 88 90 92 94 96 98 00 02 04 06 08 1084 86 88 90 92 94 96 98 00 02 04 06 08 1084 86 88 90 92 94 96 98 00 02 04 06 08 10 84 86 88 90 92 94 96 98 00 02 04 06 08 10Ep. 54TiO2 (wt%)normalized to 10 wt% MgO(Magma supply ratedoubled)2003-071983-19861986-19921992-2010MgO (wt%)CaO/Al2O3 (wt%)SiO2 (wt%)CaO/TiO2 (wt%)K2O (wt%)(b)(d)(c)(a)(f)(e)normalized to 10 wt% MgOnormalized to 10 wt% MgO(Magma supply ratedoubled)2003-07(Magma supply ratedoubled)2003-07(Magma supply ratedoubled)2003-071983-19861986-19921992-2010normalized to 10 wt% MgOnormalized to 10 wt% MgOEp. 56Figure 2. Major-element variation diagrams for Pu?u ?O?o lavas from 1983 to 2010. All major elements andratios except MgO were normalized to 10 wt % MgO [the most primitive lava erupted from Pu?u ?O?o; Gar-cia et al., 2000] by addition of equilibrium composition olivine (98.5%) and spinel (1.5%) in 0.5 mol % steps[Garcia et al., 2003; Rhodes and Vollinger, 2004]. Pu?u ?O?o lavas with <7.2 wt % MgO may have crystal-lized minerals other than olivine (e.g., clinopyroxene and plagioclase) and were not included in the olivinenormalization procedure and are not shown in all the plots, except MgO. Episode 54 (Ep. 54; 29?30 January1997) lavas involved mixing of evolved magmas stored in the rift zone and MgO-rich magma. Three intervalsof eruptive activity in legend and colors correspond with those shown in Figure 1. CaO/TiO2 and CaO/Al2O3ratios also use normalized data although are virtually unaffected by olivine fractionation. Vertical lines indi-cate nearly double magma supply rate between 2003 and 2007 [Poland et al., 2012]. Data are presented in thesupporting information. Uncertainty for analyses is described in the supporting information.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854857disequilibrium in the episode 54 lavas are thoughtto result from mixing Pu?u ?O?o magma withstored, differentiated rift zone magma [Garciaet al., 2000; Thornber et al., 2003]. In contrast,episode 56 lavas have higher MgO (8.5 versus 7.2wt %) and a relatively high 18O value (5.6 versus5.4%) but are otherwise geochemically indistin-guishable from contemporaneous Pu?u ?O?o lavas.[12] The 206Pb/204Pb ratios for Pu?u ?O?o lavasdecreased rapidly through the episodic fountainingperiod (1983?1986) and reached a minimumbetween 1989 and 1991 during the Kupaianahaphase (Figure 4b and Table 2). The rapid decreasein 206Pb/204Pb continues the longer-term trend ofdecreasing Pb isotope ratios for K?lauea lavaserupted following the 1924 collapse of the summitcaldera (Figure 4a). After 1991, the trend of206Pb/204Pb ratios in Pu?u ?O?o lavas shows cyclicvariations with two broad humps, each cycle span-ning approximately 10 years (except for a smalloffset from the overall trend between January1998 and June 1999; Figure 4b). The cyclic varia-tion in Pb isotope ratios is well shown by208Pb/206Pb ratios, which inversely mirror the206Pb/204Pb trend (Figures 4 and 5).[13] The 87Sr/86Sr ratios of Pu?u ?O?o lavas extendthe temporal trend of increasing Sr isotope ratiosfor K?lauea lavas that started following the 1924caldera collapse (Figure 4c). Overall, Pu?u ?O?olavas display an increase in 87Sr/86Sr from 1983 to2003 and a slight decrease after 2004 (Figure4d). Prior to 1999, the 87Sr/86Sr and 206Pb/204Pbratios of the lavas are not well correlated, althoughthere is an overall inverse correlation between the19831984198519861987198819891990199119921993199419951996199719981999200020012002200320042005200620072008200920100.40.50.60.70.819831984198519861987198819891990199119921993199419951996199719981999200020012002200320042005200620072008200920102.02.12.22.32.42.51.99101112161820222426 Sr/NbNb/Y La/Sm(c)(a)(d)1983-19861986-19921992-2010Ep. 54Ep. 54Ep. 5484 86 88 90 92 94 96 98 00 02 04 06 08 10 84 86 88 90 92 94 96 98 00 02 04 06 08 10(Magma supply ratedoubled)2003-07Ep. 54Zr/Nb(b)(Magma supply ratedoubled)2003-07 ?2 SE ?2 SE ?2 SE ?2 SEK?lauea summit(Sep.1982)K?lauea summit(Sep.1982)K?lauea summit(Sep.1982)K?lauea summit(Sep.1982)Figure 3. Trace-element ratios versus time for Pu?u ?O?o lavas from 1983 to 2010. Overall exponential vari-ation in trace-element ratios indicates progressive depletion of the source. In the La/Sm versus time plot, asubset of samples analyzed at PCIGR (April 1983, January 1984, September 1984, and April 2007 to January2010) are normalized to reference material Kil-93 (La/Sm of 2.09, average value from Australia National Uni-versity where most ICP-MS analyses were performed). Vertical lines indicate nearly double magma supplyrate between 2003 and 2007 [Poland et al., 2012]. Trace-element abundances in ppm (data shown in support-ing information). Average 62 bars are shown in a corner of each panel. September 1982 K?lauea summitlava composition from Pietruszka and Garcia [1999].GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854858two ratios for historical summit lavas (Figure 4and Table 2). After 1999, there is a positive corre-lation between 206Pb/204Pb and 87Sr/86Sr (Figure6), which corresponds to the second of the twomajor temporal cycles of Pb isotope ratios (Figure5). The Nd isotope ratios of lavas erupted between1983 and 2005 display no variation outside analyt-ical uncertainty [Marske et al., 2008]. Thus, nonew Nd isotopic data were collected during thisstudy.[14] Oxygen isotopic compositions of Pu?u ?O?olavas erupted over 25 years (1983?2008) show alarger range than historical K?lauea summit lavasspanning 380 years of activity (1.1 versus 0.7% ;Figure 4). Overall, O isotope ratios of Pu?u ?O?olavas have increased over time with early lavas(before 1986) having lower 18O (4.8?4.9%) thansubsequent lavas (5.0?5.6% ; Table 2). The high-est O isotope ratio observed during the Pu?u ?O?oeruption is for an episode 56 lava that erupted 6Uwekahuna BluffUwekahuna BluffHilinasubmarinelavas0.70340.70350.70360.7037198319841985198619871988198919901991199219931994199519961997199819992000200120022003200420052006200720082009201018.3718.3918.4118.4318.4518.4718.4918.5118.530.703550.703600.703651983198419851986198719881989199019911992199319941995199619971998199920002001200220032004200520062007200820092010K?laueasummitK?laueasummitPCIGR analyses (2009)Marske et al. (2008)-SDSUPCIGR analyses (2010)Time (years)Time (years)(summit)(summit)Time (years)K?laueasummitsummit calderacollapse (1924)summit calderacollapse (1924)average ?2 SEMORB mantle rangeSDSU analyses (2007)?18O18.218.418.618.81780 1820 1860 1900 1940 19801000 14001780 1820 1860 1900 1940 19801000 140016001780 1820 1860 1900 1940 198016001600Jun-99Jun-99Jun-86Jun-07(Ep. 56 uprift)(  )(  )(  )4.8 ka 3.5 ka1.7 ka4.8 ka3.5 ka1.7 ka206Pb/204Pb87Sr/ 86Sraverage ?1 SE?18O87Sr/ 86Sr206Pb/204Pb(280-130 ka)tholeiiticalkalictransitionalHilinasubmarinelavas(280-130 ka)tholeiiticalkalictransitional(episodic fountaining)Kupaianaha(gentle effusion)Pu?u ???? Pu?u ???? (mostly flank vent eruptions)(a)(c)(b)(d)Pu?u ???? Pu?u ???? Pu?u ???? (e) (f)data thisstudyMORBmantle summit calderacollapse (1924)PCIGR analyses (2009)Marske et al. (2008)-SDSUPCIGR analyses (2010)SDSU analyses (2007)4.64.85.05.25.45.65.84.64.85.05.25.45.61983198419851986198719881989199019911992199319941995199619971998199920002001200220032004200520062007200820092010no datascale changeK?lauea summit(Sep.1982)K?lauea summit(Sep.1982)Figure 4. Temporal variation of Pb, Sr, and O isotopes for K?lauea Volcano and Pu?u ?O?o lavas. (b)206Pb/204Pb in Pu?u ?O?o lavas shows cyclic variation with two broad humps, each cycle spanning approxi-mately 10 years. (d) 87Sr/86Sr in Pu?u ?O?o lavas increases from 1983 to 2003 and decreases between 2008and 2010, and is correlated with 206Pb/204Pb after 1999. (f) 18O for Pu?u ?O?o lavas shows the same range(0.7%) as historical K?lauea summit lavas and is not well correlated with 206Pb/204Pb or 87Sr/86Sr. Data sour-ces for K?lauea Volcano are Hanyu et al. [2010], Kimura et al. [2006], Marske et al. [2007], Abouchami et al.[2005], Pietruszka and Garcia [1999], Chen et al. [1996], and Garcia et al. [2008]. Data sources for previousanalyses of Pu?u ?O?o lavas areMarske et al. [2008] and Garcia et al. [1998]. For Pu?u ?O?o analyses, average62 for 206Pb/204Pb is smaller than symbol size and uncertainty for 87Sr/86Sr and 18O is shown in the panels.Data for Pu?u ?O?o lavas are presented in Table 2. September 1982 K?lauea summit lava composition fromPietruszka and Garcia [1999]. Colors for symbols in Figures 4b, 4d, and 4f) are the same as Figure 3.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854859km uprift from the Pu?u ?O?o vent in mid-June2007. The episode 56 eruption is related to intru-sion of a dike from the upper East Rift Zone, sothis lava was probably not derived from the shal-low reservoir of magma beneath Pu?u ?O?o [Mont-gomery-Brown et al., 2010]. Thus, its 18O valueis probably representative of the magma feedingthe Pu?u ?O?o eruption. It is identical to the highestvalues observed among historical summit lavas(5.6% ; Figure 4). Variations of 18O in Pu?u ?O?olavas do not correlate with Pb or Sr isotope ratios,or with other geochemical parameters, as wasnoted for previous O isotope work on lavas fromthis eruption [Garcia et al., 1998]. Therefore, Pu?u?O?o lava Pb and Sr isotope ratios were apparentlynot affected by the processes causing variable Oisotope ratios.5. Discussion[15] The high eruption rate and continuous natureof the Pu?u ?O?o eruption provide an exceptionalopportunity to use lava chemistry to evaluate thechanging roles that source, melting, and crustalprocesses play during this single prolonged erup-tion. Previous Pb isotope and trace-element studieson lavas from several multiple-year eruptions ofPiton de la Fournaise Volcano (Reunion Island)have discerned contributions from multiple com-ponents within the Reunion mantle plume and aperiodic role for shallow-level contamination [Pie-truszka et al., 2009; Vlastelic et al., 2005, 2007].Similarly, extreme Pb isotope variability in meltinclusions from Iceland basaltic lavas indicate sig-nificant source heterogeneity, with binary mixingrelationships that may result from combining sol-ids in the mantle and two stages of melt mixing (inporous mantle melt-transport channels and lowercrustal magma chambers) [Maclennan, 2008].Similarly, the geochemistry and petrography ofPu?u ?O?o lavas have been used to interpret theextent of crustal magmatic processes (olivine frac-tionation and accumulation, mixing of higher-MgO and stored rift-zone magmas, and crustalassimilation) and mantle processes (degree of par-tial melting, melt extraction and migration, andsource heterogeneity) during the Pu?u ?O?o erup-tion until 2005 [e.g., Garcia et al., 1998, 2000;Marske et al., 2008]. Here we use new high-precision Pb, Sr, and O isotope ratios, and major-and trace-element data for the entire Pu?u ?O?oeruption (1983?2010) to evaluate the causes ofcyclic and other short-term variability in the proc-esses that operate from the source to the surfacewithin K?lauea Volcano. The effects of crustalprocesses (crystal fractionation, magma mixing,and crustal contamination) on modifying Pu?u?O?o lava compositions are evaluated beforeexamining mantle source and melt transportprocesses.5.1. Magma Mixing and CrystalFractionation During Early EpisodicActivity (1983?1985)[16] The largest compositional changes in Pu?u?O?o lavas occurred from 1983 to 1985. Thesechanges mostly involved two crustal processes:crystal fractionation and magma mixing. Duringsome single eruptive episodes (5?10, 30, and 31),there were relatively large changes in MgO, Ni,and Cr, which are related to minor (3?5%) olivinefractionation in the shallow Pu?u ?O?o reservoirduring eruptive hiatuses [Garcia et al., 1992].These short-term (3?4 weeks) variations are super-imposed on longer term changes that have beenrelated to magma mixing [Garcia et al., 1992;Thornber, 2003]. The longer term variations areevident in plots of MgO-normalized major ele-ments, ratios of incompatible trace elements, andPb isotope ratios (Figures 2?4). Strontium and Oisotopes show less change during this period com-pared to their overall variation during the eruption(Figure 4). The overall progressive compositionalvariation in Pu?u ?O?o lavas from 1983?1985 hasbeen attributed to the mixing of new, relativelyMgO-rich magma (>7.5 wt %) with a decreasingproportion of hybrid, rift-zone stored differen-tiated magma (from 30% of the higher MgOmagma in March 1983 to 100% in September1984) [Garcia et al., 1992; Shamberger andGarcia, 2007].[17] The origin of the higher MgO magma compo-nent from the early phase of the Pu?u ?O?o erup-tion may have been: (1) magma from the summitreservoir, as represented by lavas from the Sep-tember 1982 summit eruption; and/or (2) newmantle-derived magma [Garcia et al., 1992;Shamberger and Garcia, 2007]. Scenario 1involves no change in the composition of thehigher MgO magma from September 1982 to1985, whereas scenario 2 requires it. The 1983?1985 Pu?u ?O?o lavas have both higher and lower206Pb/204Pb ratios than the September 1982 sum-mit lavas (Figures 4b and 7). Therefore, mixing ofa single 1982 summit magma with rift-zone storedmagma (scenario 1) cannot explain the isotopicGREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854860variation of Pu?u ?O?o lavas after 1984, when206Pb/204Pb values are lower than 1982 summitmagma (Figure 7). Ratios of some incompatibletrace elements (Sr/Nb and Zr/Nb) for some lavaswith higher MgO (>7.5 wt %) erupted after mid-1984 also are higher than those for September1982 summit lavas (Figure 3). Thus, if magmafrom the summit reservoir was supplying the Pu?u?O?o eruption, its composition must have changedafter the September 1982 eruption and prior toSeptember 1984 (Figures 4 and 7).[18] The isotopic variations for early Pu?u ?O?olavas are consistent with the eruption being sup-plied by new, compositionally variable, mantle-derived magma in addition to or instead of Sep-tember 1982 summit magma. The rate of206Pb/204Pb variation observed for the period afterthe end of early magma mixing is much faster thanduring the previous 30 years (1952?1982) ofK?lauea summit eruptions (0.016 yr1 versus0.004 yr1). These rapid variations in Pb isotopicratios suggest that magmas supplying Pu?u ?O?opartially bypassed or did not thoroughly mix withthe summit reservoir [Garcia et al., 1996]. Basedon these observations, the composition of the pa-rental magma delivered to Pu?u ?O?o from themantle is interpreted to have continually changedfor the remainder of the eruption (i.e., after 1984).The details and cause of this variation are dis-cussed in section 5.3.5.2. Oxygen Isotope Indications of CrustalContamination and Nature of MantleSource[19] Lavas from oceanic island volcanoes showwide ranges in oxygen isotopic compositions (4.5?7.5%), which have been attributed to composition-ally variable mantle-derived magmas that weremodified by oxygen exchange and/or crustal con-tamination [Harmon and Hoefs, 1995]. Our previ-ous studies revealed that some Pu?u ?O?o andK?lauea summit magmas experienced significantoxygen isotope exchange with metamorphosedK?lauea rocks [Garcia et al., 1998, 2008]. This isindicated by the disequilibrium between matrixand coexisting olivine 18O values, the relativelylow 18O values for these lavas (4.7?5.2%) andthe lack of correlation between 18O values andother geochemical parameters [Garcia et al.,1998, 2008].[20] The highest 18O value observed for any lavaduring the Pu?u ?O?o eruption is for the June 2007208Pb/206PbPCIGR analyses (2009)Marske et al. (2008)-SDSUPCIGR analyses (2010)SDSU analyses (2007)(Magma supply ratedoubled)2003-072.0592.0612.0632.0652.0672.0692.0711983198419851986198719881989199019911992199319941995199619971998199920002001200220032004200520062007200820092010Jun-99(episodic fountaining)Kupaianaha(gentle effusion)Pu?u ???? Pu?u ???? (mostly flank vent eruptions)Jun-861982summitRift-storedmagmaFigure 5. 208Pb/206Pb variation with time for Pu?u ?O?o lavas from 1983 to 2010. Previous analyses of Pu?u?O?o lavas indicated in legend are from Marske et al. [2008]. Average 62 for 208Pb/206Pb is smaller thansymbol size.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854861uprift lava (5.6% ; Figure 4). This value is identi-cal to the highest value reported for historicalK?lauea summit lavas (1820?1982) [Garcia et al.,2008] and lies within the range of normal mid-ocean ridge basalt (MORB) basalt 18O values(5.4?5.8% ; Figure 4) [Eiler, 2001]. These summitlavas (1832, 1866, 1894, and 1917?1921) wereerupted during periods of sustained lava lake ac-tivity, and are thought to be representative of theprimary uncontaminated magma feeding K?lauea[Garcia et al., 2008]. Thus, the 2007 uprift ventlava supports our previous interpretation [Garciaet al., 2008] that the 18O value for the mantlesource of K?lauea?s magma is identical to thesource for MORB.[21] The earliest Pu?u ?O?o lavas (1983?1986)show the strongest signs of disequilibriumbetween coexisting matrix mineral and olivine,and have the lowest O isotope values (<5.0%)[Garcia et al., 1998]. After the shift to continuouseffusion in July 1986, O isotope ratios are higher(5.2%) and the coexisting olivines were in equilib-rium with host matrix for about 1 year [Garciaet al., 1998]. Subsequently, the matrix O isotopevalues decreased somewhat (to 5.0% ; Figure 4)and those for olivine increased, indicating olivine-matrix disequilibrium. This O isotope disequili-brium continued for two more years, and was fol-lowed by a return to olivine-matrix equilibrium in1995?1997 [Garcia et al., 1998]. After 1997, ma-trix O isotope values are relatively low and nearlyconstant (5.36 0.1%) except for a 2005 lava(5.0% ; Figure 4), which was the most evolvedsample (analyzed for O isotopes) since 1984 (6.7wt % MgO). Thus, despite nearly 30 years of vig-orous eruptive activity (producing 4 km3 oflava), oxygen exchange with metamorphosedrocks has probably continued in the Pu?u ?O?omagmatic plumbing system. The magnitude of ox-ygen isotope exchange can be estimated assumingbulk assimilation between a parental magma (asreflected by the 2007 uprift sample with a 18Ovalue of 5.6%) and a hydrothermally alteredK?lauea rift zone lava (1.9%) [Garcia et al., 2008]as a contaminant. Pu?u ?O?o lavas erupted justbefore and after the 2007 uprift event have averageO isotope values of 5.4% (Figure 4), indicating5% bulk contamination, whereas earlier lavas(1986?2006) with average values of 5.2?5.3%,might have experienced 8?11% bulk contamina-tion. This contamination is likely to have occurredin the Pu?u ?O?o reservoir and did not have anyobvious effect on other geochemical parameters[Garcia et al., 1998] (Table 2).5.3. Cyclic Compositional VariationsFrom Mantle Processes (1985?2010)[22] Pu?u ?O?o lavas erupted after the early periodof magma mixing ended in late 1984 show cyclicvariations in several geochemical parameters thatare insensitive to olivine fractionation (e.g., CaO/TiO2, Sr/Nb, Zr/Nb,206Pb/204Pb; Figure 8). Thecyclic variations in CaO/TiO2 and K2O/TiO2ratios for Pu?u ?O?o lavas erupted between 1996and 2001 were reported to be associated with de-formation in the summit magma reservoir[Thornber, 2003]. Although the timing of thehighs and lows in these ratios are not perfectlycoincident with summit tilt changes [see Figure 8,Thornber, 2003], these geochemical cycles wereattributed to mixing of mantle-derived magma ofuniform composition (similar to Pu?u ?O?o lavasaverage ?2 SEaverage ?2 SETime (month-year)18.39018.39518.40018.40518.41018.41518.420Jan-97Jul-97Jan-98Jul-98J an-99Jul- 99Jan-00J ul -00Jan-01Jul-01Jan -02Jul-02Jan-03Jul-03J an-0 4Jul-04Jan-05J ul-05Jan-06Jul-06J an-07J ul-07Jan-08Jul-08Jan-09Jul-09J an-100.703590.703600.703610.703620.703630.703640.70365Jan-97Jul-97Jan-98Jul-98Jan-99Jul-99Jan -00Jul-00Jan-01Jul-01Jan -02Jul-02Jan-03Jul-03Jan -04Jul-04Jan-05Jul-05Jan-06Jul-06Jan-07Jul-07Jan-08Jul-08Jan -09Jul-09Jan -10206Pb/204Pb87Sr/ 86Sr(a)(b)Figure 6. Temporal variation in 206Pb/204Pb and 87Sr/86Srfor Pu?u ?O?o lavas during period of dramatic increase inmagma supply. Dashed lines indicates period of significantincrease in magma supply rate up to 0.25 km3/yr between2003 and 2007 compared to 0.1 km3/yr prior to 2003[Poland et al., 2012]. Analytical uncertainty is shown in thepanels.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854862from 1999 to 2001) with 1982 summit magma[Thornber 2003]. However, Pu?u ?O?o lavaserupted after 1999 have elevated 87Sr/86Sr ratios(at a given 206Pb/204Pb; Figure 7) compared toearlier lavas with low 206Pb/204Pb ratios, and thus,the 1999?2001 lavas cannot serve as a mixingend-members to explain the compositional trendof lavas erupted before 1999. Similar behavior isobserved on a plot of La/Yb versus 206Pb/204Pb(Figure 9), where a relative shift to lower La/Ybratios at a given 206Pb/204Pb occurred after 1999(compared to the trend of pre-1999 lavas). Theserelationships indicate that the temporal variationof Pu?u ?O?o lavas erupted after 1999 cannot beexplained simply by mixing of 1982 summitmagma with a uniform mantle-derived magma(Figure 7) within K?lauea?s shallow magmaticplumbing system. Instead, either a third magma ismixing with the other two or, as we advocatebelow, the composition of the Pu?u ?O?o magmais continually changing due to the melting ofsmall-scale compositional heterogeneities in themantle source.[23] Ratios of Pb isotopes in Pu?u ?O?o lavas showcyclic variations (Figure 5). These variations prob-ably reflect the dynamic process of melt extraction(from a heterogeneous source) over a time scale of0.703500.703550.703600.703650.7037018.36 18.38 18.40 18.42 18.44 18.46 18.48 18.50 18.52 18.54208Pb/204Pb0.703570.703580.703590.703600.703610.703620.703630.703640.7036518.38 18.39 18.40 18.41 18.42208Pb/204Pb38.04538.05538.06538.07518.38 18.39 18.40 18.41 18.4238.0038.0538.1038.1538.2018.36 18.38 18.40 18.42 18.44 18.46 18.48 18.50 18.52 18.5487Sr/ 86Sr206Pb/204PbSep-8687878889 9091909808090700030504020106929394959697 991988-911983-85Jun-9986102000-071992-992008-1096098788Oct-90Jan-90919808090700 0305040201069293949597991988-911983-85Jun-99 8610968788Mar-89Jul-89May-9096979898990005080808Jan-090909862000-071993-992008-1087Sr/ 86Sr206Pb/204Pb19831985198319851986-20101986-2010East Rift Zone(1977)East Rift Zone(1977)K?lauea summit(1982)K?lauea summit(1982)East Rift Zone(1960-69)East Rift Zone(1960-69)1984Sep.Jan.2001 composition1984Sep.Jan.?2?(b)(d)(a)(c)(b)(d)Figure 7. Pb and Sr isotopic compositions for Pu?u ?O?o lavas. Line connects samples in order of increasingeruption date in Figures 7b and 7d. Average 62 for Pb isotope ratios is smaller than symbol size. East RiftZone data is from J. Marske [personal communication, 2013]. September 1982 K?lauea summit lava composi-tion (outline in Figure 7a; orange star in Figure 7c) from Pietruszka and Garcia [1999]. K?lauea summit(1982) field in Figure 7a is new high-precision data from A. Pietruszka [personal communication, 2013]. Bluestar is 2001 composition proposed by Thornber [2003] as mixing end-member with 1982 K?lauea summitcomposition. Pu?u ?O?o lavas erupted after 1999 have elevated 87Sr/86Sr ratios (at a given 206Pb/204Pb). The1999?2001 lavas cannot serve as mixing end-members to explain the compositional trend of lavas eruptedbefore 1999.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854863years to decades rather than movement of small-scale mantle heterogeneities through the meltingzone. This interpretation is based on the hypothe-sis that buoyancy-driven upwelling through themelt-producing region beneath K?lauea occurs onlonger timescales (hundreds to thousands ofyears) than melt extraction (years to decades)[Pietruszka et al., 2006]. The highest estimatesfor solid mantle upwelling in the center of theHawaiian plume are 10 m/yr [Hauri, 1996;Pietruszka and Garcia, 1999], which wouldresult in a maximum of only 270 m of upwell-ing during the first 27 years of the Pu?u ?O?oeruption [cf. 5?10 km maximum thickness forthe zone of melting; Marske et al., 2007]. Forcomparison, estimates for solid mantle upwellingrates beneath Mauna Loa and Lo?ihi based onU-series disequilibria range from 0.4 to 1 m/yr[Sims et al., 1999] and 5?6 cm/yr [Pietruszkaet al., 2011], respectively. Melt extraction rates(or source-to-surface melt velocity) are estimatedto be on the order of 5?17 km/yr [Reiners,2002], which is extremely rapid compared tosolid mantle upwelling rates. Thus, cyclic varia-tion in Pb isotope ratios over short timescales(years) are best explained by variations in theprocess of melting of a heterogeneous source(and the transport of the melt to the surface),rather than upwelling of small-scale mantle het-erogeneities through the melting region.[24] The short-term Pb and Sr isotopic variationsin Pu?u ?O?o lavas may be generated by one ormore processes including: periodic processes ofmelting, melt extraction, or melt aggregation [e.g.,Cordier et al., 2010], changes in melt transportpathways or tapping new source areas [Marskeet al., 2008; Pietruszka et al., 2006], changes inthe volume of the melting region [Pietruszkaet al., 2001], and progressive melt extraction froma source with fine-scale heterogeneities [Garciaet al., 2000]. In the presence of small-scale hetero-geneities, changes in melt pathways over years todecades may lead to tapping compositionally dis-tinct sources and short-term isotopic variation inlavas [Marske et al., 2007]. The scale of composi-tional heterogeneities must be small relative to thesize of the melting region beneath K?lauea Vol-cano to allow for rapid (few years) variation inlava Pb isotope compositions [Pietruszka and Gar-cia, 1999]. Melt pathways within the source regionprobably migrate over years to decades [Pie-truszka et al., 2001, 2006]. Therefore, melt may besupplied from different areas of the melting region(Magma supply ratedoubled)2003-07Time (year)4.04.24.44.64.8CaO/TiO2 1986-19921992-201022232425262790 92 94 96 98 00 02 04 06 08 1090 92 94 96 98 00 02 04 06 08 1090 92 94 96 98 00 02 04 06 08 10206Pb/204PbSr/Nb(a)(d)Zr/Nb(b)(c)18.3518.3618.3718.3818.3918.4018.4118.4218.4318.44199019911992199319941995199619971998199920002001200220032004200520062007200820092010101112Figure 8. Plots of CaO/TiO2, Sr/Nb, Zr/Nb and206Pb/204Pbfor Pu?u ?O?o lavas showing cyclic variation apparent from1990 to 2010. Trace-element abundances in ppm.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854864over a relatively short period of time. These short-term geochemical fluctuations are effectivelytransported from the source to the surface becausePu?u ?O?o magmas are thought to partially bypassthe volcano?s summit magma storage reservoir,and avoid its buffering effects [Garcia et al.,2000].[25] The periodicity and rate of isotopic variationPu?u ?O?o lavas provide clues about the size anddistribution of small-scale heterogeneities in thevolcano?s mantle source. The cycles of Pb isotopicratios at Pu?u ?O?o have a peak to peak duration of10 years (Figure 10). These short-term increasesand decreases of Pb isotope ratios, at similar rateand degree, may represent melt extraction fromsmall-scale heterogeneities with a limited horizon-tal length scale (a few km or less). Modeling byFarnetani and Hofmann [2009] suggests that??filament-like?? structures derived from stretchingof deep-seated mantle heterogeneities may de-velop as the Hawaiian plume rises to the surface.Although their model was developed to explainthe long-term (>100 kyr) geochemical variationobserved for drill core from Mauna Kea [Farne-tani and Hofmann, 2010], we use the filamentmodel to explain the periodic variation in the Pbisotope ratios of Pu?u ?O?o lavas (Figure 10)because it provides a mechanism to link geochem-ical variations with the inferred deep mantle struc-ture. Other geometries have been suggested for thesmall-scale heterogeneities within Hawaiianplume, including a series of vertically stacked,elongated blobs [Blichert-Toft and Albare`de,2009], but we prefer the filament geometry toexplain the Pb isotopic variations of Pu?u ?O?olavas.[26] In this scenario, the periodic variation in Pbisotope ratios of Pu?u ?O?o lavas may reflect meltextraction from a mantle source with verticallyoriented repeating source heterogeneities, or thinstrands, on a small scale (Figure 10). The Pu?u?O?o eruption rate is thought to be greater than therate of melting, so melt must be transferred intochemically isolated channels from successivelyfurther areas within the larger melting region tosustain the eruption [Pietruszka et al., 2006]. Inthe context of the filament model of Farnetani andHofmann [2009], this process might extract meltfrom a succession of strands with different isotopiccompositions, which would potentially create theobserved periodicity in variation of the Pb isotoperatios (Figure 10). The volume of a single compo-sitional strand within the mantle tapped by thePu?u ?O?o eruption can be inferred using estimatesfor lava eruption rate (0.13 km3/yr) [Suttonet al., 2003] and melt zone porosity (1?2%) [Pie-truszka et al., 2001]. This calculation assumes that(1) there have been only two isotopically distinctcomponents since 1986 and (2) the heterogene-ities have the same melt productivity. We do notconsider the effect of melting heterogeneous lith-ologies with different melt productivities (e.g., pe-ridotite versus pyroxenite), despite the potentialsignificance for mixed lithologies in the source forHawaiian lavas [Hauri, 1996; Reiners, 2002;Sobolev et al., 2005]. Indeed, recent modeling ofincompatible trace elements suggests that Pu?u2007-1018.3818.4218.4618.5018.544.5 4.9 5.3 5.7 6.1 6.5 6.9 7.3 7.718.3818.3918.4018.4118.424.5 4.7 4.9 5.1 5.3 5.5 5.7 5.9HistoricalEast RiftZone ?2 ?summit(1982)La/Yb206Pb/204Pb206Pb/204Pb(a)(b)831988-912000-061992-99848586878889908990919394969598989708099900010304050607091092La/Ybmixingsource variationK?laueasummitFigure 9. Plots of 206Pb/204Pb versus La/Yb for Pu?u ?O?olavas with line connecting samples in eruptive order. (a) AllPu?u ?O?o lavas. (b) Smaller variations in lavas eruptedbetween 1988 and 2010 [area indicated by box with dashedline in Figure 9a]. K?lauea summit data is from Pietruszkaand Garcia [1999]. Average 62 for La/Yb shown in Figure9 a. Average 62 for Pb isotope ratios is smaller than symbolsize.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854865?O?o lavas (1986?1998) are derived from a sourcewith 13% recycled oceanic crust in a matrix ofambient depleted Hawaiian mantle [Pietruszkaet al., 2013]. However, this model was unable todistinguish if the recycled oceanic crust was pres-ent as pyroxenite or refertilized peridotite. Itshould also be noted that the compositional rangeof Pu?u ?O?o lavas from 1986 to 2010 is smallcompared to the overall range for Hawaiian volca-noes [e.g., Jackson et al., 2012; Ren et al., 2009;Weis et al., 2011], so the melt productivities of theend-member sources are probably similar.[27] The duration of a single cycle of Pb isotopevariation is approximately 10 years (Figure 10),which suggests melt is extracted from one compo-sitional strand in 5 years (before the trendreverses when melt from a different strand isencountered). Estimates for the height of K?lauea?smelting region range from <5 km [Marske et al.,2008] to 55 km [Watson and McKenzie, 1991].We assume the magma supply rate is roughlyequivalent to the eruption rate given that overallmagma storage in K?lauea has been decreasingslightly since 1983 [Poland et al., 2012]. The meltDepth (km) 13017010 50 90 130 170 210Length(km)Plate motionMelting zoneK?lauea Plate motion9 cm/yr100 0 100 200Length(km)Depth (km)15025050350Melting rate (10-11 kg m-3 s-1)0 2 4 6 8250 225 200 175 150 125?T (?C)50Depth (km)13017020 80 50Depth (km)13017020 80 50Depth (km)13017020 80Depth (km)50Depth (km)13017020 8019902.0652.0672.0692.071208 Pb/206 Pb1987198819891990199119921993199419951996199719981999200020012002200320042005200620072008200920101994     1998 2004(a)(b)(c)(d)Figure 10. Cartoon model for the Hawaiian mantle plume to explain the isotopic variation of Pu?u ?O?olavas, based on assumptions described in the text. (a) 208Pb/206Pb variation with time for Pu?u ?O?o lavas from1990 to 2010 showing cyclic variation. (b) Vertical section of the Hawaiian plume adapted from Farnetaniand Hofmann [2010]. Purple shades indicate the melting rates inside the melting zone, shown in legend.Dashed yellow lines are flow trajectories. Dashed black box is magma capture zone for K?lauea. (c and d)Sketches of the changing melting zone during Pu?u ?O?o eruption. Lithosphere is not shown in Figure 10c.Melting zone from Farnetani and Hofmann [2010].GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854866volume produced in 5 years is 0.5 km3. If meltzone porosity is 2% (estimates based on U-seriesstudies range from 1 to 3%) [Pietruszka et al.,2001; Sims et al., 1999], the total volume of a sin-gle heterogeneity tapped over a 5 year periodwould be 25 km3. If individual filament-shapedheterogeneities extend over the height of the melt-ing zone, then this 25 km3 volume would translateto a diameter of 1?3 km for melting zone heightsof 55?5 km, respectively.5.4. Effects of Doubling of Magma SupplyRate on Lava Composition (2003?2007)[28] One enigmatic question of ocean island volca-nism is whether variations in lava compositions arecorrelated with magma supply rate [e.g., Vlastelicet al., 2005]. Wide variations in magma supplyrate have occurred historically at K?lauea (0.01?0.18 km3/yr between 1840 and 1983) [Dvorak andDzurisin, 1993]. A marked decrease in lava effu-sion rate during the 19th to early 20th century atK?lauea (0.10?0.01 km3/yr) was accompanied byan increase in the ratios of highly over moderatelyincompatible trace-element abundances [Pie-truszka and Garcia, 1999], and in the modal abun-dance of clinopyroxene and plagioclase in thelavas indicating eruption of more fractionated andcooler magma [Garcia et al., 2003]. This change inlava composition is believed to be a direct result ofa decrease in the melt fraction (10?5%) and aswitch to a more depleted source [Pietruszka andGarcia, 1999]. A dramatic short-term increase ineffusion rate was observed for the Pu?u ?O?o erup-tion between 2003 and 2007 [Poland et al., 2012].Here we explore the results of this magma supplysurge on the composition of Pu?u ?O?o lavas.[29] Estimates of lava effusion rate for the Pu?u?O?o eruption prior to 2003 are based on geologicmapping, and measurements of very low fre-quency electromagnetic profiling and gas emis-sions. These techniques indicate an average rate ofmagma supply of 0.13 km3/yr [Sutton et al.,2003]. Lava effusion rate (considered to be aproxy for the magma supply rate by Poland et al.[2012]) was estimated to have increased between2003 and 2007 and to have doubled in 2005 (to0.25 km3/yr), before returning to the previousrate by 2008 [Poland et al., 2012]. An increase inmagma supply is normally expected to result inhigher MgO contents in erupted lavas as magmaundergoes less cooling prior to eruption. This rela-tionship was inferred for Pu?u ?O?o lavas eruptedfrom 1986 to 1992, when changes in tilt were fol-lowed 3 weeks later by changes in MgO [e.g.,Garcia et al., 1996]. However, during the 2003?2007 surge in magma supply, Pu?u ?O?o lavashave consistently lower MgO contents (<7.5 wt%; Figure 2a) than any period since the start ofcontinuous effusion in mid-1986, except duringepisode 54. The lower MgO contents of 2003?2007 lavas was interpreted to have resulted fromthe stirring and flushing of cooler magma withinthe volcano?s shallow magma storage system byan influx of new, hotter more primitive magma[Poland et al., 2012]. Mineralogical evidence (twopopulations of olivine) was noted in support ofthis claim, although no data were presented byPoland et al. [2012]. Our previous study of olivinecompositions in lavas erupted before and duringthe surge found no evidence for two populationsof olivines in any of the lavas, and that olivines inthese weakly phyric rocks are in Fe-Mg equilib-rium with the whole rock [Marske et al., 2008].We re-examined thin sections for these lavas andfound no textural evidence indicating magma mix-ing. In contrast, Pu?u ?O?o lavas from 1983 to1984 and episode 54 display obvious disequili-brium features from magma mixing [Garcia et al.,1989, 2000]. If mixing with a stored, coolermagma was important during the 2003?2007 surgein magma supply, the stored component must nothave differentiated very far beyond olivine control(unlike the situation for Pu?u ?O?o lavas from1983 to 1984 and episode 54).[30] A small increase in lava MgO contentoccurred after the surge, although values werevariable and overlap with those during the surge(7.0?8.1 after versus 6.7?7.4 wt % MgO during;Figure 2). The higher post-surge MgO values wereinterpreted to be a result of the heightened magmasupply from 2003 to 2007 [Poland et al., 2012].By November 2008, MgO dropped to values simi-lar to and lower than during the surge (6.5?7.2 wt%; Figure 2). Since 2000, the MgO content ofPu?u ?O?o lava has been declining with no signifi-cant change of this overall trend during the 2003?2007 surge in magma supply (Figure 2), except forless scatter in MgO content which may simplyreflect fewer interruptions in effusion during thistime. Also, it is noteworthy that the highest sus-tained MgO values were observed for Pu?u ?O?olavas from 1988 to 1993 (Figure 2), a period whenno increase in magma supply was recorded[Poland et al., 2012; Sutton et al., 2003]. Thus,the increase in magma supply from 2003 to 2007appears to have had limited impact on the varia-tion in the MgO content of Pu?u ?O?o lavas.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854867[31] Some of the other compositional features ofthe lavas erupted during the 2003?2007 magmasurge (e.g., slight offsets to lower MgO-normalizedSiO2 contents, Zr/Nb ratios and higher normalizedK2O and TiO2) could be considered to be represen-tative of flushing of differentiated magma similarto the Pu?u ?O?o lavas the late 1990s to early 2000s(Figures 2 and 3). There is also a reversal in87Sr/86Sr and 206Pb/204Pb in 2004 toward values by2007 similar to lavas from 2000 (Figure 6).These trends might be explained by the flushing ofmagma stored since the late 1990s to early 2000sduring the 2003?2007 surge in magma supply.However, based on the close time correlations(weeks to few months) between changes in summittilt and lava composition in earlier Pu?u ?O?o lavas[Garcia et al., 1996; Thornber, 2003], it is hard toimagine why 4? years were required to flushcooler magma from the Pu?u ?O?o system as pro-posed by Poland et al. [2012]. Thus, it is our inter-pretation that the magma erupted during the 2003?2007 surge was new mantle-derived magma andnot stored magma flushed from K?lauea?s shallowcrustal plumbing system. Thus, we interpret theisotopic variations during the Pu?u ?O?o eruption(since 1984) to be generated by the changingcomposition of melts coming from the mantle.5.5. Comparison of Pu?u ?O?o Lavas Withthe Long-Term Isotopic Evolution of OtherHawaiian Volcanoes[32] Studies of the long-term geochemical varia-tion of lavas from individual volcanoes in the Ha-waiian Islands provide understanding of thechemical structure of the Hawaiian mantle plume[e.g., Loa and Kea trends, Abouchami et al., 2005;Ren et al., 2009; Weis et al., 2011], and the vari-ability within single shield volcanoes [e.g., Bryceet al., 2005; Chen and Frey, 1985; Eisele et al.,2003; Marske et al., 2007; Nobre Silva et al.,2013; Rhodes and Hart, 1995; Weis et al., 2011].Here we compare the short-term Pb and Sr iso-topic variation for Pu?u ?O?o lavas to the longerterm variations for lavas from K?lauea and nearby,well-studied shield volcanoes to better understandthe rate and cause of long-term fluctuationsobserved at Hawaiian volcanoes.[33] The Pu?u ?O?o isotopic range covers a rela-tively large part of the long-term variationobserved for Mauna Kea and K?lauea volcanoes(Figures 11 and 12). The isotopic variation ofnearby Mauna Kea volcano was well documentedfor 300 kyr of shield growth using the HSDP2drill core. Lavas from the Pu?u ?O?o eruption afterthe period of magma mixing (early 1985) span30% of the total range of 206Pb/204Pb variationrecorded for HSDP2 (0.07 versus 0.22). Com-pared to K?lauea summit lavas, Pu?u ?O?o lavaserupted since 1985 span 25% of the Pb isotoperange since 950 AD and 47% of the range of Srisotope ratios (Figures 4 and 12). Thus, this singleeruption, which represents <1% of the time cov-ered by the HSDP2 core and 3% of thethousand-year period for K?lauea summit lavas,shows remarkable short-term isotopic variations.However, as seen for K?lauea summit lavas, iso-topic variation in Hawaiian shield lavas is cyclic,with each volcano showing a narrow but com-monly distinctive range (compared to neighboringvolcanoes) as seen by the relatively tight fields forPb and Sr isotope ratios in lavas from K?lauea,Lo?ihi, Mauna Kea and Mauna Loa (Figure 12).Thus, the Pu?u ?O?o eruption may represent thebest known expression of the small-scale compo-sitional heterogeneity of the Hawaiian plume. Asdiscussed above, the rapid rates of isotopic fluctu-ation found in Pu?u ?O?o lavas require a heteroge-neous source (on a scale of less than severalkilometers) that is tapped in only 5 years ofmelt extraction. How does the Pu?u ?O?o sourcecompare with those for nearby, well studied vol-canoes Lo?ihi and Mauna Kea?[34] Compared to other Hawaiian volcanoes, the Srand 206Pb/204Pb isotope ratios for Pu?u ?O?o lavasare most similar to Lo?ihi (Figure 12), althoughLo?ihi lavas have higher 208Pb/204Pb at a given206Pb/204Pb, like other Loa trend volcanoes (Figure11). Pu?u ?O?o lavas overlap with the Kea-mid8 Pbisotope array, the most common lava type inHSDP2 drill core (Figures 11c and 11d) [Eiseleet al., 2003; Nobre Silva et al., 2013]. However, Srisotopic compositions of Pu?u ?O?o lavas eruptedsince 1988 do not overlap those of Mauna Keaand trend orthogonally to the overall inverse arrayfor Hawaiian shield volcanoes and for K?laueasummit lavas (Figure 12). Thus, the Pu?u ?O?osource is isotopically distinct from other Hawaiianvolcanoes and the Pu?u ?O?o data set shows thatindividual eruptions may have trends orthogonal towhat are considered the primary source end-members for Hawaiian shield volcanoes.6. Conclusions[35] The temporal geochemical variation of Pu?u?O?o lavas from 1983 to 2010 provides insights onGREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854868the chemical structure of the Hawaiian mantleplume, and the dynamics of melt transport andmixing within the mantle. Our new results show:[36] 1. The Pu?u ?O?o eruption is being suppliedby new, compositionally variable, mantle-derivedmagma, which is being modified by various crustalprocesses including crystal fractionation (mainlyolivine), magma mixing (during 1983?1984 andepisode 54), and oxygen isotope exchange with orassimilation of altered K?lauea rocks.[37] 2. The episode 56 fissure lava has the highest18O value (5.6%) of any Pu?u ?O?o lava and thehighest MgO of any lava erupted since July 2000.Thus, it may be representative of the magma feed-ing this eruption as of June 2007. The episode 56fissure was fed from an upper East Rift Zone dikerather than the shallow Pu?u ?O?o reservoir. Coe-val lavas erupted from the Pu?u ?O?o vent havelower 18O values (5.36 0.1%) suggesting theywere contaminated, probably in the shallow reser-voir under the vent. Pu?u ?O?o lavas show no cor-relation of 18O values with other geochemicalparameters. For that reason, we suggest that the18O ratios were lowered by oxygen exchangewith or assimilation of altered K?lauea wall rock.[38] 3. Contrary to expectations, the dramaticincrease in magma supply between 2003 and 2007for the Pu?u ?O?o eruption was not accompanied byhigher MgO contents. Instead, lavas erupted duringthe 2003?2007 surge have lower MgO indicative ofgreater cooling of the magma prior to eruption,continuing the long-term trend for the eruption.[39] 4. Rapid and remarkably systematic variationsin Pb and Sr isotopic ratios are present in Pu?u37.737.837.938.038.138.238.338.418.0 18.1 18.2 18.3 18.4 18.5 18.6 18.72.042.052.072.082.092.100.825 0.830 0.835 0.840 0.845 0.850 0.855 0.860L??ihi2.0637.9037.9538.0038.0538.1038.1538.2038.2518.30 18.35 18.40 18.45 18.50 18.55 18.60 18.65 18.702.052.062.072.080.830 0.835 0.840 0.845208Pb/204Pb206Pb/204Pb208Pb/206Pb207Pb/206Pb208Pb/204Pb206Pb/204Pb208Pb/206Pb207Pb/206PbMauna LoaMauna LoaMauna KeaHistorical K?lauea summitPu?u ???? (1983-2010) (b)(d)(c)(a)1986Uwekahuna Bluff (summit)(Pietruszka and Garcia, 1999)Prehistoric K?lauea(Abouchami et al., 2005)L??ihi19831986Hilina BenchHilina Bench(Marske et al., 2007)upper endmemberkea-hi8kea-mid8kea-low8Mauna KeaMauna Kea (HSDP-2)Kea-hi8 arrayKea-mid8 arrayKea-low8 arrayarea of panel c)area of panel d)Historical K?lauea summitLoaKeaLoaKea1983Figure 11. Pb isotopic compositions for Pu?u ?O?o lavas compared to some other Hawaiian shield volca-noes. Top panels are different scales than bottom panels, shown by area of dashed boxes. Colored symbols arelavas from K?lauea Volcano. Fields are for Mauna Loa, Lo?ihi, Mauna Kea, and submarine prehistoricK?lauea (Hilina Bench; >40 ka). References for data sources are listed in the supporting information. Keaarray end-members are from Eisele et al. [2003]. Dashed line with yellow stars in Figure 11c is a best fit linefor Pu?u ?O?o lavas from 1988 to 2010.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854869?O?o lavas. Two cycles of Pb isotopic ratio varia-tion with 10 year periods were found. Thesecycles may be related to extraction of melt from asource with a pattern of vertically oriented sourceheterogeneities, or thin strands. These strands maybe 1?3 km in diameter to explain the scale ofisotopic variation for the Pu?u ?O?o eruption.[40] 5. The Pb isotopic variation of Pu?u ?O?olavas spans 25% of the range observed for the last1000 years of K?lauea summit lavas and 30% for300,000 years of shield volcanism for MaunaKea volcano. There is considerable Pb and Sr iso-topic overlap between Pu?u ?O?o lavas and lavasfrom Mauna Kea and Lo?ihi volcanoes. However,the Pb-Sr isotopic trend for the later Pu?u ?O?olavas (1988?2010) is oblique to the array definedby Hawaiian shield lavas. Thus, each Hawaiianvolcano appears to have an isotopically distinctsource.0.70320.70330.70340.70350.70360.70370.70380.70390.704018.0 18.1 18.2 18.3 18.4 18.5 18.6 18.7 18.80.703450.703500.703550.703600.703650.7037018.2 18.3 18.4 18.5 18.6 18.7 18.8Mauna LoaL??ihiMauna KeaMauna LoaMauna KeaL??ihiSalt Lake Crater xenolithsPu?u ???? (1983-2010) Mauna Kea (HSDP-2)Kea-hi8 arrayKea-mid8 arrayKea-low8 array206Pb/204Pb87Sr/ 86Sr206Pb/204Pb87Sr/ 86Sr198319851986Hilina BenchHilina BenchHistorical K?lauea summitUwekahuna Bluff (summit)(Pietruszka and Garcia, 1999)Prehistoric K?lauea(Abouchami et al., 2005)(Marske et al., 2007)area of panel b)Uwekahuna Bluff (summit)200320100.4-5 kaHistorical and PrehistoricK?lauea summit not shownHistorical K?lauea summit(b)(a)Figure 12. Plots of 206Pb/204Pb and 87Sr/86Sr for Pu?u ?O?o lavas compared to some other Hawaiian shieldvolcanoes. Figure 12b Expanded scale of dashed box in Figure 12a. Colored symbols are lavas from K?laueaVolcano. Fields are for Mauna Loa, Lo?ihi, Mauna Kea and submarine K?lauea (Hilina Bench; >40 ka).References for data sources are listed in the supporting information.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854870Acknowledgments[41] We thank Jane Barling, Bruno Kieffer, and Vivian Laifor their assistance with analyses at PCIGR at UBC, KyleTanguichi and Adonara Murek for sample preparation andcuration at the University of Hawai?i, and J. M. Rhodes forXRF analyses at University of Massachusetts. Claude Maer-schalk assisted with Pb and Sr column chemistry for a subsetof samples. Daniel Heaton provided assistance with severalisotope analyses at San Diego State University. We appreciatereviews by Julie Prytulak, Joel Baker, and Christoph Beier.This research was supported by grants from the National Sci-ence Foundation to M. Garcia (EAR11?18741) and A. Pie-truszka (EAR11?18738). This paper is SOEST ContributionNo. 8939.ReferencesAbouchami, W., S. J. G. Galer, and A. W. Hofmann (2000),High precision lead isotope systematics of lavas from theHawaiian Scientific Drilling Project, Chem. Geol., 169, 187?209, doi:10.1016/S0009?2541(00)00328-4.Abouchami, W., A. W. Hofmann, S. J. G. Galer, F. A. Frey, J.Eisele, and M. Feigenson (2005), Lead isotopes reveal bilat-eral asymmetry and vertical continuity in the Hawaiian man-tle plume, Nature, 434, 851?856, doi:10.1038/nature03402.Albare`de, F., B. Luais, G. Fitton, M. Semet, E. Kaminski, B. G.J. Upton, P. Bache`lery, and J.-L. Cheminee (1997), The geo-chemical regimes of Piton de la Fournaise Volcano (Reunion)during the last 530,000 years, J. Petrol., 38, 171?201.Blichert-Toft, J., and F. Albare`de (2009), Mixing of isotopicheterogeneities in the Mauna Kea plume conduit, EarthPlanet. Sci. Lett., 282, 190?200, doi:10.1016/j.epsl.2009.03.015.Blichert-Toft, J., D. Weis, C. Maerschalk, A. Agranier, and F.Albare`de (2003), Hawaiian hot spot dynamics as inferredfrom the Hf and Pb isotope evolution of Mauna Kea volcano,Geochem. Geophys. Geosyst., 4(2), 8704, doi:10.1029/2002GC000340.Bryce, J. G., D. J. DePaolo, and J. C. Lassiter (2005), Geo-chemical structure of the Hawaiian plume: Sr, Nd, and Osisotopes in the 2.8 km HSDP-2 section of Mauna Kea vol-cano, Geochem. Geophys. Geosyst., 6, Q09G18,doi:10.1029/2004GC000809.Caroff, M., R. C. Maury, P. Vidal, G. Guille, C. Dupuy, J. Cot-ten, H. Guillou, and P.-Y. Gillot (1995), Rapid temporalchanges in ocean island basalt composition: Evidence froman 800 m deep drill hole in Eiao Shield (Marquesas), J. Pet-rol., 36(5), 1333?1365, doi:10.1093/petrology/36.5.1333.Chen, C. Y., and F. A. Frey (1985), Trace element and isotopicgeochemistry of lavas from Haleakala Volcano, east Maui,Hawai?i : Implications for the origin of Hawaiian basalts, J.Geophys. Res., 90, 8743?8768.Chen, C.-Y., F. A. Frey, J. M. Rhodes, and R. M. Easton(1996), Temporal geochemical evolution of K?lauea Vol-cano: Comparison of Hilina and Puna basalt, in Earth Proc-esses, Reading the Isotopic Code, edited by A. Basu and S.R. Hart, pp. 161?181, AGU, Washington, D. C.Cordier, C., M. Benoit, C. Hemond, J. Dyment, B. Le Gall, A.Briais, and M. Kitazawa (2010), Time scales of melt extrac-tion revealed by distribution of lava composition across aridge axis, Geochem. Geophys. Geosyst., 11, Q0AC06,doi:10.1029/2010gc003074.DePaolo, D. J., and E. M. Stolper (1996), Models of Hawaiianvolcano growth and plume structure: Implications from theHawai?i Scientific Drilling Project, J. Geophys. Res., 101,11,643?11,655.Dvorak, J. J., and D. Dzurisin (1993), Variations in magmasupply rate at K?lauea Volcano, Hawai?i, J. Geophys. Res.,98, 22,255?22,268.Eaton, J. P., and K. J. Murata (1960), How volcanoes grow[Hawai?i], Science, 132, 925?938.Eiler, J. M. (2001), Oxygen isotope variations of basaltic lavasand upper mantle rocks, Rev. Mineral. Geochem., 43(1),319?364, doi:10.2138/gsrmg.43.1.319.Eisele, J. R., W. Abouchami, S. J. G. Galer, and A. W. Hof-mann (2003), The 320 kyr Pb isotope evolution of MaunaKea lavas recorded in the HSDP-2 drill core, Geochem. Geo-phys. Geosyst., 4(5), 8710, doi:10.1029/2002GC000339.Farnetani, C. G., and A. W. Hofmann (2009), Dynamics andinternal structure of a lower mantle plume conduit, EarthPlanet. Sci. Lett., 282(1?4), 314?322, doi:10.1016/j.epsl.2009.03.035.Farnetani, C. G., and A. W. Hofmann (2010), Dynamics andinternal structure of the Hawaiian plume, Earth Planet. Sci.Lett., 295, 231?240, doi:10.1016/j.epsl.2010.04.005.Garcia, M. O., R. A. Ho, J. M. Rhodes, and E. W. Wolfe(1989), Petrologic constraints on rift-zone processes:Results from episode 1 of the Pu?u ?O?o eruption of K?laueavolcano, Hawai?i, Bull. Volcanol., 52, 81?96.Garcia, M. O., J. M. Rhodes, E. W. Wolfe, G. E. Ulrich, and R.A. Ho (1992), Petrology of lavas from episodes 2?47 of thePu?u ?O?o eruption of K?lauea Volcano, Hawai?i : Evaluationof magmatic processes, Bull. Volcanol., 55(1), 1?16.Garcia, M. O., J. M. Rhodes, F. A. Trusdell, and A. J. Pie-truszka (1996), Petrology of lavas from the Pu?u ?O?o erup-tion of K?lauea Volcano: III The Kupaianaha episode(1986?1992), Bull. Volcanol., 58, 359?379.Garcia, M. O., E. Ito, J. M. Eiler, and A. J. Pietruszka (1998),Crustal contamination of K?lauea Volcano magmas revealedby oxygen isotope analyses of glass and olivine from Pu?u?O?o eruption lavas, J. Petrol., 39, 803?817.Garcia, M. O., A. J. Pietruszka, J. M. Rhodes, and K. Swanson(2000), Magmatic processes during the prolonged Pu?u ?O?oeruption of K?lauea Volcano, Hawai?i, J. Petrol., 41(7),967?990.Garcia, M. O., A. J. Pietruszka, and J. M. Rhodes (2003), Apetrologic perspective of K?lauea Volcano?s summit magmareservoir, J. Petrol., 44(12), 2313?2339, doi:10.1093/petrol-ogy/egg079.Garcia, M. O., E. Ito, and J. M. Eiler (2008), Oxygen isotopeevidence for chemical interaction of K?lauea historical mag-mas with basement rocks, J. Petrol., 49(4), 757?769,doi :10.1093/petrology/egm034.Hanyu, T., J.-I. Kimura, M. Katakuse, A. T. Calvert, T. W. Sis-son, and S. I. Nakai (2010), Source materials for inceptionstage Hawaiian magmas: Pb-He isotope variations for earlyK?lauea, Geochem. Geophys. Geosyst., 11, Q0AC01,doi:10.1029/2009gc002760.Harmon, R. S., and J. Hoefs (1995), Oxygen isotope heteroge-neity of the mantle deduced from global 18O systematics ofbasalts from different geotectonic settings, Contrib. Mineral.Petrol., 120(1), 95?114.Hauri, E. H. (1996), Major-element variability in the Hawaiianmantle plume, Nature, 382, 415?419.Heliker, C., and T. N. Mattox (2003), The first two decades ofthe Pu?u ?O?o -Kupaianaha eruption; chronology andselected bibliography, in The Pu?u ?O?o -Kupaianaha Erup-tion of K?lauea Volcano, Hawai?i: The First 20 Years, editedGREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854871by C. Heliker et al., pp. 1?28, U.S. Geol. Surv. Prof. Pap.1676.Hofmann, A. W., M. D. Feigenson, and I. Raczek (1984), Casestudies on the orgin of basalt : III Petrogenesis of the MaunaUlu eruption, K?lauea, 1969?1971, Contrib. Mineral. Pet-rol., 88, 24?35.Holcomb, R. T. (1987), Eruptive history and long-term behav-ior of K?lauea Volcano, in Volcanism in Hawai?i, edited byR. W. Decker et al., pp. 261?350, U.S. Geol. Surv. Prof. Pap.1350.Jackson, M. G., D. Weis, and S. Huang (2012), Major elementvariations in Hawaiian shield lavas: Source features and per-spectives from global ocean island basalt (OIB) systematics,Geochem. Geophys. Geosyst., 13, Q09009, doi:10.1029/2012GC004268.Johnson, D. J., A. A. Eggers, and M. Bagnardi (2010), Shallowmagma accumulation at K?lauea Volcano, Hawai?i, revealedby microgravity surveys, Geology 38(12), 1139?1142,doi:10.1130/G31323.1.Kimura, J.-I., T. W. Sisson, N. Nakano, M. L. Coombs, and P.W. Lipman (2006), Isotope geochemistry of early K?laueamagmas from the submarine Hilina bench: The nature of theHilina mantle component, J. Volcanol. Geotherm. Res.,151(1?3), 51?72, doi :10.1016/j.jvolgeores.2005.07.024.Kurz, M. D., W. J. Jenkins, and S. R. Hart (1982), Helium iso-topic systematics of oceanic islands and mantle heterogene-ity, Nature, 297, 43?47.Macdonald, G. A., A. T. Abbott, and F. L. Peterson (1983),Volcanoes in the Sea: The Geology of Hawai?i, 517 pp.,Univ. of Hawaii Press, Honolulu, Hawaii.Maclennan, J. (2008), Lead isotope variability in olivine-hosted melt inclusions from Iceland, Geochim. Cosmochim.Acta, 72(16), 4159?4176, doi:10.1016/j.gca.2008.05.034.Marske, J. P., A. J. Pietruszka, D. Weis, M. O. Garcia, and J.M. Rhodes (2007), Rapid passage of a small-scale mantleheterogeneity through the melting regions of K?lauea andMauna Loa volcanoes, Earth Planet. Sci. Lett., 259, 34?50,doi:10.1016/j.epsl.2007.04.026.Marske, J. P., M. O. Garcia, A. J. Pietruszka, J. M. Rhodes,and M. D. Norman (2008), Geochemical variations duringK?lauea?s Pu?u ?O?o eruption reveal a fine-scale mixture ofmantle heterogeneities within the Hawaiian plume, J. Pet-rol., 49(7), 1297?1318, doi:10.1093/petrology/egn025.Montgomery-Brown, E. K., D. K. Sinnett, P. S. M. Poland, T.Orr, H. Zebker, and A. Miklius (2010), Geodetic evidencefor en echelon dike emplacement and concurrent slow slipduring the June 2007 intrusion and eruption at K?lauea vol-cano, Hawai?i, J. Geophys. Res., 115, B07405, doi:10.1029/2009JB006658.Nobre Silva, I. G., D. Weis, and J. S. Scoates (2013), Isotopicsystematics of the early Mauna Kea shield phase and insightinto the deep mantle beneath the Pacific Ocean, Geochem.Geophys. Geosyst., 14(3), 659?676, doi:10.1029/2012GC004373.Pietruszka, A. J., and M. O. Garcia (1999), A rapid fluctuationin the mantle source and melting history of K?lauea Volcanoinferred from the geochemistry of its historical summit lavas(1790?1982), J. Petrol., 48(8), 1321?1342.Pietruszka, A. J., K. H. Rubin, and M. O. Garcia (2001),226Ra-230Th-238U disequilibria of historical K?lauea lavas(1790?1982) and the dynamics of mantle melting within theHawaiian plume, Earth Planet. Sci. Lett., 186, 15?31.Pietruszka, A. J., E. H. Hauri, R. W. Carlson, and M. O. Garcia(2006), Remelting of recently depleted mantle within theHawaiian plume inferred from the 226Ra-230Th-238U disequi-libria of Pu?u ?O?o eruption lavas, Earth Planet. Sci. Lett.,244(1?2), 155?169, doi:10.1016/j.epsl.2006.01.039.Pietruszka, A. J., E. H. Hauri, and J. Blichert-Toft (2009),Crustal contamination of mantle-derived magmas within Pi-ton de la Fournaise Volcano, Reunion Island, J. Petrol.,50(4), 661?684, doi :10.1093/petrology/egp016.Pietruszka, A. J., M. D. Norman, M. O. Garcia, J. P. Marske,and D. H. Burns (2013), Chemical heterogeneity in the Ha-waiian mantle plume from the alteration and dehydration ofrecycled oceanic crust, Earth Planet. Sci. Lett., 361, 298?309, doi:10.1016/j.epsl.2012.10.030.Poland, M., A. Miklius, T. Orr, A. J. Sutton, C. Thornber, andD. Wilson (2008), New episodes of volcanism at K?laueaVolcano, Hawai?i, Eos Trans. AGU, 89(5), 37?38,doi :10.1029/2008EO050001.Poland, M. P., A. Miklius, A. J. Sutton, and C. R. Thornber(2012), A mantle-driven surge in magma supply to K?laueaVolcano during 2003?2007, Nat. Geosci., 5, 295?300,doi :10.1038/NGEO1426.Reiners, P. W. (2002), Temporal-compositional trends inintraplate basalt eruptions: Implications for mantle hetero-geneity and melting processes, Geochem. Geophys. Geo-syst., 3(2), 1011, doi:10.1029/2001GC000250.Ren, Z.-y., T. Hanyu, T. Miyazaki, Q. Chang, H. Kawabata, T.Takahashi, Y. Hirahara, A. R. L. Nichols, and Y. Tatsumi(2009), Geochemical differences of the Hawaiian shieldlavas: Implications for melting process in the heterogeneousHawaiian plume, J. Petrol., 50(8), 1553?573, doi:10.1093/petrology/egp041.Rhodes, J. M., and S. R. Hart (1995), Episodic trace elementand isotopic variations in Mauna Loa lavas: Implications formagma and plume dynamics, in Mauna Loa Revealed:Structure, Composition, History and Hazards, edited by J.M. Rhodes and J. P. Lockwood, pp. 263?288, AGU, Wash-ington, D. C.Rhodes, J. M., and M. J. Vollinger (2004), Composition of ba-saltic lavas sampled by phase-2 of the Hawai?i ScientificDrilling Project: Geochemical stratigraphy and magmatypes, Geochem. Geophys. Geosyst., 5, Q03G13,doi:10.1029/2002GC000434.Rhodes, J. M., S. Huang, F. A. Frey, M. Pringle, and G. Xu(2012), Compositional diversity of Mauna Kea shield lavasrecovered by the Hawai?i Scientific Drilling Project: Infer-ences on source lithology, magma supply, and the role ofmultiple volcanoes, Geochem. Geophys. Geosyst., 13,Q03014, doi:10.1029/2011GC003812.Ryan, M. P. (1987), Elasticity and contractancy of Hawaiianolivine tholeiite and its role in the stability and structuralevolution of subcaldera magma reservoirs and rift systems,in Volcanism in Hawai?i, edited by R. W. Decker et al., pp.1395?1448, U.S. Geol. Surv. Prof. Pap. 1350.Shamberger, P. J., and M. O. Garcia (2007), Geochemicalmodeling of magma mixing and magma reservoir volumesduring early episodes of K?lauea Volcano?s Pu?u ?O?o erup-tion, Bull. Volcanol., 69, 345?352, doi:10.1007/s00445-006-0074-5.Sims, K. W. W., M. T. Murrell, D. J. DePaolo, W. S. Bal-dridge, S. J. Goldstein, D. Clague, and M. Jull (1999), Poros-ity of the melting zone and variations in the solid mantleupwelling rate beneath Hawai?i : Inferences from238U-230Th-226Ra and 235U-231Pa disequilibria, Geochim.Cosmochim. Acta, 63(23), 4119?4138.Sobolev, A. V., A. W. Hofmann, S. Sobolev, and I. K. Nikogo-sian (2005), An olivine-free mantle source of Hawaiianshield basalts, Nature, 434, 590?595.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854872Sutton, J. A., T. Elias, and J. Kauahikaua (2003), Lava-effu-sion rates for the Pu?u ?O?o -Kupaianaha eruption derivedfrom SO2 emissions and very low frequency (VLF) meas-urements, in The Pu?u ?O?o -Kupaianaha Eruption ofK?lauea Volcano, Hawai?i: The First 20 Years, edited byC. Heliker et al., pp. 137?148, U.S. Geol. Surv. Prof. Pap.1676.Thornber, C. R. (2003), Magma-reservoir processes revealedby geochemistry of the Pu?u ?O?o-Kupaianaha eruption, inThe Pu?u ?O?o -Kupaianaha Eruption of K?lauea Volcano,Hawai?i: The First 20 Years, edited by C. Heliker et al., pp.121?136, U.S. Geol. Surv. Prof. Pap. 1676.Thornber, C. R., C. Heliker, D. R. Sherrod, J. P. Kauahikaua,A. Miklius, P. G. Okubo, F. A. Trusdell, J. R. Budahn, W. I.Ridley, and G. P. Meeker (2003), K?lauea East Rift Zonemagmatism: An episode 54 perspective, J. Petrol., 44(9),1525?1559, doi:10.1093/petrology/egg048.Tilling, R. I., and J. J. Dvorak (1993), Anatomy of a basalticvolcano, Nature, 363, 125?133.Vlastelic, I., T. Staudacher, and M. Semet (2005), Rapidchange of lava composition from 1998 to 2002 at Piton de laFournaise (Reunion) inferred from Pb isotopes and trace ele-ments: Evidence for variable crustal contamination, J. Pet-rol., 46(1), 79?107, doi :10.1093/petrology/egh062.Vlastelic, I., A. Peltier, and T. Staudacher (2007), Short-term(1998?2006) fluctuations of Pb isotopes at Piton de la Four-naise volcano (Reunion Island): Origins and constraints onthe size and shape of the magma reservoir, Chem. Geol.,244, 202?220, doi:10.1016/j.chemgeo.2007.06.015.Watson, S., and D. McKenzie (1991), Melt generation byplumes: A study of Hawaiian volcanism, J. Petrol., 32(3),501?537.Weis, D., M. O. Garcia, J. M. Rhodes, A. M. Jellinek, and J. S.Scoates (2011), Role of the deep mantle in generating thecompositional asymmetry of the Hawaiian mantle plume,Nat. Geosci., 4, 831?838, doi:10.1038/ngeo1328.Wolfe, E. W., M. O. Garcia, D. B. Jackson, R. Y. Koyanagi, C.A. Neal, and A. T. Okamura (1987), The Pu?u ?O?o eruptionof K?lauea Volcano, episodes 1?20, January 3, 1983 to June8, 1984, in Volcanism in Hawai?i, edited by R. W. Deckeret al., pp. 471?508, U.S. Geol. Surv. Prof. Pap. 1350.Wright, T. L. (1971), Chemistry of K?lauea and Mauna Loalava in space and time, U.S. Geol. Surv. Prof. Pap., 735,1?45.GREENE ET AL. : K?LAUEA?S PU?U ?o?o ERUPTION (1983?2010) 10.1002/ggge.202854873 1  Sample Preparation and Analytical Methods  A1. University of Massachusetts XRF Analytical Methods New XRF analyses are given in the Auxiliary Materials (ts1) with previous major-element analyses. New XRF trace element analyses are also given (and indicated in table) for samples erupted prior to 1998, when a new, more precise XRF instrument became available. Fresh pieces of quenched lava were crushed in a tungsten carbide jaw crusher to ~2 mm diameter and cleaned in a beaker with deionized water and dried in an oven at 70?C for 24 hours to drive off excess water. Lava fragments were then powdered using a Rocklabs WC shatterbox for 1 ? 2 minutes. All XRF analyses were performed at the University of Massachusetts XRF Laboratory where whole-rock, major-element abundances were measured using the procedures of Rhodes and Vollinger [2004]. One sigma accuracy and precision estimates for the XRF data are ~0.5% for major elements [Rhodes, 1988]. Five grams of a powdered sample were heated in a muffle furnace at 1020?C for 10 minutes to limit the amount of ferrous iron formation [Rhodes and Vollinger, 2004]. The weight loss is LOI. Major elements were measured on a fused La-bearing lithium borate glass disc using a Siemens MRS-400 spectrometer with a Rh X-ray tube operating at 2700 W (the voltage was 45 kV and 60mA and the count times were 200 seconds). Trace element concentrations (Rb, Sr, Ba, La, Ce, Nb, Zr, Y, Zn, Ga, Ni, Cr, V) were measured on a separate powder pellet using a Philips PW2400 sequential spectrometer with a Rh X-ray tube operating at 3000 W (the voltage was 60kV and 50mA and 2.5 hour total count time). Precision and accuracy estimates for the trace element data are described by Rhodes [1996] and Rhodes and Vollinger [2004]. Results for each sample are the average of two separate analyses (2 disks for one sample) for major elements, but one analysis (one pellet run once) for trace elements. A summary of XRF trace element values of USGS standards analyzed with lavas from K?lauea?s Pu?u ???? eruption since 2002 are given in the Auxiliary Materials (ts3).  A2. PCIGR and SDSU Trace Element and Isotopic Analytical Methods Forty-one Pu?u ???? lava samples from 1983 to 1999 were selected for Pb and Sr isotopic analysis at the Pacific Centre for Isotopic and Geochemical Research (PCIGR) at the University of British Columbia (UBC; Table 2). The analytical methods for thirteen additional Pb isotopic analyses of Pu?u ???? lavas collected between 2006 and 2010, and fourteen analyses from Marske et al. [2008] that are used in this study, are described in Marske et al. [2007] (samples 2  are indicated in Table 2). Thirteen samples without trace-element data were selected for high-precision trace-element analysis and results are shown in Auxiliary Materials (ts2) with previous trace-element analyses for Pu?u ???? lavas (corrected data is also shown, as described in table caption). Isotopic ratios of Pb for these samples were measured with a Nu Plasma (Nu Instruments) 1700 Multiple Collector-Inductively Coupled Plasma-Mass Spectrometer (MC-ICP-MS) at San Diego State University (SDSU). Sr isotopic ratios for 1998-2005 lava samples from Marske et al. [2008] were measured using a VG Sector 54 thermal ionization mass spectrometer (TIMS) at SDSU. Lava samples from 2006 to 2010 were analyzed for Sr isotopic ratios at the PCIGR, according the the methods described below.  Samples were prepared for trace-element analysis at the PCIGR by the technique described by Pretorius et al. [2006] and Carpentier et al. [2013] on unleached lava fragments. Lava fragments <2 mm in diameter (~100 mg) were weighed in 7 mL screw-top Savillex? beakers and dissolved in 1 mL ~14N HNO3 and 5 mL 48% HF on a hotplate for 48 hours at 130?C with periodic ultrasonication. Samples were dried and redissolved in 6 mL 6N HCl on a hotplate for 24 hours and then dried and redissolved in 1 mL concentrated HNO3 for 24 hours before final drying. Trace element abundances were measured with a Thermo Finnigan Element2 High Resolution-Inductively Coupled Plasma-Mass Spectrometer (HR-ICP-MS) within 24 hours of redissolution, following the procedures described by Pretorius et al. [2006] and Carpentier et al. [2013]. High field strength elements (HFSE) and large ion lithophile elements (LILE) were measured in medium resolution mode at 2000x dilution using a PFA teflon spray chamber washed with Aqua Regia for 3 minutes between samples. Rare earth elements (REE) were measured in high resolution mode, and U and Pb in low resolution mode, at 2000x dilution using a glass spray chamber washed with 2% HNO3 between samples. Total procedural blanks and reference material (Kil93) was analyzed with the batch of samples. Indium was used as an internal standard in all samples and standard solutions. Background and standard solutions were analyzed after every 5 samples to detect memory effects and mass drift.  For isotope chemistry at PCIGR, fresh pieces of crushed lava (<2mm diameter; ~0.6-0.8 g total) were carefully picked using a binocular microscope to avoid signs of post-eruptive alteration. 3  Each sample was ultra-sonicated (15 minutes) and rinsed three times in ultra-pure water in a clean Savillex? beaker prior to rinsing with room temperature 0.5 M HBr for 15 seconds to remove any surface contamination. The sample was washed again in ultra-pure water, dried, and weighed. Sample digestion for purification of Pb and Sr by column chemistry involved dissolving ~100-250 mg of the cleaned, crushed lava in 1 mL ~14N HNO3 and 10 mL 48% HF on a hotplate for 48 hours at 130?C with periodic ultrasonication. Samples were dried and redissolved in 6 mL 6N HCl on a hotplate for 24 hours and then dried. Pb was separated and purified in two passes through anion exchange columns and the discard was used for Sr separation. Detailed procedures for column chemistry for separating Pb and Sr at the PCIGR are described in Weis et al. [2006] and Nobre Silva et al. [2009]. Sr isotope ratios were measured on a Thermo Finnigan Triton Thermal Ionization Mass Spectrometer (TIMS) in static mode with relay matrix rotation on a single Ta and double Re-Ta filament, respectively. Four to 5 filaments per barrel of 21 were occupied by standard NIST SRM 987 for each barrel where samples were run. Sample Sr isotopic compositions were corrected for mass fractionation using 86Sr/88Sr = 0.1194. Each sample was then normalized using the barrel average of the reference material relative to the values of 87Sr/86Sr = 0.710248 [Weis et al., 2006]. During the period when the samples were analyzed, the NIST SRM 987 standard gave an average of 0.710243 ? 13 (n = 18; 2? error is reported as times 106). PCIGR internal reference material from the K?lauea summit eruption in 1919 (Kil1919) was processed with the samples and yielded Sr isotopic ratios of 0.703468 ? 8. Twenty-six previous analyses of Kil1919 at SDSU average 0.703471 ? 13. Four chemistry duplicates of Menehune standard (an in-house glass standard collected from a Pu?u ???? lava flow of K?lauea Volcano that was quenched on June 24, 2006), processed with the other lava samples, had Sr isotopic compositions of  0.703608 ? 7, 0.703605 ? 8, 0.703611 ? 8, and 0.703624 ? 9. Twenty-six previous analyses of Menehune standard performed at SDSU averaged 0.703617 ? 13.  Pb isotopic compositions at PCIGR were analyzed by static multi-collection on a Nu Plasma MC-ICP-MS. The detailed analytical procedure for Pb isotopic analyses at the PCIGR is described in Weis et al. [2005]. The configuration for Pb analyses allows for collection of Pb, Tl, and Hg together. Tl and Hg are used to monitor instrumental mass discrimination and isobaric overlap, repectively. All sample solutions were analyzed with approximately the same Pb/Tl 4  ratio (~4) as the reference material NIST SRM 981. To accomplish this, a small aliquot of each sample solution from the Pb columns was analyzed on the Nu Plasma to determine the amount of Pb available. The SRM 981 standard was run after every two samples on the Nu Plasma. During the time samples were run, analyses of the SRM 981 Pb reference material gave values of 206Pb/204Pb = 16.9430 ? 19, 207Pb/204Pb = 15.4997 ? 18, and 208Pb/204Pb = 36.7221 ? 58 (n = 38; 2? error is reported as times 104); these values are in excellent agreement with reported TIMS triple-spike values of Galer and Abouchami [1998]. Results were further corrected by the sample-standard bracketing method described by White et al. [2000]. Reference material Kil1919 analyzed at PCIGR yielded Pb isotopic ratios of 206Pb/204Pb = 18.6556 ? 8, 207Pb/204Pb = 15.4924 ? 8, and 208Pb/204Pb = 38.2148 ? 21. Thirty-three analyses of Kil1919 performed at SDSU over the course of this study averaged 206Pb/204Pb = 18.6552 ? 27, 207Pb/204Pb = 15.4897 ? 20, and 208Pb/204Pb = 38.2068 ? 57. Four chemistry duplicates of Menehune standard analyzed at PCIGR averaged 206Pb/204Pb = 18.4062 ? 28, 207Pb/204Pb = 15.4724 ? 26, and 208Pb/204Pb = 38.0662 ? 59. Sixty-eight previous analyses of Menehune standard performed at SDSU over the course of this study averaged 206Pb/204Pb = 18.4073 ? 16, 207Pb/204Pb = 15.4714 ? 18, and 208Pb/204Pb = 38.0627 ? 60.  A3. Glass and matrix oxygen isotopes Glass and holocrystalline matrix material were separated from Pu?u ???? lavas by hand picking from coarsely crushed but otherwise untreated samples. Oxygen isotope compositions of c. 2 mg aliquots of these separates were determined by laser fluorination, using a 50W CO2 laser and BrF5 as reagent [Sharp, 1990; Valley et al., 1995]. Product O2 was converted to CO2 by reaction with hot graphite; CO2 was then analyzed for its isotopic composition by dual-inlet gas source mass spectrometry on a Thermo Finnegan Delta XL gas source isotope ratio mass spectrometer at California Institute of Technology. Data are reported in units of per mil versus the VSMOW standard. Analyses were standardized by comparison with measurements of Gore Mountain garnet standard [Valley et al., 1995]. This standard was analyzed between two and five times each day of analyses, and the data for unknowns analyzed on that day were corrected by the average difference between measured and accepted values for that standard. The external precision of repeat measurements of separate splits of unknown samples averaged 0.05? (1?). Seven analyses of Gore Mtn. Garnet were run concurrently with these samples, with a standard 5  deviation of ?0.06 ? (1?; averaged over all days).  This is comparable with the typical external precision for replicate measurements of silicate standards for this laboratory and technique [e.g., Bindeman et al., 2006; Eiler et al., 1995; Eiler et al., 1996], suggesting that the glass separates analyzed in this study are homogeneous in ?18O at the scale of c. 2 mg aliquots. The results reported here are relative to ?18OVSMOW [Coplen, 1988]. Four analyses of Menehune standard performed with the other Pu?u ???? lavas yielded 5.22 ? 0.03 ? (1?).  References  Bindeman, I. N., O. Sigmarsson, and J. M. Eiler (2006), Time constraints on the origin of large volume basalts derived from O-isotope and trace element mineral zoning and U-series disequilibria in the Laki and Grimsvotn volcanic system, Earth Planet. Sci. Lett., 245, 245-259. Carpentier, M., D. Weis, and C. Chauvel (2013), Large U loss during weathering of upper continental crust: The sedimentary record. , Chemical Geology (Isotope Geosciences Section), 340, 91-104. Coplen, T. B. (1988), Normalization of oxygen and hydrogen isotope data, Chemical Geology (Isotope Geosciences Section)  72, 293-297. Eiler, J. M., K. A. Farley, J. W. Valley, E. M. Stolper, E. H. Hauri, and H. Craig (1995), Oxygen isotope evidence against bulk recycled sediment in the mantle sources of Pitcairn Island lavas, Nature, 377, 138-141. Eiler, J. M., J. W. Valley, and E. M. Stolper (1996), Oxygen isotope ratios in olivine from the Hawaii Scientific Drilling Project, J. Geophys. Res., 101(B5), 11,807-11,813. Galer, S. J. G., and W. Abouchami (1998), Practical application of lead triple spiking for correction of instrumental mass discrimination, Mineral. Mag., 62A, 491-492. Marske, J. P., A. J. Pietruszka, D. Weis, M. O. Garcia, and J. M. Rhodes (2007), Rapid passage of a small-scale mantle heterogeneity through the melting regions of Kilauea and Mauna Loa volcanoes, Earth Planet. Sci. Lett., 259, 34-50. Marske, J. P., M. O. Garcia, A. J. Pietruszka, J. M. Rhodes, and M. D. Norman (2008), Geochemical variations during Kilauea's Pu'u 'O'o eruption reveal a fine-scale mixture of mantle heterogeneities within the Hawaiian plume, J. Petrol., 49(7), 1297-1318. Nobre Silva, I. G., D. Weis, J. Barling, and J. S. Scoates (2009), Leaching systematics and matrix elimination for the determination of high-precision Pb isotope compositions of ocean island basalts, Geochem. Geophys. Geosyst., 10(8), Q08012. Pretorius, W., D. Weis, G. Williams, D. Hanano, B. Kieffer, and J. S. Scoates (2006), Complete trace elemental characterization of granitoid (USGSG-2,GSP-2) reference materials by high resolution inductively coupled plasma-mass spectrometry, Geost. and Geoanaly. Res., 30(1), 39-54. Rhodes, J. M. (1988), Geochemistry of the 1984 Mauna Loa eruption: Implications for magma storage and supply, J. Geophys. Res., 93, 4,453-454,466. Rhodes, J. M., and M. J. Vollinger (2004), Composition of basaltic lavas sampled by phase-2 of the Hawaii Scientific Drilling Project: Geochemical stratigraphy and magma types, Geochem. Geophys. Geosyst., 5(3). 6  Sharp, Z. D. (1990), A laser-based microanalytical method for the in situ determination of oxygen isotope ratios of silicates and oxides, Geochim. Cosmochim. Acta, 54(5), 1353-1357. Valley, J. W., N. Kitchen, and M. J. Kohn (1995), Strategies for high-precision oxygen isotope analysis by laser fluorination, Geochimica Cosmoschimica Acta, 59, 5223-5231. Weis, D., B. Kieffer, C. Maerschalk, W. Pretorius, and J. Barling (2005), High-precision Pb-Sr-Nd-Hf isotopic characterization of USGS BHVO-1 and BHVO-2 reference materials, Geochem. Geophys. Geosyst., 6, (Q02002). Weis, D., B. Kieffer, C. Maerschalk, J. Barling, J. de Jong, G. A. Williams, D. Hanano, N. Mattielli, J. S. Scoates, A. Goolaerts, R. A. Friedman, and J. B. Mahoney (2006), High-precision isotopic characterization of USGS reference materials by TIMS and MC-ICP-MS, Geochem. Geophys. Geosyst., 7(Q08006). White, W. M., F. Albar?de, and P. T?louk (2000), High-precision analysis of Pb isotope ratios by multi-collector ICP-MS, Chem. Geol., 167, 257-270.     Temporal Geochemical Variations in Lavas from K?lauea?s Pu?u ???? Eruption (1983-2010): Cyclic Variations from Melting of Source Heterogeneities  Andrew R. Greene Dept. of Natural Sciences, Hawai?i Pacific University, K?ne`ohe, HI, 96744 USA  Michael O. Garcia  Dept. of Geology and Geophysics, University of Hawai?i, Honolulu, HI, 96822 USA  Aaron J. Pietruszka  Dept. of Geological Sciences, San Diego State University, San Diego, CA, 92182 USA   Dominique Weis Pacific Centre for Isotopic and Geochemical Research, Dept. of Earth and Ocean Sciences, University of British Columbia, Vancouver, BC, V6T 1Z4 Canada  Jared P. Marske Dept. of Terrestrial Magnetism, Carnegie Institution, Washington, D.C., 20015 USA   Michael J. Vollinger Ronald B. Gilmore XRF Lab, Dept. of Geosciences, University of Massachusetts, Amherst, MA, 01003 USA  John Eiler Planetary and Geological Sciences Institute, California Institute of Technology, Pasadena, CA, 91125 USA    Five supplementary files include descriptions of analytical methods, a full table of XRF major and trace element data, a full table of ICP-MS trace element data, a summary of XRF trace element values of USGS standards analyzed with Pu?u ???? lavas since 2002, and a list of references for data used in figures 11 and 12. The tables are Microsoft Excel data files. The name and contents of individual files are listed below.   2013G3Greene-methods  Sample Preparation and Analytical Methods   2013G3Greene-ts1-XRF_major_trace_elements  XRF Major Element (wt% oxide) and Trace Element (ppm) Abundances in Lavas from K?lauea?s Pu?u ???? Eruption (1983-2010)  2013G3Greene-ts2-ICP-MS_trace_elements  ICP-MS Trace Element (ppm) Abundances in Lavas from K?lauea?s Pu?u ???? Eruption (1983-2010)  2013G3Greene-ts3-XRF_trace_USGS-standards  Summary of XRF Trace Element Values of USGS Standards Analyzed with Lavas from K?lauea?s Pu?u ???? Eruption Since 2002  2013G3Greene-references-figs11-12  List of References for Data Used in Figures 11 and 12     List of References for Data Used in Figures 11 and 12  Mauna Loa  Abouchami, W., S. J. G. Galer, and A. W. Hofmann (2000), High precision lead isotope systematics of lavas from the Hawaiian Scientific Drilling Project, Chem. Geol., 169, 187-209. Abouchami, W., A. W. Hofmann, S. J. G. Galer, F. A. Frey, J. Eisele, and M. Feigenson (2005), Lead isotopes reveal bilateral asymmetry and vertical continuity in the Hawaiian mantle plume, Nature, 434, 851-856, doi:10.1038/nature03402.  Blichert-Toft, J., D. Weis, C. Maerschalk, A. Agranier, and F. Albar?de (2003), Hawaiian hot spot dynamics as inferred from the Hf and Pb isotope evolution of Mauna Kea volcano, Geochem. Geophys. Geosyst., 4, 2, 1-27, doi:10.1029/2002GC000340.  Kurz, M. D., J. Curtice, D. E. Lott, and A. Solow (2004), Rapid helium isotopic variability in Mauna Kea shield lavas from the Hawaiian Scientific Drilling Project, Geochem. Geophys. Geosyst., 5, 4, 1-35, Q04G14, doi:10.1029/2002GC000439.  Weis, D., M. O. Garcia, J. M. Rhodes, A. M. Jellinek, and J. S. Scoates (2011), Role of the deep mantle in generating the compositional asymmetry of the Hawaiian mantle plume, Nature Geoscience, 4  831?838, doi:10.1038/ngeo1328.  Mauna Kea  Eisele, J. R., W. Abouchami, S. J. G. Galer, and A. W. Hofmann (2003), The 320 kyr Pb isotope evolution of Mauna Kea lavas recorded in the HSDP-2 drill core, Geochem. Geophys. Geosyst., 4, 5, doi:10.1029/2002GC000339.  Bryce, J. G., D. J. DePaolo, and J. C. Lassiter (2005), Geochemical structure of the Hawaiian plume: Sr, Nd, and Os isotopes in the 2.8 km HSDP-2 section of Mauna Kea volcano, Geochem. Geophys. Geosyst., 6, Q09G18, doi:10.1029/2004GC000809.  K?lauea  Abouchami, W., A. W. Hofmann, S. J. G. Galer, F. A. Frey, J. Eisele, and M. Feigenson (2005), Lead isotopes reveal bilateral asymmetry and vertical continuity in the Hawaiian mantle plume, Nature, 434, 851-856, doi:10.1038/nature03402.  Chen, C.-Y., F. A. Frey, J. M. Rhodes, and R. M. Easton (1996), Temporal geochemical evolution of Kilauea Volcano: Comparison of Hilina and Puna basalt, in Earth Processes, Reading the Isotopic Code, edited by A. Basu and S. R. Hart, pp. 161-181, AGU.   Hanyu, T., J.-I. Kimura, M. Katakuse, A. T. Calvert, T. W. Sisson, and S. i. Nakai (2010), Source materials for inception stage Hawaiian magmas: Pb-He isotope variations for early Kilauea, Geochem. Geophys. Geosyst., 11, Q0AC01, doi:10.1029/2009gc002760.  Kimura, J.-I., T. W. Sisson, N. Nakano, M. L. Coombs, and P. W. Lipman (2006), Isotope geochemistry of early Kilauea magmas from the submarine Hilina bench: The nature of the Hilina mantle component, J. Volcanol. Geotherm. Res., 151, 1-3, 51-72.  Marske, J. P., A. J. Pietruszka, D. Weis, M. O. Garcia, and J. M. Rhodes (2007), Rapid passage of a small-scale mantle heterogeneity through the melting regions of Kilauea and Mauna Loa volcanoes, Earth Planet. Sci. Lett., 259, 34-50.  Marske, J. P., M. O. Garcia, A. J. Pietruszka, J. M. Rhodes, and M. D. Norman (2008), Geochemical variations during Kilauea's Pu'u 'O'o eruption reveal a fine-scale mixture of mantle heterogeneities within the Hawaiian plume, J. Petrol., 1-22, doi:10.1093/petrology/egn025.  Pietruszka, A. J., and M. O. Garcia (1999), A Rapid Fluctuation in the Mantle Source and Melting History of Kilauea Volcano Inferred from the Geochemistry of its Historical Summit Lavas (1790-1982), J. Petrology, 40, 8, 1321-1342, doi:10.1093/petroj/40.8.1321.  Loih?i  Garcia, M. O., D. J. P. Foss, H. B. West, and J. J. Mahoney (1995), Geochemical and isotopic evolution of Loihi Volcano, Hawaii, J. Petrol., 36, 6, 1647-1674.  Abouchami, W., A. W. Hofmann, S. J. G. Galer, F. A. Frey, J. Eisele, and M. Feigenson (2005), Lead isotopes reveal bilateral asymmetry and vertical continuity in the Hawaiian mantle plume, Nature, 434, 851-856, doi:10.1038/nature03402.   

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.52383.1-0075983/manifest

Comment

Related Items