Petrogenesis of Tholeiitic Lavas from the Submarine Hana Ridge

Petrogenesis of Tholeiitic Lavas from theSubmarine Hana Ridge, Haleakala Volcano,Hawaii

ZHONG-YUAN REN1*, EIICHI TAKAHASHI1, YUJI ORIHASHI2 ANDKEVIN T. M. JOHNSON3

1EARTH AND PLANETARY SCIENCES, TOKYO INSTITUTE OF TECHNOLOGY, 2-12-1 OOKAYAMA, MEGUROKU,

152-8551, JAPAN

2EARTHQUAKE RESEARCH INSTITUTE, THE UNIVERSITY OF TOKYO, 1-1 YAYOI, BUNKYO-KU, TOKYO 113-0032,

JAPAN

3DEPARTMENT OF GEOLOGY AND GEOPHYSICS, UNIVERSITY OF HAWAII, HONOLULU, HI 96822, USA

RECEIVED JUNE 2, 2003; ACCEPTED JUNE 8, 2004ADVANCE ACCESS PUBLICATION AUGUST 12, 2004

Hana Ridge, the longest submarine rift zone in the Hawaiian island

chain, extending from Maui 140 km to the ESE, has a complex

morphology compared with other Hawaiian rift zones. A total of 108

rock specimens have been collected from the submarine Hana Ridge by

six submersible dives. All of the rocks (76 bulk rocks analyzed) are

tholeiitic basalts or picrites. Their major element compositions,

together with distinctively low Zr/Nb, Sr/Nb, and Ba/Nb, overlap

those of Kilauea lavas. In contrast, the lavas forming the subaerial

Honomanu shield are intermediate in composition between those of

Kilauea and Mauna Loa. The compositional characteristics of the

lavas imply that clinopyroxene and garnet were important residual

phases during partial melting. The compositions of olivine and glass

( formerly melt) inclusions imply that regardless of textural type

(euhedral, subhedral–undeformed, deformed) olivine crystallized

from host magmas. Using the most forsteritic olivine (Fo90�6) and

partition coefficients Kol�meltDFe�Mg and D�ol�melt

CaO , the primary magma

composition is constrained to have �16�7% MgO and �8�4 wt %

CaO. Modeling calculations using MELTS show that olivine first

crystallized at 1380–1390�C and 0�1–0�3 GPa, under slightly

hydrous conditions (0�5–1 wt % water).

KEY WORDS: Hawaii; Haleakala volcano; submarine Hana Ridge;

petrogenesis; tholeiitic lavas

INTRODUCTION

Haleakala volcano on Maui Island, the third largest vol-cano of the youngest edifices formed by the Hawaiian

hotspot, has a well-developed shield that is overlapped bypost-shield and rejuvenated stages. Four formations havebeen recognized in Haleakala volcano (Stearns &Macdonald, 1942; Macdonald, 1978) from the oldest toyoungest: Honomanu Formation, shield-building tho-leiite and minor interbedded alkalic basalts; KumuiliahiFormation, transitional alkali basalts and hawaiite;Kula-postshield Formation, alkali-series lavas; Hana For-mation, rejuvenated alkalic lavas. However, the shield-building Honomanu lavas are exposed only along thenorthern coast and in deep valleys on the northern andsouthern flanks of Haleakala. Most of the subaerial shieldis covered by postshield and rejuvenated stages alkaliclavas. The exposed Honomanu lavas represent only thefinal stage of Haleakala shield building, and their compo-sition may have already shifted to transitional and alkalicbasalts (Chen & Frey, 1985; Chen et al., 1990, 1991).

Many studies have been carried out on the subaeriallavas of Haleakala volcano (e.g. Stearns & Macdonald,1942; Chen & Frey, 1983, 1985; Macdonald et al. 1983;West & Leeman, 1987, 1994; Chen et al., 1990, 1991;Sherrod et al., 2003). However, until very recently, littlework has been done on the submarine portion ofHaleakala volcano. Except for a few dredges on thesubmarine Hana Ridge (the submarine East Rift Zoneof Haleakala volcano; Moore et al., 1990), the mainshield-building stage of the huge Haleakala volcano hasbeen relatively unstudied. Moore et al. (1990) mapped thesubmarine Hana Ridge with GLORIA, and discussed

*Corresponding author. Telephone: +81-3-5734-2338. Fax: +81-3-

5734-3538. E-mail: [email protected]

Journal of Petrology 45(10) # Oxford University Press 2004; all rights

reserved

JOURNAL OF PETROLOGY VOLUME 45 NUMBER 10 PAGES 2067–2099 2004 doi:10.1093/petrology/egh076D

ownloaded from

https://academic.oup.com

/petrology/article-abstract/45/10/2067/1577187 by guest on 18 Novem

ber 2018

subsidence, the age of the slope, and measured glass rimcompositions of some dredged samples. Only a few geo-chemical analyses have been reported for the HanaRidge dredged samples (e.g. Chen et al., 1991; Wagneret al., 1998). Clague et al. (2000) discussed the formation ofsubmarine flat-topped volcanic cones including somefrom Hana Ridge. A detailed bathymetric map of thesubmarine Hana Ridge was acquired for the first timeduring the JAMSTEC Cruise in 1999 by Smith et al.(2002), who discussed its structure, volcanic morphology,and processes of construction.During the joint Japan–US Hawaii cruises in 2001 and

2002, six dives were carried out on the submarine HanaRidge covering its deepest portion (5300m deep) to theshallow ridge crest (2200m deep) (see Fig. 1). This paperreports the petrology and geochemistry of 108 rock speci-mens recovered from the submarine Hana Ridge, repre-senting the main shield-building stage of Haleakalavolcano. A significant new result is that the lavas forming

Haleakala’s main shield stage are very similar to the lavasforming Kilauea volcano in major and trace elementcomposition. In contrast, the lavas forming the subaerialHonomanu portion are intermediate in compositionbetween those of Kilauea and Mauna Loa. We alsoestimate the primary magma composition using the par-tition coefficients K ol-melt

DFe-Mg and D *ol-meltCoa using the most

forsteritic olivine. The compositional characteristics ofthe submarine Hana Ridge lavas imply that clinopyrox-ene and garnet were important residual phases duringpartial melting in the magma source region.

SAMPLE LOCALITY AND

PETROGRAPHY

The 140 km long submarine Hana Ridge (Fig. 1), extend-ing from the SE corner of Maui Island (Hana village), isthe largest identified rift zone in Hawaii (Moore et al.,

Fig. 1. Bathymetric map of the Hana Ridge showing locations of the Kaiko dives in 2001 (K212, K214, and K216) and Shinkai dives in 2002 (S686,S687, and S691).

2068

JOURNAL OF PETROLOGY VOLUME 45 NUMBER 10 OCTOBER 2004D

ownloaded from



ber 2018

1990). Hana Ridge is twice as long and nearly twice aswide as Puna Ridge, the submarine east rift zone ofKilauea volcano. Compared with other submarine rifts,Hana Ridge has complex topography comprising at leastthree partially overlapped ridge axes and a large (22 kmwide) arcuate structure at its east end. The shallowerportion of the Hana Ridge with smooth topography isthe subsided subaerial volcanic edifice capped by a seriesof reefs (Moore et al., 1990). Along the western end of theHana Ridge, a series of point-topped cones, interpretedas postshield-stage alkalic lavas, are located close to theshore (1000–1400m water depth) (Reynolds et al., 1998;Clague et al., 2000). Farther down the rift, there arenumerous flat-topped pancake cones composed oftholeiitic lavas, probably erupted during the shield-build-ing stage (Clague et al., 2000). Smith et al. (2002) identified73 volcanic constructions along the broad 8–13 km widecrest of the distal rift zone (see Smith et al., 2002, fig. 5c).Chen et al. (1991) studied a stratigraphic section of

Haleakala volcano. This 250m section of Honomanubasalts exposed in Honomanu Gulch consists of interca-lated tholeiitic and alkalic lavas with K–Ar ages from�1�1 to 0�97Ma. These ages may represent the end ofthe Haleakala shield stage; the submarine Hana Ridgelavas would be older than the Honomanu basalts.During the 2001 and 2002 JAMSTEC Cruises, six

dives (K212, K214, K216, S686, S687, and S691) wereconducted on the Hana Ridge. Local bathymetric mapsof each dive site are summarized in Figs 1 and 2. Twodives were carried out inside the arcuate structure (K212and S686), and three dives were conducted on the flankof the Hana Ridge (K214, K216 and S691). In addition,two cones formed on the shoulder of the Hana Ridgewere visited by dive S687. Dives K216, S687 and S691were carried out on the northernmost rift segment,whereas K214 was on the southern rift.The lavas from the submarine Hana Ridge vary dra-

matically in mineralogy, ranging from weakly phyric(<2 vol. % phenocrysts) to extremely olivine phyric (upto 40 vol.%). Themodalmineral compositions of the lavasare variable, but there are only four phenocryst assem-blages: (1) olivine; (2) olivine þ augite; (3) olivine þaugite þ plagioclase; (4) olivine þ augite þ plagioclase þorthopyroxene. Olivine occurs as phenocrysts and micro-phenocrysts in all the samples. The population of olivinephenocrysts was subdivided into three types: euhedral(>1mm wide with euhedral shape); subhedral–undeformed (>1mm wide with subhedral shape but nodeformation); deformed (>1mm wide with subhedralshape, at least one subgrain boundary or resorbed mar-gins). Microphenocrysts are 0�1–1�0mmwide and usuallyeuhedral in shape.Phenocrysts other than olivine are usually <1 vol. %.

Some clinopyroxene forms glomeroporphyritic clusterswith plagioclase. Plagioclase is generally subhedral to

euhedral, although rarely, rounded or embayed crystalsare present. Both types of plagioclase occur as glomero-crysts with augite. Some plagioclase contains glass (for-merly melt) inclusions. Many of these glass inclusions alsocontain gas bubbles or spinel crystals, or both. Diametersof the melt inclusions range from several to several hun-dred micrometers. Orthopyroxene is rare, elongated andembayed, and some orthopyroxene grains host plagio-clase inclusions.Quenched glassy rims are found on some pillow lavas.

Normally, the texture of the groundmass changes gradu-ally from glass, microlites of plagioclase and hyalopilitictexture to intersertal or intergranular in the pillow inter-ior. Also, the color of the glass changes from yellow–brown (sideromelane) in pillow rims to black (tachylite)in the pillow interior. The groundmass is commonlycomposed of magnetite, plagioclase, clinopyroxene andglass (black, tachylite), and has hyalopilitic, hyalophitic,intersertal, or intergranular textures.

ANALYTICAL TECHNIQUES

For the bulk-rock chemical analysis, least altered rockswere chosen after thin-section inspection. The sampleswere wrapped with plastic film and broken into chips ofseveral millimeters size using a hydraulic press, and thenthe chips were carefully handpicked to avoid weatheredparts. The fresh chips were washed in deionized waterand reduced to powder using an agate mortar and pestle,and an agate ring mill. The powders were placed in aglass bottle with deionized water in an ultrasonic bath for10min, and then the water was removed carefully byhand after the particles of the powder had settled. Thiscycle was repeated three times. Rock powder (400mg,dried at 110�C for 24 h) was mixed with 4 g of Li2B4O7

flux and fused at 1200�C for 7min in a Pt crucible foranalysis of major elements. Because the fusion takes placein air, the samples were oxidized, so total iron is reportedas Fe2O3. The whole-rock major element compositionswere determined by X-ray fluorescence (XRF) with aRigaku-3550 spectrometer equipped with a Rh X-raytube at the Tokyo Institute of Technology. Analyticalprecision for the XRF analyses is as described byShinozaki et al. (2002).For analysis of trace elements, XRF and laser ablation

inductively coupled plasma source mass spectrometry(LA-ICP-MS) at the Earthquake Research Institute, theUniversity of Tokyo, were used. A split of 1�8 g of rockpowder, prepared as above, was mixed with 3�6 g oflithium metaborate–tetraborate flux and 0�54 g of lithiumnitrate and fused at 1200�C to make a glass bead. Majorand some trace elements were analyzed by XRF accord-ing to the procedure described by Tani et al. (2002). Ba,Ta, Hf, U, Th, Pb and rare earth elements (REE) were

2069

REN et al. HANA RIDGE LAVAS, HAWAIID

ownloaded from



ber 2018

determined by LA-ICP-MS following the proceduresdescribed by Orihashi & Hirata (2003). Analytical uncer-tainty is 10% for Gd, <5% for other elements; for all ofthe trace elements replicate analyses were within 3–5%difference.Compositions of olivine, glass rims on pillow lavas, and

glassy inclusions enclosed in olivine phenocrysts wereanalyzed by electron probe microanalysis (EPMA) witha JEOL-8800 instrument at the Tokyo Institute of Tech-nology. A defocused electron beam (3 mm for glassy

inclusions, 10mm for pillow glass rim) was used for glassanalyses. The beam current was 1�20 � 10�8 A. Count-ing times were 20 s for major elements (Si, Al, Ca, Mgand Fe) and 30–60 s for minor elements (Mn, P, K, Ti,and S). Na was analyzed first in each analysis for 10 s tominimize its possible loss during analysis. To monitormachine drift, an internal glass standard ( JB-2) wasanalyzed before and after each batch analysis. Forminor elements in olivine (i.e. Ca, Ni and Mn), a longercounting time (50–60 s) was used. Analytical uncertainty

Fig. 2. Dive track maps and sample localities. Dives K212, K214, and K216 were carried out with R.O.V. Kaiko in 2001, and dives S686, S687,and S691 were carried out with manned submersible Shinkai 6500 in 2002. Bathymetric map is by SEABEAM-2112 on research vessels Kairei andYokosuka of JAMSTEC.

2070


ownloaded from



ber 2018

is 1–2% for major elements, and 5–10% for minorelements.

RESULTS

Olivine compositions

By EPMA we analyzed 31, 20, and 21 core compositionsof olivine with euhedral, subhedral–undeformed, anddeformed textures, respectively (Table 1, Fig. 3a and c).In addition, we analyzed the compositions of cores fromabout 350 unclassified olivine phenocrysts (given in theElectronic Appendix, which may be downloaded fromhttp://www.petrology.oupjournals.org; Fig. 3b). Olivinehas a wide range of core compositions: the Mg/(Mg þFe) (Mg-number) (cation ratio) ranges from 0�783 to0�906 (Fig. 3b), and CaO ranges from 0�14 to 0�33 wt %(Fig. 3b). The CaO contents of olivine from the submar-ine Hana Ridge lavas are very similar to those of Kilauea(Norman & Garcia, 1999), and significantly differentfrom those in Hawaiian xenoliths (normally, CaO con-tents from Hawaiian xenoliths are <0�15 wt %; Sen,1988; Garcia et al., 1995; Norman & Garcia, 1999)(Fig. 3a and b). The NiO contents of olivine range from0�16 to 0�69 wt %. With increasing Mg-number, NiOcontents of olivine increase. These trends in CaO (wt %)and NiO (wt %) of olivine phenocrysts as a function oftheir Mg-number show no apparent correlation with thecrystal type (see Fig. 3a and c).

Major element compositions of bulk rocksand pillow glass rims

All of the submarine Hana Ridge lavas analyzed in thisstudy (76 bulk rocks analyzed; Table 2) are fresh in thinsection. The K2O/P2O5 of Hawaiian tholeiitic basaltshas been used to evaluate the effect of subaerial weath-ering because K is easily leached whereas P is not (e.g.Frey et al., 1994). Except for sample K216-12B, all thestudied rocks from the submarine Hana Ridge haveK2O/P2O5 >1 (Fig. 4), indicative of freshness. TheK216-12B sample has lower K2O/P2O5 (�1) and alower SiO2 content compared with other submarineHana Ridge lavas (Fig. 6), implying that this sample wasslightly affected by low-temperature alteration. All ofthe samples from the six submarine dive localities aretholeiitic basalts or picrites (Fig. 5) with a compositionalrange from 6�6 to 28�9 wt % MgO, and from 43�3 to50�4 wt % SiO2 (Fig. 6a). Except for Fe2O3, all majoroxides increase with decreasing MgO and show littlescatter. At a given MgO in Fig. 6, other oxide contentsare similar to those of lavas from Kilauea. The ratios ofAl2O3/CaO and TiO2/Na2O of submarine lavas arealso similar to those of Kilauea lavas (Fig. 6) and showno correlation with water depth and dive locality (Fig. 7).

Some lavas from the northern lineation (S691-4, 5A, B)have lower SiO2 and higher K2O þ Na2O, and plot astransitional basalt (Figs 5 and 6). However, their Al2O3/CaO and TiO2/Na2O are identical to those of othersubmarine Hana Ridge rocks (Fig. 6h).Relative to these submarine lavas, the compositions of

most subaerial Honomanu lavas have been extensivelymodified by subaerial weathering. Normally, subaeriallavas tend to have lower SiO2 and K2O, higher Al2O3

and Fe2O3, and K2O/P2O5 <1 at a given MgO content(Figs 4 and 6) (Chen et al., 1991; Frey et al., 1994). TheTiO2/Na2O ratios in some subaerial lavas are higherthan in submarine lavas, and plot between the MaunaLoa and Kilauea fields. Some subaerial samples (HO-3,HO-21, and C122) are less altered, having K2O/P2O5

>1 (Fig. 4). Samples HO-3 and HO-21 have Kilauea-likemajor element compositions, whereas sample C122 isintermediate between Kilauea and Mauna Loa composi-tions (Fig. 6).Pillow glass rims from the six Hana Ridge dives have

6�0–8�2 wt % MgO and 49�5–52�6 wt % SiO2 (Table 3;Fig. 8) similar to previously dredged submarine samples.Previously dredged submarine samples are all tholeiiticbasalts except for one alkalic glass fragment recoveredabove the H terrace (smooth section of the shallowerportion of the Hana Ridge) (Moore et al., 1990), andhave a wider range in MgO (5�3–10�2 wt %) than thoseof the present study (Fig. 8) (Moore et al., 1990). Majorelement compositions of pillow glass rims from this studyoverlap the Kilauea and Mauna Loa fields (Fig. 8a and d)(e.g. Frey et al., 1994). However, in an Al2O3/CaO vsTiO2/Na2O diagram, most of the pillow glass rims plotin the Kilauea field with some samples plotting betweenthe Kilauea and Mauna Loa fields (Fig. 6h). Most pillowglass rims from this study contain high S (>0�056 wt %),indicating eruptions deeper than a few hundred metersbelow sea level (Moore & Clague, 1987; Moore &Thomas, 1988; Garcia et al., 1995).

Compositions of melt inclusions fromdifferent type of olivines

Melt inclusions occur in all types (euhedral, subhedral–undeformed and deformed) of olivine phenocrysts. Wehave analyzed, by EPMA, 31, 19, and 18 un-remeltedmelt inclusions in euhedral, subhedral–undeformed anddeformed olivine phenocrysts, respectively (Table 4). Wedetect no major element differences between melt inclu-sions from the different types of olivine hosts (Fig. 8). Thecompositional ranges of the inclusions are wider thanthose of the glass rims. Notably, the MgO contents ofmelt inclusions (2�2–7�8 wt %) are systematically lowerthan those of pillow glass rims (6�0–8�2 wt %). However,melt inclusions and pillow glass rims plot broadly on thesame compositional trend defined by the bulk rocks

2071


ownloaded from



ber 2018

http://www.petrology.oupjournals.org

Table 1: Microprobe analyses of euhedral, subhedral–undeformed, and deformed olivines from the submarine

Hana Ridge lavas

SiO2 FeO MnO MgO CaO NiO Total

Euhedral olivine

K216-12B 40.55 13.65 0.18 44.89 0.22 0.35 99.85

K216-12B 40.47 14.86 0.18 44.41 0.26 0.3 100.49

K216-12B 40.83 13.69 0.15 45.72 0.22 0.4 101

K214-1 39.91 9.49 0.12 48.57 0.2 0.52 98.8

K214-1 39.11 14.14 0.16 45.95 0.24 0.3 99.89

K214-3 40.06 11.36 0.13 47.75 0.23 0.39 99.92

K214-3 39.59 13.28 0.13 45.9 0.22 0.32 99.43

K214-3 39.24 14.12 0.17 45.97 0.21 0.32 100.02

K214-5B 39.95 9.71 0.1 48.39 0.2 0.45 98.8

K214-5B 40.06 10.02 0.12 47.96 0.16 0.48 98.81

K214-9 39.69 11.51 0.11 47.75 0.21 0.39 99.66

K214-9 39.87 10.53 0.12 47.8 0.18 0.4 98.89

K214-9 39.74 12.53 0.16 46.46 0.19 0.4 99.49

K214-11 39.1 15.94 0.21 43.94 0.29 0.22 99.7

K214-11 39.15 15.56 0.2 43.98 0.27 0.32 99.49

K214-11 39.1 15.99 0.21 45.1 0.27 0.3 100.98

K214-15A 39.21 14.5 0.16 46.08 0.25 0.28 100.48

K214-15B 40.44 10.29 0.12 48.71 0.21 0.42 100.18

K214-15B 39.83 13.75 0.15 46.3 0.26 0.22 100.51

K214-15C 39.81 14.04 0.14 45.85 0.24 0.35 100.43

K214-15C 39.61 13.42 0.17 45.45 0.23 0.31 99.18

K214-15C 39.31 12.93 0.13 46.32 0.23 0.39 99.29

K214-15C 39.27 15.11 0.16 43.99 0.27 0.3 99.11

K214-15C 39.64 11.94 0.12 46.87 0.2 0.42 99.19

K214-15C 39.51 14.19 0.14 45.79 0.27 0.32 100.22

K214-15C 40.19 10.88 0.1 47.81 0.18 0.4 99.56

K212-2A 39.16 14.89 0.19 44.42 0.3 0.33 99.28

K212-2A 39.41 11.34 0.16 47.03 0.22 0.44 98.6

K212-4 39.1 13.56 0.18 45.84 0.28 0.36 99.32

K212-4 39.89 9.81 0.17 48.76 0.24 0.46 99.32

K212-4 39.96 9.56 0.11 48.95 0.22 0.51 99.3

Subhedral—undeformed olivine

K214-1 39.73 11.8 0.13 47.76 0.2 0.27 99.89

K214-1 39.73 11.8 0.13 47.76 0.2 0.27 99.89

K214-3 40.16 9.34 0.12 49.16 0.2 0.52 99.5

K214-3 39.98 10.16 0.14 48.59 0.21 0.5 99.56

K214-3 39.04 16.72 0.2 43.48 0.24 0.29 99.96

K214-3 39.16 14.95 0.18 44.48 0.23 0.28 99.27

K214-3 39.14 15.38 0.17 44.61 0.25 0.29 99.83

K212-5B 39.93 10.26 0.11 47.91 0.22 0.41 98.83

K214-9 39.93 10.15 0.14 47.65 0.19 0.38 98.44

K214-9 39.93 10.15 0.14 47.65 0.19 0.38 98.44

K214-11 39.07 16.8 0.21 43.97 0.26 0.28 100.59

2072


ownloaded from



ber 2018

(Fig. 8). This implies that the melts of olivine inclusionsand pillow glass rims were derived from the same par-ental magma, but the former may have evolved morethan the glass rims through crystallization of the hostolivine and some other phases, such as chromite, withinthe inclusion (Clague et al., 1995).

Trace element compositions of bulkrocks

The trace element contents of 39 bulk rocks analyzed byXRF and those of 32 bulk rocks analyzed by LA-ICP-MS

are listed in Table 2 and plotted against MgO in Fig. 9.The abundance of Ni correlates positively with MgO,whereas Sc, Sr, and Zr show coherent but inverse corre-lations with MgO (Fig. 9). Abundances of incompatibleelements Th, Ba, La, Ce, and P are positively correlatedwith each other (not shown). The transitional basaltsfound in the northern lineation (S691-4, 5A, B) havehigher Sr concentrations (Fig. 9a), whereas their traceelement ratios (e.g. Zr/Nb, Sr/Nb, Ba/Nb) aresimilar to those of other Hana Ridge lavas (Fig. 11).Both transitional and tholeiitic lavas from the HanaRidge may have been derived from a similar source,and the differences in major and trace elements contents

SiO2 FeO MnO MgO CaO NiO Total


K214-15A 39.52 11.49 0.16 47.29 0.26 0.33 99.06

K214-15A 39.75 11.25 0.13 48.49 0.24 0.35 100.21

K214-15A 39.34 12.93 0.13 46.69 0.23 0.41 99.73

K214-15A 39.34 12.93 0.13 46.69 0.23 0.41 99.73

K214-15A 39.87 9.8 0.13 48.72 0.21 0.47 99.2

K212-2A 39.26 14.91 0.19 44.46 0.25 0.23 99.29

K212-2A 39.4 13.72 0.16 45.42 0.24 0.37 99.3

K212-2A 39.26 16.06 0.21 43.9 0.29 0.22 99.93

K212-2A 39.77 10.17 0.12 47.74 0.21 0.44 98.45

Deformed olivine

K214-1 39.99 9.91 0.14 49.03 0.19 0.46 99.73

K214-1 39.28 15.48 0.17 44.97 0.25 0.34 100.48

K214-3 39.06 13.6 0.15 46.06 0.19 0.25 99.3

K214-3 39.6 12.88 0.09 46.79 0.24 0.36 99.95

K214-3 39 15.14 0.18 44.43 0.26 0.21 99.21

K214-3 39.32 15.34 0.18 44.71 0.23 0.31 100.09

K214-3 38.78 14.24 0.16 45.01 0.23 0.33 98.76

K214-3 39.15 14.54 0.16 45.34 0.23 0.32 99.74

K214-5B 39.59 10.73 0.12 47.62 0.19 0.41 98.65

K214-5B 39.74 12.31 0.13 46.29 0.21 0.32 99

K214-5B 38.69 15.46 0.2 43.87 0.25 0.26 98.73

K214-5B 39.71 9.68 0.1 48.29 0.22 0.42 98.41

K214-9 40.28 9.44 0.14 47.7 0.21 0.42 98.18

K214-9 39.46 13.12 0.17 45.44 0.26 0.34 98.78

K214-9 39.7 10.08 0.12 47.4 0.2 0.39 97.88

K214-15A 39.85 11.14 0.13 47.91 0.23 0.35 99.61

K214-15B 39.79 13.46 0.16 46.13 0.27 0.34 100.15

K214-15B 40.13 11.77 0.14 47.9 0.21 0.38 100.54

K214-15C 40.11 11.19 0.11 48.06 0.21 0.35 100.03

K214-15C 39.75 14.44 0.19 45.45 0.27 0.31 100.4

K212-2A 39.94 10.18 0.09 47.89 0.2 0.46 98.76

2073


ownloaded from



ber 2018

between transitional and tholeiitic lavas may reflect adifference in the degree of partial melting or a differencein melting depth.The ratios of highly incompatible trace elements invol-

ving immobile elements such as Sm/Nd, La/Ce, and

La/Nb are relatively constant, similar to those of subaer-ial Honomanu lavas, and show no significant correlationwith sample locality or water depth (Fig. 10). However,there is a distinction between the submarine lavas and thesubaerial Honomanu stage lavas (e.g. Zr/Nb, Sr/Nb,Ba/Nb; Fig. 11). The Hana Ridge lavas have Sr/Nband Zr/Nb similar to those of Kilauea lavas (Fig. 11aand b), whereas samples from the subaerial Honomanustage tholeiitic lavas overlap the Kilauea–Mauna Loafields in terms of these geochemical parameters (Frey &Rhodes, 1993) (Fig. 11a and b). Overall, the less alteredsubaerial samples HO-21 and C122 have Kilauea-likecompositions, whereas HO-3 has a Kilauea–MaunaLoa intermediate composition (Figs 6 and 11). There islittle variation in Sm/Nd, La/Ce, and La/Nb (seeFig. 10); ratios such as Zr/Nb, Sr/Nb and Ba/Nb alsoshow limited variation and are uncorrelated with MgO(Fig. 11) in Hana Ridge lavas, suggesting that the magmasource was relatively homogeneous in chemical composi-tion. The combined major and trace element character-istics indicate that Haleakala volcano originally tapped amagma source that was chemically similar to that ofKilauea lavas, and the source composition changedfrom Kilauea-type (submarine Hana Ridge) to Kilauea–Mauna Loa intermediate-type composition duringfurther growth of the subaerial volcano. Secular variationin source composition during the growth history of otherHawaiian shields has been reported for Koolau (Jacksonet al., 1999; Shinozaki et al., 2002; Tanaka et al., 2002),and Mauna Kea (Eisele et al., 2003).

DISCUSSION

Origin of olivine phenocrysts

To make clear the nature of olivine crystals in thesubmarine Haleakala lavas, we determined compositionsof olivine crystals and inclusions from the three olivinetypes; i.e. euhedral, subhedral–undeformed, and de-formed olivine phenocrysts. Compositions among thethree types of olivine as well as olivine of mantle perido-tite were compared, and no compositional difference wasdetected among the olivine phenocryst types (e.g. CaOand NiO vs Fo diagrams, Fig. 3). CaO in olivine fromlavas is typical of magmatic values (0�15–0�41 wt %,Garcia et al., 1995; Norman & Garcia, 1999) and signifi-cantly higher than that of olivine from Hawaiianxenoliths (Sen, 1988) (Fig. 3a and b). Furthermore, thecompositions of melt inclusions show scatter but there isno systematic difference with regard to the host olivinetypes (Fig. 8), and the melt inclusion compositions plot onthe same trends as pillow glass rims and bulk rocks. Basedon these features, we conclude that the various olivinemorphological types all crystallized from a magma ratherthan being mantle-derived xenocrysts.

Upper

Fig. 3. Chemical composition of olivine phenocrysts in Hana Ridgebasalts as a function of (a) CaO (wt %)–Mg-number for different typeof olivines. The line is an estimate of the boundary between olivinecrystallized at upper- and lower-crustal depths (Garcia et al., 1995). (b)CaO (wt %)–Mg-number for texturally unclassified olivines. (c) NiO(wt %)–Mg-number for the three types of olivines. It should be notedthat CaO in Hana Ridge olivine phenocrysts is significantly higherthan CaO in olivines from Hawaiian xenoliths (Sen, 1988). No com-positional (CaO–Mg-number, NiO–Mg-number) difference wasdetected among the olivine phenocryst types.

2074


ownloaded from



ber 2018

Table 2: Major and trace element compositions for the lavas from submarine Hana Ridge

Northern rift cone

1 2 3 4 5 6 7 8 9 10

Sample no.: K216-7A K216-7B K216-8A K216-9 K216-10 K216-11A K216-11B K216-11C K216-12B K216-12C

XRF

SiO2 48.99 49.26 49.05 49.09 49 49.76 49.25 48.6 47.15 48.29

TiO2 3 2.95 2.98 2.88 2.33 2.85 2.72 2.85 2.81 2.84

Al2O3 12.51 12.49 12.45 12.28 11.91 12.79 12.48 11.96 12.49 12.37

Fe2O3 12.59 12.43 12.5 12.51 12.72 12.07 12.18 12.63 12.85 12.61

MnO 0.23 0.19 0.18 0.17 0.17 0.16 0.16 0.25 0.34 0.33

MgO 10.13 10.31 10.37 9.97 11.93 9.65 10.85 10.91 11.72 10.6

CaO 9.47 9.48 9.6 9.11 9.55 9.97 9.75 9.69 9.92 9.82

Na2O 2.1 2.16 2.12 2.31 1.95 2.14 2.12 2.05 1.96 2.07

K2O 0.44 0.42 0.45 0.49 0.36 0.44 0.41 0.39 0.3 0.39

P2O5 0.32 0.3 0.31 0.31 0.22 0.32 0.3 0.31 0.3 0.3

Total (%) 99.77 99.99 100.01 99.11 100.15 100.14 100.21 99.64 99.84 99.61

XRF

Sc 27 29

V 298 286

Cr 604 1011

Co 55 62

Ni 418 544

Zn 112 107

Ga 20 19

Rb 10 6.1

Sr 307 280

Y 33 25

Zr 181 128

Nb 18 12

Ba 109 65

LA-ICP-MS

La 15.8 11.35

Ce 36 26.78

Pr 5.07 3.57

Nd 24.6 17.86

Sm 6.55 4.93

Eu 2.29 1.76

Gd 6.73 4.42

Tb 1.07 0.78

Dy 6.21 4.32

Ho 1.16 0.86

Er 2.87 2.08

Tm 0.39 0.3

Yb 2.32 1.95

Lu 0.35 0.25

Hf 4.54 2.82

Ta 0.94 0.59

Pb 0.77 1.19

Th 1.12 0.88

U 0.33 0.25

2075


ownloaded from



ber 2018

Table 2: continued

Northern rift cone

11 12 13 14 15 16 17 18 19 20

Sample no.: K216-13 S687-1 S687-3 S687-5 S687-6 S687-9 S687-11 S687-13 S687-15 S687-20

XRF

SiO2 48.82 49.57 48.96 48.99 49.26 49.19 49.03 48.77 48.85 48.81

TiO2 2.78 2.7 2.44 2.44 2.48 2.52 2.48 2.43 2.42 2.63

Al2O3 12.17 12.36 11.22 11.22 11.46 11.73 11.35 11.14 11.32 11.59

Fe2O3 12.53 12.23 12.47 12.39 12.67 12.45 12.39 12.72 12.93 13.14

MnO 0.21 0.16 0.17 0.16 0.17 0.17 0.16 0.17 0.21 0.18

MgO 9.91 10.32 13.43 13.38 12.76 12.26 12.99 13.44 13.4 12.52

CaO 9.46 9.38 8.5 8.46 8.65 8.89 8.66 8.57 8.8 9.04

Na2O 2.21 2.23 2.05 2.07 2.07 2.13 2.08 1.87 1.95 2.04

K2O 0.47 0.44 0.4 0.4 0.41 0.42 0.41 0.35 0.34 0.32

P2O5 0.3 0.3 0.27 0.27 0.27 0.28 0.27 0.23 0.24 0.26

Total (%) 98.86 99.96 100.2 99.78 100.21 100.31 99.81 99.69 100.71 100.8

XRF

Sc 30 28 27 26 28 28 25 27 27 28

V 302 301 271 264 280 286 270 279 292 312

Cr 978 982 1062 1061 1300 835 962 1031 842 813

Co 67 55 63 63 65 60 63 66 76 64

Ni 527 524 753 725 810 576 662 731 645 612

Zn 110 113 105 104 107 112 105 106 110 117

Ga 20 21 18 18 19 19 18 18 19 19

Rb 8 7.8 7.7 7.3 8.4 8.1 7.6 5.8 5.6 4.7

Sr 308 310 279 277 285 296 284 260 270 281

Y 31 31 27 27 28 30 28 25 27 29

Zr 171 169 153 152 157 159 154 132 135 142

Nb 16 16 14 15 15 15 14 12 12 13

Ba 101 94 93 93 90 87 92 76 73 76

LA-ICP-MS

La 14.2 12 12.1 12.61 11

Ce 34.8 30 30.9 31.17 26.2

Pr 4.7 4.11 4.24 4.34 3.57

Nd 22.8 19.3 22.6 20.18 17.7

Sm 6.38 5.02 6.22 6.08 4.8

Eu 2.26 1.88 1.81 1.82 1.72

Gd 6.04 4.91 5.65 5.28 4.67

Tb 0.98 0.82 0.88 0.88 0.81

Dy 5.55 4.95 4.87 4.4 4.55

Ho 1.09 0.95 1.04 0.9 0.83

Er 2.58 2.33 2.38 2.09 2.1

Tm 0.34 0.31 0.34 0.29 0.26

Yb 2.19 1.98 2.19 1.75 1.63

Lu 0.29 0.28 0.32 0.27 0.21

Hf 4.16 3.44 3.96 3.48 3.05

Ta 0.78 0.71 0.82 0.69 0.6

Pb 2.97 0.63 2.35 0.66 0.43

Th 0.98 0.8 0.91 0.82 0.64

U 0.35 0.3 0.29 0.26 0.2

2076


ownloaded from



ber 2018

Northern rift cone Northern rift

21 22 23 24 25 26 27 28 29 30

Sample no.: S687-21 S687-23 S691-2A S691-3 S691-4B S691-5A S691-5B K216-2A K216-2B K216-3

XRF

SiO2 48.65 48.88 48.94 48.47 46.19 47.44 47.24 49.68 48.44 49.81

TiO2 2.46 2.38 2.66 2.32 2.73 2.66 2.63 2.84 2.94 2.45

Al2O3 10.94 11.41 11.43 10.01 10.96 11.79 11.65 12.63 12.08 12.66

Fe2O3 13.09 12.57 12.45 12.54 13.03 13.07 13.13 12.35 12.84 12.13

MnO 0.17 0.17 0.16 0.17 0.16 0.17 0.17 0.17 0.17 0.19

MgO 13.78 13.21 12.48 16.36 14.7 12.82 12.96 10.4 11.27 10.19

CaO 8.74 8.9 8.96 7.38 8.58 9.21 9.23 9.79 9.48 9.87

Na2O 1.89 1.9 1.95 1.96 2.08 2.1 2.1 2.16 1.9 2.1

K2O 0.34 0.34 0.45 0.43 0.54 0.5 0.52 0.42 0.45 0.4

P2O5 0.23 0.22 0.26 0.28 0.3 0.28 0.28 0.3 0.31 0.23

Total (%) 100.27 100.26 99.74 99.93 99.27 100.04 99.9 100.74 99.88 100.02

XRF

Sc 28 28 28 23 24 28 27 29

V 289 274 279 250 249 270 266 287

Cr 1039 1011 833 1326 1034 825 842 790

Co 67 63 60 77 71 64 64 61

Ni 761 685 559 992 766 570 599 403

Zn 110 104 105 105 112 111 112 103

Ga 19 18 19 18 18 19 19 19

Rb 5.4 5.3 6.2 7.4 8.4 8.3 8.1 8.7

Sr 259 262 278 236 373 373 371 286

Y 26 25 29 29 25 25 26 28

Zr 132 126 153 165 166 158 156 135

Nb 12 11 14 15 19 17 16 14

Ba 80 72 83 92 124 109 102 84

LA-ICP-MS

La 11.73 12.2 13.3 15.18 14.1 13.74 13.3

Ce 23.85 31.2 33 36.37 33.2 33.89 28.6

Pr 3.33 4.37 4.31 4.81 4.37 4.45 3.94

Nd 16.05 21.4 20.7 22.31 20.2 20.42 19.5

Sm 4.59 5.95 5.6 5.42 5.41 5.31 5.14

Eu 1.71 2.05 1.93 1.99 1.83 1.9 1.93

Gd 4.07 5.67 5.61 4.8 4.75 4.77 5.45

Tb 0.69 0.86 0.85 0.75 0.75 0.71 0.85

Dy 4.27 4.88 4.77 4.09 4.05 4.12 5.1

Ho 0.86 0.97 0.91 0.75 0.74 0.8 0.95

Er 2.03 2.33 2.28 1.75 1.83 1.71 2.33

Tm 0.25 0.3 0.31 0.23 0.23 0.22 0.28

Yb 1.67 2.21 2.03 1.53 1.53 1.55 2.08

Lu 0.22 0.32 0.27 0.21 0.19 0.2 0.29

Hf 2.84 3.72 3.84 3.35 3.12 3.17 3.44

Ta 0.57 0.78 0.67 0.87 0.77 0.78 0.71

Pb 0.62 0.6 0.88 0.73 0.81 0.71 1.88

Th 0.58 0.83 0.97 1.06 0.96 0.91 0.92

U 0.21 0.26 0.31 0.4 0.28 0.33 0.26

2077


ownloaded from



ber 2018

Table 2: continued

Northern rift Amphitheater

31 32 33 34 35 36 37 38 39 40

Sample no.: K216-4 K216-5A K216-5B K216-6 K212-1B K212-2A K212-3B K212-3A K212-3B K212-4B

XRF

SiO2 49.49 49.4 49.72 48.92 46.98 46.94 47.78 47.4 47.14 46.87

TiO2 2.6 2.59 2.63 2.54 2 2.01 2.22 2.38 2.31 2.2

Al2O3 12.69 13.21 12.92 12.67 9.8 9.74 10.64 11 10.74 10.33

Fe2O3 12.37 12.02 12.26 12.3 12.4 12.62 12.41 12.45 12.41 12.65

MnO 0.17 0.23 0.17 0.17 0.17 0.19 0.17 0.17 0.17 0.17

MgO 10.03 9.61 9.89 9.51 17.35 17.64 15.69 14.68 15.79 15.83

CaO 10.28 10.27 10.23 10.13 7.93 7.84 8.63 9.24 8.94 8.62

Na2O 1.96 2.05 1.95 2.14 1.64 1.58 1.75 1.63 1.67 1.71

K2O 0.41 0.38 0.36 0.39 0.32 0.31 0.4 0.4 0.38 0.33

P2O5 0.24 0.24 0.25 0.23 0.2 0.2 0.22 0.23 0.23 0.21

Total (%) 100.23 99.99 100.38 99.01 98.78 99.06 99.89 99.59 99.76 98.92

XRF

Sc 29 24 22 27 27

V 282 225 226 262 249

Cr 769 1156 1520 1170 1534

Co 55 76 82 71 77

Ni 381 935 1095 884 1083

Zn 105 98 99 102 101

Ga 20 8 16 17 18

Rb 8.1 12 4.6 6.1 5

Sr 293 233 232 264 262

Y 29 23 23 24 23

Zr 142 113 120 125 125

Nb 15 12 11 12 13

Ba 82 67 65 72 71

LA-ICP-MS

La 12.9 10 10.2 12 12.6

Ce 28.3 23.9 25.6 25.4 24.8

Pr 4.08 3.3 3.34 3.67 3.4

Nd 19.8 15.8 16.4 17.8 16.6

Sm 5.62 4.32 4.68 4.88 4.39

Eu 2.09 1.54 1.6 1.74 1.65

Gd 5.83 4.64 4.25 4.98 4.34

Tb 0.96 0.72 0.7 0.75 0.7

Dy 5.29 3.96 4.15 4.33 3.83

Ho 1.04 0.79 0.79 0.86 0.78

Er 2.51 1.92 1.89 2.02 1.95

Tm 0.32 0.25 0.28 0.29 0.24

Yb 2.1 1.65 1.69 1.75 1.64

Lu 0.3 0.21 0.22 0.27 0.22

Hf 3.58 2.88 2.95 3.32 2.88

Ta 0.74 0.54 0.54 0.65 0.62

Pb 0.7 0.93 2.29 0.73 0.48

Th 0.9 0.69 0.89 0.73 0.67

U 0.25 0.22 0.2 0.21 0.18

2078


ownloaded from



ber 2018

Amphitheater

41 42 43 44 45 46 47 48 49 50

Sample no.: K212-6A K212-6B K212-7A K212-8 K212-9 K212-10 K212-11A K212-11C K212-12A K212-12B

XRF

SiO2 47.91 47.64 49.83 49.65 50 50.5 49.88 49.47 49.58 48.37

TiO2 2.47 2.39 2.6 2.78 2.71 2.63 2.7 2.76 2.74 2.33

Al2O3 11.42 11.19 13.05 13.22 13.58 13.07 13.48 13.13 13.28 11.55

Fe2O3 12.53 12.57 12.05 12.14 11.91 12.49 12.08 12.63 12.47 12.65

MnO 0.17 0.17 0.16 0.18 0.16 0.17 0.2 0.3 0.28 0.17

MgO 13.86 13.47 9.31 8.4 7.58 7.92 7.65 8.44 8.62 12.89

CaO 9.48 9.18 10.5 10.63 10.83 10.46 10.86 10.32 10.28 9.64

Na2O 1.68 1.88 1.96 1.96 2.16 2.1 1.98 1.91 1.98 1.76

K2O 0.35 0.43 0.37 0.38 0.41 0.4 0.38 0.33 0.36 0.38

P2O5 0.24 0.23 0.25 0.27 0.26 0.25 0.26 0.26 0.27 0.23

Total (%) 100.1 99.15 100.08 99.6 99.59 99.99 99.46 99.55 99.84 99.97

XRF

Sc 28 31

V 269 313

Cr 1139 952

Co 67 53

Ni 787 426

Zn 103 104

Ga 18 21

Rb 5.7 7

Sr 284 300

Y 25 29

Zr 136 147

Nb 13 13

Ba 75 80

LA-ICP-MS

La 11.4 13

Ce 26.6 28.7

Pr 3.69 4.05

Nd 17.8 20.5

Sm 4.73 5.43

Eu 1.78 2.07

Gd 4.67 5.92

Tb 0.74 0.9

Dy 4.32 5.19

Ho 0.81 1.05

Er 1.99 2.63

Tm 0.27 0.35

Yb 1.77 2.29

Lu 0.24 0.28

Hf 3.09 3.86

Ta 0.66 0.76

Pb 0.6 0.68

Th 0.74 0.73

U 0.23 0.21

2079


ownloaded from



ber 2018

Table 2: continued

Amphitheater Southern rift

51 52 53 54 55 56 57 58 59 60

Sample no.: K212-12C K212-13A K212-13B K214-1 K214-2 K214-3 K214-4B K214-5B K214-6A K214-6B

XRF

SiO2 49.95 50.28 48.07 44.66 44.24 44.51 44.69 44.55 48.89 48.37

TiO2 2.68 2.72 2.62 1.16 1.23 1.28 1.27 1.26 2.66 2.52

Al2O3 13.46 13.78 12.76 6.8 6.49 6.89 6.84 6.81 12.48 11.81

Fe2O3 12.22 11.71 11.99 12.81 12.97 13.12 13.06 13.22 12.34 12.49

MnO 0.17 0.16 0.17 0.17 0.18 0.18 0.17 0.17 0.17 0.17

MgO 7.66 7.34 7.14 26.62 27.53 26.22 26.74 26.65 10.22 11.26

CaO 10.84 10.96 10.48 5.69 5.81 6.07 6.02 5.99 10.17 9.64

Na2O 2 2.13 2.1 1.07 0.93 0.99 0.96 0.95 1.92 1.98

K2O 0.35 0.41 0.41 0.19 0.16 0.17 0.17 0.16 0.44 0.4

P2O5 0.26 0.26 0.24 0.12 0.14 0.14 0.14 0.14 0.26 0.24

Total (%) 99.59 99.76 95.98 99.28 99.65 99.56 100.05 99.91 99.55 98.88

XRF

Sc 31 18 29

V 298 153 284

Cr 693 1991 1005

Co 47 108 59

Ni 293 1674 574

Zn 103 92 103

Ga 20 11 19

Rb 5.5 2.5 5.4

Sr 302 140 286

Y 29 15 28

Zr 148 71 145

Nb 14 6.3 14

Ba 76 38.3 77.2

LA-ICP-MS

La 11.9 5.5 11.4

Ce 27.9 13.9 26.9

Pr 3.98 1.92 3.87

Nd 19.6 9.26 18.7

Sm 5.39 2.86 5.23

Eu 2.01 0.97 1.88

Gd 5.13 2.69 5.15

Tb 0.85 0.47 0.8

Dy 4.97 2.75 4.73

Ho 0.96 0.57 0.89

Er 2.32 1.36 2.22

Tm 0.34 0.19 0.29

Yb 2.1 1.09 1.85

Lu 0.29 0.17 0.26

Hf 3.4 1.99 3.31

Ta 0.63 0.32 0.69

Pb 1.06 0.97 0.56

Th 0.79 0.44 0.8

U 0.23 0.12 0.24

2080


ownloaded from



ber 2018

Southern rift

61 62 63 64 65 66 67 68 69 70

Sample no.: K214-7 K214-8 K214-9 K214-10 K214-11 K214-13 K214-14C K214-15A K214-15B K214-15C

XRF

SiO2 48.5 50.37 45.39 49.24 48.73 48 43.78 44.79 43.97 43.89

TiO2 2.54 2.7 1.26 2.93 2.4 2.32 1.19 1.08 1.21 1.05

Al2O3 11.91 13.97 7.21 13.54 11.84 11.4 6.12 6.46 6.34 6.13

Fe2O3 12.44 12.1 13.19 12.95 12.4 12.64 13.83 13.31 13.72 13.19

MnO 0.17 0.17 0.17 0.24 0.17 0.17 0.2 0.17 0.18 0.17

MgO 12.12 6.93 25.77 6.48 11.83 12.93 27.44 28.09 27.61 28.65

CaO 9.63 10.84 6.02 10.45 9.95 9.54 6.03 5.36 5.76 5.15

Na2O 1.84 2.21 1.12 2.23 1.8 1.78 0.96 0.99 0.91 0.91

K2O 0.38 0.36 0.2 0.41 0.37 0.36 0.14 0.19 0.14 0.16

P2O5 0.25 0.25 0.13 0.28 0.24 0.23 0.13 0.11 0.13 0.11

Total (%) 99.78 99.91 100.45 98.76 99.73 99.37 99.8 100.55 99.96 99.42

XRF

Sc 18 32 17 17

V 167 315 146 138

Cr 1876 681 1889 2288

Co 106 62 111 117

Ni 1500 338 1602 1779

Zn 96 115 95 92

Ga 11 21 10 10

Rb 3.3 7.4 3.7 3.5

Sr 159 333 142 134

Y 15 33 13 13

Zr 70 166 61 59

Nb 6.3 15 5.2 5.5

Ba 47 88 35 37.5

LA-ICP-MS

La 5.3 14.7 5.3 4.2

Ce 13.2 33.6 11.6 11.8

Pr 1.85 4.54 1.67 1.58

Nd 9.3 22.8 8.1 7.76

Sm 2.73 6.29 2.37 2.31

Eu 0.97 2.28 0.89 0.92

Gd 2.98 6.03 2.51 2.57

Tb 0.48 1 0.41 0.41

Dy 2.73 5.73 2.31 2.18

Ho 0.51 1.08 0.5 0.45

Er 1.24 2.8 1.01 1.16

Tm 0.15 0.38 0.13 0.18

Yb 1.04 2.38 1 1.02

Lu 0.14 0.33 0.12 0.14

Hf 1.98 4 1.66 1.48

Ta 0.3 0.79 0.27 0.24

Pb 0.52 5.49 0.36 0.51

Th 0.33 1.06 0.3 0.33

U 0.11 0.31 0.09 0.11

2081


ownloaded from



ber 2018

Table 2: continued

Southern rift

71 72 73 74 75 76

Sample no.: K214-15D S686-1 S686-2A S686-2B S686-3A S686-3B

XRF

SiO2 44.44 49.94 50.15 50.19 48.21 47.74

TiO2 1.35 2.73 2.68 2.69 2.07 2.01

Al2O3 7.19 13.09 13.17 13.19 10.7 10.2

Fe2O3 13.56 12.45 12.34 12.23 12.12 12.3

MnO 0.18 0.17 0.17 0.17 0.16 0.3

MgO 25.18 8.04 8.49 8.17 15.77 16.69

CaO 6.3 10.45 10.31 10.41 8.85 8.48

Na2O 1.02 2.08 2.13 2.14 1.7 1.57

K2O 0.17 0.45 0.47 0.48 0.31 0.28

P2O5 0.14 0.26 0.26 0.26 0.2 0.2

Total (%) 99.52 99.66 100.16 99.93 100.38 99.75

XRF

Sc 31 31 34 26 26

V 313 304 309 249 245

Cr 700 516 626 1159 1404

Co 50 48 49 68 106

Ni 325 263 289 798 991

Zn 107 107 105 99 99

Ga 20 20 20 17 16

Rb 8.2 7.6 7.8 5 6.3

Sr 308 309 310 246 240

Y 30 30 29 23 22

Zr 156 154 155 114 112

Nb 15 14 14 10 10

Ba 78 75 76 59 68

LA-ICP-MS

La 12.6 12.5 13.1 9.4

Ce 29.1 29.6 29.3 29.7

Pr 4.23 4.26 4.31 3.89

Nd 21 20.7 21.4 16.6

Sm 5.78 5.67 5.95 4.28

Eu 2.13 2.2 2.09 1.44

Gd 5.63 5.56 5.78 4.13

Tb 0.96 0.86 0.95 0.69

Dy 5.3 5.09 5.23 4.35

Ho 1.01 0.99 0.99 0.72

Er 2.48 2.6 2.49 1.8

Tm 0.32 0.33 0.32 0.26

Yb 2.08 1.97 2.18 1.63

Lu 0.27 0.27 0.28 0.27

Hf 3.74 3.66 3.7 2.56

Ta 0.7 0.7 0.72 0.56

Pb 0.65 0.68 0.57 6.96

Th 0.82 0.9 0.88 1.13

U 0.2 0.28 0.21 0.23

2082


ownloaded from



ber 2018

Primary magma composition

There have been many estimates of the MgO composi-tions of primary magmas delivered to Kilauea, rangingfrom as low as 8–13% (Maaloe, 1979) to as high as 20–25 wt % (Wright, 1984). The discovery of high-MgO(up to 15 wt %) glass sands at the foot of Kilaueavolcano (Clague et al., 1991) provided direct evidencefor the existence of magnesian (at least 15 wt % MgO)primary magma in Kilauea. In contrast to Kilauea, thecomposition of the primary magma of Haleakala

volcano has been estimated by only a few researchers.Chen (1993) suggested that the Haleakala primarymagma has a MgO content of 16–17 wt % based onthe analysis of the Honomanu Gulch suite. Wagner et al.(1998) studied trace element abundances of high-MgOtholeiite glasses fromKilauea,Mauna Loa, andHaleakalavolcanoes. They estimated that the primary magmacompositions of Haleakala shield range between 16�7and 17�6 wt %MgO. Previous primary magma estimatesare usually based on the assumption of chemical equili-brium of bulk rocks or pillow glasses with olivine using anFe–Mg partition coefficient (e.g. Chen, 1993; Clagueet al., 1995; Wagner at al., 1998). In this study, based onnew compositional data for fresh samples from thesubmarine Hana Ridge, we constrain primary magmacompositions using partition coefficients K ol�melt

DFe�Mg and

D�ol�meltCaO with the most forsteritic olivine. Our estimation

method and procedures are stated below.We assume that the most forsteritic olivine core should

represent the earliest crystallized part that may have beenin equilibrium with the most primitive (high Mg-number)magma. The compositional relationship between olivineand melt, given as K ol�melt

DFe�Mg, can be used to calculate theFe/Mg of melt equilibrated with olivine (Roeder &Emslie, 1970; Roeder, 1974; Takahashi, 1978).Calcium is one of the most significant minor elements

in natural magmatic olivine, and the calcium concentra-tion of the olivine provides useful information about theevolution of the melt phase during crystallization(Libourel, 1999). According to Libourel (1999), Ca parti-tioning is independent of temperature and fO2 at lowand intermediate pressures. When applied to naturalolivine, this model reproduces Ca contents (where meltcomposition is known), to within �10%. The CaO parti-tion coefficient between olivine and basaltic melt is for-mulated as

D�ol�meltCaO ¼ X ol

CaO=a�meltCaO

where X olCaO ¼ CaO=ðCaOþMgOÞ (molar) in olivine.

a�meltCaO is a pseudo-activity of CaO in the melt, calculatedas follows:

a�meltCaO ¼ ðX 0melt

CaOÞ2=ðXmelt

SiO2þ Xmelt

TiO2Þ

with X 0meltCaO ¼ Xmelt

CaO þ XmeltNa2O

þ XmeltK2O

, and where X 0meltMO

stands for the molar fraction of the MO oxide in themelt.From the bulk-rock composition, a�melt

CaO values are cal-culated assuming that bulk rocks represent melt composi-tions. Calculated a�melt

CaO values of bulk rocks are plottedagainst FeO/MgO (open circles in Fig. 12). The D�ol�melt

CaO

partition coefficient can be fitted by an empirical

40 42 44 46 48 50 52 54

Fig. 5. Alkali–silica diagram showing that all of the lavas from thesubmarine Hana Ridge are tholeiitic basalts and picrites. The data arenormalized volatile-free, with all Fe as FeO. Macdonald & Katsura(1964) alkalic–tholeiitic boundary and IUGS picrite–tholeiitic bound-ary are shown for reference. Data for subaerial Honomanu tholeiiticand alkalic basalts from Chen et al. (1991). The three basalt samplesfrom S691-4B and S691-4A, B have higher total alkali contents. Errorbars (2s) are indicated.

Fig. 4. P2O5–K2O (wt %) diagram showing the effect of low-tempera-ture alteration of Hawaiian basalts. Data for subaerial Honomanutholeiitic and alkalic basalts from Chen et al. (1991).

2083


ownloaded from



ber 2018

polynomial as a function of forsterite mole fraction ofolivine (XFo):

lnD�ol�meltCaO ¼ lnðX ol

CaO=a�meltCaO Þ

¼ �1� 24X 3Fo þ 3�33X 2

Fo� 6�55XFo þ 2�05:

Using the above equation, a�meltCaO that equilibrated with

magnesian (Fo >90�0) olivine is calculated. The FeO/MgO ratios of melts that equilibrated with magnesianolivine are also calculated using a value of 0�3 � 0�03for the K

Fe�MgD of olivine–melt (Roeder & Emslie, 1970;

Takahashi, 1978) and assuming an Fe2þ/FeTotal ratio of

Fig. 6. Major element vs MgO variation diagrams (a–g) and Al2O3/CaO vs TiO2/Na2O3 diagram (h) for the Haleakala shield lavas. Data forsubaerial Honomanu lavas from Chen et al. (1991). Major element totals are normalized to 100wt %. Fields for Kilauea and Mauna Loa lavas arefrom numerous literature sources, available from the GEOROC Database (http://georoc.mpch-mainz.gwdg.de/). Error bars (2s) are indicated.

2084


ownloaded from



ber 2018

http://georoc.mpch-mainz.gwdg.de/

0�9 in the melt (e.g. Clague et al., 1995). Calculatedresults (a�melt

CaO and FeO/MgOmelt) are shown as filleddiamonds in Fig. 12. The bulk-rock FeO/MgO ratioswere calculated from the bulk-rock compositions. Theresults show that the calculations based on the olivinecompositions (filled diamonds) and those based on thebulk rocks (open circles) are in excellent agreement inthe range of the large circle (Fig. 12), indicating thatmagnesian olivine equilibrated with melts representedby the bulk-rock composition at around FeO/MgO�0�70 or a�melt

CaO �2�3. This is equivalent to a primarymagma composition with the most forsteritic olivine(Fo90�6) containing �16�7 wt % MgO and �8�4 wt %CaO (Fig. 12). The other major element compositions ofthe primary magma can be determined from the coher-ent bulk chemical trends of the Hana Ridge lavas. Ourestimate of this primary magma composition is listed inTable 5.

Water content of primary magma andpressure of crystallization

To explore the pressure and H2O content during thecrystallization stage of Haleakala tholeiite magma, simu-lations with the MELTS program (Ghiorso & Sack 1995)were performed. In this study, the MELTS simulationswere started with the primary magma composition esti-mated above and the following parameter ranges were

tested: temperature from 1450�C to 1000�C, pressurefrom 0�1 to 0�3GPa, and H2O contents of 0�1, 0�5, and1 wt %. The fO2 of the magma is assumed to be equiva-lent to the synthetic fayalite–magnetite–quartz (FMQ)buffer during crystallization.Calculated liquid composition lines obtained by

MELTS are shown in Fig. 13. Simulations using 0�1 wt %H2O at 0�2GPa and 0�3GPa, and 0�5 wt % H2O at0�3GPa yielded SiO2 contents lower than those of theactual submarine Hana Ridge lavas. A simulation using0�1 wt %H2O at 0�1GPa yielded Al2O3, TiO2, and FeO*

trends inconsistent with the submarine Hana Ridge lavas(Fig. 13).Mineral assemblages produced by the MELTS pro-

gram at 0�1 wt % H2O at 0�1, 0�2 and 0�3GPa, and at0�5 wt % H2O and 0�3GPa are not consistent with thephenocryst assemblages observed in the Hana Ridgerocks. For example, we observed the following pheno-cryst assemblages in the Hana Ridge rocks: (1) olivine;(2) olivine þ augite; (3) olivine þ augite þ plagioclase;(4) olivine þ augite þ plagioclase þ orthopyroxene. Thisimplies that the crystallization sequence of the submarineHana Ridge lavas was: olivine, augite, plagioclase,and finally orthopyroxene. However, in the MELTSsimulation with relatively small amounts of H2O(0�1 wt % at 0�1, 0�2 and 0�3GPa; 0�5 wt % H2O at0�3GPa), orthopyroxene always starts to crystallize ear-lier than clinopyroxene. Among the calculated liquidlines of descent and mineral crystallization sequences

northern rift conearcuate tipsubaerial TH

northern riftsouthern riftsubaerial AB

Fig. 7. Al2O3/CaO and TiO2/Na2O vs sea-water depth for Haleakala shield lavas. Data for subaerial Honomanu basalts fromChen et al. (1991).

2085


ownloaded from



ber 2018

produced by MELTS, those with 0�5 wt % H2O at0�1GPa pressure, and those with H2O contents of1 wt % (at pressures of 0�1, 0�2 and 0�3GPa), are ingeneral agreement with the crystallization sequenceobserved in lavas and bulk-rock compositional trends(Fig. 13).MELTS simulations suggest that the primary magma

had about 0�5–1 wt % water. This is consistent withmeasured water contents in melt inclusions from KilaueaIki picrites (Wallace & Anderson, 1998). Melt inclusionsfrom the first eruptive episode, before any drainbackoccurred, have an average H2O content of 0�7 �0�2 wt %. Anderson & Brown (1993) reported four inclu-sions with >0�8 wt % H2O from the same 1959 KilaueaIki eruption. The H2O/K2O of the primary magma ofthe submarine Hana Ridge estimated in this study is�1�7, which is slightly higher than that of typicalHawaiian tholeiitic magmas (�1�3) (Wallace &Anderson, 1998), but the same as that of submarineglasses from Kilauea (Puna Ridge) (�1�7) (Wallace &Anderson, 1998). Considering the effect of degassingfrom magma in the reservoir at lower pressures(e.g. Clague et al., 1995), 0�5–1 wt % of water in theprimary magma is a reasonable estimation.The results of the MELTS calculation indicate that the

submarine Hana Ridge magmas have undergone crystal-lization at pressures of 0�1–0�3GPa, which correspond to

a depth of 3–9 km assuming an average crustal densityof 3 g/cm3. This depth is comparable with estimatedmineral fractional crystallization depths of 3–6 km forPuna Ridge lavas ( Johnson et al., 2002).Delaney et al. (1990) proposed that the Kilauea rifts

may be underlain by an extensive ‘near vertical dike-likemagma system’ at depths of 3–9 km. Ryan et al. (1981)proposed that during the main shield-building stage, ashallow magma chamber fed by a central conduit ismaintained at a depth of �2–7 km beneath Hawaiianshield volcanoes. The shield grows as magmas erupt atthe summit or are injected into the prominent rift zonesthat radiate from the summit. Magmas migrate laterallyfrom the summit reservoir into rift zones, and whenmagma supply rate decreases, isolated magma chamberscan form in rift zones at depths of 2–7 km (Yang et al.,1999). Our results suggest that the primitive magma inthe Hana Ridge was also stored in shallow magmachamber(s), and that crystallization and accumulationoccurred at a depth similar to those for modern Hawaiianshields.

Crystallization and magma mixing

In the MELTS simulation under the conditions estimatedabove (0�1–0�3GPa and 0�5–1 wt % H2O), olivine begins

Table 3: Representative microprobe analyses of pillow glass rims for the submarine Hana Ridge

Sample Na2O K2O FeO MgO Al2O3 SiO2 CaO P2O5 TiO2 MnO SO3 Total

K212-3B* (14)y 2.37 0.45 10.77 6.66 13.89 51.18 10.66 0.33 2.91 0.16 0.22 99.62

K212-6B(21) 2.35 0.46 10.99 6.57 13.7 51.2 10.61 0.34 2.96 0.14 0.23 99.55

K214-1 (15) 2.13 0.34 9.79 7.75 13.08 52.26 11.37 0.26 2.29 0.16 0.14 99.56

K214-2 (13) 2.09 0.34 9.77 8.18 13.35 52.05 11.3 0.27 2.2 0.14 0.15 99.82

K214-4B (16) 2.11 0.33 9.7 7.83 13.33 52.33 11.15 0.26 2.25 0.14 0.16 99.6

K214-5B (17) 2.13 0.35 9.92 7.92 13.57 52.3 11.1 0.27 2.2 0.15 0.16 100.06

K214-7 (14) 2.23 0.4 10.5 7.17 13.51 51.78 11.07 0.31 2.74 0.16 0.24 100.1

K214-15C (12) 1.99 0.35 10.17 7.54 13.88 52.51 12.03 0.27 2.28 0.15 0.15 101.32

K216-2 (16) 2.43 0.53 10.72 6.79 13.13 51.65 10.71 0.36 3.19 0.16 0.19 99.87

K216-6 (11) 2.44 0.44 11.44 6.09 13.27 51.52 10.02 0.33 3.06 0.17 0.22 98.99

K216-7B (17) 2.55 0.5 10.58 6.07 13.67 51.81 10.15 0.37 3.2 0.17 0.18 99.25

S687-5 (18) 2.47 0.49 10.19 6.79 13.24 52.34 10.58 0.35 3 0.16 0.15 99.75

S687-6 (13) 2.44 0.48 10.04 6.76 13.39 52.57 10.62 0.36 3.01 0.16 0.15 99.97

S687-11 (14) 2.47 0.48 10.12 6.95 13.07 52.19 10.52 0.35 2.93 0.16 0.15 99.37

S687-13 (16) 2.4 0.44 10.78 6.21 13.62 51.76 10.36 0.34 3.13 0.18 0.2 99.4

S691-4B (20) 2.77 0.63 10.67 6.03 14 49.57 10.96 0.45 3.5 0.15 0.19 98.92

S691-5A (24) 2.56 0.55 10.84 6.45 13.51 49.76 11.11 0.38 3.24 0.16 0.25 98.83

*Average of measured.yNumber of analyses.

2086


ownloaded from



ber 2018

to crystallize at �1390–1380�C; clinopyroxene crystal-lizes from �1140 to 1120�C and from 6 to 5�5 wt %MgO; plagioclase crystallizes from 1170 to 1080�C andfrom 5 to 2�6 wt % MgO, and finally orthopyroxeneappears at temperatures <1060�C and <3�5 wt %MgO, respectively.However, the temperature range of quenched lavas

calculated from the MgO contents of the pillow glassrims using the geothermometer of Helz & Thornber(1987) is from �1179�C to �1137�C, which is higherthan that of clinopyroxene crystallization. Accordingly,clinopyroxene, plagioclase, and orthopyroxene pheno-crysts in the bulk rocks may not have crystallized fromthe host magmas directly. A possible interpretation is thatthese minerals crystallized from a more evolved ‘cool’magma (i.e. local magma reservoir in the Hana Ridge)and may have been entrained in later magmas. Theoccurrence of pyroxene–plagioclase glomeroporphyriticaggregates in Hana Ridge lavas supports this inter-pretation.

We observed petrographic evidence that some min-erals are in disequilibrium with the host magma. Someolivine is rounded and resorbed and has reaction rims.Some partially resorbed olivine is commonly rimmed bynecklaces of small augite crystals. Rounded orthopyrox-ene phenocrysts are sheathed in augite and someorthopyroxene has pigeonite reaction rims. Roundedphenocrysts of plagioclase sometimes occur with thin,more Ca-rich rims. Some plagioclase has spongy coresriddled with glass inclusions that are sometimes sur-rounded by clean plagioclase. These resorbed and dis-equilibrium crystals occur side by side with euhedralphenocrysts that appear to have been in equilibriumwith the host magma. Mixing of primitive and evolvedliquids could account for the occurrence of these dis-equilibrium phenocrysts.Although magma mixing has probably occurred, frac-

tional crystallization and olivine accumulation are thedominant processes accounting for the compositionalvariations of submarine Hana Ridge lavas. This is

euhedral O1subhedral-undeformed O1deformed O1glass rimbulk rockdredge glass

Fig. 8. MgO variation diagrams comparing melt inclusions from the three types of olivine phenocrysts, pillow glass rims, and bulk rocks. Data fordredged samples from Moore et al. (1990). Major element totals are normalized to 100wt %. Fields for glass compositions from Frey et al. (1994).Error bars (2s) are indicated.

2087


ownloaded from



ber 2018

Table 4: Microprobe analyses of melt inclusions from euhedral, subhedral–undeformed, and deformed olivines in

Submarine Hana Ridge lavas

Euhedral olivine

K216-12B* (4)y K216-12B (2) K216-12B (2) K214-1 (4) K214-1 (9) K214-3 (4) K214-3 (5) K214-3 (6)

SiO2 52.54 53.84 52.57 52.53 51.77 54.64 53.99 53.29

TiO2 3.35 2.57 2.84 2.52 2.09 1.86 1.68 2.33

Al2O3 15.04 15.77 15.52 15.12 13.73 15.35 14.77 14.82

FeO 9.08 6.59 6.9 6.96 9.99 7.07 10.45 8.83

MnO 0.12 0.09 0.1 0.12 0.13 0.1 0.13 0.12

MgO 3.67 3.97 3.95 5.84 7.01 4.26 4.41 4.56

CaO 11.37 13.5 13.16 12.26 11.73 12.85 11.57 12.36

Na2O 2.66 2.37 2.48 2.28 2.02 2.44 2.23 2.37

K2O 0.41 0.31 0.34 0.29 0.28 0.33 0.3 0.32

P2O5 0.13 0.13 0.12 0.26 0.21 0.1 0.09 0.12

SO3 0.17 0.09 0.2 0.23 0.26 0.21 0.37 0.31

NiO 0.01 0.04 0.02 0.04 0.03 0.01 0.01 0.03

Total 98.55 99.24 98.17 98.43 99.25 99.23 99.97 99.44

K214-5B (4) K214-5B (3) K214-9 (7) K214-9 (4) K214-9 (2) K214-11 (1) K214-11 (1) K214-11 (1)

SiO2 52.49 53.8 54.85 54.83 54.61 51.53 53.53 53.53

TiO2 2.24 2.54 2.41 2.64 2.96 2.48 2.8 2.86

Al2O3 13.94 15.62 16.2 16.11 15.28 14.12 15.08 15.07

FeO 7.27 6.08 5.98 6.05 7.39 9.33 8.45 8.45

MnO 0.13 0.1 0.09 0.08 0.14 0.13 0.14 0.09

MgO 7.74 4.64 3.03 3.9 3.45 6.45 4.28 4.28

CaO 11.59 12.54 13.31 13.14 13.18 11.7 11.96 11.95

Na2O 2.19 2.44 2.82 2.55 2.9 2.35 2.59 2.48

K2O 0.35 0.39 0.39 0.45 0.43 0.35 0.44 0.39

P2O5 0.26 0.26 0.13 0.13 0.11 0.29 0.23 0.23

SO3 0.25 0.33 0.24 0.22 0.2 0.24 0.07 0.1

NiO 0.02 0.01 0.02 0.02 0.03 0.01 0 0

Total 98.46 98.74 99.45 100.11 100.69 98.98 99.57 99.42

K214-15A (3) K214-15B (4) K214-15B (3) K214-15C (6) K214-15C (4) K214-15C (4) K214-15C (5) K214-15C (2)

SiO2 52.97 51.98 52.4 52.91 52.66 53.01 50.97 53.3

TiO2 2.64 2.49 2.41 2.27 2.11 2.5 1.98 3.56

Al2O3 15.5 14.53 14.53 14.26 15.89 14.42 14.04 15.62

FeO 8.1 8.04 9.14 9.45 8.55 8.62 9.8 7.39

MnO 0.1 0.12 0.14 0.12 0.12 0.13 0.15 0.11

MgO 3.33 6.73 5.2 6.54 4.17 6.4 7.84 3.05

CaO 13.58 12.57 14 12 12.57 13.27 12.4 13.71

Na2O 2.55 2.14 1.99 2.12 2.41 1.95 1.8 2.24

K2O 0.41 0.36 0.34 0.36 0.47 0.33 0.36 0.26

P2O5 0.12 0.24 0.27 0.13 0.12 0.12 0.1 0.12

SO3 0.3 0.26 0.34 0.29 0.35 0.27 0.24 0.28

NiO 0.01 0.04 0.04 0 0.01 0.04 0.02 0.03

Total 99.59 99.5 100.8 100.44 99.42 101.05 99.7 99.66

2088


ownloaded from



ber 2018

Euhedral olivine

K214-15C (3) K214-15C (3) K212-2A (3) K212-2A (2) K212-4 (6) K212-4 (3) K212-4 (6)

SiO2 53.26 53.47 55.24 52.73 51.35 51.07 51.93

TiO2 2.22 3 1.82 2.06 2.13 2.62 2.4

Al2O3 15.66 14.98 15.76 14.92 13.98 14.5 14.7

FeO 9 7.66 8.53 6.73 10.65 7.3 7.84

MnO 0.13 0.09 0.13 0.07 0.18 0.16 0.16

MgO 3.05 5.7 2.82 4.74 5.48 7.26 5.99

CaO 13.35 13.13 11.12 12.54 11.9 14.47 13.19

Na2O 2.42 2.37 2.6 2.33 2.09 2.22 2.45

K2O 0.36 0.27 0.5 0.36 0.31 0.31 0.36

P2O5 0.11 0.12 0.27 0.24 0.11 0.16 0.14

SO3 0.31 0.28 0.35 0.31 0.2 0.15 0.25

NiO 0.04 0.01 0.01 0.03 0.01 0.03 0.01

Total 99.91 101.07 99.15 97.06 98.38 100.24 99.41


K214-1 (3) K214-1 (5) K214-3 (4) K214-3 (4) K214-3 (2) K214-3 (4) K214-3 (3) K214-9 (3)

SiO2 51.37 51.63 54.18 54.45 49.79 52.23 53.6 53.54

TiO2 1.91 1.95 2.44 2.47 2.31 2.46 2.45 2.67

Al2O3 13.41 13.45 15.62 15.21 13.97 15.06 15.23 14.78

FeO 10.42 9.61 7.92 7.23 10.73 8.61 8.85 7.59

MnO 0.15 0.13 0.13 0.11 0.14 0.13 0.12 0.11

MgO 7.6 7.47 3.11 4.29 4.71 3.38 3.43 5.11

CaO 11.07 11.3 12.81 12.71 15.16 12.31 12.68 13.53

Na2O 2.12 2.1 2.67 2.52 1.92 2.44 2.47 2.19

K2O 0.26 0.26 0.39 0.33 0.25 0.36 0.36 0.33

P2O5 0.21 0.19 0.11 0.11 0.13 0.11 0.11 0.12

SO3 0.29 0.26 0.19 0.28 0.26 0.31 0.26 0.15

NiO 0.03 0.02 0.01 0 0.01 0.02 0.02 0.03

Total 98.82 98.36 99.57 99.71 99.39 97.42 99.58 100.16

K214-9 (5) K214-11 (7) K214-15A (3) K214-15A (5) K214-15A (4) K214-15A (6) K214-15A (6) K212-2A (5)

SiO2 53.81 52 52.13 53.51 53.56 52.24 53.28 52.41

TiO2 2.39 2.56 2.96 2.81 3.17 3.19 1.81 2.49

Al2O3 14.95 14.27 15.23 15.46 15.79 15.47 14.57 13.93

FeO 7.92 10.35 7.36 6.77 7.02 8.45 8.65 9.97

MnO 0.1 0.16 0.11 0.09 0.11 0.12 0.13 0.15

MgO 5.88 5.07 3.99 3.99 3.29 3.51 5.57 4.54

CaO 12.95 11.18 13.62 13.54 13.37 13.89 12.64 13.57

Na2O 2.16 2.43 2.15 2.04 2.62 2.41 2.22 2.09

K2O 0.27 0.38 0.48 0.39 0.32 0.35 0.25 0.38

P2O5 0.1 0.27 0.18 0.13 0.12 0.12 0.09 0.27

SO3 0.14 0.24 0.25 0.31 0.3 0.29 0.24 0.32

NiO 0.03 0.01 0 0.01 0.01 0.01 0.01 0

Total 100.71 98.9 98.47 99.05 99.68 100.04 99.45 100.11

2089


ownloaded from



ber 2018

Table 4: continued

Subhedral—undeformed olivine Deformed olivine

K212-2A (2) K212-2A (3) K212-2A (3) K214-1 (4) K214-1 (5) K214-3 (5) K214-3 (4) K214-3 (4)

SiO2 52.65 51.8 52.71 53.18 51.3 53.86 51.37 56.32

TiO2 2.2 2.14 2.39 2.14 2.04 2.26 2.31 1.96

Al2O3 14.76 13.73 14.98 14.81 13.7 14.99 14.21 15.76

FeO 9.08 9.83 9.07 6.73 9.71 8.38 11.1 7.21

MnO 0.11 0.12 0.14 0.09 0.12 0.11 0.14 0.1

MgO 4.21 6.19 3.95 7.45 7.28 3.66 3.86 2.64

CaO 13 12.14 12.86 12.05 11.44 12.94 13.77 11.3

Na2O 2.18 2.15 2.17 2.51 2.16 2.38 1.97 2.6

K2O 0.48 0.42 0.49 0.27 0.3 0.33 0.27 0.38

P2O5 0.29 0.27 0.28 0.22 0.23 0.11 0.1 0.11

SO3 0.29 0.26 0.3 0.33 0.15 0.3 0.25 0.19

NiO 0.01 0.03 0.02 0.02 0.02 0.01 0.01 0

Total 99.25 99.08 99.35 99.81 98.47 99.32 99.36 98.56

Deformed olivine

K214-3 (3) K214-5B (4) K214-5B (5) K214-5B (3) K214-5B (3) K214-9 (3) K214-9 (3) K214-9 (5)

SiO2 54.52 52.77 53.09 51.33 53.36 56.59 52.86 54.53

TiO2 2.3 1.88 1.61 1.93 2.46 2.66 2.79 2.63

Al2O3 15.93 14.04 14.53 13.43 15.05 17.01 16.43 15.86

FeO 7.21 7.2 8.19 9.76 8.89 4.03 7.53 6.08

MnO 0.13 0.11 0.09 0.16 0.11 0.07 0.1 0.1

MgO 2.74 7.28 6.11 7.76 3.96 2.46 2.91 4.46

CaO 12.33 12.18 13.13 12.72 12.15 12.11 13.62 13.68

Na2O 2.83 2.02 2.26 1.96 2.37 2.97 2.63 2.78

K2O 0.42 0.27 0.27 0.25 0.38 0.57 0.47 0.36

P2O5 0.12 0.21 0.22 0.24 0.24 0.13 0.17 0.11

SO3 0.14 0.29 0.32 0.3 0.22 0.22 0.19 0.19

NiO 0 0.03 0 0.01 0.02 0.01 0.04 0.01

Total 98.68 98.29 99.83 99.85 99.22 98.82 99.74 100.77

K214-15A (1) K214-15B (3) K214-15C (3) K214-15C (4) K212-2A (4)

SiO2 55.29 52.84 52.92 50.58 52.54

TiO2 2.65 2.51 2.41 2.05 2.2

Al2O3 16.5 15.02 15.14 13.9 14.37

FeO 6.67 7.46 9.44 9.27 6.92

MnO 0.1 0.11 0.17 0.11 0.11

MgO 2.24 5.11 3.86 7.68 6.15

CaO 11.59 14.17 13.53 13.43 12.41

Na2O 2.84 2.44 2.22 1.83 2.21

K2O 0.47 0.39 0.36 0.25 0.35

P2O5 0.11 0.31 0.1 0.1 0.28

SO3 0.3 0.33 0.23 0.2 0.23

NiO 0 0.02 0.01 0.02 0.01

Total 98.74 100.7 100.39 99.41 97.76

*Average of measured.yNumber of analyses.

2090


ownloaded from



ber 2018

because very small amounts of pyroxene and plagio-clase are present in Hana Ridge lavas (each of theseminerals is usually <1 vol. % in rocks). Mixing wouldalso fractionate Sm/Sr and Ti/Eu, because these

element ratios are sensitive indicators of plagioclaseand Fe–Ti oxide crystallization. These ratios are nearlyconstant in the submarine Hana Ridge magma (Sm/Sr ¼ 0�019 � 0�0021; Ti/Eu ¼ 7795�97 � 434�17),

S691-4B

Fig. 9. Abundances (ppm) of trace elements Sr, Zr, Sc, and Ni vs MgO content (wt %) in lavas from the submarine Hana Ridge lavas. Error bars(2s) are indicated.

Fig. 10. Sm/Nd, La/Ce, and La/Nb variations of bulk rocks against water depth. Subaerial data for Honomanu lavas from Chen et al. (1991).Error bars (2s) are indicated.

2091


ownloaded from



ber 2018

indicating that magma mixing was not volumetricallyimportant.MELTS simulations, as stated above (see Fig. 13),

indicate that a significant amount of the compositionalvariation of the submarine Hana Ridge lavas can beexplained by olivine fractionation and accumulation.The compositional trends in other elements also implythat the compositional variation of submarine HanaRidge lavas can be explained by olivine fractionation

and accumulation. The Ni–MgO trend defined by thewhole rocks intersects near the middle of the Ni–MgOfield of the olivines (Fig. 14), and we infer that thewhole-rock trend reflects olivine control with a meancomposition of Fo87 (i.e. �46�95 wt % MgO, 13%FeO, 39�4% SiO2, and 0�38 wt % NiO; see Fig. 14).Clague et al. (1995) also found a similar result in FeO*,Ni vs MgO plot for lavas dredged from the KilaueaPuna Ridge, and calculated that Fo87�7 is the averagecomposition of olivine crystallized from the primarymagma. It appears that magmatism during the mainshield-building stage of Haleakala volcano (i.e. sub-marine Hana Ridge lavas) was very similar to that inthe modern Kilauea volcano.The olivine from the submarine Hana Ridge has

compositional characteristics indicative of crystallizationfrom magma. However, the kink-banded olivine pheno-crysts could not have formed in a liquid. Deformation ofolivine phenocrysts in the submarine Hana Ridge rockscould be explained by the mechanism proposed byClague et al. (1995). According to those workers, asmuch as 22 wt % of olivine should have crystallizedfrom primary Kilauea magma, but these olivine crystalsare only rarely included in the magma upon eruption.The olivine in fact accumulated at the base of themagma chamber, and may have deformed during flowof the still-hot dunite body, prior to entrainment in latermagmas. The olivine crystals might have been deformedduring flow of a nearly solid crystal mush within therift zone.

Magma source mineralogy

Given our estimate of the primary magma composition,and analytical data for submarine Hana Ridge lavas, itis possible to evaluate the residual mineralogy that con-trolled the composition of the primary melts. To approx-imate the primary magma composition, the measuredabundances of incompatible trace elements are correctedto 16�7 wt % MgO by addition or subtraction of Fo87olivine. Ratios of incompatible trace elements also pro-vide important constraints without correction for olivineaccumulation (Norman & Garcia, 1999).It is well known, based on trace element character-

istics, that the Hawaiian tholeiitic lavas were derivedfrom a source containing residual garnet (e.g. Clague& Frey, 1982; Frey & Roden, 1987; Frey & Rhodes,1993; Norman & Garcia, 1999). Trace element patternscorrected to 16�7 wt % MgO (Fig. 15) in submarineHana Ridge lavas show significant fractionation of thelight REE (LREE) but little fractionation of the heavyREE (HREE). The variations in LREE abundances mayreflect variations in the degree of melting, whereas thelack of variation in the HREE abundances reflects theirhigh compatibility in the source because of the presence

Fig. 11. Zr/Nb, Sr/Nb, and Ba/Nb vs MgO (wt %) for bulk rocks.Discriminant boundaries for Kilauea and Mauna Loa lavas from Frey& Rhodes (1993). Subaerial data for Honomanu lavas from Chen et al.(1991). Error bars (2s) are indicated.

2092


ownloaded from



ber 2018

of residual garnet (Frey et al., 1980; Hofmann et al.,1984). Other features of the trace element systematicsare inconsistent with the persistence of apatite andamphibole in the source region.Although residual garnet is required, the lavas also

show evidence for residual clinopyroxene in their source.Following Wagner et al. (1998), Sm/Sr is plotted againstSr (corrected to 16�7 wt %MgO, Fig. 16a). The Sm/Sr ofthe lavas is nearly constant whereas the abundances of

Sm and Sr each show variation in correlated data, imply-ing that the two elements have partition coefficientsbetween source material and melt that are nearly ident-ical. When garnet is present as the sole residual phase,the D for Sr in garnet is very low, �0�0023 ( Johnson,1998; Green et al., 2000), and results in DSm/DSr of �78for garnet, which is too high to buffer Sm/Sr in the melts.To maintain a constant Sm/Sr (see Fig. 16a) duringpartial melting, a residual phase that has similar DSm

Fig. 12. Plot of a�meltCaO ----FeO=MgO for bulk rocks and melts equilibrated with the most forsteritic (Fo >90%) olivines. *, values calculated for the

bulk rocks. The ordinate (a�meltCaO ) represents the pseudo-activity for CaO in the melt as calculated from the bulk rocks using Libourel’s (1999)

equations. ^, a�meltCaO ----FeO=MgO of melt values calculated for olivine phenocrysts with Fo >90%. For this calculation, K

Fe�MgD of olivine–melt is

assumed to be 0�30 and Fe2þ/FeTotal of the melt that equilibrated with olivine is assumed to be 0�9. (See text for further explanation.)

Table 5: Estimated primary magma composition for Hana Ridge

SiO2 TiO2 Al2O3 FeO* MnO MgO CaO Na2O K2O P2O5 H2O

47.8 2.3 10.5 11.5 0.17 16.7 8.4 1.6 0.3 0.22 0.5—1

2093


ownloaded from



ber 2018

and DSr is necessary. D for Sr is moderately high in thebulk solid–melt and D for Sr is between that of Nd andSm, based on calculations using the Hofmann & Feigen-son (1983) inversion model. This constraint rules outolivine, orthopyroxene and spinel, which have very lowD values for Sr and Sm. Only clinopyroxene has a mod-erately high D of Sr, of �0�11 (Johnson, 1998; Greenet al., 2000), which is significantly higher than that of theother phases, and results in a DSm/DSr of �2 (Hauri et al.,1994; Johnson, 1998; Green et al., 2000). This indicatesthat the clinopyroxene/garnet ratio in the sourcemay be high.A Tb/Yb vs Th diagram also indicates the presence of

clinopyroxene in the source (Frey et al. (2000; Fig. 16b).The D of garnet/melt for Yb is greater than that for Tb(Green et al., 2000), whereas Yb and Tb partition intoclinopyroxene nearly equally ( Johnson, 1998; Greenet al., 2000). If garnet is dominant in the residue, theTb/Yb ratios of melts should decrease dramatically as

the degree of melting of garnet increases. Because Thdecreases with increasing degree of melting (Frey et al.,2000; Sisson et al., 2002), so melting of a source in thegarnet peridotite stability field produces a positive cor-relation of Tb/Yb with Th. However, the submarineHana Ridge rocks show nearly constant Tb/Yb vs Th(correlated to 16�7 wt % MgO) (Fig. 16b), also indicatingthat the clinopyroxene/garnet ratio may be high in thesource.The presence of a clinopyroxene-enriched source is

consistent with the slightly negative correlation betweenTh content and CaO content, whereas the Al2O3 con-tent is relatively constant (Fig. 16c and d). This can beexplained by the presence of a residual high-CaO clino-pyroxene in the source. Th decreases with increasingdegree of melting whereas CaO increases because CaOis more likely to behave as a compatible componentduring partial melting of clinopyroxene-rich source(Yasuda et al., 1994; Kogiso et al., 1998). On the other

Fig. 13. SiO2, CaO, Al2O3, and FeO vs MgO (wt %) showing calculated fractional crystallization trends using the MELTS program (Ghiorso &Sack, 1995). þ, Submarine Hana Ridge bulk rocks. (See text for further explanation.)

2094


ownloaded from



ber 2018

hand, the Al2O3 contents of the lavas (normalized toMgO 16�7 wt %) may be relatively constant becauseAl2O3 is buffered by residual garnet (e.g. Kinzler, 1997;Walter, 1998).We apply the trace elements inversion model of

Hoffmann & Feigenson (1983) to evaluate the sourcemineralogy quantitatively. The results also indicate thatthe clinopyroxene/garnet ratio is high (>2) in thesubmarine Hana Ridge source.The primary magma composition for submarine Hana

Ridge is plotted in Fig. 17, showing the melt compositions

derived from partial melting of various peridotites (Hirose& Kushiro, 1993; Kushiro, 1996; Walter, 1998). None ofthe starting materials used for melting experiments, suchas KLB-1, PHN1611, KR4003, and HK-66, can yieldprimary basaltic magma similar to the submarine HanaRidge. The Hana Ridge primary magma has much lowerAl2O3 and CaO than experimentally generated melts at agiven MgO content (Fig. 17). This result was noted byTakahashi et al. (1993), who found that FeO and TiO2

are significantly higher, and CaO and Al2O3 are lower, inthe Hawaiian tholeiitic lavas than in experimentally gen-erated melts (KLB-1) at a given MgO content. They alsoproposed that the source materials for Hawaiian tholeiitemagmas are significantly different from the normal mag-nesian peridotite predominant in mantle xenoliths andtectonic blocks.Peridotite melting experiments have demonstrated that

concentrations of FeO and incompatible elements (Ti, K,etc.) in partial melts depend strongly on the source peri-dotite composition (Takahashi & Kushiro, 1983; Kogisoet al., 1998). On the other hand, SiO2 content is ratherinsensitive to source composition but depends on pres-sure (Hirose & Kushiro, 1993). Because CaO and Al2O3

are higher and TiO2 is lower in all experimentally gen-erated peridotite partial melts than Hawaiian magmawhen compared at similar MgO, a source materialother than normal peridotite is necessary.

CONCLUSION

All the rock samples from the six submarine divesanalyzed in this study are tholeiitic basalts or picrites,similar to the lavas of the Kilauea shield-building stage.The compositions of melt inclusions and olivine indicatethat all olivine (regardless of its morphology) crystallizedfrom the host magmas. The primary magma composi-tion is estimated to have had �16�7 wt % MgO and�8�4 wt % CaO. Major and trace elements character-istics and simulations with the MELTS program implythat fractional crystallization and accumulation werethe dominant processes in the evolution of submarineHana Ridge lavas, and conditions of fractional crystal-lization were 0�1–0�3GPa pressure and 0�5–1 wt %H2O in the melt. The trace element characteristics,together with major element compositions, indicatethat Haleakala volcano originally had a source compo-sition similar to Kilauea. During the growth history ofHaleakala, the magma source changed from Kilauea-type in the submarine Hana Ridge towards Kilauea–Mauna Loa intermediate-type in the subaerial Hono-manu stage. Major element and trace element charac-teristics of the lavas imply that both clinopyroxene andgarnet were important residual phases during partialmelting.

Fig. 14. Ni (ppm), FeO (wt %), and SiO2 (wt %) vs MgO (wt %)diagram for submarine Hana Ridge lavas and olivine phenocrystscontained in these. The whole-rock compositional trends reflect olivinecontrol with a mean composition of �46�95wt % MgO, �13% FeO,�39�4% SiO2, and �0�38wt % NiO (Fo87).

2095


ownloaded from



ber 2018

Fig. 16. Correlation of some major and trace elements in submarine Hana ridge lavas. (a) Sm/Sr vs Sr (ppm); (b) Tb/Yb vs Th (ppm); (c) CaO(wt %) vs Th (ppm); (d) Al2O3 (wt %) vs Th (ppm). The lava compositions are corrected to 16�7wt % MgO.

Fig. 15. Primitive mantle (Sun & McDonough, 1989) normalized trace element patterns of submarine Hana Ridge lavas. Trace elementconcentrations of lavas are corrected to 16�7wt % MgO by adding or subtracting Fo87 olivine. The high-Sr lavas (S691-4B, S691-5A, B)correspond to the high total alkali lavas shown in Fig. 5. (See text for explanation.)

2096


ownloaded from



ber 2018

ACKNOWLEDGEMENTS

We thank the officers, crew, particularly CaptainO. Yukawa of the R.V. Kairei and Captain H. Tanakaof the Yokosuka, T. Fukui and Y. Imai (commanders of theR.O.V. Kaikou and Shinkai 6500), and the operation teamssupport during the Japan–US Hawaiian cruise in2001–2002. We also thank the scientific team for ship-board assistance and subsequent discussion. We thankF. A. Frey and M. O. Garcia for their many valuablecomments. We are grateful to Dr Kogiso Tetsu for dis-cussions and suggestions. This research was supported byGrant 12002006 from the Ministry of Education andScience to E.T. This study was also supported by theEarthquake Research Institute of the University ofTokyo co-operative research program. Constructivereviews by F. Frey and M. Garcia, and editorial revisionsby R. Arculus, M. Wilson and Editorial AssistantA. Lumsden are much appreciated.

SUPPLEMENTARY DATA

Supplementary data for this paper and available on Jour-

nal of Petrology online.

REFERENCES

Anderson, A. T., Jr & Brown, G. G. (1993). CO2 and formation

pressures of some Kilauea melt inclusions. American Mineralogist 78,

794–803.

Chen, C.-Y. (1993). High-magnesium primary magmas from Halea-

kala Volcano, east Maui, Hawaii: petrography, nickel, and major-

element constraints. Journal of Volcanology and Geothermal Research 55,

143–153.

Chen, C.-Y. & Frey, F. A. (1983). Origin of the Hawaiian tholeiite and

alkalic basalt. Nature 302, 785–789.

Chen, C.-Y. & Frey, F. A. (1985). Trace element and isotopic

geochemistry of lavas from Haleakala Volcano, East Maui, Hawaii:

implications for the origin of Hawaiian basalts. Journal of Geophysical

Research 90, 8743–8768.

Chen, C.-Y., Frey, F. A. & Garcia, M. O. (1990). Evolution of alkalic

lavas at Haleakala Volcano, East Maui, Hawaii. Contributions to

Mineralogy and Petrology 105, 197–218.

Chen, C.-Y., Frey, F. A., Garcia, M. O., Dalrymple, G. B. & Hart, S. R.

(1991). The tholeiite to alkalic basalt transition at Haleakala Volcano,

Maui, Hawaii. Contributions to Mineralogy and Petrology 106, 183–200.

Clague, D. A. & Frey, F. A. (1982). Petrology and trace element

geochemistry of Honolulu, Oahu: implications for the oceanic

mantle below Hawaii. Journal of Petrology 23, 447–504.

Clague, D. A., Weber, W. & Dixon, J. E. (1991). Picritic glasses from

Hawaii. Nature 353, 553–556.

Clague, D. A., Moore, J. G., Dixon, J. E. & Friesen, W. B. (1995).

Petrology of submarine lavas from Kilauea’s Puna Ridge, Hawaii.

Journal of Petrology 36, 299–349.

Clague, D. A., Moore, J. G. & Reynolds, J. R. (2000). Formation of

submarine flat-topped volcanic cones in Hawaii. Bulletin of Volcanology

62, 214–233.

Delaney, P. T., Fiske, R. S., Miklius, A., Okamura, A. T. & Sako, M. K.

(1990). Deep magma body beneath the summit and rift zones of

Kilauea Volcano, Hawaii. Science 247, 1311–1316.

Eisele, J. Abouchami, W., Galer, S. J. G. & Hofmann, A. W. (2003).

The 320 kyr Pb isotope evolution of Mauna Kea lavas recorded in

the HSDP-2 drill core. Geochemistry, Geophysics, Geosystems 4, No. 5,

8710, doi:10.1029/2002GC000339,2003.

Fig. 17. MgO variation diagrams (in wt %) for melts experimentally generated in peridotite partial melting. Data for partial melting experimentsfrom Hirose & Kushiro (1993), Kushiro (1996) and Walter (1998). Primary magma of submarine Hana Ridge lavas as estimated in Fig. 10 isplotted as large gray circles. The average compositions of seven Kilauea MgO-rich glasses (Clague et al., 1991) are shown for comparison.

2097


ownloaded from



ber 2018

Frey, F. A. & Rhodes, J. M. (1993). Intershield geochemical differences

among Hawaiian volcanoes: implications for source compositions,

melting process and magma ascent paths. Philosophical Transactions of

the Royal Society of London, Series A 342, 121–136.

Frey, F. A. & Roden, M. F. (1987). The mantle source for the Hawaiian

Islands: constraints from the lavas and ultramafic inclusions. In:

Menzies, M. A. & Hawkesworth, C. J. (eds) Mantle Metasomatism.

London: Academic Press, pp. 423–463.

Frey, F. A., Roden, M. F. & Zindler, A. (1980). Constraints on mantle

source compositions imposed by phosphorus and the rare-earth

elements. Contributions to Mineralogy and Petrology 75, 165–173.

Frey, F. A., Garcia, M. O. & Roden, M. F. (1994). Geochemical

characteristics of Koolau Volcano: implications of intershield

differences among Hawaiian volcanoes. Geochimica et Cosmochimica

Acta 58, 1441–1462.

Frey, F. A., Clague, D., Mahoney, J. J. & Sinton, J. M. (2000).

Volcanism at the edge of the Hawaiian plume: petrogenesis of

submarine alkalic lavas from the North Arch volcanic field. Journal of

Petrology 41, 667–691.

Garcia, M. O., Hulsebosch, T. P. & Rhodes, J. M. (1995). Olivine-rich

submarine basalts from the southwest rift zone of Mauna Loa

Volcano: implications for magmatic processes and geochemical

evolution. In: Rhodes, J. M. & Lockwood, J. P. (eds) Mauna Loa

Decade Volcano. Geophysical Monograph, American Geophysical Union 92,

219–239.

Ghiorso, M. S. & Sack, R. O. (1995). Chemical mass transfer in

magmatic processes. IV. A revised and internally consistent

thermodynamic model for the interpolation and extrapolation of

liquid–solid equilibria in magmatic systems at elevated and

pressures. Contributions to Mineralogy and Petrology 119, 197–212.

Green, T. H., Blundy, J. D., Adam, J. & Yaxley, G. M. (2000). SIMS

determination of trace element partition coefficients between garnet,

clinopyroxene and hydrous basaltic liquids at 2–7�5GPa and

1080–1200�. Lithos 53, 165–187.Hauri, E. H., Wagner, T. P. & Grove, T. L. (1994). Experimental and

natural partitioning of Th, U, Pb and other trace elements between

garnet, clinopyroxene and basaltic melts. Chemical Geology 117,

149–166.

Helz, R. T. & Thornber, C. R. (1987). Geothermometry of Kilauea Iki

lava lake, Kilauea Volcano, Hawaii. Bulletin of Volcanology 49,

651–668.

Hirose, K. & Kushiro, I. (1993). Partial melting of dry peridotites at

high pressures: determination of compositions of melts segregated

from peridotite using aggregates of diamond. Earth and Planetary

Science Letters 114, 477–489.

Hofmann, A. W. & Feigenson, M. D. (1983). Case studies on the origin

of basalt. I. Theory and reassessment of Grenada basalts. Contributions

to Mineralogy and Petrology 84, 382–389.

Hofmann, A. W., Feigenson, M. D. & Raczek, I. (1984). Case studies

on the origin of basalt. III. Petrogenesis of the Mauna Ulu eruption,

Kilauea, 1969–1971. Contributions to Mineralogy and Petrology 88, 24–35.

Jackson, M. C., Frey, M. O., Garcia, M. O. & Wilmoth, R. A. (1999).

Geology and petrology of basaltic lavas and dikes of the Koolau

Volcano in the Trans-Koolau exploratory tunnels. Bulletin of

Volcanology 60, 381–401.

Johnson, K. T. M. (1998). Experimental determination of partition

coefficients for rare earth and high-field-strength elements between

clinopyroxene, garnet, and basaltic melt at high pressures. Contribu-

tions to Mineralogy and Petrology 133, 60–68.

Johnson, K. T. M., Reynolds, J. R., Vonderhaar, D., Smith, D. &

Kong, L. S. L. (2002). Petrological systematics of submarine basalt

glasses from the Puna Ridge, Hawaii: implications for rift zone

plumbing and magmatic processes. In: Takahashi, E., Lipman, P. T.,

Garcia, M. O., Naka, J. & Aramaki, S (eds) Hawaiian Volcanoes: Deep

Underwater Perspectives. American Geophysical Union Monograph 128,

143–159.

Kinzler, R. J. (1997). Melting of mantle peridotitie at pressure

approaching the spinel to garnet transition: application to mid-

ocean ridge basalt petrogenesis. Journal of Geophysical Research 102,

853–874.

Klein, F. W., Koyanagi, R. W., Nakata, J. S. & Tanigawa, W. R.

(1987). The seismicity of Kilauea’s magma system. US Geological

Survey Professional Paper 1350, 1019–1185.

Kogiso, T., Hirose, K. & Takahashi, E. (1998). Melting experiments on

homogeneous mixing of peridotitic and basalt: application to the

genesis of ocean island basalts. Earth and Planetary Science Letters 162,

45–61.

Kushiro, I. (1996). Partial melting of a fertile mantle peridotite at

high pressures: an experimental study using aggregates of diamond.

In: Basu, A. & Hart, S. (eds) Earth Processes: Reading the Isotopic

Code. Geophysical Monograph, American Geophysical Union 95,

109–122.

Libourel, G. (1999). Systematics of calcium partitioning between olivine

and silicate melt: implications for melt structure and calcium content

of magmatic olivines. Contributions to Mineralogy and Petrology 136,

63–80.

Maaloe, P. (1979). Compositional range of primary tholeiitic magmas

evaluated from major element trends. Lithos 12, 59–72.

Macdonald, G. A. (1978). Geological map of the crater section of

Haleakala Nature Park, Maui, Hawaii. US Geological Survey,

Miscellaneous Investigations Series Map I-1088.

Macdonald, G. A. & Katsura, T. (1964). Chemical composition of

Hawaiian lavas. Journal of Petrology 5, 82–113.

Macdonald, G. A., Abbott, A. T. & Peterson, F. L. (1983). Volcanoes in

the Sea: the Geology of Hawaii, 2nd edn. Honolulu, HI: University of

Hawaii Press.

Moore, J. G. & Clague, D. A. (1987). Coastal lava flows from Mauna

Loa and Hualalai Volcanoes, Kona, Hawaii. Bulletin of Volcanology

49, 752–764.

Moore, J. G. & Thomas, D. M. (1988). Subsidence of Puna, Hawaii

inferred from sulfur content of drilled lava flows. Journal of Volcanology

and Geothermal Research 3510, 165–171.

Moore, J. G., Clague, D. A., Ludwig, K. R. & Mark, R. K. (1990).

Subsidence and volcanism of the Haleakala Ridge, Hawaii. Journal

of Volcanology and Geothermal Research 42, 273–284.

Nakamura, N. (1978). Determination of REE, Ba, Fe, Mg, Na and K in

carbonaceous and ordinary chondrites. Geochimica et Cosmochimica Acta

38, 757–775.

Norman, M. D. & Garcia, M. O. (1999). Primitive magmas and

source characteristics of the Hawaiian plume: petrology and

geochemistry of shield picrites. Earth and Planetary Science Letters 168,

27–44.

Orihashi, Y. & Hirata, T. (2003) Rapid quantitative analysis of Y and

REE abundances in XRF glass bead for selected GSJ reference rock

standards using UV laser ablation ICP-MS. Geochemical Journal 37,

401–412.

Reynolds, J. R., Clague, D. A., Maher, N. & Hatcher, G. (1998).

Evolutionary sequence of submarine volcanic rift zones in

Hawaii (abstract). EOS Transactions, American Geophysical Union 79,

F825.

Roeder, P. L. (1974). Activity of iron and olivine solubility in basaltic

liquids. Earth and Planetary Science Letters 23, 397–410.

Roeder, P. L. & Emslie, R. F. (1970). Olivine–liquid equilibrium.

Contributions to Mineralogy and Petrology 29, 275–289.

Ryan, M. P., Koyanagi, R. Y. & Fiske, R. S. (1981). Modeling the

three-dimensional structure of macroscopic magma transport

2098


ownloaded from



ber 2018

systems: application to Kilauea volcano. Journal of Geophysical Research

86, 7111–7129.

Sen, G. (1988). Petrogenesis of spinel lherzolite and pyroxenite suite

xenoliths from the Koolau shield, Oahu, Hawaii: implications for

petrology of the post-eruptive lithosphere beneath Oahu. Contribu-

tions to Mineralogy and Petrology 100, 61–91.

Sherrod, D. R., Nishimitsu, N. & Tagami, T. (2003). New K–Ar

ages and the geologic evidence against rejuvenated-stage volcanism

at Haleakala, East Maui, a postshield-stage volcano of the

Hawaiian island chain. Geological Society of America Bulletin 6,

683–694.

Shinozaki, K., Ren, Z.-Y. & Takahashi, E. (2002). Geochemical and

petrological characteristics of Nuuanu and Wailau landslide blocks.

In: Takahashi, E., Lipman, P. T., Garcia, M. O., Naka, J. &

Aramaki, S. (eds) Hawaiian Volcanoes: Deep Underwater Perspectives.

American Geophysical Union Monograph 128, 297–310.

Sisson, T. W., Lipman, P. W. & Naka, J. (2002). Submarine alkalic

through tholeiitic shield-stage development of Kilauea Volcano,

Hawaii. In: Takahashi, E., Lipman, P. T., Garcia, M. O., Naka, J. &

Aramaki, S. (eds) Hawaiian Volcanoes: Deep Underwater Perspectives.

American Geophysical Union Monograph 128, 193–219.

Smith, J. R., Satake, K., Morgan, J. K. & Lipman, P. W. (2002).

Submarine landslide and volcanic features on Kohala and Mauna

Kea volcanoes and the Hana Ridge, Hawaii. In: Takahashi, E.,

Lipman, P. T., Garcia, M. O., Naka, J. & Aramaki, S. (eds) Hawaiian

Volcanoes: Deep Underwater Perspectives. American Geophysical Union

Monograph 128, 11–28.

Stearns, H. T. & Macdonald, G. A. (1942). Geology and ground-water

resources of the island of Maui, Hawaii. Hawaii. Division of

Hydrography Bulletin 7, 344 pp. (Part 1, General geology and

ground-water resources of Maui, pp. 1–222; Part 2, Geology and

ground-water resources of the Nahiku area, West Maui, pp. 223–

274; Part 3, Petrography of Maui, pp. 275–334; 2 folded maps in

pocket, scale 1:62 500.)

Sun, S.-s. & McDonough, W. F. (1989). Chemical and isotope

systematics of oceanic basalts: implications for mantle composition

and processes. In: Saunders, A. D. & Norry, M. J. (eds) Magmatism

in the Ocean Basins. Geological Society, London, Special Publications 42, 313–

345.

Takahashi, E. (1978). Partitioning of Ni2þ, Co2þ, Fe2þ, Mn2þ, and

Mg2þ between olivine and silicate melt: compositional

dependence of partition coefficient. Geochimica et Cosmochimica Acta

42, 1829–1844.

Takahashi, E. & Kushiro, I. (1983). Melting of a dry peridotite at

high pressure and basalt magma genesis. American Mineralogist 68,

859–879.

Takahashi, E., Shimazaki, T., Tsuzaki, Y. & Yoshida, H. (1993).

Melting study of a peridotite KLB-1 to 6�5GPa, and the origin of

basaltic magmas. Philosophical Transactions of the Royal Society of London,

Series A 342, 105–120.

Tanaka, R., Nakamura, E., & Takahashi, E. (2002). Geochemical

evolution of Koolau Volcano, Hawaii. In: Takahashi, E., Lipman,

P. T., Garcia, M. O., Naka, J. & Aramaki, S. (eds) Hawaiian Volcanoes:

Deep Underwater Perspectives. American Geophysical Union Monograph 128,

311–332.

Tani, K., Orihashi, Y. & Nakada, S. (2002). Major and trace

components analysis of silicate rocks by X-ray fluorescence spectro-

meter using fused glass beads: evaluation of analytical precision of

three, six, eleven times dilution fused glass beads methods. Technical

Research Report, Earthquake Research Institute, University of Tokyo 8, 26–36

(in Japanese with English abstract).

Wagner, T. P., Clague, D. E., Hauri, E. H. & Grove, T. L. (1998).

Trace element abundances of high-MgO glasses from Kilauea,

Mauna Loa and Haleakala Volcanoes, Hawaii. Contributions to

Mineralogy and Petrology 131, 13–21.

Wallace, P. J. & Anderson, A. T., Jr (1998). Effects of eruption and lava

drainback on the H2O contents of basaltic magmas at Kilauea

Volcano. Bulletin of Volcanology 59, 327–344.

Walter, M. J. (1998). Melting of garnet peridotite and the origin of

komatiite and depleted lithosphere. Journal of Petrology 39, 29–60.

West, H. B. & Leeman, W. P. (1987). Isotopic evolution of lavas

from Haleakala Crater, Hawaii. Earth and Planetary Science Letters 84,

211–225.

West, H. B. & Leeman, W. P. (1994). The open-system geochemical

evolution of alkalic cap lavas from Haleakala Crater, Hawaii, USA.

Geochimica et Cosmochimica Acta 58, 773–796.

Wright, T. (1984). Origin of Hawaiian tholeiite: a metasomatic model.

Journal of Geophysical Research 89, 3233–3252.

Yang, H.-J., Frey, F. A., Clague, D. A. & Garcia, M. O. (1999).

Mineral chemistry of submarine lavas from Hilo Ridge, Hawaii:

implications for magmatic processes within Hawaiian rift zones.

Contributions to Mineralogy and Petrology 135, 355–372.

Yasuda, A., Fuji, T. & Kurita, K. (1994). Melting phase relations of an

anhydrous mid-ocean ridge basalt from 3 to 20GPa: implications for

the behavior of subducted oceanic crust in the mantle. Journal of

Geophysical Research 99, 9401–9414.

2099


ownloaded from



ber 2018

Documents

Petrogenesis of Tholeiitic Lavas from the Submarine Hana Ridge