Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Geochemistry of post-shield lavas from Kea- and Loa-trend Hawaiian volcanoes : constraints on the origin… Hanano, Diane 2008

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

Item Metadata

Download

Media
24-ubc_2008_fall_hanano_diane.pdf [ 3.72MB ]
Metadata
JSON: 24-1.0052325.json
JSON-LD: 24-1.0052325-ld.json
RDF/XML (Pretty): 24-1.0052325-rdf.xml
RDF/JSON: 24-1.0052325-rdf.json
Turtle: 24-1.0052325-turtle.txt
N-Triples: 24-1.0052325-rdf-ntriples.txt
Original Record: 24-1.0052325-source.json
Full Text
24-1.0052325-fulltext.txt
Citation
24-1.0052325.ris

Full Text

GEOCHEMISTRY OF POST-SHIELD LAVAS FROM KEA- AND LOA-TREND HAWAIIAN VOLCANOES: CONSTRMNTS ON THE ORIGN AND DISTRIBUTION OF HETEROGENEITIES IN THE HAWAIIAN MANTLE PLUME  by DIANE HANANO B.Sc.H., The University of British Columbia, 2004  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (GEOLOGICAL SCIENCES)  THE UNIVERSITY OF BRITISH COLUMBIA July 2008  © Diane Hanano, 2008  ABSTRACT The alteration mineralogy, major and trace element chemistry, and Sr-Nd-Pb-Hf isotopic compositions of post-shield lavas from Mauna Kea, Kohala, and Hualalai on the island of Hawaii in the Pacific Ocean are used to constrain the origin and distribution of heterogeneities in the Hawaiian mantle plume. Ocean island basalts contain a variety of secondary minerals that must be removed by acid-leaching to achieve high-precision Pb isotopic compositions, a powerful geochemical tracer of variation in plume source composition. Post-shield lavas range from transitional/alkalic basalt to trachyte and are enriched in incompatible trace elements (e.g. LaN/ThN = 6.0-16.2) relative to shield stage tholeiites. Post-shield lavas are characterized by a limited range of Sr-Nd-Hf isotopic compositions Sr/ 87 = 0.70343-0.70365; Nd/’”Nd ( Sr 86 43 = 0.51292-0.51301; Hf” ‘ 76 = ‘ Hf 77 0.28311-0.28314) and have Pb isotopic compositions Pb/ 206 = 17.89-18.44; ( Pb 204 Pb 204 Pb/ 207  15.44-15.49; 208 Pb = 37.68-38.01) that belong to their respective Kea or 204 Pb/  Loa side of the Pb-Pb boundary. Mauna Kea lavas show a systematic shift to less radiogenic Pb isotopic compositions from the shield to post-shield stage and trend to low Sr towards compositions characteristic of rejuvenated stage lavas. Hualalai post 86 Sr/ 87 shield lavas lie distinctly above the Hf-Nd Hawaiian array (spj = + 12 to + 13;  Nd 8  =  +5.5 to  +6.5) and have some of the least radiogenic Pb isotopic compositions (e.g. 206 Pb = 204 Pb/ 17.89-18.01) of recent Hawaiian volcanoes. In contrast, comparison of Kohala with the adjacent Mahukona shows that lavas from these volcanoes become more radiogenic in Pb during the late stages of volcanism. The Sr-Nd-Pb-Hf isotope systematics of the post shield lavas cannot be explained by mixing between the Kea and Koolau end-members or by assimilation of Pacific lithosphere and are consistent with the presence of ancient  recycled lower oceanic crust and sediments in their source. More than one depleted component is sampled by the post-shield lavas and these components are long-lived features of the Hawaiian plume that are present in both the Kea and Loa source regions. The geochemistry of the post-shield lavas provide evidence for a bilaterally zoned plume, where the compositional boundary between the Kea and Loa sources is complex and vertical components of heterogeneity are also significant.  III  TABLE OF CONTENTS ABSTRACT TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES ACKNOWLEDGEMENTS CO-AUTHORSHIP STATEMENT.  CHAPTER 1: GEOCHEMICAL EVOLUTION OF THE HAWAIIAN ISLANDS AND THE IMPORTANCE OF POST-SHIELD VOLCANISM 1 1.1 INTRODUCTION 1.2 OVERVIEW OF THE THESIS  7  1.3 REFERENCES  11  CHAPTER 2: ALTERATION MINERALOGY AN]) THE EFFECT OF ACIDLEACHING ON THE Pb ISOTOPE SYSTEMATICS OF OCEAN ISLAND BASALTS 15 2.1 INTRODUCTION  16  2.2 SAMPLE DESCRIPTION  17  2.3 ANALYTICAL TECHNIQUES  21  2.4 RESULTS  22  2.4.1 Characterization of alteration phases  22  2.4.1.1 Mauna Loa, Hawaii (J2-019-04)  22  2.4.1.2 Mauna Kea, Hawaii (SR0954-8.00)  22  2.4.1.3 Mont des Ruches, Kerguelen Archipelago (BY96-27)  28  2.4.1.4 Site 1140, Northern Kerguelen Plateau (1140A 31R-1 57-61)  28  2.4.2 Removal of alteration phases by acid-leaching 2.5 DISCUSSION  33 36  2.5.1 Alteration phases in oceanic basalts and the effectiveness of acid-leaching  36  2.5.2 Effect of alteration phases on Pb isotopes  38  2.6 ACKNOWLEDGEMENTS  43  2.7 REFERENCES  44  iv  CHAPTER 3: TRACE ELEMENT AN]) ISOTOPE GEOCHEMISTRY OF POSTSHIELD LAVAS FROM MAIJNA KEA, KOHALA, AN]) HUALALAI: EVIDENCE FOR ANCIENT DEPLETED COMPONENTS IN THE HAWAIAN PLUME 47 3.1 INTRODUCTION  -  3.2 SAMPLE DESCRIPTION AND GEOLOGIC SETTING 3.3 ANALYTICAL TECHNIQUES 3.4 RESULTS  64  3.4.1 Major and trace elements  64  3.4.2 Sr-Nd-Hf-Pb isotopic compositions  70  3.5 DISCUSSION  80  3.5.1 Sr-Nd-Pb-Hf isotope systematics of Hawaiian lavas  80  3.5.1.1 Shield to post-shield transition  80  3.5.1.2 Mixing relationships  84  3.5.2 Constraints on the origin and extent of depleted components in Hawaiian lavas  85  3.5.2.1 The role of oceanic lithosphere: Ambient Pacific crust or recycled plume component’  86  3.5.2.2 Identification of depleted signatures at other Hawaiian volcanoes  88  3.5.3 Spatial distribution of heterogeneities in the Hawaiian plume  90  3.5.3.1 Horizontal zoning: Constraints from the shield to post-shield transition  90  3.5.3.2 Vertical heterogeneity: Constraints from consecutive volcano pairs  93  3.6 CONCLUSIONS  94  3.7 ACKNOWLEDGEMENTS  96  3.8 REFERENCES  97  CHAPTER 4: IMPLICATIONS FOR THE NATURE OF GEOCHEMICAL HETEROGENEITIES IN THE HAWAIIAN MANTLE PLUME  103  4.1 SUMMARY AND CONCLUSIONS  104  4.2 REFERENCES  107  APPENDICES  109  APPENDIX A: ANALYTICAL QUALITY ASSURANCE-QUALITY CONTROL (QA-QC)  110  Al. Trace element abundances determined by XRF and ICP-MS  111  A2. Trace element abundances of reference materials  111  A3. Reproducibility of Pb and Hf isotopic analyses  115  A4. References  120  V  APPENDIX B: SUPPLEMENTARY SAMPLES  .122  Bi. Mahukona basalts  123  B2. Submarine Hawaiian basalts from Hualalai, Kohala, and Penguin Bank  133  B3. References  136  vi  LIST OF TABLES  Table 2.1:  Sample characteristics and alteration indicators of studied Hawaiian and Kerguelen basalts 18  Table 2.2:  Summary of alteration phases identified in studied Hawaiian and Kerguelen basalts 37  Table 2.3:  Pb isotopic compositions of studied Hawaiian and Kerguelen basalts  39  Table 3.1:  Sampling and vent locations of post-shield lavas  52  Table 3.2:  Major and trace element abundances of Hawaiian post-shield lavas  55  Table 3.3:  Hf, Sr, and Nd isotopic compositions of Hawaiian post-shield lavas  60  Table 3.4:  Pb isotopic compositions of Hawaiian post-shield lavas  62  VII  LIST OF FIGURES Figure 1.1: Bathymetric map of the seafloor around the Hawaiian Islands  3  Figure 1.2: Geological map of the island of Hawaii  6  Figure 2.1: Photomicrographs of thin sections showing representative regions of Hawaiian and Kerguelen basalts  19  Figure 2.2: Bulk powder X-ray diffraction patterns for unleached whole rock sample powders showing identified minerals 23 Figure 2.3: Backscattered electron images and qualitative energy-dispersive spectra from regions of basalt sample J2-0 19-04 (Mauna Loa, Hawaii) 26 Figure 2.4: Backscattered electron images and qualitative energy-dispersive spectra from regions of basalt sample SR0954-8.00 (Mauna Kea, Hawaii) 27 Figure 2.5: Backscattered electron images and qualitative energy-dispersive spectra from regions of basalt sample BY96-27 (Mont des Ruches, Kerguelen Archipelago) 29 Figure 2.6: Backscattered electron images and qualitative energy-dispersive spectra from regions of basalt sample 1 140A 31R-1 57-6 1 (Site 1140, Northern Kerguelen Plateau) 30 Figure 2.7: Bulk powder X-ray diffraction patterns comparing unleached (black lines) and leached (red lines) whole rock powders for the Kerguelen basalt samples 34 Figure 2.8: Comparison of Pb isotope results from unleached and leached whole rock powders of Hawaiian and Kerguelen basalts with potential sources of contamination 41 Figure 3.1: Geological map of the island of Hawaii  51  Figure 3.2: Total alkalis vs. silica classification diagram  65  Figure 3.3: MgO variation diagrams of selected major element oxides and compatible trace elements for the Hawaiian post-shield lavas 66 Figure 3.4: Nb variation diagrams of selected trace elements (in ppm) for the Hawaiian post-shield lavas 68 Figure 3.5: Chondrite-normalized rare earth element abundances of the Hawaiian post shield lavas 69 Figure 3.6: Primitive mantle-normalized incompatible trace element abundances of the Hawaiian post-shield lavas 71 Figure 3.7: Isotopic (Sr-Nd-Pb-Hf) co-variation diagrams of the Hawaiian post-shield lavas 72  VIII  Figure 3.8:  Nd/”Nd vs. 87 Sr and (b) s vs. 8 56 Sr! (a) 143 Nd for the post-shield lavas from Mauna Kea, Kohala, and Hualalai compared to selected Hawaiian shield stage lavas 74  Figure 3.9:  Pb and (b) 87 204 Pb/ (a) e vs. 206 Sr vs. 206 86 SrJ Pb for the post-shield lavas 204 Pb/ Kea, Kohala, from Mauna and Hualalai compared to Hawaiian shield and rejuvenated stage lavas 76  Figure 3.10: Pb/ Pb for the post-shield lavas from Mauna Kea, Kohala, 204 Pb/ 208 vs. 206 Pb 204 and Hualalai compared to Hawaiian shield and rejuvenated stage lavas 78 Figure 3.11: Pb/ Pb for shield and post-shield lavas from consecutive 204 Pb/ 208 vs. 206 Pb 204 pairs of Hawaiian volcanoes: (a) Mahukona and Kohala and (b) Hualalai and Mauna Kea 82 Figure 3.12: Schematic comparison of the two main models for the spatial distribution of Loa and Kea source compositions in the Hawaiian plume: (a) concentric and (b) bilateral zonation 91  ix  ACKNOWLEDGEMENTS There have been many people who have helped me over the years that have led up to the completion of this thesis. Firstly, I would like to thank my supervisors Dominique Weis and James Scoates, whose enthusiasm for science and dedication to their students have been a great source of inspiration. Since the day I began working with them as an undergraduate student just over 5 years ago, they have continued to support and encourage me. Thanks also to Kelly Russell for being part of my supervisory committee. I thank the staff of the Pacific Centre for Isotopic and Geochemical Research (PCIGR) including Jane Barling, Bruno Kieffer, Bert Mueller, and Wilma Pretorius for their help with MC-ICP MS, TIMS, and HR-ICP-MS analyses, as well as Claude Maerschalk and Rich Friedman for support in the lab. Mati Raudsepp and Sasha Wilson are also thanked for assistance with SEM and XRD analyses. I would like to thank Don DePaolo for giving me the opportunity to work on the Hawaiian post-shield lavas. This project interested me from the start and I am grateful to have been a part of it. I would also like to thank Sarah Aciego for her assistance with the post-shield samples. Mike Garcia and Mark Kurz are thanked for allowing me to work on the Mahukona basalts, and Dave Clague is also thanked for providing the submarine basalts from Hualalai, Kohala, and Penguin Bank. These additional samples brought a unique perspective to this project that would not have otherwise been possible. I am thankful for the support that my family and friends have given me throughout this process. My parents have always encouraged my studies and I thank them for their love and guidance. My fellow students of the PCIGR have shared in the emotional ups and downs of being a graduate student and have become my good friends. I would especially  x  like to thank Ryan Fisher for being there for me from the very beginning. For understanding when I had to study, for driving me to UBC every morning even though I made him late for work, and for his unconditional love and support even during the most difficult times, I will always be grateful.  xi  CO-AUTHORSHIP STATEMENT  All research, analytical work (except as indicated below), and manuscript preparation was carried out by the author with contributions from my co-authors of each manuscript. As co-authors, my supervisors Dominique Weis and James Scoates provided research advice and ideas, careful editing of both manuscripts, as well as financial support.  CHAPTER 2 Alteration mineralogy and the effect of acid-leaching on the Pb isotope systematics of ocean island basalts Authors: Diane Hanano, James S. Scoates, Dominique Weis CHAPTER 3 Trace element and isotope geochemistry of post-shield lavas from Mauna Kea, Kohala, and Hualalai: Evidence for ancient depleted components in the Hawaiian plume Authors: Diane Hanano, Dominique Weis, Sarah Aciego, James S. Scoates, Donald J. DePaolo Donald DePaolo and Sarah Aciego (University of California, Berkeley) carried out field work and sample collection. Sarah Aciego provided the major and trace element XRF analyses as well as the Sr and Nd isotopic compositions for the post-shield samples. Donald DePaolo was responsible for the development of the research project.  XII  CHAPTER 1  GEOCHEMICAL EVOLUTION OF THE HAWAIIAN ISLANDS AND THE IMPORTANCE OF POST-SHIELD VOLCANISM  1.1 INTRODUCTION This study focuses on the geochemistry of post-shield lavas from Mauna Kea, Kohala, and Hualalai volcanoes on the island of Hawaii, part of the Hawaiian-Emperor Seamount Chain in the Pacific Ocean (Figure 1.1). The -6O0O km long chain of at least 129 islands, atolls, and seamounts extends from ‘-4 9°N to the Aleutian subduction zone at 55°N. The chain displays a systematic age progression from >76 Ma in the north at Meiji and Detroit seamounts to active volcanism in the south at Kilauea volcano and Loihi seamount [Duncan and Keller, 2004]. The prominent bend in the chain at -5O Ma records a change in the direction of Pacific plate motion from N10°W to N65°W and is thought to be due to a major reorganization of Pacific spreading centres [Sharp and Clague, 2006]. Hawaii is the archetype of intraplate ocean island volcanism, whereby an age-progressive hotspot track is created as the lithosphere migrates over a relatively stationary source (i.e., a plume) [Wilson, 1963; Morgan, 19711. The Hawaiian mantle plume in particular is one of the most robust plumes on Earth and is inferred to have an origin in the deep mantle [e.g., Sleep, 1990; Courtillot et al., 2003; Lei and Zhao, 2006; Montelli et al., 2006; Nolet et aL, 2007]. The Hawaiian Islands, at the southern end of the Hawaiian-Emperor Seamount Chain, preserve the last ‘--5 million years of volcanism related to the Hawaiian mantle plume (Figure 1.1). A striking feature of the Hawaiian Islands is the two geographically distinct chains of volcanoes [Dana, 1849; Jackson et aL, 1972]. These two chains, termed Loa and Kea after their largest volcanoes (Mauna Loa and Mauna Kea), are sub-parallel to each other and separated by a distance of approximately 40 km. Lavas erupted from volcanoes belonging to each chain are characterized by persistent differences in major elements, trace  2  Figure 1.1: Bathymetric map of the seafloor around the Hawaiian Islands [Eakins et a!., 20031. Subaerial topography is represented by shades of grey and historical lava flows are shown in red. Broad terraces that surround the islands represent paleo-coastlines and prominent ridges extending from the islands represent submarine rift zones. Fields of blocky debris were created by landslides and slumps. The numerous seamounts distributed across the seafloor are Late Cretaceous in age (80 Ma) and are unrelated to the Hawaiian plume. The inset map shows the bathymetry of the northwest Pacific Ocean and the —6000 km long Hawaiian-Emperor Seamount Chain.  3  elements, and isotope ratios [e.g., Tatsumoto, 1978; Frey and Rhodes, 1993; Hauri, 1996; Lassiter et al., 1996; Abouchami et al., 2005]. Variations in chemistry are also observed in lavas from a single volcano, which has led to the recognition that Hawaiian volcanoes evolve through four distinct stages during their —‘600,000 year lifetime [e.g., Macdonald, 1968; Moore et al., 1982; Macdonald et al., 1983; Clague and Dairymple, 1987]. The initial submarine pre-shield stage is characterized by a heterogeneous mixture of mostly alkalic lavas and accounts for only 3% of the total volume of the volcano. The majority of the volcano (95-98%) is produced by the eruption of tholeiitic basalts that build the volcano above sea level during the subsequent shield stage. As the volcano migrates away from the plume axis, magma supply and eruption rates decrease and the volcano enters the post-shield stage, where lavas typically form a thin veneer of alkalic basalts, hawaiites and mugearites that contribute —1% of the volcano’s volume. After a period of subsidence, extensive erosion, and volcanic quiescence that may last between 0.5 and 2.5 million years, some volcanoes experience rejuvenated (or “post-erosional” or “secondary”) volcanism, which is volumetrically minor (<<1%) and characterized by strongly alkalic lavas. The geochemical variability observed in Hawaiian lavas requires involvement of at least three compositionally distinct source components [e.g., Staudigel et al., 1984; Stille et al., 1986; West et al., 1987; Eiler et aL, 1996; Hauri, 1996]. Recent studies have shown that a depleted component may also be required to explain the depleted isotopic signatures observed in some Hawaiian lavas [e.g., Mukhopadhyay et al., 2003; Frey et al., 2005]. However, much debate currently exists over the origin of this component as being related to either the Pacific lithosphere [e.g., Chen and Frey, 1985; Gaffiiey et a!., 2004] or material intrinsic to the plume [e.g., Frey et al., 2005; Fekiacova et al., 2007]. Furthermore, the  4  geochemical structure of the plume, commonly inferred to be concentrically or bilaterally zoned, is another controversial aspect of Hawaiian volcanism [e.g., DePaolo et a!., 2001 Abouchami eta!., 2005; Bryce et al., 2005; Xii eta!., 2007]. Post-shield lavas, which are derived from low degrees of partial melting and small volumes within the melting region, are especially useful for helping to resolve such questions and are the focus of this study. The island of Hawaii, the youngest in the Hawaiian chain, is made up of five coalesced volcanoes; two additional volcanoes (Mahukona and Loihi) that are presently below sea level also contribute to the island’s submarine base (Figure 1.2). Mauna Loa and Kilauea, which make up the southern part of the island, are in the shield-building stage of growth, whereas the older Hualalai, Mauna Kea, and Kohala volcanoes to the north have experienced post-shield volcanism. A fair amount of previous work has been carried out on post-shield lavas from Mauna Kea and Kohala [e.g., Spengler and Garcia, 1988; West et a!., 1988; Frey et al., 1990; Kennedy eta!., 1991]. Relatively less work has been carried out on post-shield lavas from Hualalai; the isotopic compositions of several lavas from Hualalai were determined as part of a regional survey of ocean island basalts [Park, 1990] and in a recent study of post-shield trachytes [Cousens eta!., 2003]. However, in the older studies, the ages and vent locations of the studied samples were not well-constrained, and very few, if any, Hf isotopic data were reported. Furthermore, recent advances in analytical techniques, particularly the development of MC-ICP-MS, permit the isotope ratios of numerous elements (including Hf and Pb) to be determined with significantly improved precision [e.g., Albarède et al., 2004]. Specifically, the ability of the plasma source to more efficiently ionize Hf has resulted in external precision better than 40 ppm [Blichert-Toft et a!., 1997; Weis et a!., 2007]. For Pb, the use of Ti to correct for instrumental mass  5  Figure 1.2: Geological map of the island of Hawaii [Wolfe and Morris, 1996]. Different colours and patterns represent distinct lithologic units (mainly lavas flows of varying age). The thick grey lines mark the approximate boundaries between lavas from Kilauea, Mauna Loa, Hualalai, Mauna Kea, and Kohala volcanoes. The approximate locations of the submarine summits of Mahukona and Loihi volcanoes are also shown. The inset map shows the location of the island of Hawaii with respect to the other Hawaiian Islands in the Pacific Ocean.  6  fractionation, combined with enhanced acid-leaching techniques, have allowed for external precision in the 100 ppm range [e.g., White et al., 2000; Woodhead, 2002; Weis et at., 2006].  1.2 OVERVIEW OF TILE THESIS This thesis comprises two distinct studies, which are presented in Chapters 2 and 3. The main objectives of this thesis were to (1) characterize the geochemistry of lavas from the post-shield stage of Hawaiian volcanism, (2) determine the extent and origin of depleted components in Hawaiian lavas, and ultimately (3) provide constraints on the geochemical structure of the Hawaiian plume. Due to the potential susceptibility of Pb isotopes to modification by post-magmatic alteration, and their importance in helping to answer the objectives of this research, we first undertook a mineralogical study of the alteration phases present in ocean island basalts, the results of which are presented in Chapter 2. Chapter 3 represents the central work of this thesis and involves the trace element and isotope geochemistry of post-shield lavas from Mauna Kea, Kohala, and Hualalai volcanoes on the island of Hawaii. The results of this work have been presented by the author at a number of international conferences including the 2005 and 2006 AGU Fall Meetings (Hanano et a!., 2005; Hanano et at., 2006a), the 2006 GAC-MAC Joint Annual Meeting (Hanano et a!., 2006b), and the  th 17  Annual Goldschmidt Conference (Hanano et al., 2007). Both chapters  were prepared in manuscript format appropriate for submission to international scientific journals. Chapter 2 is the revised version of the manuscript submitted to American Mineralogist after undergoing peer-review. Chapter 4 presents a summary of the major findings of the thesis.  7  Chapter 2 explores the alteration mineralogy of four weakly altered basalts from two major ocean island volcanic systems, Hawaii and Kerguelen, using scanning electron microscopy and X-ray diffraction. This work was initiated as part of a research project for EOSC 521 (Microbeam Diffraction Methods for the Characterization of Minerals and Materials) taught by Mati Raudsepp at UBC, and the initial results were included in a final report required for partial completion of the course. Many studies have documented the changes in Pb isotopic compositions resulting from acid-leaching, but very few have tried to link these changes with the alteration mineralogy of the samples. A wide variety of secondary minerals are identified, and the effectiveness of their removal by acid-leaching is evaluated. These results are then integrated with Pb isotope data on the same samples (from  Nobre Silva et al., in revision, 2008) to identify possible sources of contamination and to determine the effect of incomplete removal of alteration phases on the reproducibility of Pb isotopic compositions. The results of this study have important implications for highprecision Pb isotope studies of ocean island basalts. Chapter 3 documents the geochemistry of 32 post-shield lavas from Mauna Kea, Kohala, and Hualalai (Figure 1.2). The samples were collected by Don DePaolo and Sarah Aciego of the University of California, Berkeley as part of a larger project to better define the geochemical structure of the Hawaiian mantle plume. Of the data reported in this chapter, the trace element concentrations and Pb and Hf isotopic compositions were acquired by the author, whereas the major element and Sr and Nd isotope analyses were carried out by Sarah Aciego at the University of California, Berkeley. The major and trace element results provide the framework for investigating the origin and source of the post shield lavas. The Sr, Nd, Pb, and Hf isotopic compositions allow for identification of  8  discrete mantle components sampled by each volcano and have implications for the origin and distribution of small-scale heterogeneities and large-scale zoning within the Hawaiian plume. Literature data from other Hawaiian volcanoes were compiled in an effort to determine if the post-shield lavas were derived from the same sources as shield lavas and to provide the appropriate context for interpretation of the post-shield results. The post-shield lavas are compared to potential depleted sources to identif’ the origin of the depleted signature observed in the post-shield lavas. The post-shield lavas are also compared to lateshield, post-shield, and rejuvenated stage lavas from other Hawaiian volcanoes to constrain the extent of the depleted component related to Hawaiian volcanism. Lastly, the isotope systematics of two consecutive pairs of Hawaiian volcanoes (Mahukona-Kohala and Hualalai-Mauna Kea) are used to evaluate models of horizontal zoning and to determine the importance of vertical heterogeneity within the plume. Appendix A reports the analytical quality assurance, quality control (QA-QC) results for the chemical analyses reported in this study including: a comparison between trace element abundances determined by XRF and ICP-MS (Al), the trace element abundances of USGS reference materials (A2), and the reproducibility of the Pb and Hf isotopic analyses (A3). Appendix B is a brief summary and preliminary interpretation of the isotopic results for offshore submarine Hawaiian basalts. The Sr, Nd, Pb, and Hf isotopic compositions were determined for 13 samples from Mahukona (Bi) and 10 samples from Penguin Bank, Hualalai, and Kohala (B2). These supplementary samples, provided by Dave Clague, Mike Garcia, and Mark Kurz, were analyzed in an effort to better understand the onshore post shield volcanism and to increase the geographical sampling area. The isotopic compositions of the submarine basalts are used in Chapter 3 to provide additional constraints on the large  9  scale geochemical structure of the Hawaiian plume. The isotopic results for Mahukona will form part of a larger manuscript on the geology of this volcano (Garcia et a!., in preparation, 2008). The geochemistry of this submerged Hawaiian shield volcano is not well characterized and is particularly important to our understanding of Hawaiian volcanism because it fills the gap between Hualalai and Kahoolawe in the paired sequence of Hawaiian volcanoes.  10  1.3 REFERENCES  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, 85 1-856. Albarède, F., P. Telouk, J. Blichert-Toft, M. Boyet, A. Agranier, and B. Nelson (2004), Precise and accurate isotopic measurements using multiple-collector ICPMS, Geochimica et Cosmochimica Acta, 68(12), 2725-2744. Blichert-Toft, J., C. Chauvel, and F. Albarède (1997), Separation of Hf and Lu for highprecision isotope analysis of rock samples by magnetic sector-multiple collector ICP-MS, Contributions to Mineralogy and Petrology, 127, 248-260. 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, Geochemistry Geophysics Geosystems, 6(9), doi: 10. 1029/2004GC000809. Chen, C. Y., and F. A. Frey (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. Clague, D. A., and G. B. Dalrymple (1987), The Hawaiian-Emperor volcanic chain, Part 1: Geologic evolution, in Volcanism in Hawaii, edited by R.W. Decker, T.L. Wright, and P.H. Stauffer, pp. 1667, US Geological Survey Professional Paper 1350, Denver. Courtillot, V., A. Davaille, J. Besse, and J. Stock (2003), Three distinct types of hotspots in the Earth’s mantle, Earth and Planetary Science Letters, 205, 295-3 08. Cousens, B. L., D. A. Clague, and W. D. Sharp (2003), Chronology, chemistry, and origin of trachytes from Hualalai Volcano, Hawaii, Geochemistry Geophysics Geosystems, 4(9), 1078, doi: 10.1 029/2003GC000560. Dana, J. D. (1849), Geology, in United States Exploring Expedition, 1838-1842, vol. 10, pp. 756, C. Sherman, Philadelphia. DePaolo, D. J., J. G. Bryce, A. Dodson, D. L. Shuster, and B. M. Kennedy (2001), Isotopic evolution of Mauna Loa and the chemical structure of the Hawaiian plume, Geochemistry Geophysics Geosystems, 2, doi: 10.1 029/2000GC000 139. Duncan, R. A., and R. A. Keller (2004), Radiometric ages for basement rocks from the Emperor Seamounts, ODP Leg 197, Geochemistry Geophysics Geosystems, 5, doi: 10.1 029/2004GC000704. Eakins, B. W., J. E. Robinson, T. Kanamatsu, J. Naka, J. R. Smith, E. Takahashi, and D. A. Clague (2003), Hawaii’s volcanoes revealed, Map 1-2809, United States Geological Survey, Denver, CO. Eiler, J. M., K. A. Farley, J. W. Valley, A. W. Hofinanri, and B. M. Stolper (1996), Oxygen isotope constraints on the sources of Hawaiian volcanism, Earth and Planetary Science Letters, 144, 453-468. Fekiacova, Z., W. Abouchami, S. J. G. Galer, M. 0. Garcia, and A. W. Hofmann (2007), Origin and temporal evolution of Koolau volcano, Hawaii: Inferences from isotope data on the Koolau Scientific Drilling Project (KSDP), the Honolulu Volcanics and ODP Site 843, Earth and Planetary Science Letters, 261, 65-83.  11  Frey, F., and J. Rhodes (1993), Intershield geochemical differences among Hawaiian volcanoes: Implications for source compositions, melting process and magma ascent paths, Philosophical Transactions Royal Society ofLondon Part A, 342, 12 1-136. Frey, F. A., S. Huang, J. Blichert-Toft, M. Regelous, and M. Boyet (2005), Origin of depleted components in basalt related to the Hawaiian hot spot: Evidence from isotopic and incompatible element ratios, Geochemistry Geophysics Geosystems, 6(1), doi: 10.1 029/2004GC000757. Frey, F. A., W. S. Wise, M. 0. Garcia, H. B. West, S.-T. Kwon, and A. K. Kennedy (1990), Evolution of Mauna Kea volcano, Hawaii: Petrologic and geochemical constraints on postshield volcanism, Journal of Geophysical Research, 95(B2), 127 1-1300. Gaffiiey, A. M., B. K. Nelson, and J. Blichert-Toft (2004), Geochemical constraints on the role of oceanic lithosphere in intra-volcano heterogeneity at West Maui, Hawaii, Journal ofPetrology, 45(8), 1663-1687. Hanano, D., D. Weis, S. Aciego, J. S. Scoates, and D. J. DePaolo (2005), Geochemical systematics of Hawaiian post-shield lavas: Implications for the chemical structure of the Hawaiian mantle plume, Eos Transactions, AGU, 86(5 2), Fall Meeting Supplement, Abstract V41D-1497. Hanano, D., D. Weis, S. Aciego, J. S. Scoates, and D. J. DePaolo (2006a), Geochemistry of post-shield lavas and submarine tholeiites from the Kea and Loa trends of Hawaiian volcanoes, Eos Transactions, AGU, 87(52), Fall Meeting Supplement, Abstract V13B-0673. Hanano, D., D. Weis, S. Aciego, J. S. Scoates, and D. J. DePaolo (2006b), Pb isotopic systematics of Mauna Kea, Kohala, and Hualalai post-shield lavas: Heterogeneity at the periphery of the Hawaiian plume, GAC-MAC Program with Abstracts, 31, 63. Hanano, D., D. Weis, S. Aciego, J. S. Scoates, and D. J. DePaolo (2007), Geochemistry of post-shield lavas from paired Loa- and Kea-trend Hawaiian volcanoes, Geochimica et Cosmochimica Acta, 71-15, Supplement 1: A375. Hauri, E. H. (1996), Major-element variability in the Hawaiian mantle plume, Nature, 382, 415-419. Jackson, E. D., E. A. Silver, and G. B. Dalrymple (1972), Hawaiian-Emperor chain and its relation to Cenozoic circum-Pacific tectonics, Geological Society ofAmerica Bulletin, 83, 601-618. Kennedy, A. K., S.-T. Kwon, F. A. Frey, and H. B. West (1991), The isotopic composition of postshield lavas from Mauna Kea volcano, Hawaii, Earth and Planetary Science Letters, 103, 339-353. Lassiter, J. C., D. J. DePaolo, and M. Tatsumoto (1996), Isotopic evolution of Mauna Kea Volcano: results from the initial phase of the Hawaii Scientific Drilling Project, Journal of Geophysical Research, 101(B5), 11,769-11,780. Lei, J., and D. Zhao (2006), A new insight into the Hawaiian plume, Earth and Planetary Science Letters, 241, 438-453. Macdonald, G. A. (1968), Composition and origin of Hawaiian lavas, Geological Society of America Memoir, 116, 477-522. Macdonald, G. A., A. T. Abbott, and F. L. Peterson (1983), The Geology of Hawaii, in Volcanoes in the Sea, pp. 517, University of Hawaii Press, Honolulu.  12  Montelli, R., G. Nolet, F. A. Dahien, and G. Masters (2006), A catalogue of deep mantle plumes: New results from finite-frequency tomography, Geochemistry Geophysics Geosystems, 7(11), doi: 10.1 029/2006GC00 1248. Moore, J. G., D. A. Clague, and W. R. Normark (1982), Diverse basalt types from Loihi Seamount, Hawaii, Geology, 10, 88-92. Morgan, W. J. (1971), Convection plumes in the lower mantle, Nature, 230, 42-43. Mukhopadhyay, S., J. C. Lassiter, K. A. Farley, and S. W. Bogue (2003), Geochemistry of Kauai shield-stage lavas: Implications for the chemical evolution of the Hawaiian plume, Geochemistry Geophysics Geosystems, 4(1), 1009, doi: 10.1 029/2002GC000342. Nobre Silva, I. G., D. Weis, J. Barling, and J. S. Scoates (2008), Leaching systematics for the determination of high-precision Pb isotope compositions of ocean island basalts, Geochemistry Geophysics Geosystems, paper # 2007GC00 1891 (in revision). Nolet, G., R. Allen, and D. Zhao (2007), Mantle plume tomography, Chemical Geology, 241, 248-263. Park, K.-H., Strontium, neodymium, and lead isotopic studies of ocean island basalts: Constraints on their evolution and origin, Ph.D. thesis, Columbia University, Palisades, N.Y., 1990. Sharp, W. D., and D. A. Clague (2006), 50-Ma initiation of Hawaiian-Emperor bend records major change in Pacific plate motion, Science, 313, 1281-1284. Sleep, N. H. (1990), Hotspots and mantle plumes: Some phenomenology, Journal of Geophysical Research, 95, 6715-6736. Spengler, S. R., and M. 0. Garcia (1988), Geochemistry of the Hawi lavas, Kohala volcano, Hawaii, Contributions to Mineralogy and Petrology, 99, 90-104. Staudigel, H. A., A. Zindler, S. R. Hart, T. Leslie, C.-Y. Chen, and D. Clague (1984), The isotope systematics of a juvenile intraplate volcano: Pb, Nd and Sr isotope ratios of basalts from Loihi Seamount, Hawaii, Earth and Planetary Science Letters, 69, 1329. Stille, P., D. M. Unruh, and M. Tatsumoto (1986), Pb, Sr, Nd, and Hf isotopic constraints on the origin of Hawaiian basalts and evidence for a unique mantle source, Geochimica et Cosmochimica Acta, 50(10), 2303-2319. Tatsumoto, M. (1978), Isotopic composition of lead in oceanic basalt and implication to mantle evolution, Earth and Planetary Science Letters, 38, 63-87. Weis, D., B. Kieffer, D. Hanano, I. N. Silva, J. Barling, W. Pretorius, C. Maerschalk, and N. Mattielli (2007), Hf isotope compositions of U.S. Geological Survey reference materials, Geochemistry Geophysics Geosystems, 8(6), doi: 10.1 029/2006GC00 1473. Weis, D., B. Kieffer, C. Maerschalk, J. Barling, J. deJong, G. A. Williams, D. Hanano, W. Pretorius, N. Mattielli, J. S. Scoates, A. Goolaerts, R. M. Friedman, and J. B. Mahoney (2006), High-precision isotopic characterization of USGS reference materials by TIMS and MC-ICP-MS, Geochemistry Geophysics Geosystems, 7(8), doi: 10.1 029/2006GC00 1283. West, H. B., M. 0. Garcia, F. A. Frey, and A. Kennedy (1988), Nature and cause of compositional variation among the alkalic cap lavas of Mauna Kea volcano, Hawaii, Contributions to Mineralogy and Petrology, 100, 3 83-397.  13  West, H. B., D. C. Gerlach, W. P. Leeman, and M. 0. Garcia (1987), Isotopic constraints on the origin of Hawaiian lavas from the Maui volcanic complex, Hawaii, Nature, 330, 216-220. White, W. M., F. Albarède, and P. Telouk (2000), High-precision analysis of Pb isotope ratios by multi-collector ICP-MS, Chemical Geology, 167, 257-270. Wilson, J. T. (1963), A possible origin of the Hawaiian Islands, Canadian Journal of Physics, 41(6), 863-870. Wolfe, E. W., and J. Morris (1996), Geologic map of the island of Hawaii, Map 2524-A, United States Geological Survey, Denver, Co. Woodhead, J. D. (2002), A simple method for obtaining highly accurate Pb-isotope data by MC-ICPMS, Journal ofAnalytical Atomic Spectrometry, 17, 1381-1385. Xu, G., F. A. Frey, D. A. Clague, W. Abouchami, J. Blichert-Toft, B. Cousens, and M. Weisler (2007), Geochemical characteristics of West Molokai shield- and postshield stage lavas: Constraints on Hawaiian plume models, Geochemistry Geophysics Geosystems, 8, doi: 10.1 029/2006GC00 1554.  14  CHAPTER 2  ALTERATION MINERALOGY AN]) THE EFFECT OF ACID-LEACHING ON THE Pb ISOTOPE SYSTEMATICS OF OCEAN ISLAND BASALTS  15  2.1 INTRODUCTION’  The Pb isotopic compositions of ocean island basalts are a powerful tool for the characterization of their mantle source components, especially with recent improvements in precision and reproducibility offered by the triple spike method [e.g., Abouchami et a!., 2000] and by multiple collector inductively coupled plasma mass spectrometry (MC-ICP MS) [e.g., White et a!., 2000; Weis et a!., 2005]. However, Pb isotopic compositions of ocean island basalts are highly susceptible to modification by non-magmatic Pb from seawater, hydrothermal alteration, aeolian dust, subaerial weathering, as well as by contamination during sample recovery and processing [e.g., McDonough and Chauvel, 1991; Thirlwall, 2000; Weis et al., 2005]. Acid-leaching of basaltic rock powders or chips may not remove the foreign Pb component completely or in a reproducible way [e.g., Abouchami et a!., 2000; Eisele et al., 2003], resulting in erroneous values, data scatter, and poor reproducibility, which reduces the effectiveness of the Pb isotopic system as a geochemical tracer. It is therefore critical to understand the types, quantities, and distribution of alteration phases in ocean island basalts that are analyzed for Pb isotope ratios. Most research concerning the alteration of basalt has focused on strongly altered samples [e.g., Böhlke et a!., 1980; Alt and Honnorez, 1984; Banerjee et a!., 2004], which are typically not selected for geochemical analysis in igneous studies. In this study, we present a mineralogical investigation of alteration phases in a select group of weakly altered basalts from two major ocean island volcanic systems (Hawaii and Kerguelen) based on scanning  A version ofthis chapter has been submitted for publication. Hanano, D., J. S. Scoates, and D. Weis (2008), Alteration mineralogy and the effect of acid-leaching on the Pb isotope systematics of ocean island basalts, American Mineralogist, paper # 2845R.  16  electron microscopy (SEM), X-ray diffraction (XRD), and acid-leaching experiments that were specifically designed to test the effectiveness of leaching at removing alteration phases  and the resulting effect on Pb isotope systematics.  2.2 SAMPLE DESCRIPTION Basalts from four volcanic centers from Hawaii and Kerguelen were examined to cover a range of ages, eruptive environments, lava compositions, and alteration styles (Table 2.1; Figure 2.1). The basalts selected for this study display only minor to moderate alteration, and are typical samples analyzed for geochemical studies. Loss-on-ignition (LOl), a commonly used parameter for helping to assess the degree of alteration in a sample and thus its suitability for geochemical work, varies from 0-2 wt% in the four samples (Table 2.1), within the range of values that are considered acceptable. The two Hawaiian basalts are from the two largest Hawaiian volcanoes, Mauna Loa (J2-0 19-04) and Mauna Kea (SR0954-8.00), which belong to the “Loa” and “Kea” trends, respectively [Dana, 1849; Jackson et a!., 1972]. Basaltic lavas from the two chains are distinguished both geographically and geochemically, exhibiting systematic differences in radiogenic isotopic compositions [e.g., Tatsumoto, 1978; Lassiter et a!., 1996; Abouchami et a!., 2005]. Alteration of Hawaiian basalts mainly involves the partial to complete replacement of glass and olivine by secondary minerals, including clays (mostly smectite), zeolites (most commonly chabazite and phillipsite), as well as Ca-silicate minerals, gypsum, and rare pyrite [e.g., Walton and Shffinan, 2003; Garcia eta!., 20071. The two Kerguelen basalts are from Mont des Ruches on the Kerguelen Archipelago (BY96-27) and Site 1140 on the Northern Kerguelen Plateau (1 140A 31R-l 57-61), which  17  Table 2.1: Sample characristics and alteration indicators of studied Hawaiian and Kerguelen basalts Hawaii Sample  J2-019-04  Location  Mile High Section, Southwest Rift Zone, Mauna Los (“Loa-trend”)  Sampling Method Depth/Elevationu Age” Eruption Environment Composition MgO (wt%)°  Kerguelen SR0954-8.00  BY96-27  1140A 31R-l 57-61  HSDP-2, Mauna Kea (“Kea-trend”)  Mont des Ruches, Loranchet Peninsula, Kerguelen Archipelago  ODP Leg 183 Site 1140, Northern Kerguelen Plateau  ROV Jason2  drilling  hammer  drilling  1986 mbsl —450 ka submarine tholeiitic basalt  3009.3 mbsl —550 ka submarine tholeiitic basalt (picritic)  455 masl —28 Ma subaerial transitional basalt  270.07 mbsf —34 Ma submarine tholeiitic basalt  14.2  18.2  10.7  5.7  Olivine Phenocryst Abundance (vol%)  —25  —19  —12  —20 (plagioclase, clinopyroxene, olivine)  General Alteration Description  trace groundmass alteration  minor alteration of olivine; moderate groundmass alteration  moderate alteration of olivine; moderate groundmass alteration  complete iddingsite replacement of olivine; secondaty minerals in vesicles  0  0.85  2.01  1.65  7  6  34.8 170  32.2  15 51.0  15 50.4  254  288  359  LOT (wt%)° 1 LeachingSteps Leaching WeightLoss(%f Reproducibility (ppm)°  ‘mbsl = meters below sea level; mast = meters above sea level; mbsf meters below the sea floor. J2-019-04: B. singer(personal communication); 5R0954-8.00:DePaoloandsrolper [19961; BY96-27: Dosceretal. [20021; 1140A 31K-i 57-61: Duncan etal. [2002]. 5 2-019-04: D. Weis (unpublished data); 5R0954-5.00:Rhodes and Vollinger [2004]; BY96-27: Doucet etal. [2002]; 1140A 31K-i 57-61: Wet, and Frey [2002]. dHawali: NoIre Silvaetal. (in revision, 2008); leaching wisg,ht loss is the mean of 2-3 separate powder aliquots. Ksrguslen: this stody; leaching weight loss is the mean of 6 separate powder aliquots. Keproducibility of the MC-ICP-MS Pb isotopic analyses is represented by the external reproducibility (251)/mean x i() of zosPb,uPb on acid-leached ntsole rock powders.  18  Hawaii  Mauna Kea  Mont des Ruches  Figure 2.1: Photomicrographs of thin sections showing representative regions of Hawaiian and Kerguelen basalts examined in this study. Photomicrographs in plane-polarized light. Abbreviations: ol = olivine; plag = plagioclase. (a) Sample J2-0 19-04 (Mauna Loa, Hawaii) is characterized by large unaltered olivine phenocrysts in a fme-grained groundmass with trace alteration. (b) Sample SR0954-8.00 (Mauna Kea, Hawaii) shows minor alteration of the olivine phenocrysts and moderate alteration of the groundmass (yellow-green patches). (c) Sample BY96-27 (Mont des Ruches, Kerguelen Archipelago) contains highly fractured and altered olivine phenocrysts and moderate groundmass alteration. (d) Sample 1 140A 3 1R-1 57-61 (Site 1140, Northern Kerguelen Plateau) is dominated by plagioclase phenocrysts set in a fine-grained groundmass. Olivine phenocrysts are small (<0.5 mm) and completely altered. Secondary minerals fill vesicles see text for description. —  19  belong to the Kerguelen large igneous province in the Southern Indian Ocean. The Kerguelen basalts are significantly older and have a style of alteration that is markedly different from that in the Hawaiian basalts (Table 2.1). The subaerial flood basalts of the Kerguelen Archipelago have been affected by hydrothermal alteration and contain a wide variety of secondary minerals, including carbonates, oxides, sulfides, quartz, clays, epidotes, and zeolites [e.g., Nougier eta!., 1982; Verdier and Native!, 1988; Verdier, 1989]. In contrast, the alteration of submarine basalts from the Northern Kerguelen Plateau does not involve zeolites and most commonly includes smectite (± hydromica), chlorite, and carbonate [Kurnosov eta!., 2003]. These basalts represent a sub-selection of samples from the study of Nobre Silva et a!. (Nobre Silva, I. G., D. Weis, J. Barling, and J. S. Scoates, Leaching systematics for the determination of high-precision Pb isotope compositions of ocean island basalts, manuscript in revision for Geochemistry Geophysics Geosystems, 2008; hereinafter referred to as Nobre Si!va et a!., in revision, 2008). The Nobre Silva et a!. (in revision, 2008) study identifies changes in the Pb isotope compositions and reproducibility of Hawaiian and Kerguelen basalts via leaching and column chemistry experiments, whereas this study seeks to link these changes to the alteration mineralogy of the samples. The Pb isotope analyses were carried out using whole rock powders of basalt because glass is commonly not available or present in sufficient quantity (especially for older and subaerially erupted basalts). However, to ensure only the freshest material was powdered, surface alteration was first removed from the whole rocks; powders were subsequently prepared using agate and a crushing procedure specifically designed to limit sample exposure to tungsten carbide.  20  2.3 ANALYTICAL TEChNIQUES Alteration phases were initially identified based on their morphology and optical properties using conventional optical microscopy, and further characterized by scanning electron microscopy (SEM) and X-ray diffraction (XRD) at the University of British Columbia. Backscattered electron (BSE) imaging and qualitative energy dispersive spectrometry (EDS) on carbon-coated polished petrographic thin sections were carried out on a Phillips XL-30 scanning electron microscope equipped with a Princeton-Gamma-Tech energy-dispersion spectrometer. An operating voltage of 15 Ky was used with a spot diameter of 6 I.tm and peak count time of 30 s. Whole rock sample powders for XRD were ground under ethanol with an agate mortar-and-pestle, smear-mounted onto glass slides, and then placed in aluminum holders. XRD data were collected over the range 3-80° 20 using CoKcL radiation on a Siemens D5000 X-ray powder diffractometer equipped with a diffracted-beam graphite monochromator crystal and operating at 35 Ky and 40 Ma. After identification of the alteration phases, we carried out acid-leaching experiments on the sample powders to determine how effectively these minerals are removed. The premise of leaching is that any alteration phases or contaminants will be more susceptible to acid attack, leaving behind the more resistant primary (i.e., magmatic) minerals. Due to substantial weight loss during leaching (typically --30-60%) and to avoid possible “nugget” effects, sufficient whole rock powder for XRD was only available for the two Kerguelen basalts (BY96-27 and 11 40A 31 R- 1 57-61). Following the sequential leaching procedure of Weis et al. [20051, sample powders were repeatedly acid-leached in screw-top Teflon® vessels with quartz-distilled 6N HC1 in a warm (--50 °C) ultrasonic bath for 20-minute intervals until the supernatant fluid became clear. The use of hot or boiling HC1 could  21  reduce the total leaching time and workload, but significantly increases the likelihood of dissolving primary minerals and is therefore not recommended. Both sample powders required 15 leaching steps and lost ‘-5O% of their initial weight (Table 2.1). Leached powders were rinsed twice with 18 M cm water to remove any residual traces of acid, taken to dryness, and analyzed by XRD in addition to the unleached powders.  2.4 RESULTS 2.4.1 Characterization of alteration phases 2.4.1.1 Mauna Loa, Hawaii (J2-019-04) Olivine phenocrysts are unaltered and there is only minor alteration of the groundmass. No secondary minerals were detected by XRD (Figure 2.2a). Manganeseoxides (pyrolusite) fill interstitial voids in the groundmass and display fine compositional banding (Figure 2.3a). Orange-yellow, transparent and birefringent fibro-palagonite [Zhou and Fyfe, 1989] lines the walls of originally glassy vesicles (Figure 2.3b) and is likely composed of poorly crystalline smectite [Zhou et a!., 1992]. Minor iron-oxides (goethite) and clays (smectite) were also identified in the groundmass.  2.4.1.2 Mauna Kea, Hawaii (SR0954-8.OO) Olivine phenocrysts are altered along rims and fractures to talc, characterized by brownish fibrous aggregates with subparallel alignment and high birefringence (third-order) (Figure 2.4a). Interstitial voids and vesicles are filled with material that occurs as fibrous bundles with lower aggregate birefringence (first-order) and is likely a variety of Mg-rich smectite. The broad peak between 5100 20 that is visible in the original diffractogram  22  a 1200  12001  Mauna Loa, HawaH (J2-019-04)  1000•  1000  :i’jj  &  10  20  2-The1 -0oIe 00  x  400  .  —  20: 3  10  2 2-Theta Sca e -  b 1400 1200  1400  Mauna Kea, Hawaii (SR0954-8.OO)  12001  smectite  1000 —  0  (0  800  0  3  10  00  2-Th-&.I 00  -ö  600  :: 0  E  0)  smeite?jj  20  j’  —  2-Theta Scale -  Figure 2.2  23  70  80  C 2400 2000  Mont des Ruches Kerguelen Archipelago  2000  (BY96-27)  1 600 F  zeolite mineral (aerinite, cowlesite?) d = 14.97 A clay mineral (dickite?) d=1O.09A serpentine mineral (lizardite?) d 7.36A  10  0  2-The-SIo  —  .._-1 200 •  ><  —  800  40: rn  ‘‘‘  ,,  J 2-Theta Scale -  2000-  d  I 2000’  0) -  1600•  1eoo  Site 1140, Northern Kerguelen Plateau  l2O0  (1140A31R-1 57-61)  I  1200 2-Thou SIo  D  0  -  -  x  -  —  smectite mineral  no’)  1’,  4 L&.%  2-Theta Scale -  Figure 2.2 (continued)  24  Figure 2.2: Bulk powder X-ray diffraction patterns for unleached whole rock sample powders showing identified minerals. Inset diagrams: raw X-ray diffraction patterns prior to background subtraction. (a) Sample J2-019-04 (Mauna Loa, Hawaii) contains olivine, clinopyroxene, plagioclase, and no alteration minerals. (b) Sample SR0954-8.00 (Mauna Kea, Hawaii) contains olivine, clinopyroxene, plagioclase, and ilmenite. The broad hump between 51O0 29 likely reflects the presence of clay minerals such as smectite. (c) Sample BY96-27 (Mont des Ruches, Kerguelen Archipelago) contains olivine, clinopyroxene, plagioclase, and ilmenite. Additional peaks indicate the presence of zeolite, clay, and serpentine group minerals. (d) Sample 1 140A 31R-1 57-5 1 (Site 1140, Northern Kerguelen Plateau) contains clinopyroxene and plagioclase, as well as smectite ± celadonite.  25  2 fibro-palagonite (smectite)  0.0  2.0  4.0  key  6.0  8.0  10.0  Figure 2.3: Backscattered electron images and qualitative energy-dispersive spectra from regions of basalt sample J2-019-04 (Mauna Loa, Hawaii). Inset photomicrographs in plane-polarized light show the general area (i.e. similar scale) of above BSE images; this applies to all subsequent figures (Figs. 3-5). Abbreviations: ol = olivine; plag plagioclase; cpx = clinopyroxene. The numbers (1-2) indicate the locations of EDS analyses. (a) Finely banded manganese-oxide (pyrolusite) in groundmass adjacent to olivine phenocryst. The lighter layers contain a greater proportion of Mn, Mg, and Al. (b) Fibro-palagonite (smectite) replacing vesicle-lining glass adjacent to olivine phenocryst.  26  Figure 2.4: Backscattered electron images and qualitative energy-dispersive spectra from regions of basalt sample SR0954-8.OO (Mauna Kea, Hawaii). Black spots are marker points for microprobe traverses. Inset photomicrographs in plane-polarized light. Abbreviations as in Fig. 2; urn = ilmenite. The numbers (1-2) indicate the locations of EDS analyses. (a) Talc alteration along fracture within olivine phenocryst. (b) Irregularly shaped patches of barite/celestite (bright white regions) within interstitial smectite.  27  without background subtraction (Figure 2.2b inset) indicates the presence of poorly-  ordered clay minerals such as smectite. Patches of barite/celestite (Figure 2.4b) and minor secondary pyrite occur within the smectite alteration.  2.4.1.3 Mont des Ruches, Kerguelen Archipelago (BY96-27) Olivine phenocrysts are distinctly altered to red-brown “iddingsite,” composed primarily of iron-oxides, smectite and serpentine (Figure 2.5 a). Colorless radiating bundles or “flakes” with low birefringence (first-order) are commonly observed and likely composed of zeolite finely intergrown with phyllosilicates (Figure 2.5a). The large well-defined XRD peak with a d-spacing of’—l 5 A (Figure 2.2c) confirms the presence of a zeolite mineral, possibly aerinite or cowlesite. Additional XRD peaks occur at d-spacings of-V- 10.1 A and —7.4 A and reflect the presence of clay and serpentine group minerals such as dickite and lizardite, respectively. The groundmass is characterized by widespread patchy brown alteration that is typically associated with vesicles and is likely altered and devitrified glass. A calcium silicate mineral (apophyllite) was also identified extending into this groundmass alteration along with zeolite (Figure 2.5b).  2.4.1.4 Site 1140, Northern Kerguelen Plateau (1 140A 31 R- 1 57-61) Plagioclase and clinopyroxene were the only primary minerals identified by XRD (Figure 2.2d). Olivine phenocrysts have been completely replaced by red-brown and yellow-green secondary minerals, including iron-oxides (goethite) and clay minerals (primarily smectite ± chlorite) (Figure 2.6a). Vesicles are typically filled with several smectite layers of varying composition (mostly in Fe, Mg, and Al) and thickness (pore  28  —  ç  j -  IN i’AhIt.1i.i  I zeolite (+ phyllo silicates?)  3 altered and devitrified glass I  •1  A  0.0  2.0  4.0  keV  6.0  8.0  2 “iddingsite” (iron-oxides serpentine + smectite)  0.0  10.0  .  2.0  4.0  +  p  keV  6.0  8.0  10.0  4 Ca-silicate (apophyllite)  g  ,‘ 0.0  .4__.___ 2.0  4.0  key  } 6.0  \.__ .  8.0  10.0  0.0  2.0  4.0  key  6.0  8.0  10.0  Figure 2.5: Backscattered electron images and qualitative energy-dispersive spectra from regions of basalt sample BY96-27 (Mont des Ruches, Kerguelen Archipelago). Abbreviations as in Fig. 2; urn = ilmenite; sp = spinel. The numbers (1-4) indicate the locations of EDS analyses. (a) “Iddingsite” alteration of olivine phenocryst: iron-oxides + serpentine + smectite. Flakes of zeolite (finely intergrown with phyllosilicates?) at upper left. Inset photomicrograph in plane-polarized light. (b) Ca-silicate mineral (apophyllite) surrounded by zeolite as well as altered and devitrified glass. Inset photomicrograph in crossed-polars.  29  ‘A31R-1 57-61  iron-oxide (goethite)  0.0  2.0  4.0  keV  6.0  8.0  0.0  10.0  2.0  2 iron-oxide + smectite/chiorite  4.0  key  6:0  8:0  10.0  5 green smectite + mica (celadonite)  I  —.  0.0  2.0  4.0  keV  6.0  8.0  10.0  0.0  2.0  3 green smectite + mica (celadonite)  4.0  keV  6.0  8.0  10.0  6 Mg-rich pore-lining smectite  ____ 0.0  2.0  4.0  key  6.0  8.0  10.0  0.0  2.0  4.0  key  6.0  Figure 2.6  30  8.0  10.0  7 green smectite mica (celadoni 8 more Mg &Al, / less Fe  +  IL 0.0  2.0  4.0  keV  9 iron-oxide smectite  0.0  2.0  4.0  8eV  6.0  8.0  10.0  8.0  10.0  8.0  10.0  +  6.0  10 Mg-rich pore-lining smectite  0.0  2.0  4.0  keV  6.0  Figure 2.6 (continued)  31  Figure 2.6: Backseattered electron images and qualitative energy-dispersive spectra from regions of basalt sample 1 140A 3 1R-1 57-61 (Site 1140, Northern Kerguelen Plateau). Abbreviations as in Fig. 2. The numbers (1-14) indicate the locations of EDS analyses. (a) Complete “iddingsite” replacement of olivine phenocrysts: iron-oxide + chlorite + smectite. Inset photomicrograph in plane-polarized light. (b) Vesicle lined with smectite of varying composition and filled with euhedral crystals of dolomite that show zoning due to variable Fe content. Inset photomicrograph in crossed-polars. (c) Vesicle filled with smectite and mica (celadonite) layers of varying composition as well as patches of iron-rich material. Inset photomicrograph in plane-polarized light. (d) Vesicle lined with smectite and mica (celadonite) layers of varying composition and filled with an iron-rich center dominantly composed of goethite. Inset photomicrograph in plane-polarized light.  32  lining layers are —P5 j.im thick and pore-filling layers are —‘50 p.m thick) (Figure 2.6b-d). The red-brown variety represents a mixture of iron-oxides and smectite. The presence of appreciable amounts of K, as well as the higher birefringence of the yellow-green variety, suggests that it may contain a mica, which is tentatively identified as celadonite. The broad XRD peak approximately centered at d = 13.3 A (Figure 2.2d) reflects the presence of smectite and celadonite that is intergrownllayered, which is common in fine-grained phyllosilicates formed at low grades [Peacor, 19921 and expected with members of the celadonite family [Li et a!., 1997]. Vesicles also contain dolomite, showing perfect rhombohedral cleavage and compositional zoning (Figure 2.6b), as well as minor disseminated grains of secondary chalcopyrite. Palagonite (partially altered to smectite) lines unfilled glassy vesicles.  2.4.2 Removal of alteration phases by acid-leaching  Comparison of the X-ray diffractograms of unleached and leached whole rock powders for the two Kerguelen basalts allows for identification of the minerals removed by the Weis et a!. [2005] acid-leaching procedure. For the subaerial Kerguelen basalt from Mont des Ruches (BY96-27), leaching removed the zeolite, clay, and serpentine alteration (Figure 2.7a). An unexpected consequence of the leaching experiment was the disappearance of olivine in the leached powder. The displacement of the leached XRD spectrum to lower intensity (Figure 2.7a  —  inset) reflects the decreased iron radiation due to  removal of iron (in olivine) via leaching. For the submarine Kerguelen basalt from Site 1140 (1 140A 31-1 57-61), the leached and unleached powders show excellent agreement in terms of the primary minerals present (Figure 2.7b). The XRD pattern for the leached  33  a 2400:  2400  Mont des Ruches,  2000-  C D  2000:  Kerguelen Archipelago  isoo  (BY96-27)  cioo:  0  16OO  0  o  3  unleached  /  /leached  10  20  30  40 50 2-Theta &ale  60  70  8Ô  -  -  N  c_)  -  -  1 200-  : : iJ  JJtL  5O’J’7O 2-Theta  2000-  -  b 2000  -  -  1600-  1200 C  Site  1140,  1600  Northern Kerguelen Plateau  12O0  (11 40A 31 R-1 57-61)  500  unleached  /  /eached  J/  -  0 -  20  0  50 2-Thol Soolo -  800 -  80  Scale  a) C)  40:  2-Theta  -  Scale  Figure 2.7  34  6Ô  0  80  Figure 2.7: Bulk powder X-ray diffraction patterns comparing unleached (black lines) and leached (red lines) whole rock powders for the Kerguelen basalt samples. Inset diagrams: raw X-ray diffraction pattern prior to background subtraction. Minerals that were removed via leaching are highlighted with colored bars. (a) Sample BY96-27: leaching effectively removed the zeolites, clays, and serpentine alteration, as well as olivine. (b) Sample 1 140A 3 1R-1 57-51: (incomplete?) removal of the smectite ± celadonite alteration.  35  powder does not show the broad peak caused by the presence of smectite ± celadonite, indicating that this alteration was largely removed through leaching. However, the noise in the leached X-ray diffractogram between 6-9° 20 (Figure 2.7b) may indicate that not all of the alteration was removed. Microscopic examination of the leached residue revealed the presence of altered grains in varying shades of green, which is characteristic of celadonite.  2.5 DISCUSSION 2.5.1 Alteration phases in oceanic basalts and the effectiveness of acid-leaching The four Hawaiian and Kerguelen basalts contain a wide variety of secondary minerals (Table 2.2), most commonly clay mineral mixtures (primarily smectite) and ironoxides. These are typical alteration products of basaltic glass and magnesium silicate minerals resulting from interaction with seawater [e.g., Böhlke et al., 1980; Giorgetti et at., 20011. The subaerial Kerguelen basalt from Mont des Ruches (BY96-27) also contains zeolite, which is characteristic of hydrothermally altered flood basalts from the Kerguelen Archipelago [e.g., Nativel and Nougier, 1983; Nativel et at., 19941. The complete removal of the zeolite, clay, and serpentine alteration from this sample (Figure 2.7a) demonstrates the effectiveness of the Wets et a!. [2005] leaching procedure. The additional removal of olivine was unintended because the leaching procedure is not designed to remove primary minerals. However, the olivine phenocrysts in this basalt were moderately altered, which suggests that the leaching process cannot selectively remove just the alteration from olivine. Despite aggressive leaching (15 steps), the presence of some smectite ± celadonite alteration in the leached residue of the submarine Kerguelen basalt from Site 1140 (1 140A 3 1R-1 5761) suggests that the Wets et at. [2005] leaching procedure may not be effective at  36  Table 2.2: Summary of alteration phases identified in studied Hawaiian and Kerguelen basalts  Hawaii Sample  SR0954-8,00  BY96-27  1 140A 31R-l 57-61  none  tale smectite  zeolite smeetite serpentine goethite  smectite celadonite chlorite goethite  pyrolusite smeetite goethite  barite/celestite pyrite  apophyllite  dolomite ehalcopyrite  Major (>0.5 wt%) alteration phases  Minor (<<0.5 wt%) alteration phases  Kerguelen  J2-019-04  37  completely removing this type of alteration. This could be due to the abundance or distribution of the alteration phases within this basalt, or because some secondary minerals are not readily soluble in HCI. The incomplete removal of alteration phases, which may have a distinct Pb isotopic signature, has important implications for Pb isotope studies of oceanic basalts.  2.5.2 Effect of alteration phases on Pb isotopes The external reproducibility of the Pb isotopic compositions of the four Hawaiian and Kerguelen basalts correlates with their degree of alteration (based on extent of phenocryst and matrix alteration, LOl, number of required leaching steps and associated weight loss due to leaching) (Table 2.1). External reproducibility is calculated as the 2 standard deviation on the mean of multiple Pb isotope analyses of separately processed powder aliquots of the same sample. After leaching, the least altered basalt (J2-0 19-04) has the best 206 Pb reproducibility (2 SD/mean x 106 = 170 ppm), whereas the most altered 204 Pb/ basalt (1 140A 31R-1 57-61) has the poorest reproducibility (359 ppm) (Table 2.3). The relatively poor reproducibility of the most altered basalt is likely related to the incomplete removal of the smectite ± celadonite alteration from this sample by leaching. In general, unleached powders have better Pb isotope reproducibility than the leached powders, by a factor of more than two (Table 2.3). This indicates that the alteration phases may be homogeneously distributed throughout the unleached sample powders, but that leaching does not remove these minerals in an entirely reproducible way [Eisele et a!., 2003; Fekiacova et a!., 2007]. The poorer reproducibility of the leached powders may reflect variable extents of dissolution (between separate powder aliquots) of secondary  38  Table 2.3: Pb isotopic compositions of studied Hawaiian and Kerguelen basaitsa Sample  pb 204 pb/ 206  2G  Pb 4 Pb?° 207  2cr  Pb 204 Pb/ 208  2cr  15.5703 15.4583  14 8  38.4676 37.9751  42 22 18 106(278)  Ha-wail 12 8  J2-019-O4unL J2-019-04L1 J2-019-04L2  18.4951 18.1796 18.1774  J2-019-O4Lmean  18.1785  7 31 (170)  15.4558 15.4570  6 35 (227)  37.9677 37.9714  SR0954-8.OOunLl SR0954-8.OOunL2 SR0954-8.OOunL3 SR0954-8.O0unLmean SR0954-8.OOL1 SR0954-8.00 L2 SR0954-8.00L3 SR0954-8.00 Lmean  18.6336 18.6346 18.6323 18.6335 18.5986 18.6031 18.6020 18.6012  14 12 10 23(122) 12 13 14 47(254)  15.4921 15.4921 15.4908 15.4916 15.4811 15.4832 15.4804 15.4816  12 ii ii 15(98) 16 14 13 29(186)  38.1946 38.1943 38.1898 38.1929 38.1574 38.1641 38.1561 38.1592  BY96-27unLi BY96-27unL2 BY96-27unL3 BY96-27unLmean BY96-27L1 BY96-27L2 BY96-27L3 BY96-27L mean  18.2551 18.2543 182534 18.2542 182539 182542 182495 182525  Kerguelen 6 11 11 17(95) 14 10 6 53 (288)  15.5272 15.5273 15.5258 15.5268 15.5201 15.5211 15.5167 15.5193  5 10 11 17(110) 16 10 6 47 (301)  38.8273 38.8260 38.8210 38.8248 38.7853 38.7885 38.7737 38.7825  35 67(171) 55 31 21 156(402)  1140A31R-i 57-61 unL 1140A31R-i 57-61 Li 114OA31R-157-6iL2 ii4OA3lR-i 57-61 L3 ii4OA3lR-i 57-61 Lmean  18.5451 18.5583 18.5519 18.5567 18.5556  15.5619 15.5637 15.5570 15.5631 15.5613  9 10 15 8 74 (475)  38.9056 38.9384 38.9214 38.9349  29 27 46 23  38.9316  180 (461)  8 12 15 9 67 (359)  34 33 34 54(142) 43 50 41 86(225)  14 29  aAfl Pb isotope data are from Nobre Silva et al. (in revision, 2008). Pb isotope ratica were determined by MG ICP-MS and have been normalized to the NBS981 triple spike values of Galer andAbouchami [19981. Multiple analyses correspond to separate aliquots of unleached (unL) and leached (L) powders. The 2cr error is the absolute error value of an individual sample analysis (internal error) and applies to the last decimal place(s). For the nan values, the error reported is the 2 standard deviation (external error). Numbers in brackets represent the external reproducibility (2SD/mean x 1(P) reported in ppm.  39  minerals that are not readily soluble in HC1 or have differing grain sizes. Alternatively, the poorer reproducibility could reflect discrepancies in the leaching procedure; in particular, the point at which leaching is discontinued (i.e., after how many steps) is based on when a “clear” solution is obtained and is thus inherently subjective. However, such inconsistencies can be greatly reduced if leaching is carried out by the same analyst. The Pb isotope differences between unleached and leached powders (Figure 2.8) are consistent with the presence of distinct secondary mineral assemblages (e.g., celadonite, Mn-oxides, barite, zeolite). The observed alteration assemblages are unique to each basalt and reflect differences in their eruption environment, age, and sampling method (Table 2.1). The two Kerguelen basalts from Mont des Ruches and Site 1140 exhibit opposite Pb isotope relationships between unleached and leached powders (Figure 2.8), indicating discrete sources of contamination that are likely related to their eruption in subaerial and submarine environments, respectively. Although the Pb isotope differences between unleached and leached powders of these Kerguelen basalts are relatively small, they are resolvable outside of the analytical error (Table 2.3). Despite the significantly younger ages (ka vs. Ma) and lower degrees of alteration of the two Hawaiian basalts, the unleached powders are considerably more radiogenic than the leached powders and indicate contamination by highly radiogenic components (Figure 2.8). For the Mauna Loa basalt, the presence of manganese-oxides (Figure 2.3a) suggests that the source of contamination is likely to be seawater (approximated by Pacific Fe-Mn deposits). Fe-Mn deposits form through precipitation out of seawater and occur widely on the ocean floor, and are thus an important consideration for submarine Hawaiian basalts. For the Mauna Kea basalt, the presence of barite/celestite (Figure 2.4b) indicates that drilling mud is the probable contaminant. Barite  40  39.5  HSDP-2 bore hole mud X  39.3 39.1 .  0  Pacific Fe-Mn deposits  38.9  -I  HSDP-2 X slag pond mud  0  38.7  a  I  0 c.1  Hawaii D • J2-019-04 • SR0954-8.OO Kerguelen o. BY96-27 A 1140A31R-1 57-61 L... unleached leached  38.3 38.1 37.9 18.0  —  18.2  18.4  18.6  18.8  19.0  19.2  19.4  19.6  19.8  Pb 204 PbJ 206 Figure 2.8: Comparison of Pb isotope results from unleached and leached whole rock powders of Hawaiian and Kerguelen basalts examined in this study with potential sources of contamination indicated. Pb isotopic compositions determined by MC-ICP-MS from Nobre Silva et al. (in revision, 2008). Unleached compositions are shown with black-filled symbols, leached with white-filled symbols. The ± 2 SD error bars on the mean of multiple analyses are smaller than the symbol sizes. Dashed lines connect the unleached and leached compositions of the same sample and point towards a probable contaminant. Pb isotopic compositions of Pacific Fe-Mn deposits from Abouchami and Galer [1998] (field from Fekiacova et al. [2007]); HSDP-2 mud compositions from Eisele et al. [2003].  41  is commonly used to add weight to drilling mud, which is a significant source of contamination for samples recovered in the Hawaii Scientific Drilling Project [Abouchami et a!., 2000; Eisele et aL, 2003]. The least altered basalt (J2-0 19-04) exhibits the largest difference between leached and unleached Pb isotopic compositions (Figure 2.8), demonstrating that acid-leaching prior to Pb isotopic analysis may be necessary even for basalts that appear to be relatively unaltered. This study demonstrates the importance of complete removal of alteration phases by acid-leaching for high-precision Pb isotope studies of ocean island basalts. The variable results of the leaching experiment indicate that one leaching procedure may not be effective for all ocean island basalts and that leaching procedures may have to be modified depending of the style of alteration encountered. Therefore, studies analyzing the Pb isotopic compositions of ocean island basalts would benefit from a cursory investigation of the degree and type of alteration of the basalts and evaluation of the effectiveness of the particular leaching procedure. A selection of basalts from each sample suite should be examined both macroscopically and by petrographic methods. Samples that contain celadonitic clay mixtures should be treated with caution, and possibly subjected to a stronger leaching procedure. The leached residues of more altered basalts should also be examined microscopically. Furthermore, we recommend that at least two complete procedural duplicates (separate powder aliquots) be analyzed as part of standard quality control protocols to evaluate the external reproducibility of the Pb isotope ratios. Recent improvements in analytical techniques developed for MC-ICP-MS and TIMS, using either thallium or a double/triple spike to correct for instrumental mass bias, have allowed for increased precision (external reproducibility <100 ppm) of Pb isotope analyses. In  42  agreement with previous authors [e.g., Thirlwall et a!., 2000; Eisele et al., 2003; Baker et a!., 2004, 2005; A!barède et a!., 2005; Weis eta!., 2005], the incomplete removal of alteration phases and contaminants by acid-leaching represents one of the main sources of uncertainty of Pb isotope measurements and may be the ultimate limitation on highprecision Pb isotopic compositions of ocean island basalts.  2.6 ACKNOWLEDGEMENTS The author would like to thank Inês Garcia Nobre Silva for providing leaching statistics and Pb isotopic data, as well as Mike Garcia and Njoki Gitahi for sending the Hawaiian samples. Sasha Wilson and Mati Raudsepp are thanked for SEM and XRD sample preparation and instrument training. Andrew Greene, Robert Frei, an anonymous reviewer, and American Mineralogist associate editor Peter Dahl are thanked for their constructive reviews. D. Hanano was supported by an NSERC Canada Graduate Scholarship (CGS-M). Funding for this research was provided by NSERC Discovery Grants to J. S. Scoates and D. Weis.  43  2.7 REFERENCES Abouchami, W., and S.J.G. Galer (1998), The provinciality of Pb isotopes in Pacific Fe-Mn deposits, Mineralogical Magazine, 62A, 1—2. Abouchami, W., S.J.G. Galer, and A.W. Hofmann (2000), High precision lead isotope systematics of lavas from the Hawaiian Scientific Drilling Project, Chemical Geology, 169, 187—209. Abouchami, W., A.W. Hoflnann, 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, 85 1-856. Albarêde, F., A. Stracke, V.J.M. Salters, D. Weis, J. Blichert-Toft, P. Telouk, and A. Agranier (2005), Comment to “Pb isotopic analysis of standards and samples Pb double spike and thallium to correct for mass bias with a double204 Pbusing a 207 focusing MC-ICP-MS” by Baker et al., Chemical Geology, 217, 171—174. Alt, J.C., and J. Honnorez (1984), Alteration of the upper oceanic crust, DSDP site 417: mineralogy and chemistry, Contributions to Mineralogy and Petrology, 87, 149—169. Baker, J., D. Peate, T. Waight, and C. Meyzen (2004), Pb isotopic analysis of standards Pb double spike and thallium to correct for mass bias 204 Pband samples using a 207 with a double-focusing MC-ICP-MS, Chemical Geology, 211, 275—303. Baker, J.A., D.W. Peate, T.E. Waight, and M.F. Thirwall (2005), Reply to the Comment Pb double spike 204 Pbon “Pb isotopic analysis of standards and samples using a 207 and thallium to correct for mass bias with a double-focusing MC-ICP-MS” by Baker et al., Chemical Geology, 217, 175—179. Banerjee, N.R., J. Honnorez, and K. Muehlenbachs (2004), Low-temperature alteration of submarine basalts from the Ontong Java Plateau, in Origin and Evolution ofthe Ontong Java Plateau, edited by J.G. Fitton, J.J. Mahoney, P.J. Wallace, and A.D. Saunders, pp. 259—273, Geological Society, Special Publications, London. Böhlke, J.K., J. Honnorez, and B.-M. Honnorez-Guerstein (1980), Alteration of Basalts From Site 396B, DSDP: Petrographic and Mineralogic Studies, Contributions to Mineralogy and Petrology, 73, 341—364. Dana, J.D. (1849), Geology, in United States Exploring Expedition, 1838-1842, vol. 10, 756 pp., C. Sherman, Philadelphia. DePaolo, D.J., and E.M. Stolper (1996), Models of Hawaiian volcano growth and plume structure: Implications of results from the Hawaii Scientific Drilling Project, Journal ofGeophysical Research, 1O1(B5), 11643—11654. Doucet, S., D. Weis, J.S. Scoates, K. Nicolayson, F.A. Frey, and A. Giret (2002), The depleted mantle component in Kerguelen Archipelago basalts: petrogenesis of tholeiitic-transitional basalts from the Loranchet Peninsula, Journal ofPetrology, 43, 1341—1366. Duncan, R.A. (2002), A time frame for construction of the Kerguelen Plateau and Broken Ridge, Journal ofPetrology, 43, 1109—1119. Eisele, J., W. Abouchami, S.J.G. Galer, and A.W. Hoflnann (2003), The 320 kyr Pb isotope evolution of Mauna Kea lavas recorded in the HSDP-2 drill core, Geochemistry Geophysics Geosystems, 4, doi: 10. 1029/2002GC000339. Fekiacova, Z., W. Abouchami, S.J.G. Galer, M.O. Garcia, and A.W. Hofmann (2007), Origin and temporal evolution of Koolau volcano, Hawaii: Inferences from  44  isotope data on the Koolau Scientific Drilling Project (KSDP), the Honolulu Volcanics and ODP Site 843, Earth and Planetary Science Letters, 261, 65—83. Garcia, M.O., E.H. Haskins, E.M. Stolper, and M. Baker (2007), Stratigraphy of the Hawaii Scientific Drilling Project core (HSDP2): Anatomy of a Hawaiian shield volcano, Geochemistry Geophysics Geosystems, 8, doi: 10.1 02912006GC00 1379. Giorgetti, G., P. Marescotti, R. Cabella, and G. Lucchetti (2001), Clay mineral mixtures as alteration products in pillow basalts from the eastern flank of Juan de Fuca Ridge: a TEM-AEM study, Clay Minerals, 36, 75-91. Jackson, E.D., E.A. Silver, and G.B. Dalrymple (1972), Hawaiian-Emperor chain and its relation to Cenozoic circum-Pacific tectonics, Geological Society ofAmerica Bulletin, 83, 601—618. Kurnosov, V., B. Zolotarev, A. Artamonov, S. Garanina, V. Petrova, V. Eroshchev-Shak, and A. Sokolova (2003), Data Report: Alteration of Basalts from the Kerguelen Plateau, in Proceedings ofthe Ocean Drilling Program, Scientfic Results, vol. 183, edited by F.A. Frey, M.F. Coffin, P.J. Wallace, and P.G. Quilty, pp. 1—40, Ocean Drilling Program, College Station, Texas. Lassiter, J.C., D.J. DePaolo, and M. Tatsumoto (1996), Isotopic evolution of Mauna Kea volcano: Results from the initial phase of the Hawaii Scientific Drilling Project, Journal of Geophysical Research, 1 01(B5), 11769—11780. Li, G., D.R. Peacor, D.S. Coombs, and Y. Kawachi (1997), Solid solution in the celadonite family: The new minerals ferroceladonite, 4 K F S 3 o 2 O 8 , (OH) e i and ferroaluminoceladonite, 4 F K A S o 2 , (OH) igO e l American Mineralogist, 82, 503— 511. McDonough, W.F., and C. Chauvel (1991), Sample contamination explains the Pb isotopic composition of some Rurutu island and Sasha seamount basalts, Earth and Planetary Science Letters, 105, 397—404. Nativel, P., and J. Nougier (1983), Les faciês zëolitiques des basalts des plateaux des Iles Kerguelen (T.A.A.F.); implications volcanologiques, Bulletin de la Societe Géologique de France, 6, 957-96 1. Nativel, P., 0. Verdier, and A. Giret (1994), Nature et diversitë des zéolites de Kerguelen, Mémoires de la Societe Géologique de France, 166, 31-45. Nobre Silva, I. G., D. Weis, J. Barling, and J.S. Scoates (2008), Leaching systematics for the determination of high-precision Pb isotope compositions of ocean island basalts, Geochemistry Geophysics Geosystems, paper # 2007GC001 891 (in revision). Nougier, J., R. Ballestracci, and B. Blavoux (1982), Les manifestations post-volcaniques dans les Iles Australes françaises (TAAF); zones fumerolliennes et sources thermo-minërales, Comptes Rendus de l’Academie des Sciences, Paris, 295, 389— 392. Peacor, D.R. (1992), Diagenesis and low-grade metamorphism of shales and slates, Reviews in Mineralogy and Geochemistry, 27, 335-380. 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, Geochemistry Geophysics Geosystems, 5, doi: 10.1 029/2002GC000434. Tatsumoto, M. (1978), Isotopic composition of lead in oceanic basalt and implication to mantle evolution, Earth and Planetary Science Letters, 38, 63—87.  45  Thiriwall, M.F. (2000), Inter-laboratory and other errors in Pb isotope analyses investigated using a 207 Pb double spike, Chemical Geology, 163, 299—322. 204 Pb— Verdier, 0., Champs gëothermiques et zéolitisation des Iles Kerguelen: Implications géologiques (Terres Australes et Antarctiques Francaises, Ocean Indien Austral), thse Doctorat, 271 pp., Université Paris, France, 1989. Verdier, 0., and P. Nativel (1988), Les activités geothermiques des Iles Kerguelen (TAAF). Les champs fumerolliens actuels et la zéolitisation, in Les lies Océaniques et le Voicanisme des Oceans, pp. 96—100, Colloque France-Danemark, Sp. Pub. Université Paris-Sud, Orsay. Walton, A.W., and P. Schiffman (2003), Alteration of hyaloclastites in the HSDP 2 Phase 1 Drill Core 1. Description and paragenesis, Geochemistry Geophysics Geosystems, 4, doi: 10.1029/2002GC000368. Weis, D., and F.A. Frey (2002), Submarine basalts of the Northern Kerguelen Plateau: Interaction between the Kerguelen Plume and the Southeast Indian Ridge revealed at ODP site 1140, Journal ofPetrology, 43, 1287—1309. 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 B}TVO-2 reference materials, Geochemistry Geophysics Geosystems, 6, doi: 10.1 029/2004GC000852. White, W.M., F. Albarède, and P. Télouk (2000), High-precision analysis of Pb isotope ratios by multi-collector ICP-MS, Chemical Geology, 167, 257—270. Zhou, Z., and W.S. Fyfe (1989), Palagonitization of basaltic glass from DSDP Site 335, Leg 37: Textures, chemical composition, and mechanisms of formation, American Mineralogist, 74, 1045—1053. Zhou, Z., W.S. Fyfe, K. Tazaki, and S.J. Vandergaast (1992), The structural characteristics of palagonite from DSDP Site 335, Canadian Mineralogist, 30, 75—8 1.  46  CHAPTER 3  TRACE ELEMENT AND ISOTOPE GEOCHEMISTRY OF POST-SHIELD LAVAS FROM MATJNA KEA, KOHALA, AND HITALALAI: EVIDENCE FOR ANCIENT DEPLETED COMPONENTS IN THE HAWAIIAN MANTLE PLUME  47  3.1 INTRODUCTION 1 The -—6000 km long Hawaiian-Emperor Seamount Chain represents the surface expression of the Hawaiian mantle plume and is the classic example of intraplate volcanism. The Hawaiian plume is one of the longest-lived and hottest plumes with the largest buoyancy flux [Sleep, 1990] and is inferred to have a deep mantle origin [e.g., Courtillot et a!., 2003; Montelli et a!., 2006]. Volcanoes at the young end (<5 Ma) of the Hawaiian chain define two parallel geographic trends, termed Loa and Kea [Dana, 1849; Jackson eta!., 1972]. Individual Hawaiian volcanoes evolve through four different growth stages [Clague and Dairymple, 1987]: (1) a small volume (—3%) alkalic pre-shield, (2) the main tholeiitic shield stage during which the majority (95-98%) of the volcano is built, (3) a small volume (-—1%) alkalic post-shield, and following 0.5-2.5 myr of quiescence, (4) a volumetrically minor (<<1%) strongly alkalic rejuvenated stage. The geochemistry of Hawaiian lavas can provide important constraints on the structure and composition of the Hawaiian plume [e.g., DePaolo et a!., 2001; Blichert-Toft et al., 2003; Eisele et a!., 2003; Abouchami et a!., 2005; Bryce eta!., 2005; Marske eta!., 2007]. In particular, systematic geochemical variations are observed between lavas erupted from the two geographic trends and during different growth stages. These variations are generally explained by inferring that the Hawaiian plume is either concentrically zoned [e.g., Hauri, 1996; Kurz et a!., 1996; Lassiter et at., 1996] or bilaterally zoned [Abouchami et at., 2005] in cross-section. However, vertical heterogeneities within the upwelling plume are also likely to be an important factor in creating the observed geochemical variability [Blichert-Toft eta!., 2003; Marske et at., 2007]. 1 A version of this chapter will be submitted for publication. Hanano, D., D. Weis, S. Aciego, J. S. Scoates, and D. J. DePaolo (2008), Geochemistry of post-shield lavas from Mauna Kea, Kohala, and Hualalai: Evidence for ancient depleted components in the Hawaiian plume, Geochemistry Geophysics Geosystems.  48  The compositional range of Hawaiian shield stage lavas is typically accounted for by mixing of at least three isotopically distinct end-members, referred to as Loihi, Kea, and Koolau [e.g., Staudigel et al., 1984; Stille et aL, 1986; West et aL, 1987; Eiler et al., 1996; Hauri, 1996]. However, the depleted isotopic signatures of post-shield and rejuvenated stage lavas require contributions from an additional component that may be intrinsic to the plume [e.g., Frey et al., 2005] or derived at least in part from external sources, including entrained upper mantle and the underlying Pacific lithosphere [e.g., Chen and Frey, 1985; Gaffiiey et al., 2004]. The distribution of heterogeneities in the Hawaiian plume, and in particular the depleted component involved in the post-shield and rejuvenated stages, remains a controversial unresolved aspect of Hawaiian volcanism. Post-shield lavas occur on many Hawaiian volcanoes including Haleakala, West Maui, East and West Molokai, Waianae, and Kauai. At other Hawaiian volcanoes, the postshield stage is either minor (e.g., Kahoolawe and Niihau) or absent (e.g., Lanai and Koolau). Post-shield lavas are the target for this research because they are derived from low degrees of partial melting (i.e., a few percent) and can thus provide fmer resolution of variation in source composition than shield lavas. The lower magma supply during the post-shield stage (<0.005 km /yr; one-tenth to one-hundredth of the rate during the shield stage) prevents a 3 single long-lived conduit from being maintained, and each eruption is likely to be associated with a separate vent and path through the lithosphere [e.g., Frey et a!., 1990; Wolfe et a!., 1997; Hieronymus and Bercovici, 20011. Assuming that melt can be extracted from small regions (e.g., length scale <10 km) within the melting zone without substantial subsequent chemical modification [Mars/ce et a!., 2007], post-shield lavas permit identification of smallscale heterogeneities within the Hawaiian plume. Furthermore, because post-shield lavas  49  erupt after the volcano has migrated to the periphery of the plume, the geochemistry of these lavas has implications for the large-scale structure of the Hawaiian plume. We report high-precision trace element concentrations by HR-ICP-MS and Sr, Nd, Pb and Hf isotopic compositions of 32 post-shield lavas from Mauna Kea, Hualalai, and Kohala volcanoes on the island of Hawaii (Figure 3.1; Table 3.1). Systematic temporal trends are identified within these volcanoes as they evolve from the shield to post-shield stage. We then evaluate if the distinct isotopic compositions of the post-shield lavas can be explained by mixing between the shield end-members or by contributions from the underlying Pacific lithosphere, shallow asthenosphere, or components within the plume. The post-shield lavas from this study are compared with late-shield, post-shield, and rejuvenated stage lavas from other Hawaiian volcanoes to determine the extent and origin of depleted components in Hawaiian lavas. Additionally, comparison with Mahukona volcano allows the geochemistry of two consecutive Loa-Kea volcano pairs to be evaluated, which provides both spatial and temporal constraints on the geochemical structure of the Hawaiian plume.  3.2 SAMPLE DESCRIPTION AND GEOLOGIC SETTiNG Kohala is the oldest volcano on the island of Hawaii and has completed post-shield activity. Extensive landsliding and erosion have carved deep canyons in the flanks of the volcano, exposing the older stratigraphy of the shield-stage Polulu Volcanics. Three lateshield lavas from the Polulu Volcanics were included in this study, two of which have been dated at 450 ± 40 ka and 375 ±22 ka (Aciego, S. M., F. Jourdan, D. J. DePaolo, B. M. Ar geochronology of post-shield 39 Ar/ Kennedy, and P. R. Renne, Combined U-Th!He and 40  50  Figure 3.1: Geological map of the island of Hawaii [Wolfe and Morris, 1996]. The vent locations of the post-shield samples from Hualalai, Mauna Kea, and Kohala volcanoes are indicated. Different colours and patterns represent distinct lithologic units (mainly lavas flows of varying age). The thick grey lines mark the approximate boundaries between lavas from Kilauea, Mauna Loa, Hualalai, Mauna Kea, and Kohala volcanoes. The approximate locations of the submarine summits of Mahukona and Loihi volcanoes are also shown. The inset map shows the location of the island of Hawaii with respect to the other Hawaiian Islands in the Pacific Ocean.  51  Table 3.1: Sampling and Vent Locations of Hawaiian Post-shield Lavasa  Sample  Vent Name  O2AHIJ-1 O2AHIJ-2 O2AHIJ-3 02A1{U-4 O2AHIJ-5 O2AHU-6 O2AHU-7 02AH(J-8 O2AHU-9 O2AHU-10 O2AHU-11 O2AHU-12 O2AHU-13  WahaPele PuuIkaaka Poikahi no vent no vent no vent unnamed vent Puu Waawaa Kuainiho PuuNahaha PuuNahaha unnamed vent unnamedvent  O2AMK-1 O2AMK-2 O2AMK-3 O2AMK-4 O2AMK-5 O2AMK-6 O2AMK-7 O2AMK-8 O2AMK-10 O2AMK-1i O2AMK-12 O2AMK-13  PuuoKauha no vent novent no vent Puulo Kaluamakani unnamedvent Kalepa Aahuwela PuuHinai PuuPa PuuPapapa  O2AKA-l O2AKA-2 O2AKA-3 O2AKA-4 O2AKA-5 O2AKA-6 O2AKA-7  Kaiwaiwai PuuPili  Age (ka)  0.71 7.25 2.25 29 , 2 b  12.951) 2.25 4•71)  114 3C 7.5 4 4 2.25 2.25  Elevation (masl)  PuuHonu  175  PuuMakela no vent PuuKamalii PuuoNale  137±5e d 450 ± 40 400 375±22’  Longitude  Vent Latitude  Longitude  155.94233 155.8156 155.83253 155.94355 155.99557 155.93588 155.97395 155.83072 155.77017 156.01565 156.01565 155.81035 155.80987  19.64333 19.62713 19.68167 19.62833 19.69 19.65 19.73333 19.77768 18.83842 19.775 19.775 19.74972 19.74733  155.82833 155.8156 155.83253 155.835 155.865 155.83333 155.935 155.83072 155.77017 155.975 155.975 155.81035 155.80987  Volcano 19.76283 19.77262 19.77272 19.77285 19.98213 19.95283 19.94378 19.91 19.7766 19.94015 20.00908 19.90715  155.57828 155.4725 155.47212 155.47188 155.56375 155.50212 155.47448 155.34382 155.35068 155.83792 155.81372 155.7051  19.7975 19.82167 19.82333 19.825 19.98213 19.90417 19.9125 19.88 19.785 19.935 19.98917 19.89333  155.56667 155.47 155.47167 155.4 7333 155.56375 155.49083 155.47167 155.36333 155.365 155.77667 155.70417 155.68917  Kohala Volcano 302 20.04422 3052 20.11615 3250 20.10255 3428 20.07263 442 20.04913 187 20.04182 20,20917 1559  155.73507 155.78838 155.77872 155.76115 155.83075 155.83263 155.83312  20.04422 20.115 20.105 20.07263 20.08333 20.04917 20,20083  155.73507 155,76 155.76833 155.76115 155.71 667 155.82167 155.84333  MaunaKea 5923 9913 9990 10068 3962 4866 123±5” 5312 45 5730 45 6742 19±4” 259 239±84’ 362 142±22” 2790  O± 9 l O 2 d  Latitude  Hualalai Volcano 19.5692 5428 19.62713 6820 19.67532 1352 19.5791 137 19.64175 1281 19.553 1979 19.74867 19.77768 2594 18.83842 372 19.77888 19.77888 4100 19.74972 4118 19.74733  45 160 d 155± 11 30 225 45  175  Sample  masl = meters above sea level; latitude and longitude expressed in decimal degrees; no vent indicates that the sample could not be traced back to a specific vent due to overlying flows the corresponding vent coordinates (in italics) are approximate; where sample and vent coordinates are the same, the sample was taken directly from the vent; ages are approximate and based on the geologic mapping of Wolfe andMorris [19961 unless otherwise specified. bWO and Morris [1996]. C-14 age. Cousens et al. [2003]. °Ar/ Ar isochron age with analytical uncertainty given at 2a. 39 dAciego et al. [submitted, 2008]. Ar/ Ar age with analytical uncertainty given at 2o. 39 °McDougall [1969]. K-Ar age with analytical uncertainty given at 2a. -  52  lavas from the Mauna Kea and Kohala volcanoes, Hawaii, manuscript submitted to Earth and Planetary Science Letters, 2008; hereinafter referred to as Aciego et al., submitted, 2008). Four samples were collected from the overlying lava flows of the post-shield Hawi Volcanics. The majority of the post-shield vents are located along Kohala’s northwest rift zone. The post-shield lavas are significantly younger, with ages of samples in this study ranging from 190  ±  20 ka [Aciego et a!., submitted, 2008] to 137 ± 5 ka [McDougall, 1969].  Mauna Kea is the second largest Hawaiian volcano (after Mauna Loa) and rises to 4205 m above sea level. In contrast to other Hawaiian volcanoes, where vents are largely confined to the rift zones, the post-shield vents at Mauna Kea are scattered over the surface of the volcano. Post-shield volcanism on Mauna Kea is divided into an earlier basaltic substage (Hamakua Volcanics) and a later hawaiitic substage (Laupahoehoe Volcanics) [Wolfe eta!., 1997]. Six samples from each sub-stage were selected for this study. The basaltic samples range in age from 239 ± 84 ka to 123  ±  5 ka, and the hawaiitic samples are  as young as 19 ±4 ka [Aciego eta!., submitted, 2008]. Hualalai is the youngest of the three volcanoes in this study and is still actively in the post-shield stage, having last erupted in 1801. Most of the surface of the volcano is covered by the post-shield lavas of the Hualalai Volcanics, which form a thin veneer over the shieldstage lavas. The post-shield vents are primarily located along the well-developed northwest rift zone as well as a south-southeast trending rift zone. Thirteen samples were collected, most of which range in age from 13 ka to 2.3 ka [Wolfe and Morris, 19961. One sample is from the young (0.71 ka) Waha Pele flow on the southwestern flank of the volcano, and one sample is from the Puu Waawaa pumice cone (Waawa.a Trachyte Member) on the northern slope, dated at 113.5  ±  3.2 ka [Cousens eta!., 2003].  53  A total of 32 subaerial late-shield and post-shield lavas were collected from Kohala, Mauna Kea, and Hualalai volcanoes on the northern part of the island of Hawaii (Figure 3.1). The samples were taken from gulches and road cuts on the flanks of these volcanoes from lava flows that could be traced back to a specific vent (Table 3.1). The collected samples are fresh; olivine phenocrysts are unaltered and there is only rare minor alteration of the groundmass.  3.3 ANALYTICAL TECHNIQUES Sample preparation of the 32 post-shield lavas was carried out at the Center for Isotope Geochemistry (CIG) at the University of California, Berkeley. Whole rocks were cut using a diamond-embedded saw and then abraded with sandpaper to eliminate saw traces. The samples were coarse-crushed (to —5 mm diameter) in a hydraulic piston crusher between tungsten carbide (WC) plates and subsequently ground to a fine powder in a WC mill for XRF analyses and an agate mill for ICP-MS and isotope analyses (the effect of WC contamination on trace elements and Pb and Hf isotopes is discussed in Appendix A3). The major element oxides and some trace elements (Co, Ni, Cu, Zn, Ga,, Ge, As, Rb, Sr, Y, Zr, Nb, Mo, Ba, La, Ce, Pr, Nd, Sm, Pb, Th and U) were determined by X-ray fluorescence (XRF) using a Philips PW2400 spectrometer at the CIG, University of California, Berkeley. Measurements were calibrated against USGS reference materials AGV, BHVO, and BCR. The major elements were measured on glass discs (fused with lithium tetraborate flux) and the trace elements were measured on pressed powder pellets. XRF compositions are the mean of duplicate analyses and are reported in Table 3.2. Complete trace element characterization by high-resolution inductively coupled  54  Table 3.2: Major and Trace Element Abundances of Hawaiian Post-shield Lavasa Hualalai Volcano O2AHU-1 Major elements 2 Si0 2 Ti0 3 0 2 A1 * 3 0 2 Fe  O2AHIJ-2  02AHU3b  O2AHU-4  O2AHU-5  O2AHU-6  O2AHLJ-7  02A[{LJ-8  (wt%) 46.74 1.99 13.63 13.90 0.18 10.55 10.39 2.76 0.81 0.26 101.21  45.39 1.53 9.76 13,29 0.18 16.74 10.08 1.93 0.55 0.19 99.65  49.46 2.45 14.99 12.65 0.18 6.30 8.42 3.76 1.40 0.45 100.07  47.51 1.82 12.61 13.13 0.18 12.00 10.43 2.45 0.68 0.21 101.02  47.33 2.03 12.70 13.03 0.18 10.36 11.38 2.48 0.69 0.23 100.40  46.63 2.25 13.49 13.67 0.18 9.30 10.06 2.93 0.93 0.29 99.73  47.35 2.90 14.41 14.06 0.19 5.80 9.87 3.38 1.15 0.38 99.49  62.20 0.39 17.28 4.54 0.31 0.49 0.88 7.26 4.91 0.15 98.42  0.63 0.71  0.73 0.12  0.52 1.29  0.67 -0.02  0.64 0.09  0.60 1.04  0.48 1.44  0.19 3.59  Trace elements (ppm) ScTC’ 25.4 VIC? 301 470 60.4 252 Ni Zn 99 21 Ga 0.131 Cs’ RbKP 19.2 SrXI 427 Ba 249 Y 21.1 ZrXl 125 Hi’ 2.77 Nb 20.9 Ta 0.93 W’°’ 0.22  28.9 278 1513 80.6 648 95 15 0.052 10.5 319 183 16.8 101 2.28 16.1 0.83 0.14  18,8 250 163 40.6 96 103 21 0.294 33.9 549 443 29.6 209 4.55 39.3 1.80 0.45  27.0 279 778 62.4 362 96 18 0.039 13.2 364 224 19.2 109 2.44 18.3 0.74 0.17  31.0 316 644 65.5 240 90 20 0.115 15.4 382 213 20.8 114 2.69 19.7 0.69 0.28  25.6 299 385 57.5 208 105 21 0.213 21.6 488 304 23.5 140 3.25 24.0 1.16 0.53  22.4 377 100 46.2 76 114 23 0.258 27.6 508 370 27.8 181 4.06 32.4 1.49 0.52  6.2 0.68 0.23 0.14 4 180 26 1.109 115.5 37 281 58.7 960 20.23 139.8 4.78 2.69  15.0 33.6 4.27 17.8 4.06 1.36 4.01 0.61 3.68 0.70 1.80 1.44 0.213  12.0 26.9 3.42 14.6 3.30 1.10 3.26 0.49 2.98 0.58 1.46 1.16 0.165  28.4 60.7 7.38 29.8 6.26 2.03 5.90 0.88 5.24 1.01 2.57 2.14 0.308  12.8 28.5 3.57 15.2 3.53 1.22 3.55 0.56 3.43 0.65 1.69 1.36 0.199  13.8 31.3 3.95 16.7 3.90 1.31 3.83 0.61 3.62 0.67 1.73 1.46 0.215  18.7 42.1 5.11 21.4 4,81 1.63 4.68 0.71 4.19 0.79 2.02 1.69 0.248  21.6 47.7 6.11 25.6 5.75 1.92 5.42 0.84 4.86 0.91 2.34 1.96 0.283  57.5 116.8 12.51 42.4 8,06 1.89 6.55 1.20 7.64 1.55 4.61 5.18 0.818  1.38 1.44 0.396  1.02 1.03 0.281  2.23 2.74 0.759  1.23 1.15 0.303  1.16 1.17 0.282  4.48 1.59 0.458  1.70 1.94 0.585  7.15 8.37 2.458  MnO MgO CaO 0 2 Na 0 2 K 5 0 2 P  Total mg-number A.I.  La Ce’ prICP  Nd ’ 1 Sm Eu’  Gd’ Tb’’  Dy Ho  ErICl’  Lu 1 Pb ThICP UICP  55  Table 3.2 (continue Mauna Kea Volcano O2AFIU-9 Major elements 2 Si0 2 Ti0 3 0 2 A1 * 3 0 2 Fe  O2AHU-10 O2AHU-11  O2AHU-12  O2AH(J-13  O2AMK-1  O2AMK-2  02AMK-3  (wi%) 47.12 2.05 13.10 12.71 0.18 9.60 11.58 2.48 0.73 0.23 99.78  47.50 1.87 12.60 13.26 0.18 11.95 10.67 2.45 0.68 0.22 101.39  45.33 2.19 12.33 13.38 0.18 11.65 10.61 2.53 0.81 0.26 99.27  47.06 2.23 14.13 13.77 0.19 8.72 10.58 2.81 0.94 0,27 100.70  47.09 2.18 14.10 13.72 0.19 9.05 10.59 2.78 0.91 0,27 100.88  48.37 2.93 16.59 12.04 0.20 4.48 7.16 4.53 1.74 0.76 98.80  47.92 3.17 14.09 13.75 0.18 5.92 10.85 2.88 0.81 0.38 99.95  47.96 3.30 14.17 14.54 0.19 5.74 10.77 2.94 0.80 0.41 100.83  0.62 0.21  0.66 -0.01  0.66 1.00  0,58 0.77  0.59 0.70  0.45 2.80  0.49 0.39  0.46 0.42  Trace elements (ppm) Sc’” 29.3 318 CrTC’ 555 Coa 55.2 Ni 228 Zn 101 Ga 21 CsK’ 0.066 Rb’’ 14.0 5 Sr 375 Ba’ 226 °° 3 Y 21.8 Zr 122 2.92 Nb 18.7 Ta 0.98 WICP 0.34  29.0 296 701 67.2 343 97 20 0.081 13.4 347 205 19.8 109 2.56 18.8 0,99 0.69  28.0 327 689 63.5 328 100 20 0.176 18.4 436 268 19.8 129 3.03 24.8 1,22 0.63  27.6 304 338 55.6 173 104 22 0.199 20.5 443 269 24.1 153 3.52 24.7 1.14 0.70  25.5 302 372 55.7 195 106 22 0.226 20.2 439 257 23.4 146 3.36 24.0 1.20 0.70  9.9 109 4.5 18.4 13 124 26 0.252 29.8 1287 533 46.3 396 8.11 55.9 2.69 1.26  23.9 280 30 42.5 74 118 26 0.110 11.8 565 209 33.0 230 4.75 25.0 2.62 0.79  25.7 298 34 46,2 64 127 26 0.105 12.1 573 231 35.0 241 5.23 29.0 1.41 0.73  14.3 32.4 4.06 17.3 4.10 1.38 4.06 0.63 3.74 0.71 1.83 1.53 0.222  12.9 28.4 3.78 15.9 3.68 1.27 3.71 0.58 3.39 0.65 1.71 1.41 0.198  16.5 36.6 4.57 19.4 4.35 1.47 4.24 0.63 3.54 0.68 1.76 1.40 0.198  18.6 41.5 5.22 21.7 4.84 1.56 4.62 0.73 4.04 0.80 2.16 1.75 0.255  17.4 38.7 4.94 20.3 4.56 1.50 4.39 0.69 3.99 0.75 2.02 1.58  0.240  39.2 92.9 12.07 51.9 11.65 3.74 10.37 1.44 7.96 1.41 3.55 2.66 0.378  19.1 45.4 6.10 26.4 6.61 2.19 6.45 0.97 5.23 0.98 2.54 1.98 0.279  22.4 52.3 7,02 31.1 7.28 2.38 6.93 1.04 5.92 1.05 2.68 2.07 0.297  1.19 1.20 0.272  1.10 1.13 0.273  1.33 1.46 0.399  1.57 1.56 0.434  1.55 1.52 0.434  2.32 3.13 0.886  1.22 1,49 0.469  1.28 1.65 0.519  MnO MgO CaO 0 2 Na  K 0 2 5 0 2 P Total mg-number A.I.  La Ce  NdP Sm Eu 1 ThICP DyTCP  Ho  Er’ Yb Lu Pb” Th” UICP  56  Table 3.2  (continued)  O2AMK-4 Major elements 2 Si0 2 Ti0 3 0 2 A1 * 3 0 2 Fe  (wt%) 49.85 2.71 16.76 11.61  02AI4K-5  O2AMK-6  b 027  O2M8b  O2AMK-10 02AMK-1 1 O2AMK-12  46.60 2,20 11,27 12.47  50.31 2.52 16.92 11.28  46.54 3.79 13.80 14.43  50.96 2.46 17.08 11.22  48.91 2.60 16.76 11.52  49.66 2.76 16.86 11.86  46.72 1.82 10.09 11.84  MnO  0.21  0.17  0.21  0.19  0.21  0.21  0.21  0.16  MgO CaO 0 2 Na 0 2 K 5 0 2 P Total  4.22 6.98 4.83 1.86 0.83 99.85  13.35 11.13 1.97 0.53 0.26 99.94  3.94 6.80 4.89 1.96 0.94 99.77  5.88 9.61 3.39 1.27 0.64 99.54  3.79 6.68 4.98 2.02 0.98 100.38  3.94 6.75 4.66 1.86 0.88 98.09  4.27 6.98 4.86 1.85 0.81 100.11  15.95 11.19 1.80 0.44 0.20 100.22  0.44 2.68  0.70 -0.31  0.43 2.67  0.47 1.87  0.43 2.57  0.43 2.85  0.44 2.77  0.75 -0.62  Trace elements (ppm) 9.2 102 Cr 8.9 Co 15.9 Ni 14 Zn° 126 Ga 25 Cs 0.300 RbICP 33.1 SrXl 1279 Ba’ 564 Y 48.8 Zr 422 H1CP 8.83 58.2 Ta’ 3.41 WIC’ 1.75  30.7 227 818 61.0 451 97 19 0.067 9.4 458 180 21.2 149 3.69 19.6 0.92 0.54  8.3 88 18 15.2 14 130 26 0.215 32.8 1271 578 52.1 461 9.56 61.4 3.13 1.53  21.3 289 166 45.2 91 139 26 0.143 21.1 724 378 40.1 341 7.35 45.0 1.76 0.99  8.2 80 0.87 14.1 5 133 25 0.097 33.3 1255 610 52.6 478 10.23 65.5 3.04 1.23  8.9 86 1.9 16.2 8 141 26 0.120 29.3 1246 573 49.0 442 9.48 61.3 3.03 1.74  10.0 83 0.75 17.2 12 115 25 0.286 32.6 1295 556 47.2 416 8.75 57.5 2.53 5.03  32.7 221 1271 70.9 539 88 16 0.063 7.7 377 135 17.8 120 2.97 14.7 0.91 0.65  43.6 103.3 13.38 56.7 12.62 3.98 11.16 1.59 8.59 1.52 3.81 2.90 0.405  16.5 38.6 4.99 21.8 5.08 1.69 4.95 0.71 4.01 0.72 1.86 1.45 0.193  49.0 115.7 14.64 61.9 13.60 4.18 11.78 1.69 9.05 1.62 4.10 3.16 0.444  36.1 81.2 10.45 44.1 9.69 3.08 8.99 1.29 7.19 1.28 3.19 2.42 0.339  52.9 121.8 15.53 66.5 14.14 4.48 12.53 1.78 9.54 1.71 4.28 3.32 0.466  46.9 104.2 13.96 60.0 13.07 4.20 11.52 1.60 8.77 1.57 3.81 2.95 0.405  45.1 103.3 13.44 57.8 12.65 4.09 11.34 1.63 8.80 1.56 3.86 2.97 0.420  12.9 28.4 3.91 17.4 4.18 1.41 4.12 0.60 3.44 0.63 1.56 1.19 0.161  2.46 3.46 1.091  0.96 1.26 0.398  2.64 3.84 1.236  1.95 2.96 0.892  2.92 4.04 1.252  2.86 3.61 1.127  2.58 3.49 1.073  0.79 1.00 0.297  mg-number  A.I. Sc VICP  LaTC Ce pICP  NdK Sm Eu Gd ThICP  DyICP Ho ErIC’ ,ICP 1 y LuW’  ThICP  UICP  57  Table 3.2 (continued)  O2AMK-13 Major elements 2 Si0 2 Ti0 3 0 2 A1 * 3 0 2 Fe  Kohala Volcano O2AKA1b O2AKA-2  O2AXA-3  O2AKA4b  O2AKASb  O2AKA-6  O2AKA-7  (wt%) 47.15 2.96 15.11 13.55 0.18 5.80 11.46 2.76 0.75 0.36 100.08  46.88 2.85 15.64 12.92 0.22 4.34 7.36 4.84 1.54 2.15 98,74  50.69 2.18 16.67 10.92 0.23 3.55 6.10 5.38 1.99 1.60 99.31  49.04 2,54 16.51 12.17 0.23 3.86 6.77 4.89 1.80 1.83 99.64  58.02 1.02 17.84 7.13 0.23 1.48 2.86 6.90 2.90 0.68 99.06  46.90 3.17 13.06 13.31 0.18 7.75 10.65 2.78 0.80 0.49 99.10  49.63 2.81 14.11 13.25 0.18 5.89 10.33 2.77 0.61 0.38 99.97  48.42 3.32 14.04 13.34 0.17 4.78 9.85 3.15 0.85 0.50 98.41  0.49 0.49  0.42 3.46  0.42 3.04  0.41 2.98  0.31 2.76  0.56 0.66  0.49 -0.55  0.44 0.51  Trace elements (ppm) 28.4 291 1 Cr 139 Coldl’ 46.0 Ni 75 Zn 108 Ga 25 t Cs 0.116 Rb 13.6 SrX 609 e T Ba 233 Y 29.6 5 Zr 203 J{fICP 5.21 Nb 25.4 TaIC’ 1.35 WICP 0.96  8.1 77 0.30 15.5 14 129 20 0.251 24.1 1832 553 62.3 320 6.62 54.0 2.60 2.94  6.2 36 0.24 9.8 10 128 22 0.337 33.3 1954 689 60.3 424 8.64 69.6 2.49 0.80  7.3 62 0.17 13.4 10 123 21 0.236 31.9 1734 623 58.0 376 7.98 61.3 3.27 1.95  2.5 3.2 0.44 2.1 7 115 21 0.243 75.9 1319 933 44.9 626 11.14 86.6 3.62 1.15  26.9 240 385 47.5 180 120 25 0.065 9.1 504 258 38.2 269 5.77 28.9 1.22 0.86  29.8 278 78 42.3 81 118 24 0.112 11.7 439 176 36.0 223 4.82 22.0 0.99 0.97  27.5 317 89 42.7 63 125 28 0.124 11.6 536 260 38.2 264 5.91 33.1 1.48 1.18  22.5 51.4 6.75 29.4 7.02 2.30 6.73 0.97 5.51 0.98 2.50 1.88 0.263  51.2 123.7 17.06 77,4 17.40 5.51 15.57 2.08 10.90 1.95 4.60 3.24 0.457  65.2 148.9 19.18 82.0 16.80 5.17 14.13 1.95 10.41 1.88 4.63 3.39 0.48 1  57.3 133.5 17.63 77.6 16.70 5.21 14.79 1.99 10.61 1.88 4.52 3.25 0.450  81.0 163.2 17.48 68.6 12.55 4.05 9.26 1.39 8.23 1.48 3.96 3,41 0.493  25.3 58.3 7.83 34.5 8.13 2.57 7.91 1.16 6.72 1.25 3.11 2.47 0.348  20.5 46.2 6.36 28.1 6.88 2.28 6.75 1.05 6.22 1.16 2.96 2.32 0.325  25.6 57.3 7.72 33.8 8.15 2.71 7.88 1.19 6.74 1.26 3.18 2.51 0.358  1.24 1.75 0.568  1.94 2.85 0.891  2.64 3.89 1.073  2.40 3.74 1.242  4.37 8.00 2.258  1.37 1.85 0.506  1.19 1.37 0.454  1.36 1.80 0.662  MnO MgO CaO 0 2 Na 0 2 K 5 0 2 P Total mg-number A.I. Sc 1t VICP  ’ 1 LaIC 1 Ce Pr  Nd Sm” EuICP ThICP DytC  5 Er yi I 0 CP 1 Lu  PbtdI’ Thn UICP  ‘Major element abundances were determined by XRF; trace element abundances were determined by either XRF or HR-ICP-MS (denoted as 0 (Na 0 + ).O.37*Si0 14.43. +K * is total Fe; mg-number = (Mg/Mg+Fe 3 0 2 ); Al. (Aikalinity Index) 2 21 ICP); Fe bICp.MS abundances are the mean of duplicate or replicate analyses. 58  plasma mass spectrometry (HR-ICP-MS) was carried out at the Pacific Centre for Isotopic and Geochemical Research (PCIGR), University of British Columbia. Comprehensive sample preparation and analytical procedures are described in Pretorius et al. [2006]. 3 in sealed Sample powders were digested in a mixture of concentrated HF and HNO Teflon® vessels and subsequently diluted (with 1% HNO 3 and 1 ppb In) to 1000 and 3000 times for the rare earth element (REE) and high field strength element (HF SE) analyses, respectively. Indium (In), which was used as an internal standard, was spiked at 1 ppb in all blank, standard, and sample solutions. The solutions were analysed on a Thermo Finnigan Element2 HR-ICP-MS using external calibration. The REE were measured in high resolution, whereas U, Pb and Th were measured in low resolution. The majority of the HFSE were measured in medium resolution, except for Sr, Zr and Ba, which were measured in high resolution to avoid overloading the detector. Trace element abundances by HR-ICP MS are reported in Table 3.2 (see Appendix Al for a comparison of XRF and HR-ICP-MS values). For most elements, values of the USGS reference materials BCR-2 and AGV-2, reported in Appendix A2, are within 2a error of literature and recommended values (Tables Al and A2). Procedural duplicates and replicate measurements show excellent agreement, with relative standard deviations (RSD) less than 5% for most elements. Procedural blank values for the REE and most other elements are in the parts-per-trillion Q,pt) range, and are considered negligible in comparison to the sample concentrations, which are in the parts per-million (ppm) range. Column blanks for Sr, Nd, Pb and Hf isotope chemistry (discussed below) were also negligible (<<1 ng). Sr and Nd isotopic compositions were determined at the CIG, University of California, Berkeley and are reported in Table 3.3. Sample powders were leached in a  59  Table 3.3: Hf, Sr, and Nd Isotopic Compositions of Hawaiian Post-shield Lavasa Sample  Volcanic Series  O2AHU-1 O2AHU-2 O2AHU-3  Hualalai Hualalai Hualalai  O2AHU-4 O2AH(J-5 O2AH(J.6 O2AHU-7 O2AHU-8 O2AHIJ-9 O2AHU-10 O2AHU-11 O2AHIJ-12 O2AHIJ-13  Hualalai Hualalai Hualalai Hualalai Hualalai” Hualalai Hualalai Hualalai Hualalai Hualalai  O2AIvIK-1 O2AMK-2 O2AMK-3 O2AMK-4 O2AMK-5 O2AMK-6 O2AMK-7 O2AMK-8  Laupahoehoe Hamakua Hamakua Laupahoehoe Hamakua Laupahoehoe Hamakua Laupahoehoe  O2AMK-10 O2AMK-11 O2AMK-12 O2AIVIK-13  Laupahoehoe Laupahoehoe Hamakua Hamakua  O2AKA-1  Hawi  O2AKA-2 O2AXA-3 O2AKA-4 O2AKA-5 O2AKA-6 O2AKA-7  Hawi Hawi Hawi Polulu Polulu Polulu  177 j 176 jf  2  Hualalai Volcano 0.283144 4 0.283135 6 0.283108 5 0,283119 5 0.283129 6 0.283134 5 0.283127 5 0.283130 5 0.283093 4 0.283123 5 0.283112 5 0.283128 6 0.283113 6 4 0.283110 0.283113 4  S7 Sr/ta  Sr  2o  Nd/’Nd 143  2a  0.703578 0.703556  8 13  0.512944 0.512934 0.512935  10 6 4  0.703564 0.703537 0.703532 0.703568 0.703713 0.703570 0.703636 0.703511 0.703587 0.703597  11 11 10 10 35 14 10 10 12 10  0.512945 0.512915 0.512954 0.512958 0.512904 0.512936 0.512949 0.512964 0.512941 0.512928  5 6 5 5 6 4 6 6 6 4  11 20 17 8 14 10 8 11  0.513003 0.5 12980 0.5 12979 0.5 12969 0.513003 0.513015 0,513009 0.512987  6 4 4 5 8 5 9 4  8 11 8 10  0.512996 0.513012 0.512987 0.512996  6 8 6 4  0.703520  8  0.512986  6  0.703534 0.703513 0.703525 0.703647 0.703654 0.703633  10 10 10 7 9 10  0.512987 0.512988 0.513008 0.512993 0.512996 0.512969  5 6 4 7 4 4  Mauna Kea Volcano 0.283130 4 0.703432 0.283 126 4 0.70355 1 4 0.703488 0.283 127 0.703455 0.283 133 5 0.703512 0.283 127 7 0.703463 0.283134 5 0.283124 0.703517 5 0.283126 4 0.703454 0.283 127 4 0.283133 0.703449 6 0.283135 0.703488 3 0.703562 0.283131 6 0.283132 4 0.703530 Kohala Volcano 0.283 133 4 0.283 133 5 0.283139 4 0.283136 4 0.283141 4 4 0.283116 0.283119 5 4 0.283119  ‘All Hf isotopic ratios were determined by MC-ICP-MS and have been normalized to JMC 475 HVmHf = 0.282160 [Vervoort and Blichert-Toft, 1999]. All Sr and Nd isotopic ratios were determined by TIMS; Sr data were nonnalized to SrItaSr = 0.7 10248; Nd data were normalized to the Berkeley Ames standard ‘ 7 Nd/Nd = 0.5 10939. The 2a 4 SRM 987 S error is the absolute error value of an individual sample analysis (internal error) and applies to the last decimal place(s). Additional analyses for the same sample represent procedural duplicates using separate powder aliquots. bWaawaa Trachyte Member  60  boiling mixture of strong HC1 and HNO 3 to remove non-magmatic Sr resulting from diagenetic processes. Sr was purified on chromatographic ion exchange columns using Eichrom Sr spec resin following the procedures of Horwitz et a!. [19911 and loaded in Ta 5 0 2 on rhenium filaments. Nd was purified using a three-column procedure following the methods outlined in Lw eta!. [1997] and DePaolo [1978] and loaded in 5 N HNO 3 on 6 mbar. Sr and Nd rhenium filaments. Nd was run as Nd0 with an oxygen pressure of--10 isotopic measurements were made on a VG Sector 54 multicollector thermal ionization mass spectrometer operating in dynamic mode. Analytical procedures were similar to those described in DePaolo and Daley [2000]. Sr and Nd isotopic ratios were normalized to Sr = 0.1194 and Nd/’ 88 Sr/ 86 46 = 0.636151, respectively, to correct for within-run ‘ Nd 42 fractionation. Nd data were further corrected for oxygen isotopes assuming that 0.0003 87 and 0.710281  ±  180/160  =  170/160  =  0.00211. The average 87 5r of the SRM 987 Sr standard was 86 5r/  0.0000 15 (2sd; n = 140) during the course of the post-shield analyses. Mean  values of the Berkeley Ames standard gave 0.5 10977 ± 0.000006 (2sd; n = 86) and BCR-1 gave Nd/’Nd 43 = 0.5 11876 ± 0.000005 (2sd; n = 10). All measured Sr and Nd isotopic ‘ Sr = 0.710248 and Berkeley Ames standard 86 Sr/ ratios were normalized to SRM 987 87 43 = 0.5 10939, respectively. ‘ Nd 44 Nd/’ Pb and Hf isotopic compositions were determined at the PCIGR, University of British Columbia and are reported in Tables 3.3 and 3.4. All of the sample powders were acid-leached to remove alteration phases prior to isotopic analysis, following the sequential leaching procedure of Weis eta!. [2005]. Leached sample powders were digested in a mixture of concentrated HF and HNO 3 in sealed Teflon® vessels and processed on chromatographic ion exchange columns. A thorough review of sample dissolution, isotopic  61  Table 3.4: Pb Isotopic Compositions of Hawaiian Post-shield Lavasa Sample  Volcanic Series  O2AHU-1 O2AHU-2 02AHU-3 O2AHU-4 O2AHU-5  Hualalai Hualalai Hualalai Hualalai Hualalai  O2AHU-6 O2AHU-7 O2AHU-8 O2AHU-9 O2AHU-10 O2AHLJ-11 O2AHU-12 O2AH(J-13  Hualalai Hualalai Hualalai” Hualalai Hualalai Hualalai Hualalai Hualalai  O2AMK-1 O2AIvIK-2 O2AMK-3 O2AMK-4 O2AMK-5 O2AMK-6 O2AMK-7 O2AMK-8 O2AMK-10 O2AMK-11  Laupahoehoe Hamakua Hamakua Laupahoehoe Hamakua Laupahoehoe Hamakua Laupahoehoe Laupahoehoe Laupahoehoe  O2AMK-12 O2AMK-13  Hamakua Hamakua  O2AKA-1 O2AKA-2 O2AKA-3 O2AKA-4  Hawi Hawi Hawi Hawi  O2AKA-5 O2AKA-6 O2AKA-7  Polulu Poluhs Polulu  ’Pb 20 Pb/ 207  2a  Pb 204 Pb/ 2tt  2o  Hualalai Volcano 37.6779 21 37.7063 27 37.7444 18 37.7532 27 37.7442 18 37.7470 16 37.7285 19 37.7134 14 37.7291 19 37.7392 36 37.7621 24 37.7483 16 37.7472 24 37.71 19 16 37.7092 18  15.4459 15.4471 15.4515 15.4479 15.4473 15.4483 15.4549 15.4505 15.4456 15.4449 15.4578 15.4524 15.4590 15.4500 15.4491  7 10 6 10 6 6 7 5 7 13 10 6 9 6 7  17.8913 17.9356 17.9690 17.9784 17.9802 17.9805 17.9572 17.9702 17.9578 17.9544 17.9992 18.0114 18.0059 17.9426 17.9417  9 12 8 11 6 5 7 6 8 17 10 7 11 6 7  Mauna Kea Volcano 40 37.9365 38.0072 15 37.9894 15 37.9259 13 37.9603 31 37.9328 18 37.9718 20 37.9310 13 37.9377 17 37.9244 29 37.9268 23 37.9608 18 37.9796 12  15,4743 15.4857 15.4806 15.4708 15.4787 15.4720 15.4771 15.4711 15.4741 15.4717 15.4717 15.4788 15.4784  7 5 6 5 12 7 7 4 6 10 8 6 6  18.3437 18.4210 18.4070 18.3431 18.3712 18.3495 18.3782 18.3500 18.3517 18,3411 18.3411 18.3496 18.4080  7 6 6 6 12 8 9 5 6 9 9 6 6  Kohala Volcano 38.0041 17 37.9908 41 37.9865 17 37.9949 13 37.9994 17 37.9045 21 37.8989 19 37.9233 14  15.4843 15.4823 15.4780 15.4808 15.4824 15.4762 15,4728 15,4763  5 15 7 5 6 8 8 5  18,4393 18.4258 18.4378 18.4394 18.4410 18.2863 18.2482 18.3134  7 18 7 6 6 8 8 6  Pb 204 Pb/ 2tt  2a  Al1 Pb isotopic ratios were deternined by MC-ICP-MS on acid-leached sample powders; Pb data were corrected for fractionation by Ti spiking and then nomalized to the NBS 981 triple spike values of Galer andAbouchami [1998] using the standard bracketing method. The 2a error is the absolute error value of an individual sample analysis (internal error) and applies to the last decimal place(s). Additional analyses for the same sample represent procedural duplicates using separate powder aliquots. Waawaa Trachyte Member 5  62  likely related to accumulated apatite, whereas the low values in the bemnoreite and trachyte reflect apatite fractionation. There are systematic variations in incompatible trace elements with Nb concentration, which ranges between 15-70 ppm for most of the post-shield lavas (Figure 3.4). Abundances of incompatible trace elements are generally well-correlated with Nb, although some subtle differences exist between the transitional/alkalic basalts and alkalic lavas as well as between lavas from each volcano. For example, the Ba and Rb abundances of basalts from Hualalai define a separate, steeper trend than the other post-shield lavas. The alkalic lavas from Kohala are also distinct, being shifted to slightly lower Zr and higher La and Ce concentrations. The elevated Y and Sr concentrations (--60 ppm Y; --1 800 ppm Sr) in the three of these lavas from Kohala are consistent with accumulated apatite. The trachyte from Hualalai (140 ppm Nb) has anomalous trace element contents, including extremely low Sr (37 ppm) and Ba (281 ppm) concentrations that are indicative of extensive feldspar (+ apatite) fractionation. Chondrite-normalized rare earth element (REE) patterns for the post-shield lavas are smooth with relatively steep slopes (LaNIYbN = 6.0-16.2), reflecting enrichment in the light REE (Figure 3.5). The overall patterns are similar, but show variations in the REE concentrations between samples from each volcano. The alkalic lavas from Kohala have the highest REE abundances (e.g., La =215-340 times chondrites), and the transitional/alkalic basalts from Mauna Kea and Hualalai have the lowest abundances (e.g., La  50-95 times  chondrites). The REE pattern of the trachyte from Hualalai is distinctive, showing enrichment in the light and heavy REE relative to the middle REE, and a small negative Eu anomaly.  63  3.4 RESULTS 3.4.1 Major and trace elements The post-shield lavas from Hualalai, Mauna Kea, and Kohala range widely in composition from basalt to trachyte (Figure 3.2). At Mauna Kea, the earlier substage of post-shield volcanism (Hamakua Volcanics) is characterized by transitional to alkalic basalts, whereas the later substage (Laupahoehoe Volcanics) contains hawaiites and mugearites. At Kohala, the late-shield lavas (Polulu Volcanics) are dominated by transitional to alkalic basalts and the post-shield lavas (Hawi Volcanics) have compositions ranging from hawaiite to benmoreite. In contrast, at Hualalai there is a large compositional gap between the alkalic basalts and trachytes of the post-shield stage (Hualalai Volcanics). 0/P 1, indicating that they have not undergone 2 K > 5 For most samples (31 of 32) 0 significant post-eruptive alteration, which results in a preferential loss of K relative to P [e.g., Feigenson eta!., 1983; Chen and Frey, 1985; Frey eta!., 1994]. The major element variations of the post-shield lavas can be divided into two groups, transitionallalkalic basalts and alkalic lavas (hawaiite to trachyte). The MgO contents of the post-shield lavas range widely from —O.5-17 wt% (Figure 3.3). For lavas with >5 wt% MgO (transitional/alkalic basalts), Ni is strongly correlated with MgO, which is consistent with the combined effects of olivine fractionation and olivine accumulation (lavas with MgO> 10 wt% have abundant olivine phenocrysts). For lavas with <5 wt% MgO (alkalic lavas), CaO and Sc abundances decrease strongly with decreasing MgO, indicating an important 2 also decrease below 5 3 and Ti0 0 2 role for dilnopyroxene fractionation. Abundances of Fe wt% MgO in the alkalic lavas, which signals the effect of fractionation of Fe-Ti oxides. The anomalously higher P 5 contents (1.6-2.2 wt%) of three alkalic lavas from Kohala are 0 2  64  14 12  08 C.’  +6  0 Z4 2 0 40  45  50  55  60  65  70  (wt%) 2 Si0 Figure 3.2: Total alkalis vs. silica classification diagram (modified from Le Bas et al.  [1986]) showing the range in composition of the post-shield lavas from basalt to trachyte. The tholeiitic-alkalic dividing line is from Macdonald and Katsura [1964]. The shaded field represents the compositions of shield tholeiites from Hualalai, Mauna Kea, and Kohala. The majority of the post-shield lavas in this study are alkalic basalts and hawaiites.  65  65  20  Maunakea A Laupahoetioe A Hamakua Kohala  OZAHV-8  60  18  •A  Haw  Aoac-4  Apojulu Hualalai 0 Hualalai  55  16  • WawTrohyto  0:14  0  A  t’.to  50  12  A  0 A 45  0 A  10  16  I  I  I  I  I  I  I  I  I  I  I  14  I  I  I  I  I  I  I  I  12  jA(%8cA  12  I  I  14  A  A  oc Acp O 3  10  10 68  4  0  A  0  C-)  .  4  A  2  2 4.0 3.5  I  2.5  ‘‘‘‘A’’’’’’’’  I  I  I  I  I  I  I  I  I  I  I  I  A  2.0 3.0  A  2.5  /0  1.5  A  620  A  0  1.5  A  A  1.0  1.0  0.5  0.5  A0 I  700  I  I  I  I  I  I  I  I  I  I  I  I  I  111111  35  I  0  A  A  500  25  A 400  E  &  300 200  0  •  0  15 10  A  100  20  A  0 0 A AQ’O  30  600  11111111111  5  A  A  0  0 0  2  4  6  8  10  12  14  16  18  0  2  4  6  8  10  12  14  16  18  MgO (wt%)  MgO (wt°Io)  Figure 3.3: MgO variation diagrams of selected major element oxides and compatible trace  elements for the Hawaiian post-shield lavas. The transitional/alkalic basalts and alkalic lavas exhibit marked differences in their major element chemistry. The anomalously high O contents of three lavas from the Kohala Hawi Volcanics likely reflect the effects of 2 P apatite accumulation.  66  likely related to accumulated apatite, whereas the low values in the benmoreite and trachyte reflect apatite fractionation. There are systematic variations in incompatible trace elements with Nb concentration, which ranges between 15-70 ppm for most of the post-shield lavas (Figure 3.4). Abundances of incompatible trace elements are generally well-correlated with Nb, although some subtle differences exist between the transitional/alkalic basalts and alkalic lavas as well as between lavas from each volcano. For example, the Ba and Rb abundances of basalts from Hualalai define a separate, steeper trend than the other post-shield lavas. The alkalic lavas from Kohala are also distinct, being shifted to slightly lower Zr and higher La and Ce concentrations. The elevated Y and Sr concentrations (-60 ppm Y; —1800 ppm Sr) in the three of these lavas from Kohala are consistent with accumulated apatite. The trachyte from Hualalai (140 ppm Nb) has anomalous trace element contents, including extremely low Sr (37 ppm) and Ba (281 ppm) concentrations that are indicative of extensive feldspar (+ apatite) fractionation. Chondrite-normalized rare earth element (REE) patterns for the post-shield lavas are smooth with relatively steep slopes (LaN/YbN = 6.0-16.2), reflecting enrichment in the light REE (Figure 3.5). The overall patterns are similar, but show variations in the REE concentrations between samples from each volcano. The alkalic lavas from Kohala have the highest REE abundances (e.g., La =215-340 times chondrites), and the transitional/alkalic basalts from Mauna Kea and Hualalai have the lowest abundances (e.g., La = 50-95 times chondrites). The REE pattern of the trachyte from Hualalai is distinctive, showing enrichment in the light and heavy REE relative to the middle REE, and a small negative Eu anomaly.  67  1200 1000 800  80  Mauna Kea A Laupahoehoe Hamakua Kohala Hawi A Polulu Hualaiai 0 Hualalai  OZHU-8  60  • WTrohyt  i 600  A  >-  A  40  400 20 200  0  1000  liii  2400  A 2000  800  I  600  400  1600  A  1200  CA  800  200  400  140  111111  10  if  iii  120  A  8 100 80  6  A  I-  60  4  oAf  40 2  F  20 100  liii  liii  ii  Iii  200  A  80  11111  A  150  iii  liii  A  A  A  60  iii  A.  -I 40  A  y  100  At ,0  50 20  0  0 0  20  40  60  80  100  120  140  160  0  20  40  60  80  100  120  140  160  Nb  Nb  Figure 3.4: Nb variation diagrams of selected trace elements (in ppm) for the Hawaiian  post-shield lavas. Positive correlations with Nb are observed, as well as subtle variations in lavas from Hualalai and Kohala. The three Kohala lavas with elevated Y and Sr concentrations have accumulated apatite.  68  1000  La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu  100  [o Mauna Kea  A Laupahoehoe Volcanics Hamakua Volcanics 1000  I  I  I  I  I  I  I  I  I  I  I  I  100  [0  1000  Kohala Hawi Volcanics APolulu Volcanics I  I  I  I  I  I  I  I  I  I  I  0  100  10 Hualalal 0 Hualalai Volcanics 1  • Vawaa Trachyte Member  La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu  Figure 3.5: Chondrite-normalized rare earth element abundances of the Hawaiian post-shield lavas. Normalizing values from McDonough and Sun [1995]. The shaded field in each panel encompasses the range of values from all three volcanoes (excluding the trachyte from Hualalai).  69  Primitive mantle-normalized patterns for the post-shield lavas show a prominent negative Pb anomaly, as well as enrichment in the light REE and high field strength elements, but with a distinctive convex-down shape defmed by lower U, Th, and especially Cs (Figure 3.6). The positive Rb anomaly and the negative Ba, Sr, and Eu anomalies in the trachyte from Hualalai reflect the effects of extensive feldspar (+ apatite) fractionation. The trachyte also has elevated Zr and Hf abundances signalling the presence of accumulated zircon. At Mauna Kea, the alkalic lavas from the late post-shield sub-stage (Laupahoehoe Volcanics) have higher trace element abundances than basalts from the early post-shield sub-stage (Hamakua Volcanics). Similarly, at Kohala, the post-shield alkalic lavas (Hawi Volcanics) are more enriched than the late-shield basalts (Polulu Volcanics). Collectively, the post-shield lavas are significantly more enriched in incompatible trace elements relative to typical shield-stage tholeiites (Figure 3.6), and demonstrate a systematic trend of trace element enrichment as each volcano evolves though the shield and post-shield stages.  3.4.2 Sr-Nd-Hf-Pb isotopic compositions The majority of the post-shield lavas are characterized by a relatively limited range 87 ( Sr 86 of Sr, Nd, and Hf isotopic compositions Sr/  43 ‘ 0.7O345-0.7O365; Nd/”Nd  -O.51292-O.513O2; 77 H17’ 7 ‘ Hf 6 0.2831 1-0.283 14) (Figure 3.7). In contrast, the Pb isotopic compositions of the post-shield lavas range to a greater extent (e.g., 206 Pb 204 Pb/  =  ‘-17.9-18.45) and show a clear separation between the Kea-trend (Mauna Kea and Kohala) and Loa-trend (Hualalai) volcanoes. The transitional/alkalic basalts and alkalic lavas from Mauna Kea and Kohala, which are clearly distinguished by their major and trace element chemistry, also exhibit differences in their Sr and Pb isotopic compositions (Figure 3.7).  70  1000  Cs Rb Ba Tb U Nb Ta La Ce Pb Pr Nd Sr Zr Hf Sm Eu Gd Tb Dy Ho Y Er Yb Lu Ga Mauna Kea ALaupahoehoe Volcanics AHamakua Volcanics  100 2  E 0.  .10 E C’) tholeute  1000  I  I  I  I  Kohala AHawi Volcanics APolulu Volcanics 100 2  0.  10  0.  E  C’)  shield tholeiite  1000  I  I  I  I  I  I  I  I  I  I  I  I  I  I  I  I  Hualatai 0 Hualalai Volcanics • Waawaa Trachyte Member  100 2  0. 0.  10  E  Cl)  shield tholeilte  Cs Rb Ba lb U Nb Ta La Ce Pb Pr Nd Sr Zr Hf Sm Eu Gd Tb Dy Ho Y Er Yb Lu Ga  Figure 3.6: Primitive mantle-normalized incompatible trace element abundances of the Hawaiian post-shield lavas. Element compatibility increases from left to right and the normalizing values are from McDonough and Sun [19951. The post-shield lavas are more enriched in incompatible trace elements compared to representative shield-stage tholeiitic basalts from each volcano: Mauna Kea tholeiite, Huang and Frey [2005]; Kohala and Hualalai tholeiites, this study.  71  0.5131  0.28320  AAA w 4 A  0.5130  ii.  0.28315  A  0.5129  0.28310  I1  120  20  0.5128 0.7034  0.7035  0.7036  0.7038  0.7037  0.28305 17.8  17.9  18.0  Sr 6 SrP 87 15.50  I  I  I  18.2  18.3  18.4  18.5  38.1  I  18.3  18.4  18.5  MaunaKea A Laupahoehoe Ramakua  15.49  A  38.0  Kohala  37.9  iai  Hawi  15.48 15.47 .  18.1  Pb 204 Pb1 206  15.46  A  37.8  15.45 37.7 15.44 IrIlIrlIlill 15.43  17.8  17.9  18.0  18.1  18.2  18.3  18.4  18.5  37.6 17.8  17.9  18.0  18.1  18.2  Pb! 6 ° 2 Pb 204  Pb 204 Pb/ 206  Figure 3.7: Isotopic (Sr-Nd-Pb-Hf) co-variation diagrams of the Hawaiian post-shield lavas. Representative 2a error bars for Sr, Nd, and Hf are indicated; for Pb, the error bars are smaller than the symbol sizes. The samples are characterized by a limited range of Sr, Nd, and Hf isotopic compositions, but vary widely with respect to Pb. The post-shield lavas from Hualalai have distinctly less radiogenic Pb isotope ratios than those from Mauna Kea and Kohala. Despite overlapping Nd and Hf isotopic compositions, the younger lavas from Mauna Kea and Kohala can be distinguished from older lavas from each volcano by their Sr and Pb isotopic compositions.  72  The trachyte from Hualalai has higher Sr and slightly lower Nd and Hf isotopic 87 = 0.70371; Nd/’Nd ( Sr 86 43 = 0.5 1290; Hf7’ ‘ 76 0.28309) than the ‘ Hf 77 compositions 5r/ other post-shield lavas, but does not have a distinctive Pb isotopic composition. The youngest sample in this study (710 years; Waha Pele, Hualalai) has the highest Hf and lowest Pb isotopic compositions of the post-shield lavas. 5r, 86 5r/ The post-shield lavas as a group lie towards the more depleted (i.e., low 87 76 end of the Hawaiian array in Nd-Sr and Hf-Nd diagrams ‘ Hf) 77 high 143 Nd, high Hf” 44 Nd/’ (Figure 3.8). Post-shield lavas from Mauna Kea and Kohala have Sr, Nd, and Hf isotopic compositions that are consistent with shield stage lavas from Mauna Kea, Kilauea, and West Sr and 86 Sr/ Maui. In contrast, post-shield lavas from Hualalai are shifted to slightly lower 87 higher Hf” 76 compared to other Loa-trend volcanoes (e.g., Mauna Loa), having values ‘ Hf 77 more similar to Kea-trend volcanoes and plotting above the  Nd 6 £—  Hawaiian array.  Post-shield lavas from Hualalai have some of the least radiogenic Pb isotopic Pb = 17.89) observed in recent (<5 Ma) 204 Pb/ compositions Pb/ 208 = 37.67; 206 ( Pb 204 Hawaiian lavas and do not overlap with any Loa- or Kea- trend volcano in Hf-Pb and Sr-Pb diagrams (Figures 3.9 and 3.10). Shield and late-shield lavas from Kohala have the least radiogenic Pb isotopic compositions of any Kea-trend volcano, whereas the post-shield lavas Pb values that are more characteristically Kea-like, consistent with 204 Pb/ have higher 206 shield lavas from Mauna Kea, Kilauea, and West Maui. Post-shield lavas from Mauna Kea 5r than most shield lavas from Kea-trend 86 Sr/ Pb and 87 204 Pb/ extend to less radiogenic 206 volcanoes, similar to the trend observed in lavas from West Maui.  73  0.5132  I  I  (a) EPR MORB  0.5131  +  I  I  rejuvenated  West aui .  0.5130 Mauria Loa  0.5129 Kahoolawe 0.5128  •. . This study: AMaunakea LKohaIa QHuaIaIai  0.5127  0.5126 0.7024  Lanai  I  I  0.7028  0.7032  I  I  0.7036  0.7040  0.7044  0.7048  Sr 86 SrI 87 16 (b) ” 1 MaunaKe  14 12  A  Mauna Loa Kahoolawe •  10  West Maui  _O  ‘4-  8 6 13  4  12  o O2AHU-8 2 0 -2  / -1  0  11  o  ENd 5  1  2  4  3  5  6  6  7  7  8  8  9  ENd Figure 3.8  74  10  Figure 3.8: (a) Nd/’ 43 vs. 87 ‘ Nd 44 Sr and (b) EHf vs. ENd for the post-shield lavas from 86 SrI Mauna Kea, Kohala, and Hualalai compared to selected Hawaiian shield stage lavas. Loa-trend volcanoes are represented by circles and Kea-trend volcanoes are represented by triangles. The 2a error bars are smaller than the symbol sizes. Also shown in (a) are fields for Hawaiian rejuvenated stage lavas (Haleakala, West Maui, East Molokai, Oahu, Kauai, North Arch) and East Pacific Rise mid-ocean ridge basalts (EPR MORB). The Hawaiian and OIB arrays in (b) are from Blichert-Toft et a!. [1999]. Data sources: Lothi Blichert-Toft eta!. [1999], Abouchami eta!. [2005]; Mauna Loa B!ichert-Toft eta!. [2003], D. Weis, unpublished data; Hualalai D. Weis, unpublished data; Kahoolawe B!ichert-Toft eta!. [1999], Abouchami et a!. [2005], Huang eta!. [2005]; Lanai Abouchami eta!. [2005], Gaffiwy et a!. [2005]; Koolau B!ichert-Toft eta!. [1999], Abouchami et a!, [2005]; Kilauea Blichert-Toft et a!. [1999], Abouchami et at. [20051; Mauna Kea B!ichert-Toft et a!. [2003], Eise!e et a!. [2003], Bryce eta!. [2005]; Kohala Feigenson eta!., 1983, Abouchami et a!. [2005], M. Garcia, unpublished data, D. Weis, unpublished data; Haleakala West and Leeman [1987], Chen eta!. [1990], Chen et al. [1991], Blichert-Toft eta!. [1999]; West Maui Gaffiiey eta!. [2004]; East Molokai -Xu et a!. [2005]; Oahu Reiners and Nelson [1998]; Kauai Reiners and Nelson [1998], Lassiter et a!. [2000]; North Arch Frey eta!. [2000]; EPR MORB Niu eta!. [1999], Rege!ous eta!. [1999], Castillo eta!. [2000]. -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  75  16  rejuvenated  (a)  L+ 4..+  West Maui  14  Kea  12 !L •  10 C8 .  • S  Kahoj.  6 Lanai  Koolau 4  2  This Study: AMaUnaKea Koha 0 Huaga  KooIau  I  0 17.6  18.0  17.8  18.4  18.2  18.8  18.6  19.0  19.2  Pb/204pb 6 O 2 0.7048  (b)  Lanai  • 0.7044  * • • •  • 4 Kah ••  MaurlaLoa  V  0.7040  Mauna Kea  0.7036 +  0. 7032  West Maui  + + rejuvenated  4+-  +  uppe re 2 Ga ni OCea CS •Pacifjc  er  a iow 2 G ,flICC  •  Paciflc midgower Crust  • •  fresh  EPRMORB 0.7024 17.7  17.8  17.9  18.0  18.1  18.2  18.3  18.4  pb 2046 b10 2 P  18.5  18.6  18.7  Figure 39  76  18.8  Figure 3.9: (a) vs. 206 Pb and (b) 87 204 Pb/ Sr vs. 206 86 Sr/ Pb for the post-shield lavas 204 Pb/ from Mauna Kea, Kohala, and Hualalai compared to Hawaiian shield and rejuvenated stage lavas. The hypothetical Kea and Koolau end members in (a) are from Blichert-Tofi et a!. [1999]. In (b) the field for EPR MORB and isotopic compositions of the Pacific mid-lower crust (gabbroic xenoliths, Hualalai: Lassiter and Hauri [1998]) and Pacific upper crust (fresh to altered basalts, ODP Site 843: King et a!. [1993]) are shown for comparison. For the Pacific crust, unleached and measured isotopic ratios were used. The blue line in (b) represents a mixing line between the 2 Ga lower and upper oceanic crust end-members from Gaffney et a!. [2004]. Symbols and data sources are the same as in Figure 3.8.  77  38.2Zi7Iauea  .. 4A 38.1  Mauna Kea •A  MaunaLoa  .•  L 38.0  Hualalai Lanai Kahooawe  .  • •  37.9 Koolau  Maui  • •  Kohala  37.8 •  37.7  LOA  This study: A Mauna Kea LKohaIa Hualalai  + rejuvenated  +  KEA  37.6 17.7 17.8 17.9 18.0 18.1  18.2 18.3 18.4 18.5 18.6 18.7 18.8  Pb 204 Pb1 206 38.2  (b)  I ‘  I  I  38.1  38.0  +  +*±++  37.9  ++ EPRMORB 37.8  +  ++  37.7 +  * 37.6 17.8  I  17.9  +  I  18.0  18.1  +  I  18.2  18.3  2O6  Pb 204 bI  I  18.4  I  I  18.5  18.6  18.7  Figure 3.10  78  Figure 3.10: 208 Pb vs. 206 204 Pb/ Pb for the post-shield lavas from Mauna Kea, Kohala, 204 Pb/ and Hualalai compared to Hawaiian shield and rejuvenated stage lavas. The thick grey line in (a) represents the Loa-Kea Pb isotopic division defined by Abouchami et aL [20051. In (b) the isotopic compositions of EPR MORB and shield stage lavas from Mauna Kea, Kohala, and Hualalai are shown for comparison. For Mauna Kea, only shield stage samples from to the “1o8” array of HSDP-2 [Eisele et a!., 2003] are included. The thin lines in (b) are regression lines through shield and post-shield lavas from the same volcano. Symbols and data sources are the same as in Figure 3.8.  79  3.5 DISCUSSION 3.5.1 Sr-Nd-Pb-Hf isotope systematics of Hawaiian lavas The isotopic variability observed within individual Hawaiian volcanoes and between volcanoes on the scale of the entire Hawaiian chain is used to infer the compositional characteristics of their mantle source region. Specifically, the change in the isotopic compositions of lavas during the critical transition from the shield to post-shield stage, and the relationship of the post-shield lavas to the inferred isotopic mixing trends defined by shield lavas, provides constraints on the geochemical structure of the plume and the sources of Hawaiian lavas. Pb isotopes in particular are one of the most powerful isotopic tracers and have a demonstrated ability in identifying heterogeneities and source components associated with the Hawaiian plume [e.g., Eisele et al., 2003; Abouchami et a!., 2005; Fekiacova et a!., 2007; Marske et a!., 2007].  3.5.1.1 Shield to post-shield transition Lavas erupted during different growth stages (i.e., pre-shield, shield, post-shield, rejuvenated) of a Hawaiian volcano have distinct geochemical characteristics. Post-shield alkalic capping lavas at Haleakala volcano have higher incompatible element abundance ratios (e.g., Ba/La, Nb/La, La/Yb) than underlying shield tholeiites, as well as higher 43 and lower 87 ‘ Nd 44 Nd/’ Sr [Chen and Frey, 1985]. Geochemical changes associated 86 Sr/ with the transition to the post-shield stage have also been documented at other Hawaiian volcanoes, including East and West Molokai [Xu et a!., 2005; Xu et a!., 2007] and Kauai [Swinnard et a!., in preparation, 2008].  80  At Mauna Kea, Hualalai, and Kohala, the enriched trace element characteristics of post-shield lavas compared to their respective shield lavas (Section 3.4.1) are accompanied Sr as well as by marked changes in Pb isotopic compositions 86 Sr/ by slightly lower 87 (Figures 3.9b and 3.11). Lavas from Mauna Kea show a systematic shift from their most radiogenic Pb isotope ratios during the shield stage to less radiogenic values during the early 208 = ( Pb 204 basaltic post-shield substage to the lowest values yet observed at Mauna Kea Pb/ 3793; 206 Pb 204 Pb/  18.34) during the late hawaiitic post-shield substage. Similarly, post-  shield lavas from Hualalai are significantly less radiogenic in Pb than the shield tholeiites. In contrast, at Kohala, late-shield lavas are more radiogenic in Pb than most shield tholeiites, and the post-shield lavas have distinctly more radiogenic Pb isotopic compositions. Shield and post-shield lavas from the same volcano form linear correlations in Pb-Pb isotope diagrams (Figure 3. lOb). Given that these linear arrays are best interpreted as binary mixing lines (rather than isochrons) [e.g., Abouchami et a!., 2000; Eisele et a!., 2003; Abouchami et a!., 2005], this indicates that shield and post-shield lavas may be derived from common source components mixed in variable proportions. Therefore, with respect to Pb, a greater contribution from a less radiogenic end-member is required during the post-shield stage at Mauna Kea and Hualalai, whereas at Kohala, a more radiogenic end-member dominates the post-shield stage (Figure 3.11). However, the Pb-Pb arrays of Mauna Kea Pb 204 Pb/ and Hualalai are sub-parallel and suggest that two separate unradiogenic (lower 208 and 206 Pb) end-members are required. The temporal Pb isotope variations, combined 204 Pb/ with the distinct Sr, Nd, and Hf isotopic compositions of the post-shield lavas, correspond to a systematic change in the proportions of discrete components sampled by each volcano during the post-shield stage.  81  38.2  I  younger transitional to weakly alkalic  older tholeiltic  -  .  I  I  Mahukona (LOA)  38.1  38.0  I  (a)  •/  post-shield  —  c-I  37.9  late-shield  -  o  shield  (‘I  Kohala (KEA)  A  37.8  37.7  I  I  I  18.0  18.1  18.2  37.6  17.8  17.9  I  I  I  18.3  18.4  18.5  18.6  18.7  Pb 204 Pb1 206 38.2  I  (b)  I  I  I  I  I  A/A  38.1 shield  38.0 shield early post-shield  Hualalai (LOA)  0  37.9  late post-shield  0 0  Mauna Kea (KEA)  37.8 post shie  37.7  37.6 17.8  0 17.9  I  I  18.0  18.1  I  I  18.2  18.3  2OG  Pb 204 b1  18.4  18.5  18.6  18.7  Figure 3.11  82  Figure 3.11: 208 Pb vs. 206 204 Pb/ Pb for shield and post-shield lavas from consecutive 4 PbP° pairs of Hawaiian volcanoes: (a) Mahukona and Kohala and (b) Hualalai and Mauna Kea. For Mauna Kea, only shield stage samples from to the “1o8” array of HSDP-2 [Eisele et al., 2003] are included. The grey arrows show the temporal geochemical trend from the shield to post-shield stage and point towards a hypothetical end-member. Symbols and data sources are the same as in Figure 3.8, except for Mahukona [Appendix B 1; Garcia et al., in preparation, 2008].  83  3.5.1.2 Mixing relationships The compositional variability in Hawaiian shield stage lavas requires the existence of at least three isotopically distinct components, referred to as Loihi, Koolau, and Kea [e.g., Staudigeletal., 1984; Stille eta!., 1986; West et at., 1987; Eileretal., 1996; Hauri, 1996]. However, principal component analysis of Sr, Nd, Pb, He, and 0 isotope ratios in Hawaiian shield lavas indicates that almost 90% of the variation can be explained by mixing between the Koolau and Kea end-members [Eiler et al., 1996]. The Koolau end-member has the highest 87 Sr, and lowest ‘ 86 Sr/ NdJ’Nd, 4 76 and Pb isotope ratios and is best ‘ Hf, 77 3 Hf7’ expressed in lavas from Koolau and other Loa-trend volcanoes. The Koolau source is thought to contain recycled oceanic crust and pelagic sediments [e.g., Lassiter and Hauri, 1998; Blichert-Tofi et a!., 1999]. In contrast, the Kea end-member is characterized by the 43 Hf” ‘ lowest 87 Sr, and highest NdJ’Nd, 86 Sr/ 76 and Pb isotope ratios and is best ‘ Hf, 77 expressed in lavas from Mauna Kea and other Kea-trend volcanoes. The origin of the Kea component is not as well characterized, and has been variously interpreted to be assimilated Pacific Ocean lithosphere, entrained asthenosphere (i.e., MORB-source mantle), or recycled oceanic crust [e.g., Eiler eta!., 1996; Lassiter eta!., 1996; Lassiter and Hauri, 1998]. An important question is whether mixing between the Koolau and Kea end-members can produce the range of isotopic compositions observed in post-shield lavas from Mauna Kea, Hualalai, and Kohala. 206 diagrams (Figure 3.9), Hawaiian shield — Pb 204 In E Pb/ 206 and 87 — Pb 204 Sr Pb/ 86 SrI lavas form strongly concave hyperbolic mixing trends between the Kea and Koolau endmembers, which are thus inferred to have very different Hf/Pb and Sr/Pb abundance ratios [Blichert-Tofi et a!., 1999; Huang eta!., 2005]. Importantly, both late- and post-shield lavas  84  from Kohala lie within the Hawaiian shield mixing hyperbola. In contrast, post-shield lavas from Hualalai are displaced to higher Hf/ 76 and lower 87 ‘ Hf 177 Sr and 206 86 Sr/ Pb than 204 Pb/ expected from mixing between the Koolau and Kea end-members, and do not lie within the mixing hyperbola. Post-shield lavas from Mauna Kea deviate slightly from the Sr-Pb hyperbola, showing a weakly positive correlation that is oblique to the shield trend and extends towards less radiogenic Sr and Pb isotopic compositions (Figure 3 .9b). The isotopic compositions of post-shield lavas from Hualalai and Mauna Kea therefore cannot be explained by binary mixing between the Koolau and Kea end-members. Specifically, the unradiogenic Pb isotopic compositions of post-shield Hualalai lavas and the positive Sr-Pb correlation defined by post-shield Mauna Kea lavas preclude their generation from these two end-members. In addition, the Nd and Hf isotopic compositions of the postshield lavas define a separate trend with a shallower slope than either the OIB or Hawaiian arrays (Figure 3.8b), indicating a source with higher Hf” 76 for a given ‘ ‘ Hf 77 NdJ’Nd. 4 3 The observed isotope systematics require a contribution from at least one additional source component with relatively unradiogenic 87 Sr and 206 86 Sr/ Pb (and possibly also more 204 Pb/ radiogenic 177 H17 7 ‘ Hf) 6 that is not expressed in Hawaiian shield stage lavas.  3.5.2 Constraints on the origin and extent of depleted components in Hawaiian lavas Most ocean island basalts are characterized by enriched geochemical signatures, but some also commonly contain trace element abundances and isotope ratios with compositional similarities to MORB (i.e., La/SmN  <  Hf, and 77 Hf” 1; high 144 Nd/ 4 ‘ Nd 3 and 176  low 87 Sr compared to primitive mantle). These depleted signatures have been observed 86 Sr/ in a number of hotspot settings, including Iceland, Kerguelen, Galapagos, and Hawaii [e.g.,  85  Blichert-Tofi and White, 2001; Doucet et a!., 2002; Fitton et a!., 2003; Mukhopadhyay et a!., 2003; Frey et a!., 2005]. In Hawaii, however, lavas with depleted isotope ratios are associated with enriched rather than depleted trace element abundances. In particular, the combination of low 87 Sr and high Nd!’ 86 Sr/ 43 with high Rb/Sr and low Sm/Nd is a ‘ Nd 44 paradox that remains largely unresolved.  3.5.2.1 The role of oceanic lithosphere: Ambient Pacific crust or recycled plume component? The origin of depleted components in Hawaiian lavas and at other ocean islands is the subject of continued debate. The depleted isotopic compositions are commonly attributed to assimilation of the underlying oceanic crust or entrained shallow asthenosphere (i.e., MORB-source mantle) [e.g., Chen and Frey, 1985; Gaffney eta!., 2004]. To explain the enrichment in trace elements, these models generally invoke a source recently enriched (i.e., metasomatized) in incompatible trace elements [e.g., Yang et a!., 2003; Shafer et a!., 2005] or very small degree melts (0.1-1%) of a depleted source [e.g., Chen and Frey, 1985]. In contrast, other studies argue that the depleted signatures are derived from within the plume itself or from entrained mantle material [e.g., Frey et a!., 2005; Fekiacova et a!., 2007]. In these models, the depleted material is ancient oceanic lithosphere (and sediments) that have been subducted, stored at the base of the mantle, and recycled into the upwelling plume. We investigate whether any of these sources could serve as a suitable end-member Sr, low86 Sr/ 87 Pb compositions observed in post-shield lavas from 204 Pb/ 206 for the lowHualalai and Mauna Kea.  86  Basalts from ODP Site 843, located 320 km west of the island of Hawaii, are representative of the ambient 110 Ma Pacific upper oceanic crust [King et at., 1993]. Gabbroic xenoliths from the 1800-180 1 Kaupuhuehue flow on Hualalai are inferred to represent the Pacific mid-lower oceanic crust [Lassiter and Hauri, 1998]. Fresh Pacific upper crust, Pacific mid-lower crust, and EPR MORB have the low 87 5r required of an 86 Sr/ end-member (Figure 3.9b). However, the 206 Pb of these sources, especially the Pacific 204 Pb/ upper crust, is too radiogenic to be a plausible depleted end-member for the Hualalai postshield lavas. Linear Pb-Pb trendlines from Hualalai and Mauna Kea are parallel to and do not intersect the EPR MORB field (Figure 3.lOb), providing further evidence against assimilation of MORB-related lithosphere or asthenosphere. The depleted component sampled by the post-shield lavas is distinct from the depleted MORB source and must therefore be closely related to the Hawaiian plume, either as material within the plume core, or as mantle material that is thermally accreted to the sides of the plume during ascent. Suitable candidates for this depleted source material include ancient recycled oceanic lithosphere and sediments. Preferential removal of the mobile elements U and Rb during subduction dehydration may lead to low U/Pb and Rb/Sr and the development of unradiogenic 206 Pb and 87 204 Pb/ Sr [McCulloch and Gamble, 86 Sr/ 1991; Weaver, 1991]. Forward modelling of the composition of 2 Ga oceanic lithosphere shows that gabbroic lower oceanic crust evolves to unradiogenic 206 Pb and 87 204 Pb/ Sr 86 Sr/ over time [Gaffney et at., 2004], and could thus account for the Sr and Pb isotopic compositions of the post-shield lavas (Figure 3.9b). Similarly, pelagic sediments develop unradiogenic 206 Pb because of long-term evolution with low U/Pb. The presence of 204 Pb/ pelagic sediments, which have high Lu/Hf [e.g., Patchett et al., 1984; Vervoort et at., 1999],  87  could also explain the Hf-Nd isotope systematics of the post-shield lavas (Figure 3 .8b). These observations are consistent with the recycling model and provide evidence for ancient lower oceanic crust and sediments in the mantle source of the post-shield lavas.  3.5.2.2 Identification of depleted signatures at other Hawaiian volcanoes The extent to which depleted components are sampled varies considerably between different volcanoes and is mainly restricted to lavas erupted after the volcano has migrated away from the plume center following the main shield-building stage. The low degrees of partial melting from which such lavas are derived correspond to small volumes within the melting region and may allow for these isotopic signatures to be more readily identified. The trend to low 87 Pb observed in post-shield lavas from Mauna 204 Pb/ Sr and low 206 86 Sr/ Kea has also been found for late-shield and post-shield lavas from West Maui [Gaffizey et a!., 2004], East Molokai [Xii eta!., 2005] and Haleakala [Chen eta!., 1991]. The presence of this isotopic signature at four consecutive Kea-trend volcanoes indicates that this depleted component is a persistent feature of the Kea source [Xii eta!., 2005]. The isotopic compositions of post-shield lavas from Hualalai represent the first example of a low Sr, low86 SrI 87 Pb signature at a Loa-trend volcano. This demonstrates that depleted 204 Pb/ 206 components are also present in the Loa source and are therefore more widely distributed within the plume than previously thought. Rejuvenated stage lavas from East Molokai, Oahu, Kauai, and the North Arch 3 and Hf7’ Volcanic Field, which have the highest ‘ Nd!’Nd 4 176 and lowest 87 Hf 77 Sr of 86 Sr/  all Hawaiian lavas, have also sampled a depleted component [e.g., Frey et a!., 2000; Lassiter et al., 2000; Xu eta!., 2005; Fekiacova eta!., 2007; Swinnard et al., in preparation, 2008].  88  However, rejuvenated stage lavas are characterized by a relatively limited range of isotopic compositions and plot almost exclusively on the Kea side of the Pb isotopic boundary [e.g., Fekiacova et a!., 2007] (Figure 3.1 Oa). Late-shield lavas from West Maui were interpreted to have a distinct source from Hawaiian rejuvenated lavas [Gaffrzey et a!., 20041, whereas at East Molokai, late-shieldlpost-shield and rejuvenated lavas were found to contain variable amounts of the same depleted component [Xu eta!., 2005]. Post-shield lavas from Mauna Kea trend towards the field for rejuvenated lavas and may thus share a common depleted component. Given that rejuvenated lavas erupt considerably later (0.25-2 Myr) than postshield lavas and at a distance significantly further from the plume centre, this depleted component is inferred to be a relatively long-lived feature of the plume. In contrast, postshield lavas from Hualalai are Loa-like and have Pb isotopic compositions that are too unradiogenic to share a source with the rejuvenated lavas. This indicates sampling of source material at Hualalai that is not expressed at any other Hawaiian volcano and implies that more than one depleted component is present within the plume. A depleted component is required to explain the isotopic variability observed in shield lavas from Kauai, and may also be present in varying proportion and to a lesser degree at other Hawaiian volcanoes [Mukhopadhyay eta!., 2003]. Despite proximity of the plume to a spreading ridge axis, the depleted component that contributed to lavas from the 76-8 1 Ma Detroit Seamount at the northern end of the Emperor seamount chain is thought to be intrinsic to the Hawaiian plume [Frey et a!., 2005]. Thus, the sampling of depleted components is not solely restricted to the late-stages (late-shield/post-shield and rejuvenated) of volcanism or to the recent (<5 Ma) volcanoes of the Hawaiian chain. Instead, these depleted components represent a fundamental part of the Hawaiian plume that  89  may only be sampled under specific melting conditions or in certain geodynamic settings. For example, melting of a depleted plume component could be facilitated under thin lithosphere [Regelous et aL, 2003] or, alternatively, in a secondary melting zone [Ribe and Christensen, 1999] or by flexural arch decompression [Bianco et a!., 2005], as has been proposed to explain Hawaiian rejuvenated volcanism.  3.5.3 Spatial distribution of heterogeneities in the Hawaiian plume The geochemical differences between lavas erupted along the Loa and Kea spatial trends are typically explained by two different models for the geochemical structure of the plume beneath Hawaii (Figure 3.12). The concentrically zoned model [e.g., Hauri, 1996; Kurz et a!., 1996; Lassiter et aL, 1996] proposes a plume core with enriched (Loa-type) compositions and an outer margin with more depleted (Kea-type) compositions. In contrast, the Pb isotope systematics of Hawaiian lavas suggest that the plume is bilaterally zoned [Abouchami eta!., 2005] into southwest and northeast halves that correspond to Loa and Kea source regions, respectively. Pb isotopes are one of the most reliable discriminants Pb for a given 204 Pb/ between lavas from the two trends, with Loa volcanoes having higher 208 Pb than Kea volcanoes [Tatsumoto, 1978; Abouchami et al., 2005]. 204 Pb/ 206  3.5.3.1 Horizontal zoning: Constraints from the shield to post-shield transition The geochemistry of the shield to post-shield transition provides the opportunity to discriminate between the concentrically and bilaterally zoned plume models (Figure 3.12). During the time period corresponding to shield and post-shield volcanism, individual volcanoes traverse the plume conduit from a more central to peripheral location due to the  90  (a)  (b)  concentric  bilateral  0 Mahukona  0 Mahukona  fN 20km  2OO ka  20km  2OO ka  Figure 3.12: Schematic comparison of the two main models for the spatial distribution of Loa and Kea source compositions in the Hawaiian plume: (a) concentric and (b) bilateral zonation. The map of the plume melting region (coloured areas) beneath the island of Hawaii is after DePaolo et al. [2001], based on the melt supply models of DePaolo and Stolper [1996] and Ribe and Christensen [1999]. The outer field represents the lateral extent of the primary melting zone; the darker inner circle corresponds to a melt supply rate of >0.05 cm/yr. In the concentric model (a), Loa source material (blue) is in the core of the plume and Kea source material (red) is restricted to the plume margin [e.g., Hauri, 1996; Kurz et al., 1996; Lassiter et al., 1996], whereas in the bilateral model (b), Loa and Kea source material is divided into southwest and northeast sides of the plume, respectively [Abouchami et al., 2005]. Superimposed on the map are the outline of the island of Hawaii and the current summits of Mahukona, Kohala, Hualalai, and Mauna Kea (white circles). Also shown are the summit locations of these volcanoes at the time of eruption of the post-shield lavas (sample/vent locations are omitted for clarity), assuming a Pacific plate velocity of 9 cm/yr. The colour of the paleo-summits corresponds to the isotopic signature of the lavas (blue = Loa-like; red = Kea-like) and show that the post-shield lavas are consistent with a bilaterally zoned plume (b).  91  motion of the Pacific lithosphere. In a concentrically zoned plume, where Loa-trend volcanoes are inferred to sample the plume centre and Kea-trend volcanoes sample the plume edge, both Loa- and Kea-trend shield volcanoes are predicted to be Kea-like during the post-shield stage. In contrast, in a bilaterally zoned plume, Loa- and Kea-trend volcanoes will retain their initial Loa- and Kea-like geochemical characteristics, respectively. Post-shield lavas from the Kea-trend volcanoes, Mauna Kea and Kohala, are Kea like, and post-shield lavas from the Loa-trend Hualalai are Loa-like in their Pb isotopic compositions (Figure 3.1 Oa). Transitional to weakly alkalic basalts from Mahukona, a submerged Hawaiian shield volcano located —-50 km west of Kohala on the Loa spatial trend, likely belong to the early post-shield stage of this volcano and are also Loa-like [Appendix B 1; Garcia et al., in preparation, 2008]. The persistence of the Kea and Loa Pb signatures from the shield to post-shield stage at these four Hawaiian volcanoes is inconsistent with the concentric model and strongly supports a bilateral plume zonation (Figure 3.12). However, these observations could also be explained by a concentrically zoned plume that has been distorted in the downstream direction by the movement of the over-riding Pacific lithosphere [DePaolo et a!., 2001; Bryce et a!., 2005]. Information from the southwest side of the plume is required to differentiate between these two models; Loa-like isotopic compositions to the southwest would support the bilateral model, whereas Kea-like compositions would support the distorted concentric model. Basalts from Penguin Bank, which lies west of the main Hawaiian Ridge (Figure B 1), have Loa-like Sr, Nd, and Hf isotopic compositions and Pb isotopic compositions that lie along the Loa-Kea Pb isotopic division [Appendix B2; Xii et al., 2007]. The dominantly  92  Loa-like isotopic compositions of lavas from Penguin Bank are consistent with a bilateral plume zonation. However, the possibility that an exclusively Kea-like volcano exists further to the southwest (Kaula?) cannot be ruled out. The oblique Pb-Pb trend of Kohala that crosses the Pb isotopic division (Figure 3.10) reflects sampling of Loa source material by a Kea-trend volcano and indicates that the division between Loa and Kea sources in most plume zoning models is oversimplified. The presence of both Loa and Kea geochemical characteristics within a single Hawaiian volcano has been documented at Mauna Kea [Eisele et al., 2003], Haleakala [Ren et a!., 2006], Kilauea [Marske et a!., 2007], West Molokai [Xu et a!., 2007] and Mahukona [Appendix B 1; Garcia et a!., in preparation, 2008]. These observations may be explained by the presence of a “mixed” zone within the plume conduit that contains both Loa and Kea source material, or if the division between the two sources fluctuates with time.  3.5.3.2 Vertical heterogeneity: Constraints from consecutive volcano pairs The previous discussion has assumed a time-invariant geochemical cross-section of the plume conduit. However, the geochemical variability observed in Hawaiian lavas may also be related to vertical heterogeneities within the upwelling plume. Due to the high upwelling velocities (e.g., >1 mlyr) predicted for the Hawaiian plume, the vertical component of heterogeneity cannot be ignored [Blichert-Toft et a!., 2003; Marske et aL, 2007]. Rather than investigating the temporal evolution of a single volcano, which mainly records a horizontal geochemical gradient across the plume conduit, we compare post-shield lavas from two consecutive pairs of Hawaiian volcanoes: Mahukona-Kohala and Hualalai Mauna Kea.  93  Lavas from the older Mahukona and Kohala volcanoes initially have similar isotopic compositions and become more radiogenic in Pb with time (Figure 3.11 a). In contrast, lavas from the younger Hualalai and Mauna Kea volcanoes become less radiogenic in Pb during the post-shield stage (Figure 3.11 b). Mahukona and Kohala were thus sampling a more Pb and 206 204 Pb/ Pb) source that was present in the melting region of 204 Pb/ enriched (high 208 the plume at that time. As these volcanoes migrated away from the plume axis, the mantle Pb) 204 Pb/ Pb and 206 204 Pb/ continued to rise and was replaced by relatively depleted (low 208 source material that was subsequently sampled by Hualalai and Mauna Kea. The sampling of isotopically similar source material at another pair of neighbouring volcanoes, Mauna Loa and Kilauea, is thought to have been caused by the passage of a small-scale heterogeneity (either a single body or the plume matrix itself) through their melting regions [Marske et aL, 2007]. Although for Mahukona-Kohala and Hualalai-Mauna Kea, the Pb-Pb arrays do not converge to a common composition, the observed Pb isotope systematics provide compelling evidence for a significant component of vertical heterogeneity within the upwelling Hawaiian plume.  3.6 CONCLUSIONS High-precision trace element concentrations and Sr, Nd, Pb and Hf isotopic compositions of Hawaiian post-shield lavas from Mauna Kea, Kohala, and Hualalai show that the shield to post-shield transition was accompanied by systematic changes in the types and proportions of discrete components sampled by each volcano. The unradiogenic Pb isotopic compositions at Hualalai, positive Sr-Pb correlation at Mauna Kea, and the shallow 8 H 6 N f— d trend  preclude generation of these lavas from mixing between the Kea and Koolau  94  end-members. These lavas require a contribution from at least one additional source Pb, as well as more radiogenic 204 Pb/ Sr and 206 86 Sr/ component with relatively unradiogenic 87 6 that is not expressed in Hawaiian shield lavas. This depleted isotopic signature Hf’ 7 ‘ Hf, 177 cannot be explained by assimilation of the underlying Pacific oceanic crust or shallow entrained asthenosphere. Basalts from ODP Site 843 and gabbroic xenoliths from Hualalai do not have sufficiently unradiogenic Pb isotopic compositions and Pb-Pb trendlines for post-shield lavas from Mauna Kea and Hualalai do not intersect the EPR MORB field. The Sr, Nd, Pb, and Hf isotopic compositions of the post-shield lavas reflect sampling of an intrinsic plume component and are consistent with the presence of ancient lower oceanic crust and sediments in their source. Post-shield lavas from Mauna Kea trend Pb towards compositions characteristic of Hawaiian rejuvenated 204 Pb/ 5r and 206 86 Sr/ to low 87 stage lavas; this trend is observed at four consecutive Kea-trend volcanoes and indicates that this depleted component is characteristic of the Kea source and is a long-lived feature of the Pb component is sampled by post204 Pb/ 206 Sr, low86 Sr/ 87 Hawaiian plume. A different lowshield lavas from Hualalai, which indicates that several depleted components may be present in the Hawaiian plume and that they are widely distributed in both the Kea and Loa sources. The post-shield lavas retain their characteristic Loa- and Kea-like Pb isotope signatures and are consistent with a bilateral or distorted concentric model of plume zonation. The Pb-Pb array for Kohala crosses the Loa-Kea Pb isotopic division and indicates that the boundary between Loa and Kea source material within the plume may fluctuate with time or that a mixed zone is present. The contrasting Pb isotope systematics of the consecutive pairs of volcanoes, in which Mahukona and Kohala become more radiogenic and Hualalai and Mauna Kea become less radiogenic, imply that the vertical component of heterogeneity is  95  significant and also contributes to the observed geochemical variation observed in Hawaiian lavas.  3.7 ACKNOWLEDGEMENTS The author would like to thank Don DePaolo and Sarah Aciego for the opportunity to work on the post-shield samples. Jane Barling, Rich Friedman, Bruno Kieffer, Claude Maerschalk, Bert Mueller, and Wilma Pretorius are thanked for their help with sample processing and HR-ICP-MS, TIMS, and MC-ICP-MS analyses at the Pacific Centre for Isotopic and Geochemical Research, University of British Columbia. D. Hanano was supported by an NSERC Canada Graduate Scholarship (CGS-M). Funding for this research was provided by NSERC Discovery Grants to J. S. Scoates and D. Weis.  96  3.8 REFERENCES Abouchami, W., S. J. G. Galer, and A. W. Hoflnann (2000), High precision lead isotope systematics of lavas from the Hawaiian Scientific Drilling Project, Chemical Geology, 169, 187-209. Abouchami, W., A. W. Hothiann, 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. Bianco, T. A., G. Ito, J. M. Becker, and M. 0. Garcia (2005), Secondary Hawaiian volcanism formed by flexural arch decompression, Geochemistry Geophysics Geosystems, 6(Q08009), doi: 10.1 029/2005GC000945. Blichert-Toft, J., C. Chauvel, and F. Albarède (1997), Separation of Hf and Lu for highprecision isotope analysis of rock samples by magnetic sector-multiple collector ICP-MS, Contributions to Mineralogy and Petrology, 127, 248-260. Blichert-Toft, J., F. A. Frey, and F. Albarêde (1999), Hf isotope evidence for pelagic sediments in the source of Hawaiian basalts, Science, 285, 879-882. Blichert-Toft, J., D. Weis, C. Maerschalk, A. Agranier, and F. Albarède (2003), Hawaiian hot spot dynamics as inferred from the Hf and Pb isotope evolution of Mauna Kea volcano, Geochemistry Geophysics Geosystems, 4(2), doi: 10.1 029/2002GC000340. Blichert-Toft, J., and W. M. White (2001), Hf isotope geochemistry of the Galapagos Islands, Geochemistry Geophysics Geosystems, 2, doi: 10.1 029/2000GC000 138. 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, Geochemistry Geophysics Geosystems, 6(9), doi: 10. 1029/2004GC000809. Castillo, P., E. Klein, J. Bender, C. Langmuir, S. Shirley, R. Batiza, and W. White (2000), Petrology and Sr, Nd, and Pb isotope geochemistry of mid-ocean ridge basalt glasses from the 11°45’N to 15°00’N segment of the East Pacific Rise, Geochemistry, Geophysics, Geosystems, 1(11), doi: 10.1029/1 999GC000024. Chen, C. Y., and F. A. Frey (1985), Trace element and isotopic geochemistry of lavas from Haleakala volcano, East Maui, Hawaii: Implications for the origin of Hawaiian basalts, Journal ofGeophysical Research, 90, 8743-8768. Chen, C. Y., F. A. Frey, and M. 0. Garcia (1990), Evolution of alkalic lavas at Haleakala volcano, East Maui, Hawaii, Contributions to Mineralogy and Petrology, 105, 197218. Chen, C.-Y., F. A. Frey, M. 0. Garcia, G. B. Dalrymple, and S. R. Hart (1991), The tholeiitic to alkalic basalt transition at Haleakala volcano, Maui, Hawaii, Contributions to Mineralogy and Petrology, 106, 183-200. Clague, D. A., and G. B. Dalrymple (1987), The Hawaiian-Emperor volcanic chain, Part 1: Geologic evolution, in Volcanism in Hawaii, edited by R.W. Decker, T.L. Wright, and P.H. Stauffer, pp. 1667, US Geological Survey Professional Paper 1350, Denver. Courtillot, V., A. Davaille, J. Besse, and J. Stock (2003), Three distinct types of hotspots in the Earth’s mantle, Earth and Planetary Science Letters, 205, 295-308.  97  Cousens, B. L., D. A. Clague, and W. D. Sharp (2003), Chronology, chemistry, and origin of trachytes from Hualalai Volcano, Hawaii, Geochemistry Geophysics Geosystems, 4(9), 1078, doi:10.1029/2003GC000560. Dana, J. D. (1849), Geology, in United States Exploring Expedition, 1838-1842, vol. 10, pp. 756, C. Sherman, Philadelphia. DePaolo, D. J., Study of magma sources, mantle structure and the differentiation of the earth Nd in igneous rocks, Ph.D. dissertation thesis, California 44 using variations of 14 Nd/’ Institute of Technology, Pasadena, 1978. DePaolo, D. J., J. G. Bryce, A. Dodson, D. L. Shuster, and B. M. Kennedy (2001), Isotopic evolution of Mauna Loa and the chemical structure of the Hawaiian plume, Geochemistry Geophysics Geosystems, 2, doi: 10.1 029/2000GC000 139. DePaolo, D. J., and E. E. Daley (2000), Neodymium isotopes in basalts of the southwest basin and range and lithospheric thinning during continental extension, Chemical Geology, 169, 157-185. DePaolo, D. J., and B. M. Stolper (1996), Models of Hawaiian volcano growth and plume structure: Implications from the Hawaii Scientific Drilling Project, Journal of Geophysical Research, 101, 11,643-11,655. Doucet, S., D. Weis, J. S. Scoates, K. Nicolaysen, F. A. Frey, and A. Giret (2002), The depleted mantle component in Kerguelen Archipelago basalts: Petrogenesis of the tholeiitic transitional basalts from the Loranchet Peninsula, Journal ofPetrology, 43(7), 1341-1366. Eiler, J. M., K. A. Farley, J. W. Valley, A. W. HofInann, and E. M. Stolper (1996), Oxygen isotope constraints on the sources of Hawaiian volcanism, Earth and Planetary Science Letters, 144, 453-468. Eisele, J., W. Abouchami, S. J. G. Galer, and A. W. Hofmann (2003), The 320 kyr Pb isotopic evolution of Mauna Kea lavas recorded in the HSDP-2 drill core, Geochemistry Geophysics Geosystems, 4(5), doi: 10.1 029/2002GC0003 39. Feigenson, M. D., A. W. Hoflnann, and F. J. Spera (1983), Case studies on the origin of basalt: II The transition from tholeiitic to alkalic volcanism on Kohala volcano, Hawaii, Contributions in Mineralogy and Petrology, 84, 390-405. Fekiacova, Z., W. Abouchami, S. J. G. Galer, M. 0. Garcia, and A. W. Hofmann (2007), Origin and temporal evolution of Koolau volcano, Hawaii: Inferences from isotope data on the Koolau Scientific Drilling Project (KSDP), the Honolulu Volcanics and ODP Site 843, Earth and Planetary Science Letters, 261, 65-83. Fitton, J. G., A. D. Saunders, P. D. Kempton, and B. S. Hardarson (2003), Does depleted mantle form an intrinsic part of the Iceland plume? Geochemistry Geophysics Geosystems, 4(3), doi: 10.1 029/2002GC000424. Frey, F., D. Clague, J. Mahoney, and J. Sinton (2000), Volcanism at the edge of the Hawaiian plume: Petrogenesis of submarine alkalic lavas from the North Arch Volcanic Field, Journal ofPetrology, 41, 667-691. Frey, F. A., M. 0. Garcia, and M. F. Roden (1994), Geochemical characteristics of Koolau Volcano: Implications of intershield geochemical differences among Hawaiian volcanoes, Geochimica et Cosmochimica Acta, 58(5), 1441-1462. Frey, F. A., S. Huang, J. Blichert-Toft, M. Regelous, and M. Boyet (2005), Origin of depleted components in basalt related to the Hawaiian hot spot: Evidence from -  ,  98  isotopic and incompatible element ratios, Geochemistry Geophysics Geosystems, 6(1), doi: 10.1029/2004GC000757. Frey, F. A., W. S. Wise, M. 0. Garcia, H. B. West, S.-T. Kwon, and A. K. Kennedy (1990), Evolution of Mauna Kea volcano, Hawaii: Petrologic and geochemical constraints on postshield volcanism, Journal of Geophysical Research, 95(B2), 127 1-1300. Gaffney, A. M., B. K. Nelson, and J. Blichert-Toft (2004), Geochemical constraints on the role of oceanic lithosphere in intra-volcano heterogeneity at West Maui, Hawaii, Journal ofPetrology, 45(8), 1663-1687. Gaffney, A. M., B. K. Nelson, and J. Blichert-Tofi (2005), Melting in the Hawaiian plume at 1-2 Ma as recorded at Maui Nui: The role of eclogite, peridotite and source mixing, Geochemistry Geophysics Geosystems, 6(10), doi: 10.1 029/2005GC000927. Galer, S. J. G., and W. Abouchami (1998), Practical application of lead triple spiking for correction of instrumental mass discrimination, Mineralogical Magazine, 62A, 491492. Hauri, B. H. (1996), Major-element variability in the Hawaiian mantle plume, Nature, 382, 415-4 19. Hieronymus, C. F., and D. Bercovici (2001), A theoretical model of hotspot volcanism: Control on volcanic spacing and patterns via magma dynamics and lithosheric stresses, Journal of Geophysical Research, 106(B 1), 683-702. Horwitz, E. P., D. M.L., and D. E. Fisher (1991), Separation and preconcentration of strontium from biological, environmental, and nuclear waste samples by extraction chromatography using a crown ether, Analytical Chemistry, 63, 522-525. Huang, S., and F. A. Frey (2005), Trace element abundances of Mauna Kea basalt from phase 2 of the Hawaii Scientific Drilling Project: Petrogenetic implications of correlations with major element content and isotopic ratios, Geochemistry, Geophysics, Geosystems, 4(6), doi: 10.1 029/2002GC000322. Huang, S., F. A. Frey, J. Blichert-Toft, R. V. Fodor, G. R. Bauer, and G. Xu (2005), Enriched components in the Hawaiian plume: evidence from Kahoolawe Volcano, Hawaii, Geochemistry Geophysics Geosystems, 6(11), doi: 10.1 029/2005GC00 1012. Jackson, E. D., E. A. Silver, and G. B. Dairymple (1972), Hawaiian-Emperor chain and its relation to Cenozoic circum-Pacific tectonics, Geological Society ofAmerica Bulletin, 83, 601-618. King, A. J., D. G. Waggoner, and M. 0. Garcia (1993), Geochemistry and petrology of basalts from Leg 136, Central Pacific Ocean, in Proceedings ofthe Ocean Drilling Program, ScientWc Results, edited by R.H. Wilkens, J. Firth, and J. Bender et al., pp. 107-118, College Station, TX. Kurz, M. D., T. C. Kenna, J. C. Lassister, and D. J. DePaolo (1996), Helium isotopic evolution of Mauna Kea volcano: First results from the 1-km drill core, Journal of Geophysical Research, 101, 11,781-11,793. Lassiter, J. C., D. J. DePaolo, and M. Tatsumoto (1996), Isotopic evolution of Mauna Kea Volcano: results from the initial phase of the Hawaii Scientific Drilling Project, Journal ofGeophysical Research, 101(B5), 11,769-11,780. Lassiter, J. C., E. H. Hauri, P. W. Reiners, and M. 0. Garcia (2000), Generation of Hawaiian post-erosional lavas by melting of a mixed lherzolite/pyroxenite source, Earth and Planetary Science Letters, 178, 269-2 84.  99  Lassiter, L., and E. Hauri (1998), Osmium-isotope variations in Hawaiian lavas: Evidence for recycled oceanic lithosphere in the Hawaiian plume, Earth and Planetaiy Science Letters, 164, 483-496. Le Bas, M. J., R. W. Le Maitre, A. Streckeisen, and B. Zanettin (1986), A chemical classification of volcanic rocks based on the total alkali silica diagram, Journal of Petrology, 27, 745-750. Th 232 ThJ’ Luo, X., M. Rehkamper, D. C. Lee, and A. N. Halliday (1997), High precision 230 and 234 U measurements using energy-filtered ICP magnetic sector multiple 238 U/ collector mass spectrometry, International Journal ofMass Spectrometry and Ion Processes, 171, 105-117. Macdonald, G. A., and T. Katsura (1964), Chemical composition of Hawaiian lavas, Journal ofPetrology, 5(1), 82-133. Marske, J. P., A. J. Pietruszka, D. Weis, M. 0. 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 and Planetary Science Letters, 259, 34-50. McCulloch, M. T., and J. A. Gamble (1991), Geochemical and geodynamical constraints on subduction zone magmatism, Earth and Planetary Science Letters, 102, 358-374. McDonough, W. F., and S.-s. Sun (1995), The composition of the Earth, Chemical Geology, 120, 223-253. McDougall, I. (1969), Potassium-argon ages on lavas of Kohala volcano, Hawaii, Geological Society ofAmerica Bulletin, 80, 2597-2600. Montelli, R., G. Nolet, F. A. Dahien, and G. Masters (2006), A catalogue of deep mantle plumes: New results from finite-frequency tomography, Geochemistry Geophysics Geosystems, 7(11), doi: 10.1 029/2006GC00 1248. Mukhopadhyay, S., J. C. Lassiter, K. A. Farley, and S. W. Bogue (2003), Geochemistry of Kauai shield-stage lavas: Implications for the chemical evolution of the Hawaiian plume, Geochemistry Geophysics Geosystems, 4(1), 1009, doi: 10.1 029/2002GC000342. Niu, Y., K. D. Collerson, R. Batiza, J. I. Wendt, and M. Regelous (1999), Origin of enriched-type mid-ocean ridge basalt at ridges far from mantle plumes: The East Pacific Rise at 1 1°20’N, Journal of Geophysical Research, 104(B4), 7067—7087. Patchett, P. J., W. M. White, H. Feldmann, S. Kielinczuk, and A. W. Hofmann (1984), Hafniumlrare earth element fractionation in the sedimentary system and crustal recycling into the Earth’s mantle, Earth and Planetary Science Letters, 69, 365-378. Pretorius, W., D. Weis, G. Williams, D. Hanano, B. Kieffer, and J. S. Scoates (2006), Complete trace elemental characterization of granitoid (USGS G-2, GSP-2) reference materials by high resolution inductively coupled plasma-mass spectrometry, Geostandards and Geoanalytical Research, 30(1), 39-54. Regelous, M., A. W. Hoflnann, W. Abouchami, and S. J. G. Galer (2003), Geochemistry of lavas from the Emperor Seamounts, and the geochemical evolution of Hawaiian magmatism from 85 to 42 Ma, Journal ofPetrology, 44(1), 113-140. Regelous, M., Y. Niu, J. I. Wendt, R. Batiza, A. Greig, and K. D. Collerson (1999), Variations in the geochemistry of magmatism on the East Pacific Rise at 10°30’N since 800 ka, Earth and Planetary Science Letters, 168, 45-63. -  100  Reiners, P. W., and B. K. Nelson (1998), Temporal-compositional-isotopic trends in rejuvenated-stage magmas of Kauai, Hawaii, and implications for mantle melting processes, Geochimica et Cosmochimica Acta, 62, 2347-2368. Ren, Z.-Y., T. Shibata, M. Yoshikawa, K. T. M. Johnson, and E. Takahashi (2006), Isotope compositions of submarine Hana Ridge lavas, Haleakala volcano, Hawaii: Implications for source compositions, melting process and the structure of the Hawaiian plume, Journal ofPetrology, 47(2), 255-275. Ribe, N. M., and U. R. Christensen (1999), The dynamical origin of Hawaiian volcanism, Earth and Planetary Science Letters, 171, 517—531. Shafer, J. T., C. R. Neal, and M. Regelous (2005), Petrogenesis of Hawaiian postshield lavas: Evidence from Nintoku Seamount, Emperor Seamount Chain, Geochemistry Geophysics Geosystems, 6, doi: 10.1 029/2004GC000 875. Sleep, N. H. (1990), Hotspots and mantle plumes: Some phenomenology, Journal of Geophysical Research, 95, 67 15-6736. Staudigel, H. A., A. Zindler, S. R. Hart, T. Leslie, C.-Y. Chen, and D. Clague (1984), The isotope systematics of a juvenile intraplate volcano: Pb, Nd and Sr isotope ratios of basalts from Loihi Seamount, Hawaii, Earth and Planetary Science Letters, 69, 1329. Stevenson, R. K., and P. J. Patchett (1990), Implications for the evolution of continental crust from Hf isotope systematics of Archean detrital zircons, Geochimica et Cosmochimica Acta, 54(6), 1683-1697. Stille, P., D. M. Unruh, and M. Tatsumoto (1986), Pb, Sr, Nd, and Hf isotopic constraints on the origin of Hawaiian basalts and evidence for a unique mantle source, Geochimica et Cosmochimica Acta, 50(10), 2303-2319. Tatsumoto, M. (1978), Isotopic composition of lead in oceanic basalt and its implication to mantle evolution, Earth and Planetary Science Letters, 38, 63-87. Vervoort, J. D., and J. Blichert-Toft (1999), Evolution of the depleted mantle: Hf isotope evidence from juvenile rocks through time, Geochimica et Cosmochimica Acta, 63, 533-556. Vervoort, J. D., P. 3. Patchett, J. Blichert-Toft, and F. Albarède (1999), Relationships between Lu—Hf and Sm—Nd isotopic systems in the global sedimentary system, Earth and Planetary Science Letters, 168, 79—99. Weaver, B. L. (1991), The origin of ocean island basalt end-member compositions: trace element and isotopic constraints, Earth and Planetary Science Letters, 104, 381-397. Weis, D., B. Kieffer, D. Hanano, I. Nobre Silva, J. Barling, W. Pretorius, C. Maerschalk, and N. Mattielli (2007), Hf isotope compositions of U.S. Geological Survey reference materials, Geochemistry Geophysics Geosystems, 8(6), doi: 10.1 029/2006GC00 1473. Weis, D., B. Kieffer, C. Maerschalk, 3. Barling, J. deJong, G. A. Williams, D. Hanano, W. Pretorius, N. Mattielli, J. S. Scoates, A. Goolaerts, R. M. Friedman, and J. B. Mahoney (2006), High-precision isotopic characterization of USGS reference materials by TIMS and MC-ICP-MS, Geochemistry Geophysics Geosystems, 7(8), doi: 10.1 029/2006GC00 1283. Weis, D., B. Kieffer, C. Maerschalk, W. Pretorius, and J. Barling (2005), High-precision PbSr-Nd-Hf isotopic characterization of USGS BHVO-1 and BHVO-2 reference materials, Geochemistry Geophysics Geosystems, 6(2), doi: 10.1 029/2004GC000852.  101  West, H. B., D. C. Gerlach, W. P. Leeman, and M. 0. Garcia (1987), Isotopic constraints on the origin of Hawaiian lavas from the Maui volcanic complex, Hawaii, Nature, 330, 216-220. West, H. B., and W. P. Leeman (1987), Isotopic evolution of lavas from Haleakala Crater, Hawaii, Earth and Planetary Science Letters, 84, 211-225. White, W. M., F. Albarêde, and P. Telouk (2000), High-precision analysis of Pb isotope ratios by multi-collector ICP-MS, Chemical Geology, 167, 257-270. Wolfe, E. W., and J. Morris (1996), Geologic map of the island of Hawaii, Map 2524-A, United States Geological Survey, Denver, CO. Wolfe, E. W., W. S. Wise, and G. B. Dairymple (1997), The geology and petrology of Mauna Kea volcano, Hawaii A study of postshield volcanism, US. Geological Survey Professional Paper 1557, 129 pp., Menlo Park, CA. Xu, G., F. A. Frey, D. A. Clague, W. Abouchami, J. Blichert-Toft, B. Cousens, and M. Weisler (2007), Geochemical characteristics of West Molokai shield- and postshield stage lavas: Constraints on Hawaiian plume models, Geochemistry Geophysics Geosystems, 8, doi: 10. 1029/2006GC001554. Xu, G., F. A. Frey, D. A. Clague, D. Weis, and M. H. Beeson (2005), East Molokai and other Kea-trend volcanoes: Magmatic processes and sources as they migrate away from the Hawaiian hot spot, Geochemistry Geophysics Geosystems, 6, doi: 10. 1029/2004GC000830. Yang, H.-J., F. A. Frey, and D. A. Clague (2003), Constraints on the source components of lavas forming the Hawaiian North Arch and Honolulu Volcanics, Journal of Petrology, 44(4), 603-627. -  102  CHAPTER 4  IMPLICATIONS FOR THE NATURE OF GEOCHEMICAL HETEROGENEITIES IN THE HAWAIIAN MANTLE PLUME  103  4.1 SUMMARY AND CONCLUSIONS This study of post-shield lavas from Mauna Kea, Kohala, and Hualalai volcanoes on the island of Hawaii uses major and trace element chemistry as well as Sr, Nd, Pb, and Hf isotopic compositions to characterize the geochemical changes that occur during the critical transition to the post-shield stage of ocean island volcanoes. Integration of the results of this study with the geochemistry of shield, post-shield, and rejuvenated lavas from other Hawaiian volcanoes, particularly submarine basalts from Mahukona, permits the identification of systematic spatial and temporal isotopic variations. The distinct isotopic signatures of the post-shield lavas have implications for the origin of depleted components related to Hawaiian volcanism and provide additional constraints on the geochemical structure of the Hawaiian mantle plume. Post-magmatic alteration can significantly modify the Pb isotopic compositions of basalts from Hawaii and other oceanic islands and, if not dealt with adequately, hinders the use of Pb isotopes as a geochemical tracer. The large difference in Pb isotopic composition between unleached and leached Hawaiian basalts indicates the presence of highly radiogenic components (e.g., Mn-oxides and barite/celestite) and suggests that even young, relatively unaltered basalts must be acid-leached prior to isotopic analysis. The incomplete removal of some secondary minerals (e.g., celadonite) by acid-leaching results in poor Pb isotope reproducibility and, with the improved precision now afforded by MC-ICP-MS [e.g., Albarède et al., 2004], may be the ultimate limitation on high-precision Pb isotopic compositions of ocean island basalts. The external Pb isotope reproducibility achieved in this study, discussed in Appendix A3, is better than —1 50 ppm (Table A3), which allows for identification of small-scale heterogeneities in the source of the post-shield lavas.  104  The enriched trace element abundances of the post-shield lavas contrast with their depleted isotopic characteristics, which require contributions from a source component with high NdJ’Nd 43 and Hf’ ‘ 76 and low 87 ‘ Hf 177 Sr and 206 86 Sr/ Pb. The isotopic variation 204 Pb/ observed in the post-shield lavas cannot be explained by mixing between the two principal Hawaiian shield stage end-members (Kea and Koolau) or by assimilation of the underlying Pacific lithosphere or shallow asthenosphere. This conclusion is consistent with a growing number of recent studies [e.g., Frey et al., 2005; Fekiacova et at., 2007; Swinnard et a!., in preparation, 2008] and provides compelling evidence that the depleted isotopic signatures observed in Hawaiian lavas are derived from within the Hawaiian plume. The unradiogenic Sr and Pb isotopic compositions of the post-shield lavas are consistent with those inferred for ancient (2 Ga) gabbroic lower oceanic crust and sediments. The Nd and Hf isotopic systematics provide further evidence for the presence of ancient pelagic sediments in the source of the post-shield lavas. The recognition of ancient oceanic crust and sediments in the Hawaiian plume source is well documented [e.g., Lassiter and Hauri, 1998; Blichert Toft et at., 1999] and supports a recycling model whereby such materials are subducted into the mantle and returned to the surface via upwelling plumes. The depleted isotopic signature identified in post-shield lavas from Mauna Kea is also present at West Maui, East Molokai, and Haleakala [Chen et a!., 1991; Gaffi2ey et a!., 2004; Xu et a!., 2005], suggesting that this component is a fundamental characteristic of the Kea source. If rejuvenated lavas are derived from the same depleted source as these four consecutive Kea-trend volcanoes, which is suggested by the isotopic systematics, then this depleted component must be a long-lived (i.e., millions of years) feature of the Hawaiian plume. The depleted component sampled by post-shield lavas from Hualalai represents the  105  first example of a depleted signature at a Loa-trend volcano. This indicates that more than one depleted component is present in the plume and that these components are more widely distributed, in both the Kea and Loa sources, than previously thought. Future research can take advantage of these findings in developing geodynamic and melting models to explain the physical processes that allow these depleted components to be sampled [e.g., Ribe and Christensen, 1999; Regelous eta!., 2003; Bianco eta!., 2005]. The presence of both Loa and Kea isotopic compositions in lavas from Mahukona and Kohala represents two more examples of a volcano with a subset of lavas that have Pb isotopic compositions characteristic of lavas from the adjacent trend [e.g., Eisele et a!., 2003; Ren eta!., 2006; Marske eta!., 2007; Xu eta!., 2007] and indicate that the compositional boundary between these two source materials is more complex than predicted by most current plume models. The Loa-like Pb isotope signatures of post-shield lavas from Hualalai and Mahukona, combined with the Kea-like signatures observed at Mauna Kea and Kohala, are inconsistent with a concentrically zoned plume and support a bilateral zonation of Pb isotope chemistry within the Hawaiian plume. The contrasting Pb isotope systematics of the consecutive volcano pairs indicates that, in addition to any horizontal zonation, vertical heterogeneity within the upwelling plume is also significant. More information from locations transverse to the axis of the Hawaiian Ridge, and particularly from the southwest side of the plume, is required to confirm these hypotheses and resolve the long standing debate over the geochemical structure of the Hawaiian plume.  106  4.2 REFERENCES Albarêde, F., P. Telouk, 3. Blichert-Toft, M. Boyet, A. Agranier, and B. Nelson (2004), Precise and accurate isotopic measurements using multiple-collector ICPMS, Geochimica et Cosmochimica Acta, 68(12), 2725-2744. Bianco, T. A., G. Ito, J. M. Becker, and M. 0. Garcia (2005), Secondary Hawaiian volcanism formed by flexural arch decompression, Geochemistry Geophysics Geosystems, 6(Q08009), doi: 10.1 029/2005GC000945. Blichert-Toft, J., F. A. Frey, and F. Albarêde (1999), Hf isotope evidence for pelagic sediments in the source of Hawaiian basalts, Science, 285, 879-8 82. Chen, C.-Y., F. A. Frey, M. 0. Garcia, G. B. Dafrymple, and S. R. Hart (1991), The tholeiitic to alkalic basalt transition at Haleakala volcano, Maui, Hawaii, Contributions to Mineralogy and Petrology, 106, 183-200. Eisele, J., W. Abouchami, S. J. G. Galer, and A. W. Hofmann (2003), The 320 kyr Pb isotopic evolution of Mauna Kea lavas recorded in the HSDP-2 drill core, Geochemistry Geophysics Geosystems, 4(5), doi: 10.1 029/2002GC000339. Fekiacova, Z., W. Abouchami, S. J. G. Galer, M. 0. Garcia, and A. W. Hofmann (2007), Origin and temporal evolution of Koolau volcano, Hawaii: Inferences from isotope data on the Koolau Scientific Drilling Project (KSDP), the Honolulu Volcanics and ODP Site 843, Earth and Planetary Science Letters, 261, 65-83. Frey, F. A., S. Huang, J. Blichert-Toft, M. Regelous, and M. Boyet (2005), Origin of depleted components in basalt related to the Hawaiian hot spot: Evidence from isotopic and incompatible element ratios, Geochemistry Geophysics Geosystems, 6(1), doi: 10.1029/2004GC000757. Gafffiey, A. M., B. K. Nelson, and J. Blichert-Toft (2004), Geochemical constraints on the role of oceanic lithosphere in intra-volcano heterogeneity at West Maui, Hawaii, Journal ofPetrology, 45(8), 1663-1687. Lassiter, L., and E. Hauri (1998), Osmium-isotope variations in Hawaiian lavas: Evidence for recycled oceanic lithosphere in the Hawaiian plume, Earth and Planetary Science Letters, 164, 483-496. Marske, J. P., A. J. Pietruszka, D. Weis, M. 0. 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 and Planetary Science Letters, 259, 34-50. Regelous, M., A. W. Hofmann, W. Abouchami, and S. J. G. Galer (2003), Geochemistry of lavas from the Emperor Seamounts, and the geochemical evolution of Hawaiian magmatism from 85 to 42 Ma, Journal ofPetrology, 44(1), 113-140. Ren, Z.-Y., T. Shibata, M. Yoshikawa, K. T. M. Johnson, and E. Takahashi (2006), Isotope compositions of submarine Hana Ridge lavas, Haleakala volcano, Hawaii: Implications for source compositions, melting process and the structure of the Hawaiian plume, Journal ofPetrology, 47(2), 255-275. Ribe, N. M., and U. R. Christensen (1999), The dynamical origin of Hawaiian volcanism, Earth and Planetary Science Letters, 171, 517—531. Xu, G., F. A. Frey, D. A. Clague, W. Abouchami, J. Blichert-Toft, B. Cousens, and M. Weisler (2007), Geochemical characteristics of West Molokai shield- and postshield stage lavas: Constraints on Hawaiian plume models, Geochemistry Geophysics Geosystems, 8, doi: 10.1 029/2006GC00 1554.  107  Xu, G., F. A. Frey, D. A. Clague, D. Weis, and M. H. Beeson (2005), East Molokai and other Kea-trend volcanoes: Magmatic processes and sources as they migrate away from the Hawaiian hot spot, Geochemistry Geophysics Geosystems, 6, doi: 10.1 029/2004GC000830.  108  601.  sEflnNacIdv  APPENDIX A  ANALYTICAL QUALITY ASSURANCE-QUALITY CONTROL (QA-QC)  110  Al. Trace element abundances determined by XR1? and ICP-MS  An estimate of the precision of trace element abundances reported in this study can be obtained by comparing the values determined by both ICP-MS (University of British Columbia) and XRF (University of California, Berkeley) for the same sample (Figure Al). The linear correlation coefficients for select trace elements including Sr, Zr, Ba, Rb, and Y are all greater than 0.97, except for Nb where it is 0.94. Furthermore, the slopes for Sr, Zr, Ba, and Rb are near unity (0.93-0.96). However, for Y and Nb, there is a discrepancy  between the two techniques, with ICP-MS values consistently lower (by -2O-30%) than XRF values. Systematically lower Y abundances have been documented in other recent ICP-MS studies and may indicate overestimated XRF values related to correction and calibration methods [e.g., Robinson et a!., 1999; Pretorius et al., 2006; Robinson et a!., 1999]. The Y concentrations determined by ICP-MS for reference materials analysed in this study are within error of the USGS recommended values as well as literature values (see Appendix A2).  A2. Trace element abundances of reference materials  The United States Geological Survey (USGS) reference materials AGV-2 and BCR 2 were run as samples to monitor the accuracy of the ICP-MS analyses during the course of this study. Values of AGV-2 show excellent agreement with both USGS [Wilson, 1998] and literature values [e.g., Raczek et a!., 2001; Wilibold and Jochum, 2005], except for Sn where the value in this study is 25% lower than that recommend by the USGS (Table Al). The BCR-2 standard has a composition that more closely matches the majority of the post-shield samples and was analysed three times (Table A2). Values of BCR-2 are within ±8%, and  111  2500  1000  $2000  800  2  2  6  61500  600  1000  400  500  200  0  0 0  500  1000  1500  2000  2500  0  200  400  600  800  1000  140  1000  120 800  Cl)  2  6  2  100  C.)  80  Ea a  60  at  600 400  40  200 20 0  0 200  0  400  600  800  1000  0  60  E a a  30  80  100  120  140  —  125  0)  2so 40  60  150  Cl)  C.)  40  ppm (XRF)  70  &  20  2  at  100  C)  E a a  20  75 50 25  10  0 0  10  20  30  40  50  60  70  0  25  50  75  100  125  150  ppm (XRF)  ppm (XRF)  Figure Al: Comparison of select trace element abundances determined by XRF and ICP-MS. R 2 is the linear correlation coefficient. Slopes of unity are shown for reference.  112  Table Al: Trace Element Abundances of USGS Reference Material AGV2a This Study  (I-IR-Icp-Ms) AGV-2 (n=1) Sc V Cr Co Ni Cu Zn Ga Rb Sr Y Zr Nb Mo Cd Sn Sb Cs Ba Hf Ta W Pb Bi Th U La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu  13 125 16 16 18 46 88 20 66.9 675 18 232 14 9.3 0.34 1.7 0.6 1.14 1135 5.10 0.84 1.6 12 0.04 6.6 2.00 41 71 8.1 30 5.5 1.53 4.44 0.62 3.4 0.67 1.80 0.25 1.6 0.25  USGS Recommended Valuesb  Raczek eta!. [200 lic  Wilibold and OOSj 2 Joch [  Mean (n=5)  Mean (n=3)  66.3 661  65.4 657 20.5 240 14.6  Mean  1SD  %RSD  13 120 17 16 19 53 86 20 68.6 658 20 230 15  1 5 2 1 3 4 8 1 2.3 17 1 4 1  7.7 4.2 11.8 6.3 15.8 7.5 9.3 5.0 3.4 2.6 5.0 1.7 6.7  2.3 0.6 1.16 1140 5.08 0.89  0.4  17.4  0.08 32 0.20 0.08  6.9 2.8 3.9 9.0  13  1  7.7  14.2  6.1 1.88 38 68 8.3 30 5.7 1.54 4.69 0.64 3.6 0.71 1.79 0.26 1.6 0.25  0.6 0.16 1 3 0.6 2 0.3 0.10 0.26 0.04 0.2 0.08 0.11 0.02 0.2 0.01  9.8 8.5 2.6 4.4 7.2 6.7 5.3 6.5 5.5 6.3 5.6 11.3 6.1 7.7 12.5 4.0  6.16 1,81 37.8 67.4 7.57 30.2 5.36 1.49 4.51 0.636 3.43 0.651 1.82 0.267 1.62 0.25 1  1130  37.9 68.6 7.68 30.5 5.49 1.53 4.52 0.641 3.47 0.653 1.81 0.259 1.62 0.247  1.31 1154 5.04 0.83 1  aAll abundances in ppm. %RSD is the relative standard deviation (1 SD/mean x 100). 50 [1998]. l 1 bW cMl abundances determined by ID-TIMS, except for Pr, Tb, Ho, and Tm, which were determined by MIC-SSMS. dAll abundances determined by combined isotope dilution (ID) sector field inductively coupled plasma-mass spectrometry (SF-ICP-MS).  113  Table A2: Trace Element Abundances of USGS Reference Material BCR2a This Study (HR-ICP-MS)  Sc V Cr Co Ni Cu Zn Ga Rb Sr Y Zr Nb Mo Cd Sn Sb Cs Ba Hf Ta W Pb Bi Th U La Ce Pr Nd Sm Eu Gd Th Dy Ho Er Tm Yb Lu  USGS Recommended Valuesb  Raczek et a!.  r200llc  Wilibold and Joch [2005]d  Mean (n=3)  Mean (n=3)  46.9 340  47.3 339 35.3 184 12.5  Mean (n=3)  1SD  %RSD  Mean  1SD  %RSD  33 417 15.00  1 18 0.07 0.05 0.2 0.2 0.3 0.1 0.8 2 3 2 0.4 9 0.04 0.07 0.005 0.01 7 0.02 0.09 0.02 1 0.004 0.2 0.08 2 5 0.5 2 0.4 0.09 0.4 0.05 0.4 0.08 0.2 0.04 0.1 0.03  3.2 4.2 0.5 0.1 1.3 1.2 0.2 0.6 1.7 0.5 7.8 1.3 3.6 3.6 4.4 4.3 2.4 1.2 1.0 0.5 13.2 3.9 10.8 8.7 2.9 4.9 7.3 8.4 7.4 5.6 6.8 4.9 6.3 5.5 5.8 6.3 6.2 7.5 4.8 6.9  33 416 18 37  2 14 2 3  6.1 3.4 11.1 8.1  19 127 23 48 346 37 188  2 9 2 2 14 2 16  10.5 7.1 8.7 4.2 4.0 5.4 8.5  248  17  6.9  1.1 683 4.8  0.1 28 0.2  9.1 4.1 4.2  11  2  18.2  12  6.2 1.7 25 53 6.8 28 6.7 2 6.8 1.07  0.7 0.2 1 2 0.3 2 0.3 0.1 0.3 0.04  11.3 11.2 4.0 3.8 4.4 7.1 4.5 5.0 4.4 3.7  1.33  0.06  4.5  3.5 0.51  0,2 0.02  5.7 3.9  5.23 1.63 25.6 55.3 6.76 28.6 6.41 2.06 6.68 1.06 6.33 1.26 3.62 0.5 12 3.36 0.504  37.44  11.6 14.9 128.1 21.3 47.7 347 33 182 10.0 247 0.90 1.57 0.209 1.13 690 4.70 0.65 0.46 9 0.04 1 6.0 1.66 26 55 6.7 27 6.4 1.85 6.3 0.98 6.2 1.26 3.5 0.49 3.1 0.48  677  24.9 52.9 6.57 28.7 6.57 1.96 6.75 1.07 6.41 1.30 3.66 0.564 3.38 0.5 19  1.07 687 4.74  aAll abundances in ppm. %RSD is the relative standard deviation (1SD/mean x 100). ’1 Wilson [1997. cpj abundances determined by ID-TIMS, except for Pr, Tb, Ho, and Tm, which were determined by MIC-SSMS. dAli abundances determined by combined isotope dilution (ID) sector field inductively coupled plasma-mass spectrometly (SF-ICP-MS). -  114  are consistently better than ±5%, of the values reported by Raczek et a!. [20011 and Wilibold and Jochum [20051. For most elements, values obtained in this study are within ±7% of the USGS recommended values [Wilson, 1997], and are well within the reported standard deviations. Notable exceptions include Cr, Cu, and Pb, for which values in this study are 15-20% lower than the USGS recommended values. Variable Pb concentrations have been identified in many USGS reference materials and may be related to contamination during processing [e.g., Weis et al., 2005; Weis eta!., 2006].  A3. Reproducibility of Pb and Hf isotopic analyses  The external reproducibility of Pb and Hf isotopic analyses was determined via several sets of complete procedural duplicates, which involved a separate leaching, digestion, chemical separation, and isotopic analysis. The external precision of Pb isotopic analyses in this study is better than 166 ppm, and consistently better than 105 ppm (Table A3). This level of precision is comparable to that achieved in other recent Pb isotope studies [e.g., Abouchami eta!., 2000; Eisele eta!., 2003]. This shows that the use of thallium to correct for instrumental mass bias, particularly when all samples are analysed with the same Pb/Tl and intensity as the standards, produces data of comparable quality to the double or triple spike techniques [e.g., White et a!., 2000; Woodhead, 2002; Albarède et a!., 2004]. For the Hf isotopic analyses, the external precision is better than 57 ppm, with duplicate 76 = <0.000003 (Table A4). Replicate ‘ Hf 77 analyses commonly agreeing within Hf7’ analyses, which involved re-running the same sample solution, were also carried out during each analytical session to monitor instrument performance, and are comparable to or better than the reproducibility of duplicate analyses.  115  a Table A3: External Reproducibility of Pb Isotopic Compositions 2a Sample 2a Pb 204 Pb/ 207 Pb 204 Pb/ 206  Pb 204 Pb/ 208  #1 #2 mean ppm O2AHU-13 #1 #2 mean ppm #1 O2AMK-11 #2 mean ppm O2AKA-4 #1 #2 mean ppm  procedural duplicates (separate digestions) 37.7442 17.9802 0.0006 0.0006 15.4473 0.0006 37.7470 17.9805 0.0005 15.4483 0.0014 37.7456 15.4478 17.9803 0.0003 105 19 88 0.0006 37.71 19 15.4500 17.9426 0.0006 0.0007 37.7092 154491 17.9417 0.0007 37.7105 15.4496 0.0013 17.9421 0.0013 101 71 82 0.0010 37.9244 15.4717 18.3411 0.0009 37.9268 15.4717 0.0008 18.3411 0.0009 37.9256 15.4717 0.00003 18.3411 0.00004 88 2 2 0.0005 37.9949 18.4394 0.0006 15.4808 0.0006 37.9994 15.4824 18.4410 0.0006 37.9972 15.4816 0.0022 18.4402 0.0024 166 128 145  WC agate mean ppm O2AHU-10 WC agate mean ppm O2AMK-12 WC agate mean ppm  17.8877 17.8913 17.8895 286 18.0013 17.9992 18.0003 166 18.3463 18.3496 18.3480 253  O2AKA-6  18.2482 18.2482 18.2482 2  O2AHU-5  O2AHU-1  WC agate mean ppm  2c  0.0018 0.0016 0.0040 0.0016 0.0018 0.0038 0.0029 0.0023 0.0033 0.0013 0.0017 0.0063  WC/agate duplicates (separate powders) 15.4434 0.0007 0.0008 15.4459 0.0007 0.0009 0.0036 0.0051 15.4447 235 15.4592 0.0008 0.0009 15.4578 0.0010 0.0010 0.0019 0.0030 15.4585 121 15.4777 0.0008 0.0009 0.0006 15.4788 0.0006 0.0016 15.4782 0.0046 100  37.6712 37.6779 37.6746 252 37.7658 37.7621 37.7640 137 37.9536 37.9608 37.9572 270  0.0019 0.0021 0.0095  15.4713 15.4728 15.4720 131  37.8962 37.8989 37.8976 102  0.0020 0.0019 0.0038  0.0009 0.0008 0.00004  0.0008 0.0008 0.0020  0.0021 0.0024 0.0052 0.0021 0.0018 0.0103  Pb isotopic ratios were determined by MC-ICP-MS on acid-leached sample powders [Weis et al. 2005]. Pb data were corrected for fractionation by Tl spiking and then normalized to the NBS 981 triple spike values of Galer and Abouchami [1998] using the standard bracketing method. The 2a error is the absolute error value of an individual sample analysis (internal error). For the mean values, the error reported is the 2 standard deviation (external error). External reproducibility is represented by the 2 standard deviation from the mean of duplicate ). 6 analyses and is reported in ppm (2SD/mean x10 ,  116  Table A4: External Reproducibility of Hf Isotopic Compositions a Sample 02AHU-3  #1 #2 mean ppm O2AMK-8 #1 #2 mean ppm O2AMK-12  20  76 ‘ Hf 177 Hf’  WC agate mean ppm  O2AMK-2  L unL mean ppm O2AMK-7 L unL  mean ppm  Sample  76 ‘ Hf 77 Hf/’  2a  procedural duplicates (separate digestions) 0.283108 0.000005 O2AHU-13 #1 0.283119 0.000005 #2 0.283113 0.000016 mean 57 ppm 0.283126 0.000004 O2AKA-1 #1 0.283127 0.000004 #2 0.283126 0.0000003 mean 1 ppm  0.283110 0.283113 0.283112 14 0.283133 0.283133 0.283133 1  0.000004 0.000004 0.000004  WC/agate duplicates (separate powders) 0.283138 0.000007 O2AKA-6 WC 0.283131 0.000006 agate 0.283135 0.000009 mean 32 ppm  0.283109 0.283119 0.283114 47  0.000004 0.000005 0.000013  leached/unleached duplicates (separate digestions) 0.283 127 0.000006 O2AMK-4 L 0.283 124 0.283126 0.000004 0.283133 unL 0.283126 0.000003 0.283128 mean 45 9 ppm 0.283 125 0.000005 0.283124 0.000005 0.283 125 0.000002 7  0.000004 0.000005 0.0000003  0.000005 0.000005 0.000013  Hf isotopic ratios were determined by MC-ICP-MS and have been normalized to JMC 475 Hf’ 76 ‘ Hf 177 0.282160 [ Vervoort and Blichert-Toft 1999]. The 2a error is the absolute error value of an individual sample analysis (internal error). For the mean values, the error reported is the 2 standard deviation (external error). External reproducibility is represented by the 2 standard deviation from the mean of duplicate analyses and is ). 6 reported in ppm (2SDImean x10 ,  117  No systematic differences in Pb isotopic composition were observed between powders that were crushed in tungsten carbide (WC) and those that were crushed in agate (Table A3), contrary to what is observed for trace elements and Hf isotopes (see below). This is likely due to the removal of potential contaminants through acid-leaching of the powders. This is consistent with Weis et a!. [2005] who found that differences between the USGS reference materials BHVO-1 (contaminated during processing) and BHVO-2 were significantly reduced after leaching. However, the difference between powders crushed in different materials (WC vs. agate) reaches up to 286 ppm, which is significantly larger than the external precision of powders that were crushed in the same material (Table A3). The poorer reproducibility of the WC/agate duplicates may reflect heterogeneity in the different powder splits or incomplete removal of contaminants via leaching. This emphasizes the importance of careful sample crushing and leaching in obtaining high-precision Pb isotopic compositions [e.g., Weis eta!., 2005; Nobre Silva eta!., in revision, 2008]. Contamination due to crushing in WC is a potential problem for Hf isotopes due to the isobaric interferences of Ta 80 and W ‘ 80 on Hf. ‘ 80 The effect of Ta ‘ 80 and W ‘ 80 can be ‘ assessed by comparing the Hf’ 80 values of the samples to the JMC 475 Hf standard, ‘ Hf 177 which gave an average Hf/’ 80 of 1.886837 ± 0.000099 (2sd; n = 175). The samples ‘ Hf 77 crushed in agate gave Hf7’ 80 1.886883 ± 0.000092 (2sd; n = 36), whereas those ‘ Hf= 77 crushed in WC gave 180 Hf= 1.892 ± 0.018 (2sd; n = 7), indicating the presence of 77 Hf7’ variably elevated amounts of Ta and W. This was confinned by their anomalously high trace element abundances (up to —46.5 ppm Ta and ‘--3100 ppm W) compared to the samples crushed in agate (average —2 ppm Ta and —1 ppm W). Some of the additional Ta and W will be removed during sample purification, and any remaining Ta and W will have a relatively  118  minor effect due to the size of the 18 °Hf peak and the low natural abundances of Ta 80 ‘ (O.O 1%) and ‘(O. 1%). Regardless, any potential interference on Hf 80 does not affect ‘ the reported ratio of 176 Hf. No systematic differences in Hf isotopic composition or 77 Hf” reproducibility were observed between the  wc- and agate-crushed powders (Table A4).  Similarly, the Hf isotopic compositions of leached and unleached powders of the same sample show excellent agreement (Table A4). This reflects the low susceptibility of Hf isotopes to modification by secondary alteration and demonstrates that, unlike for Pb and Sr, samples do not need to be acid-leached prior to Hf isotopic analysis.  119  A4. References  Abouchami, W., S. J. G. Galer, and A. W. Hofmann (2000), High precision lead isotope systematics of lavas from the Hawaiian Scientific Drilling Project, Chemical Geology, 169, 187-209. Albarède, F., P. Telouk, J. Blichert-Toft, M. Boyet, A. Agranier, and B. Nelson (2004), Precise and accurate isotopic measurements using multiple-collector ICPMS, Geochimica et Cosmochimica Acta, 68(12), 2725-2744. Eisele, J., W. Abouchami, S. J. G. Galer, and A. W. Hofmann (2003), The 320 kyr Pb isotopic evolution of Mauna Kea lavas recorded in the HSDP-2 drill core, Geochemistry Geophysics Geosystems, 4(5), doi: 10.1 029/2002GC000339. Nobre Silva, I. G., D. Weis, J. Barling, and J. S. Scoates (2008), Leaching systematics for the determination of high-precision Pb isotope compositions of ocean island basalts, Geochemistry Geophysics Geosystems, paper # 2007GC00 1891 (in revision). Pretorius, W., D. Weis, G. Williams, D. Hanano, B. Kieffer, and J. S. Scoates (2006), Complete trace elemental characterization of granitoid (USGS G-2, GSP-2) reference materials by high resolution inductively coupled plasma-mass spectrometry, Geostandards and Geoanalytical Research, 30(1), 39-54. Raczek, I., B. Stoll, A. W. Hofmann, and K. P. Jochum (2001), High-precision trace element data for the USGS reference materials BCR- 1, BCR-2, BHVO- 1, BHVO-2, AGV- 1, AGV-2, DTS-1, DTS-2, GSP-1 and GSP-2 by ID-TIMS and MIC-SSMS, Geostandards Newsletter, 25(1), 77-86. Robinson, P., A. T. Townsend, Y. Zongshou, and C. Munker (1999), Determination of scandium, yttrium and rare earth elements in rocks by high resolution inductively coupled plasma-mass spectrometry, Geostandards Newsletter: The Journal of Geostandards and Geoanalysis, 23(1), 3 1-46. Weis, D., B. Kieffer, C. Maerschalk, J. Barling, J. deJong, G. A. Williams, D. Hanano, W. Pretorius, N. Mattielli, J. S. Scoates, A. Goolaerts, R. M. Friedman, and J. B. Mahoney (2006), High-precision isotopic characterization of USGS reference materials by TIMS and MC-ICP-MS, Geochemistry Geophysics Geosystems, 7(8), doi: 10.1029/2006GC001283. Weis, D., B. Kieffer, C. Maerschalk, W. Pretorius, and J. Barling (2005), High-precision PbSr-Nd-Hf isotopic characterization of USGS BHVO-1 and BFIVO-2 reference materials, Geochemistry Geophysics Geosystems, 6(2), doi: 10.1 029/2004GC000852. White, W. M., F. Albarêde, and P. Telouk (2000), High-precision analysis of Pb isotope ratios by multi-collector ICP-MS, Chemical Geology, 167, 257-270. Willbold, M., and K. P. Jochum (2005), Multi-element isotope dilution sector field ICP-MS: A precise technique for the analysis of geological materials and its application to geological reference materials, Geostandards and Geoanalytical Research, 29(1), 63-82. Wilson, S. A. (1997), The collection, preparation, and testing of USGS reference material BCR-2, Columbia River, basalt, Open-File Report, United States Geological Survey, Denver, CO. Wilson, S. A. (1998), Data compilation and statistical analysis of intra-laboratory results for AGV-2, Open-File Report, United States Geological Survey, Denver, CO.  120  Woodhead, J. D. (2002), A simple method for obtaining highly accurate Pb-isotope data by MC-ICPMS, Journal ofAnalytical Atomic Spectrometry, 17, 1381-1385.  121  APPENDIX B  SUPPLEMENTARY SAMPLES  122  Bi. Mahukona basalts Mahukona is a submerged volcano located off the northwest coast of the island of Hawaii, --50 km west of Kohala (Figure Bi). The proposed existence of Mahukona as an independent Hawaiian shield volcano [Moore and Cambell, 1987] was not confirmed until recently [Garcia et al., 1990], and as a result much debate still exists over the location of its summit, whether or not it reached sea level, and what growth stage it was in when it became extinct [Garcia et aL, 1990; Clague and Moore, 1991]. Mahukona is of particular importance because it fills the gap on the Loa trend, between Hualalai and Kahoolawe, in the paired sequence of Hawaiian volcanoes [Dana, 1849; Jackson et al., 1972]. The location and age of Mahukona permits comparison with the adjacent Kohala volcano (on the Kea trend) and the younger Hualalai volcano (on the Loa trend), thus providing both spatial and temporal constraints on the evolution of volcanism related to the Hawaiian mantle plume. To better characterize Mahukona volcano, 13 samples from Mahukona were selected for geochemical analysis (Figure B 1; Table B 1). Two samples (MA- 12 and MA-32) were collected by dredging with the R!V Atlantis II near the summit high of the volcano [Garcia et al., 1990]. One additional sample (F2-88-HW Dl 6-6) was collected by dredging with the RIV Farnella on the west ridge of Mahukona [Clague and Moore, 1991]. The remaining samples were collected by the Pisces V submersible during three separate dives on ridges that radiate to the west-southwest, southwest, and southeast from the summit high [Garcia et al., 1990; M. Garcia, personal communication]. All of the lavas recovered from Mahukona are basalts, which can be divided into two separate groups, tholeiitic basalts and transitional to weakly alkalic basalts (Figure B2).  123  Figure Bi: (a) Map of the Hawaiian Islands showing the sample locations of the submarine lavas from Mahukona, Hualalai, Kohala, and Penguin Bank. The dashed box indicates the area shown in (b). (b) Bathymetric map of the seafloor to the northwest of the island of Hawaii (after Garcia et al. [1990] and Clague and Moore [19911) showing the locations of submarine samples from Mahukona volcano and the northwest rift zone of Hualalai volcano. Contour interval is 400 m. Prominent ridges radiating from the summit high of Mahukona are shaded along their crests.  124  7 7 17 8 7 7 7 6 6 8 6 10 11 5  15.4683 15.4820 15.4670 15.4776 15.4703 15.4771 15.4719 15.4697 15.4699 15.4669 15.4665 15.4611 15.4680 15.4752  21 19 47 20  22 22 18 16 16 23 17 29 25 15  38.0624 38.1446 37.9719 38.1066  38.0952 38.1232 38.0636 38.0494 38.0495 37.9514 37.9512 37.9395 37.9750 3 7.9879  MA-12  MA-32 P5 159-1 P5 159-4 P5 159-6 P5 160-1 P5 160-2 P5 160-5 P5 160-6 P5 161-1  18.3795 18.1879 18.3412 18.3488 18.3661 18.3076 18.2944 18.2936 18.2983 18.2980 18.2913 18.1901 18.2967  18.3147  Pb 204 Pb/ 206  8 19 8 7 8 8 6 6 9 7 11 11 6  8  2a  0.513001  0.5 13013 0.512991 0.512911 0.512974 7 7 9 10  5 6 4 8 6 7  0.283088 0.283125 0.283125 0.283126 0.283067 0.283117  9  0.703687 0.703586 0.703581 0.703569 0.703776 0.703644  4  0.283084  9 7 7  0.703689  6  0.512945 0.512938 0.5 12932 0.512925 0.512939 9  7 6 6 6 6 6 6 7 7 6 7  6 6  0.512929 0.512907 0.512929  6  0.512942  7 8  0.703674 0.703769 0.703689 0.703699 0.703696 0.703682  2a  d 3 Nd/ 4 ‘ N 1  7  7  0.703686  5 4 4 7 4 4 4 5  0.283086 0.283090 0.283089 0.283063 0.283089 0.283095 0.283089 0.283092  2  Sr t6 SrI t7  2a  Hf 177 Hf7 176  aAli Pb and Hf isotopic ratios were determined by MC-ICP-MS; Pb data were corrected for fractionation by Ti spilcing and then normalized to the NBS 981 triple spike 76 0.282160 [ Vervoort and Blichert-Toft, 1999]. ‘ Hf= 77 values of Galer andAbouchami [1998] using the standard bracketing method; Hf data were normalized to JMC 475 Hf” d 43 ‘ N 1 Sr/Sr = 0.7 10248; Nd data were normalized to La Jolla Nd/ All Sr and Nd isotopic ratios were determined by TIMS; Sr data were normalized to SRM 987 87 43 = 0.5 10939. The 2o error is the absolute error value of an individual ‘ 0.511858, except for F2-88-HW D16-6, which was normalized to the Berkeley Ames standard Nd/Nd sample analysis (internal error) and applies to the last decimal place(s). Samples labeled “D’ represent procedural duplicate analyses carried Out on separate powder aliquots.  P5 16 1-2 P5 161-4 F2-88-HW D16-6  2o  Pb 204 Pb/ 207  2a  Pb 204 Pb/ 208  Sample  Table Bi: Pb, Hf, Sr, and Nd Isotopic Compositions of Mahukona Basaitsa  10  8  0 +  z 2  0  40  45  50  55  60  2 (wt%) Si0 Figure B2: Total alkalis vs. silica classification diagram (modified from Le Bas et al. [1986]) for the submarine lavas from Mahukona, Hualalai, Kohala, and Penguin Bank. The tholeiitic-alkalic dividing line is from Macdonald and Katsura [1964]. The majority of the Hawaiian submarine lavas are tholeiitic to transitional basalts.  126  Three of the samples were selected for 40 Ar dating at Oregon State University. 39 ArI The deepest of these samples, recovered from a depth of 1970 mbsl on the WSW rift, is a tholeiite with an age of 653  ±  74 ka. The shallowest sample, collected near the summit of  Mahukona (-1200 mbsl) on the SE rift, is a transitional basalt that yields the youngest age of 351  ±  32 ka. Another transitional basalt, located midway between the other two samples at  a depth of 1685 mbsl on the SW rift, has an intennediate age of 479 ± 75 ka. The significantly younger ages of the shallower and more alkaline basalts from Mahukona support the interpretation that these lavas erupted during the post-shield stage [Clague and Moore, 1991] rather than the pre-shield stage [Garcia et al., 1990]. The observed range of ages for tholeiitic to transitional volcanism on Mahukona coincides with the period of shield to late-shield volcanism (Polulu Volcanics) on the neighbouring Kohala volcano [Spengler and Garcia, 1988], and allows for direct comparison of the geochemistry of lavas from this pair of volcanoes. Sr, Nd, Pb, and Hf isotopic compositions (Table B 1) were determined at the PCIGR, University of British Columbia, using procedures described in Section 3.3 (see also Weis et al. [2006] and Weis et al. [2007] for complete chemical and analytical protocols). The Sr, Nd, and Hf isotopic compositions of the Mahukona basalts generally overlap the field Sr and highest 86 Sr/ defmed by Mauna Loa, except for three samples with the lowest 87 Nd and Hf” 144 NcJJ 143 76 that overlap with Mauna Kea (Figure B3). These Kea-like ‘ Hf 77 samples also have distinct Pb isotopic compositions and plot on the Kea side of the Pb isotopic division in Pb-Pb space [Abouchami et al., 2005] (Figure B4). The majority of the Mahukona basalts, however, are characteristically Loa-like in their Pb isotope systematics. The transitional to weakly alkalic basalts extend to more radiogenic compositions  127  0.5132  (a)  0.5131 veated  0.5130  z  MaunaK-ea  e  t.. 0.5129  MaunaLoa 0.5128  0.5127  nMahukona OHualalai AKohala KPenguin Bank  0.5126 0.7024  i 8 iin N  I  I  I  0.7036  0.7040  0.7044  I  0.7028  0.7032  0.7048  Sr 86 Sr! 87 16  I  I  I  I  I  I  I  I  (b) 14 Mauna Kea 12 Mauna Loa 10  J  pt t  8  Lan1  .—  6 Koolau c’ DMahukona OHualalai A Kohala ‘DPenguinBank  2 0  I  -2  -1  0  I  I  I  I  1  2  3  4  5  6  7  8  9  Nd 8 Figure B3  128  10  Figure B3: (a) 143 Nd vs. 87 1 NW Sr and (b) E vs. £Nd for for the submarine basalts from 86 SrI Mahukona, Hualalai, Kohala, and Penguin Bank compared to selected Hawaiian shield stage lavas. Also shown in (a) are fields for Hawaiian rejuvenated stage lavas and EPR MORB. The Hawaiian and OIB arrays in (b) are from Blichert-Tofi et al. [1999]. Data sources are the same as in Figure 3.8.  129  206 =18.38; 208 ( Pb 204 Pb/ Pb = 38.14) than the tholeiites and overlap with the field for 204 Pb/ Loihi (Figure B4). The Loa-like isotopic compositions of the majority of the Mahukona basalts, and in particular their similarity to lavas from Mauna Loa and Loihi, confirm that Mahukona is compositionally a Loa-trend volcano, consistent with its location on the Loa spatial trend. The exclusively Loa-like Pb isotope systematics of the transitional to weakly alkalic basalts preserve the Loa-Kea Pb isotopic division [Abouchami et al., 2005] (Figure B4). If these lavas belong to the post-shield stage, as suggested by the age relationships, this implies that the Loa source extends to the plume margin and supports the bilaterally zoned plume model [Abouchami et aL, 2005; Xu et al., 2007] (Figure 3.12). This is consistent with post-shield lavas from Hualalai, Mauna Kea, and Kohala, which retain their respective Loa- and Kea like Pb isotope signatures (see Section 3.5.3 for discussion). The three anomalous Kea-like samples are tholeiites and are the deepest samples in the study (—2000-2700 mbsl), suggesting that they may belong to the adjacent Kohala volcano. This would imply that the large submarine feature interpreted to be Mahukona volcano instead corresponds to Kohala’s westward-trending rift zone. In such a scenario, the Loa-like summit high may represent a large (post-shield?) cone developed on this rift zone. Alternatively, a more plausible explanation is that the compositions of these lavas reflect sampling of the Kea source by this Loa-trend volcano. The sampling of source material from the adjacent trend has been documented at numerous other Hawaiian volcanoes [e.g., Mauna Kea: Eisele et al., 2003; Haleakala: Ren et at., 2006; Kilauea: Marske et a!., 2007; West Molokai: Xu et a!., 2007]. This indicates considerable heterogeneity in the melting region and suggests that the Hawaiian plume has more complex  130  38.3  I  I  I  I  I  38.2  I  I  Loihi Mauna Kea  38.1 Mauna Loa  0.. 38.0  ,D.  o  Hualafai 0_ 37.9 0  ala 37.8  Koo  I E Mahukona I 0 Hualalai  37.7 A 4 LO 37.6 17.7 17.8 17.9 18.0 18.1  ILsKohala Penguin Bank 18.2 18.3 18.4 18.5 18.6 18.7 18.8  Pb 204 Pb! 206 Pb vs. 206 4 PbP° Figure B4: 208 Pb for the submarine basalts from Mahukona, Hualalai, 4 PbP° Kohala, and Penguin Bank compared to selected Hawaiian shield stage lavas. The thick black line represents the Loa-Kea Pb isotopic division defined by Abouchami et al. [2005]. Data sources are the same as in Figure 3.8.  131  geochemical structure than predicted by either the concentrically or bilaterally zoned models [e.g., Hauri, 1996; Kurz et al., 1996; Lassiter et a!., 1996; Abouchami eta!., 2005]. Comparison of the Pb isotope systematics of Mahukona with Kohala, its neighbour on the Kea-trend, shows that lavas from both volcanoes become more radiogenic in Pb with time (Figure 3.11 a). In contrast, the younger pair of volcanoes, Hualalai and Mauna Kea, shows the opposite relationship and become less radiogenic in Pb during the shield to postshield transition (Figure 3.11 b). The sampling of isotopically distinct components by these consecutive pairs of volcanoes suggest that, in addition to any horizontal zoning, vertical heterogeneities within the upwelling plume also contribute to the geochemical variability observed in Hawaiian lavas [Blichert-Toft et a!., 2003]. During the late stages of volcanism at Mahukona and Kohala, a contribution from an enriched source with radiogenic Pb isotopic compositions is required. Although lavas from these two volcanoes could share one Pb 204 Pb/ common component, the linear Pb-Pb array of Mahukona trends towards higher 208 Pb than that of Kohala, indicating that each volcano sampled a different 204 Pb/ for a given 206 enriched component (Figure 3.11 a). Similarly, the sub-parallel Pb-Pb arrays of Hualalai and Mauna Kea suggest that more than one depleted component may be involved during the post-shield stage (Figure 3.1 ib). The depleted signature (relatively unradiogenic Sr and Pb isotopic compositions) identified in post-shield lavas from Hualalai and Mauna Kea is distinct from the MORB source (see Section 3.5.2 for discussion). The observed isotope systematics of lavas from these volcanoes reflect derivation from a heterogeneous source with both enriched and depleted components of varying composition from within the Hawaiian plume.  132  B2. Submarine Hawaiian basalts from Hualalai, Kohala, and Penguin Bank Ten additional samples collected from the NW rift of Hualalai, NE flank of Kohala, and Penguin Bank, were also selected for isotopic analysis (Figure B 1; Table B2). All of the submarine samples are tholeiitic basalts (Figure B2), except for one picritic sample from Hualalai that has 3O wt% MgO. Pb and Hf isotopic compositions were determined at the PCIGR, University of British Columbia, and Sr and Nd isotopic compositions were determined at the CIG, University of California, Berkeley using procedures described in Section 3.3. Submarine basalts from Hualalai and Penguin Bank have Sr, Nd, and Hf isotopic compositions that plot within the field defined by Mauna Loa, whereas basalts from Kohala coincide with those from Mauna Kea (Figure B3). This is consistent with the location of Hualalai and Penguin Bank on the Loa-trend, and Kohala on the Kea-trend. The Pb isotopic compositions of the submarine lavas show less overlap and allow for better distinction between the three volcanoes (Figure B4). Submarine basalts from Hualalai are Loa-like and similar to subaerial shield lavas from Hualalai. In contrast, submarine basalts from Kohala are more radiogenic than their subaerial counterparts and trend towards the field for Mauna Kea. The Pb isotopic compositions of submarine basalts from Penguin Bank lie along the Loa-Kea Pb isotopic division [Abouchami et a!., 2005]. Due to the location of Penguin Bank west of the main Hawaiian Ridge, the geochemical characteristics of these lavas are critical for evaluating the concentric and bilateral models of plume zonation [e.g., Hauri, 1996; Kurz et a!., 1996; Lassiter et a!., 1996; Abouchami eta!., 2005]. The Loa-like Sr, Nd, and Hf isotopic compositions of Penguin bank lavas are consistent with a bilaterally zoned plume. However, the Pb isotopic  133  7 6 5  15.4618 15.4635 15.4730  37.8579 37.8631 37.9613  P5255-7 P5255-8 P5255-Il  18.1587 18.1606 18.2985  Penguin Bank  18.5563 18.3026 18.3029 18.3867  Kohala (NE flank)  18.2048 18.2107 18.2647  18.2583 18.2402  Hualalai (NWrft)  Pb 204 Pb/ 206  9 6 5  7 9 6 10  7 6 6  11  53  2a  6 5 5  0.703813 0.703775 0.703706  24 11 8  11  0.703674 6  0.283098 0.283091 0.283087 0.283104  8 8  0.703620 0.703620  4 6  8  0.703773  6  0.283080 0.283129 0.283 122  11 8 10 10  2cs  0.703758 0.703660 0.703696 0.703742  Sr 86 Sr/ t7  6 6 5 5  2a  0.283084 0.283092 0.283090 0.283088  6 Hf/ 7 ‘ Hf 177  0.512932 0.512935 0.512968  0.512931  4 6 9  5  8 6  6  0.512906 0.512975 0.5 12950  6  6 6 11  2a  0.512931  0.512956 0.512952 0.512918  3 Nd,” 4 ‘ Nd 44  eAll Pb and Hf isotopic ratios were determined by MC-ICP-MS; Pb data were conected for fractionation by TI spiking and then normalized to the NBS 981 triple spike values 76 = 0.282160 [Vervoort and Blichert-Toft, 1999]. All Sr ‘ 1{f 77 andAbouchami [1998] using the standard bracketing method; Hf data were normalized to JMC 475 Hf7’ Galer of 43 ‘“N ‘ Sr = 0.7 10248; Nd data were normalized to the Berkeley Ames standard Nd/ 6 Sr/’ and Nd isotopic ratios were determined by TIMS; Sr data were normalized to SRM 987 87 = 0.510939. The 2a error is the absolute error value of an individual sample analysis (internal error) and applies to the last decimal place(s). Samples labeled “D” represent procedural duplicate analyses carried out on separate powder aliquots.  P5407-4 19 13 13  6 8 7 9  15.4898 15.4676 15.4695 15.4771  15 19 19 27  38.1322 37.9780 37.9816 38.0666  T302R19 P5 406-12  F2-88-HWD27-2  7 6 7  15.4676 15.4700 15.4626  19 14 19  15 10  2a  37.9998 38.0095 38.0343  F2-88-HWD26-1  15.4649  Pb 204 Pb/ 207  15.4634  37.9392 37.9897  KK 10-1 KK 14-7  2a  38 26  Pb 204 Pb/ 208  Sample  Table B2: Pb, Hf, Sr, and Nd Isotopic Compositions of Submarine Hawaiian Basaitsa  compositions are not explicitly Loa- or Kea-like and prevent a definitive distinction between the two models. This may support the hypothesis that the Loa-Kea compositional trends do not extend beyond the Molokai Fracture Zone [Abouchami et al., 20051. The Molokai Fracture Zone is a complex large-offset fracture zone that coincides with a change in the  strike of Pacific plate motion, and may have had a significant influence on Hawaiian volcanism and the plume components sampled by different Hawaiian shields [e.g., Basu and Faggart, 1996; Abouchami et al., 2005; Xu et al., 2007].  135  B3. References Abouchami, W., A. W. Hofinann, 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, 85 1-856. Basu, A. R., and B. E. Faggart (1996), Temporal isotopic variations in the Hawaiian mantle plume: The Lanai anomaly, the Molokai Fracture Zone and a seawater-altered lithospheric component in Hawaiian volcanism, in Earth Processes: Reading the Isotopic Code, edited by A.R. Basu, and S.R. Hart, pp. 149-159, AGU, Washington, D.C. Blichert-Toft, J., D. Weis, C. Maerschalk, A. Agranier, and F. Albarède (2003), Hawaiian hot spot dynamics as inferred from the Hf and Pb isotope evolution of Mauna Kea volcano, Geochemistry Geophysics Geosystems, 4(2), doi: 10.1 029/2002GC000340. Clague, D. A., and J. G. Moore (1991), Geology and petrology of Mahukona volcano, Hawaii, Bulletin of Volcanology, 53, 159-172. Dana, J. D. (1849), Geology, in UnitedStates Exploring Expedition, 1838-1842, vol. 10, pp. 756, C. Sherman, Philadelphia. Eisele, J., W. Abouchami, S. J. G. Galer, and A. W. Hofmann (2003), The 320 kyr Pb isotopic evolution of Mauna Kea lavas recorded in the HSDP-2 drill core, Geochemistry Geophysics Geosystems, 4(5), doi: 10.1 029/2002GC000339. Garcia, M., M. D. Kurz, and D. W. Muenow (1990), Mahukona: The missing Hawaiian volcano, Geology, 18, 1111-1114. Hauri, E. H. (1996), Major-element variability in the Hawaiian mantle plume, Nature, 382, 415-419. Jackson, E. D., E. A. Silver, and G. B. Dalrymple (1972), Hawaiian-Emperor chain and its relation to Cenozoic circum-Pacific tectonics, Geological Society ofAmerica Bulletin, 83, 601-618. Kurz, M. D., T. C. Kenna, J. C. Lassister, and D. J. DePaolo (1996), Helium isotopic evolution of Mauna Kea volcano: First results from the 1 -km drill core, Journal of Geophysical Research, 101, 11,781-11,793. Lassiter, J. C., D. J. DePaolo, and M. Tatsumoto (1996), Isotopic evolution of Mauna Kea Volcano: results from the initial phase of the Hawaii Scientific Drilling Project, Journal ofGeophysical Research, 101(B5), 11,769-11,780. Marske, J. P., A. J. Pietruszka, D. Weis, M. 0. 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 and Planetary Science Letters, 259, 34-50. Moore, J. G., and J. F. Cambell (1987), Age of tilted reefs, Hawaii, Journal of Geophysical Research, 92, 2641-2646. Ren, Z.-Y., T. Shibata, M. Yoshikawa, K. T. M. Johnson, and E. Takahashi (2006), Isotope compositions of submarine Hana Ridge lavas, Haleakala volcano, Hawaii: Implications for source compositions, melting process and the structure of the Hawaiian plume, Journal ofPetrology, 47(2), 255-275. Spengler, S. R., and M. 0. Garcia (1988), Geochemistry of the Hawi lavas, Kohala volcano, Hawaii, Contributions to Mineralogy and Petrology, 99, 90-104. Weis, D., B. Kieffer, D. Hanano, I. Nobre Silva, J. Barling, W. Pretorius, C. Maerschalk, and N. Mattielli (2007), Hf isotope compositions of U.S. Geological Survey  136  reference materials, Geochemistry Geophysics Geosystems, 8(6), doi: 10.1 029/2006GC00 1473. Weis, D., B. Kieffer, C. Maerschalk, J. Barling, J. deJong, G. A. Williams, D. Hanano, W. Pretorius, N. Mattielli, J. S. Scoates, A. Goolaerts, R. M. Friedman, and J. B. Mahoney (2006), High-precision isotopic characterization of USGS reference materials by TIMS and MC-ICP-MS, Geochemistiy Geophysics Geosystems, 7(8), doi: 10.1 029/2006GC00 1283. Xu,, G., F. A. Frey, D. A. Clague, W. Abouchami, J. Blichert-Toft, B. Cousens, and M. Weisler (2007), Geochemical characteristics of West Molokai shield- and postshield stage lavas: Constraints on Hawaiian plume models, Geochemistry Geophysics Geosystems, 8, doi: 10.1 029/2006GC00 1554.  137  

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.24.1-0052325/manifest

Comment

Related Items