THESES SIS/LIBRARY TELEPHONE: +61 2 6125 4631 R.G. MENZIES LIBRARY BUILDING NO:2 FACSIMILE: +61 2 6125 4063 THE AUSTRALIAN NATIONAL UNIVERSITY EMAIL: [email protected] CANBERRA ACT 0200 AUSTRALIA
USE OF THESES
This copy is supplied for purposes of private study and research only. Passages from the thesis may not be copied or closely paraphrased without the written consent of the author. COMPARATIVE GEOCHEMISTRY OF SOME VOLCANIC SUITES OF THE SOLOMON ISLANDS AND BOUGAINV ILLE: IMPLICATIONS FOR METALLOGENESIS
CROMWELL QOPOTO
BSc. University of Papua New Guinea, Pon Moresby, Papua ew Guinea
A THESIS SUBMITTED FOR THE DEGREE OF MASTER OF SCIENCE OF THE AUSTRALIAN NATIONAL UN IVERSITY
November 2002 STATEMENT
All the results presented in this thesis, and the conclusions drawn from them, are the author's own, except where otherwise indicated. The various persons who have assisted and guided the author are decailed in the acknowledgments, as are those who have helped in the preparation of this thesis. All sources are acknowledged in this thesis.
I certify that the substance of this thesis has not already been submitted for any degree, and is not currently being submitted for any degree or qualification.
Cromwell Qopoto
II COMPARATIVE GEOCHEMISTRY OF SOME VOLCANIC SUITES OF THE SOLOMON ISLANDS AND BOUGAINV ILLE: lMPLICA TIONS FOR METALLOGENESIS
TAB LE OF CONTENTS
Page number
Statement II Table of Contents 111
List of Figures VI
List of Tables IX Acknowledgements x
Abstract
Chapter 1: Introduction 3 I .0 Previous work done 7 1. l Tectonic Settings and Regional Geology 7 i Te1Tane Model l I ii. Tectonic Style 12 iii. Metallogenesis 12 1.2 Local Geology 14 l.2. 1 Bougainville 14 1.2.2 Fauro Island Group 15 1. 2.3 Choiseul 17 1.2.4 Guadalcanal 19 1.3 Petrologic and Geochemical Studies 21 1.3.J Petrologic classification of the Solomon Arc 2 1 1. Pre-Miocene Volcanic Basement 22 11. Ultramafic Basement 22 iii. Intrusive Bodies 23 iv. Plio-Pleistocene: calc-alkaline volcanic rocks 23 1.3.2 Solomon Islands Volcanic Suites r_.) l.3.3 Bougainville Volcanic Groups 26
iii C ha pter 2: Geochemical Studies of the Solomon Isla nds Arc Systems 27
2.1 Geochemistry of the Solomon Arc 27 2. 1. l Solomon Islands Arc geochemical studies 27 i. Harker Diagrams 27 ii. AFM Diagram 31 2. 1.2 Bougainvillc Volcanic Suites Geochemical Studies 31 2.2.1 Whole rock (bulk) Geochemistry 31 2.2.2 Trace Element (bulk) Geochemistry 45 2.2.2.1 alkalies, alkaline earths. transition clemen ts, and rare ea rth elements 45 2.2.2.2 REE: Chondrite-Normalised pauerns 47 2.2.2.3 Primitive Mantle-Normalised "Spider" diagrams 49
Chapter 3: Melt Inclusion Studies
3. 1 Introduction 55 3.2 Evolution and Origin of Melt Inclusions 55 3.3 Rock Associations and Petrographic Studies of M ls 59 3.4 Different Types of Mls 60 3.5 Types of data recoverable from the analysis of Mis 60 i. Trapping and closure temperature 60 ii. Constraints on maxi mun and minimum cooling rates 61 iii. Bulk composition and liquid line of descent 61 iv. Volatile content 61 3.6 Mis resulting from melt-fluid immiscibility 61 3.7 Studies of M Js in Selected Solomon Islands Volcanic Suites 62 3.7.1 Bougai nville Volcanic Suites 62 3.7.2 Major Element Compositions of M is and Host Minerals 64 3.7.3 Fauro Volcanic Suites 65 3.7.4 Major Element Compositions of MJs and Host Minerals 72 3.7.5 Choiscul Volcanic Suites 82 i. Maetambe 82 ii. Kumboro 88
iv ii. Kumboro 88 3.7.6 Major Element Compositions of Mls and Host Minerals 88 3.7.7 Gold Ridge (Guadalcanal) Volcanics 92 3.7.8 Major Element Compositions of Mls and Host Minerals 92 3.8 Comparison of major element compositions of the Mis and host crystals for Bougainville, Faure, Choiseul, and Gold Ridge 97
Chapter 4: Discussion, Conclusions, and Further Studies 4.1 The texture and nature of the Mls and their fractionation trends 105 4.2 Geochemical compositions of the Mis and matrix glasses 105 4.3 Whole rock geochemical compositions 106 4.4 Trace element geochemical compositions 106 4.5 Conclusions 107 4.6 Fu11her Studies 108
References 109
Appendix 1 119 Appendix 2 120
v List of Figures Page Number Figure l. Map showing the location of the Solomon Islands. Specific areas of study for this thesis are indicated by the aITows. 4 Figure 2. P.J. Coleman's province model for the Solomon Islands (e.g., Coleman 1965; 1966; 1970). 8 Figure 3. Terrane model of the Solomon Islands (after M. G. Petterson, 1999). 10 Figure 4. Generalised geological map of Bougainville and Buka Islands after Rogerson et al. (1989). l3 Figure 5. Aerial view and general geological map of the Fauro I sland Group. 16 Figure 6. A general geological map of Choiseul. 18 Figure 7. Map showing the location of the Gold Ridge Volcanics and Gallego. 20 Figure 8. Petrographic features of the volcanic rocks. 24 Figure 9. Harker diagrams for the Fauro sui te (after Turner and Ridgway, 1982). 28 Figure 10. Harker diagrams for selected volcanic basements of different I slands of the Solomon Islands. 30 Figure 11. A-F-M diagram for selected rock suites from the Solomon Island arc 32
Figure 12. Relationship between K 20 and Si02 wt% and locations of individual volcanoes on Bougainville. 33 Figure 13. Harker diagrams for selected volcanic suites from Savo, Gallego, Fauro, Choiseul , and Gold Ridge. 37 Figure 14. Plots of MgO wt% vs other major oxides for selected bulk analyses of suites from Savo, Gallego, Fauro, Choiseul, and Gold Ridge. 38
Figure 15. Plots of K20 wt% vs major oxides for selected bulk analyses of sui tes from Savo, Gallego, Fauro, Choiseul, and Gold Ridge. 39
Figure 16. Plots of K 20 + Na20 wt% vs major oxides for selected bulk Analyses of sui tes from Savo, Gallego. Fauro, Choiseul, and Gold Ridge. 40 Figure 17. Abundances of selected large ion lithophile, ferromagnesian and
chalcophile trace elements plotted against Si02 wt%. 41 Figure 18. Abundances of selected rare earth and high field strength elements
plotted against Si02 wt%. 42
Figure 19. K 20 vs. Si02 wt% for selected volcanic suites from the Solomon Islands and Bougainville. 43
Figure 20. Total alkalies vs. Si02 wt% for selected volcanic suites from the
VI Solomon ls lands and Bougainvi lie. 44 Figure 21. Rare earth element abundances normalised to chondritic meteorite values for selected rock types of the Fauro volcano. 46 Figure 22. Rare earth element abundances normalised to chondritic meteorite values for the calc-alkaline volcanoes of Choiseul (Kumboro and Maetambe) and si .licified volcanic breccias from the Gold Ridge Volcanics. 46 Figure 23. Rare earth element abundances normalised to chondritic meteorite Values for selected volcanic rocks from Bougainville. 48 Figure 24. Trace element abundances no1malised to "primitive mantle" for selected volcanic frocks from Fauro, Choiseul, and Gold Ridge. 5 J Figure 25. Trace element abundances no1malised to "primitive mantle" for selected volcanic frocks from Savo. 52 Figure 26. Trace element abundances normalised to " primitive mantle" for selected volcanic frocks from Gallego. 53 Figure 27. Trace element abundances normalised to "primitive mantle" for selected volcanic frocks from Bagana (Bougainville). 54 Figure 28. Photomicrographs of Mls in volcanics from Choiseul. 56 Figure 29. Photomicrographs of Mis in volcanics from Choiseul. 57 Figure 30. Photomicrographs of plagioclase- and clinopyroxene-hosted Mi s in volcanics from Choiseul. 58 Figure 31. Hypothetical outline of possible routes of magmatic differentiation. as interpreted from magmatic inclusion of melts. fluid and vapour. 63 Figure 32. Compatison of whole rocks and Mls hosted by va1ious phenocryst
Phases projected in terms of total alkalies vs Si02 wt%. 70 Figure 33 . Major oxides of Mls plotted against silica for various Bougainville volcanics. 71 Figure 34. Two well developed glass inclusions embedded inside an Fe-Ti oxide crystal. 73 Figure 35 . Mis trapped in a quartz phenocrysr. 73 Figure 36. Comparison of whole rocks, M i s hosted by various phenocryst phases, and manix glasses projected in terms of total alkalies vs
Si01 wt%, for Fauro. 80 Figure 37. Major oxide concentrations of Mis and mat1ix glasses from Fauro plotted against silica concentrations. 81
VII Figure 38. Mis hosted by various crystals wi thin a coarse grained pegmatite 87 Figure 39. MJs trapped in a plagioclase host (Maetambe). 87 Figure 40. Mis and vari ous crystalline phases (including Fe-Ti ox ides) trapped in phenocrysts of hornblende, biotite, and clinopyroxene. 89 Figure 41. Plots of major ox ide concentrations vs . silica of Mis and matrix glasses in volcanic rocks from Choiseul. 90 Figure 42. Comparison of whole rocks, Mls hosted by various phenocryst phases, and matrix glasses projected in terms of total alkalies vs SiO ~ wt%, fo r Choiseul. 9 l Figure 43. Major oxide compositions of Mis and matrix glasses plolted against si lica for Gold Ridge. 95 Figure 44. Comparison of whole rocks, Mis hosted by various phenocryst phases, and mat1ix glasses projected in terms of total alkal ies vs
Si02 wt%, for Gold Ridge. 96 Figure 45. Pl ots of major oxides vs SiO ~ wt% for Mls hosted by clinopyroxene. 98
Figure 46. Pl ots of major ox ides vs Si02 wt% for Mls hosted by plagioclase. 99
Figure 47. Pl ots of major ox ides vs Si02 wt% fo r Mis hosted by Fe-Ti oxides. 100
Figure 48. Plots of major oxides vs Si02 wt% fo r Mls hosted by quartz. lO I
Figure 49. Plots of major ox ides vs Si01 wt% for matrix glasses. 102
Figure 50. Plots of major oxides vs Si02 wt% for Mis hosted by amphibole and mica. 104
vi ii List of Tables Page Number
Table lA. Analyses of igneous rocks from Fauro. 29 Table l B. Analyses of basaltic roc ks from various areas of the Solomon Islands. 29 Table 2. Bulk major and trace element analyses of Choiseul and Gold Ridge. 35 Table 3. Bulk major and trace element analyses of Fauro. 36 Table 4. Mls trapped in clinopyrexene phenocrysts (Bougainvi lle). 66 Table 5. M l s trapped in plagioclase and quartz phenocrysts (Bougainville). 67 Table 6. MJs trapped in Fe-Ti oxide phenocrysts (Bougainville). 68 Table 7. Mls trapped in clinopyroxene phenocrysts (Fauro). 74 Table 8. Mls trapped in plagioclase phenocrysts (Fauro). 75 T able 9. Mis trapped in Fe-Ti oxide phenocrysts (Fauro). 76 Table 10. Mis trapped in sulfide phenocrysts (Faure). 77 Table 11. Mis trapped in amphibole, mica, and qua 11z phenocrysts (fauro). 78 Table L2. Represenrative analyses of matiix glasses (Faure). 79 Table L3 . M fs trapped in Fe-Ti oxide phenocryscs (Choiseul). 83 Table 14. Representative analyses of matrix glasses (Choi seul). 84 Table 15. Mls trapped in clinopyroxene and amphibole phenocrysts (Choiseul). 85 Table L6. Mls trapped in plagioclase phenocrysts (Choiseul). 86 Table 17. Mis trapped in plagioclase and quartz phenocrysts (Gold Ridge). 93 Table 18. M fs trapped in su lfide phenocrysts and matrix glasses (Gold Ridge). 94
IX ACKNOWLEDGEMENTS I would like to thank those who have supported my work and contributed many ideas that made my work and study very enjoyable. Firstly, my thanks go to Professor Richard J. Arculus for his experienced and excellent supervision; his understanding towards my work, the major editing of various draft versions of thi s thesis , and the assistance he has provided during the duration of my study; without his help, this thesis could not have been completed.
Secondly, I would like to thank Dr Michael G. Petterson, for his help and contribution of the Savo and Gallego geochemical data, hi s time in drafting a number of the diagrams, editing. commenting on, and improving this manuscript. Mike's proposals and comments have contributed significantly to my work.
Thirdly, my thanks go to the staff and students of the Department of Geology at the Australian National University, for their academic and technical support. Dr. Steve Eggins kindly provided data from the Inductively-Coupled Plasma Source Mass Spectrometer at the Resea rch School of Earth Sciences, Dr. Ian Parkinson helped with initial geochemical and petrological interpretations, Dr. Ulrich Senff and Mr. Tony Phimphisane provided high-quality X-ray fluorescence analyses. and Mr. John Vickers gave excellent technical supervision in preparing thin- and polished section .
Fourthly, my thanks goes to the Electron Microscope Unit at the Research School of Earth Sciences and the Sca nning Electron Microscope at the Research School of Biological Sciences . for providing excellent microprobe facilities. Fifthly, my thanks goes to the Australian National University for providing a Masters scholarship, Richard · Arculus (via his ARC grants) and the Department of Geology for their considerable financial supporc during my studies.
Sixthly, I would like to thank the Solomon Islands Government Department of Mines and Minerals; the landowners, and the Chief of Faure Islands and Choiseul Jslands for the understanding and support provided during my field expeditions.
And finally, but by no means least. my thanks go to my wife, Guthrie Qopoto. and two sons. Tony and Steve, and my parents for their encouragement, support, and time they have sacrificed to enable me to achieve an academic goal. and complete thi s thesis
x ABSTRACT
The Solomon Islands arc a complex collage of crustal units or tcrrancs (collectively called the Solomon Block) that have formed and accreted within an intra-oceanic environment during the Cretaceous and Tertiary Period ·. The Solomon Block has recently being subdivided into five terrane ba cd on: basement lithology and geochemistry. age, and development (or lack) of younger. arc-dominated, basaltic basement . cqucncc . A combination of terranc assessment coupled with the ongoing magmatic and tectonic developments in the Solomon Islands has direct implications for understanding a number or mctallogenctic processes.
The study of glass inclusions (or formerly melt inclu ions; Mis) in phcnocrysts provide direct information about the chemical composition of the naturally-occurring. supra subduction zone-derived melts within an overall transpressional environment. MJs have been trapped in variou phenocrysts of clinopyroxenc. plagioclase. quartz. amphibole, iron titanium oxides, and su lfides . This study was directed primarily at documenting and comparing bulk rock. MJs. and matrix glass compositions for calc-alkaline volcanoes of the islands of Bougainville. Fauro, Choiscul, and Guadalcanal.
The Mis occur a isolated blebs and sometimes in groups with preferred alignment . and range in individual Sii'.C from< I µm to 250 µm in length. Variable rates of crysral growth coupled with magmatic cooling arc involved in the formation of such glass inclusions.
Overall, the compositions of the Mrs for each island range from basaltic (<53 wt% SiO ~ ) through intermediate Si01 (53-60 wt%) to highly silicic (>60 wt% SiO ~ ). For all Mrs. the oxides of Al-Mg-Fe-Ti-Mn (and generally Ca) decrease while those of 1a and K increase as the content of Si01 increases. The Gold Ridge rocks in pnrticular have very high K concentrations.
For most or the sample su ites, the light rurc earth clements (REE) arc enriched relative lo the heavy REE. similar to many medium- to high-K volcanic arc suites world-wide. The fluid mobile clements (e.g., Sr, K , Rb. and Ba) arc enriched relative to REE or similar mell crystal incompatibilities, while the nominally fluid-immobile clement (e.g., Th, Ta. l\b. Zr, H f. Ti . Y. and Yb) arc similar in abundances to REE of equivalent incompatibility/compatibility. I n the future, more refined targe ting of potentially mineralised, volcanic rock-hosted ore occurrences wil I require knowledge or terrane (and hence basernenl geological) history, coupled with current tectonomagmatic scuings.
2 C HAPTER 1: I TnODUCT IO
The original objective of th is study was to examine a number of met<1llogcnic processes that can have a major influence on metal deposit geology in a subduction zone setting. The Solomon lslands together with Bougainville were chosen as a natural laboratory for thi study. The Solomon Islands consist of six mujor islands. trending NW-SE, in u double chain of island : Choi cul, the New Georgia Group, Santa lsabcl. Guadalcanal. Ylalaita and Makira (formerly called San Cristobal) as hown in Figure I. 13ougainville is clearly geotectonically part of this segment of the SW Pacific arc-trench systems.
The Solomon Islands arc ituatcd within the Greater Melanesian Arc, which includes Papua New Guinea (PNG), Solomon fslands, and Vanuatu. An understanding of the relationship between tectonic settings, geological evolution, and base metal depo its arc fundamental to mineral exploration. To date there have been few published studies of metal deposits in the Solomon Islands, especially those referring to fluid inclusions and M is. A notable exception however, is the monograph by Stanton ( 1994) on the behaviour of ore elements in arc lavas with specific reference to the cw Georgia Group. Stanton ( l 994) discussed the behaviour of ore and associated clements during the crystallization or arc magma serie in an attempt to determine the origin(s) of exhalative ore deposits.
For Bougainville (PNG), fluid, and Mis trapped inside phenocrysts of young volcanic rocks have been investigated and studied by several authors including Taylor ( 1974) 1 ash ( J 976) and Eastoe ( 1982). Eastoe ( 1978) for example, proposed that the Panguna porphyry copper deposit was genetically associated with a boiling salt-rich liquid of magmatic origin.
This study presents and models/interprets a range of geochemical data, including: petrographic observations: whole rock major and trace clement concentration : geochemical data derived from phenocry ts, groundmass glasses and M rs. The calc-alkaline volcanic rocks studied were from: Bagann (Bougainville I sland); rauro: Choiseul, Savo. and Guadalcanal (f-igurc I ). The nssociation of hydrothermal and ca lc-alkaline activity coupled with mineralisation (primarily Cu-Au) was the original reason for the selection of these suites. However, us the project progressed. it became clear that the primary objective should be a thorough geochemical and petrological characterisation of these suires in the context or the improved tectonic knowledge of the islands. The major active volcanic .~ \ North \ ... ~_Qntong Java reef ... ~ >
• F.wro caldora t '• . / Choiseul . .:tr ~Maotamoo volcanic ~ / Kumboto volcamc <;:> o.. .. ~~ Shortlands ~ "'o Santa Ysabel ~ '-\.© • 1l "-'eti-' "'l . -· . Ge ,{') O;,.,. ~~ <:llq G. Russels Slivo YO#cnno 'oll:0 ~ / Go/logo volcanics-~ Gold Ridgo volcllnics ~.. Guadalcanal 0 ... Maki ~ (San Cristobal) '
PNG \., Solo11 on Islands .'-, ~ ...... ~ , Rennel & Bellona Vanuatu Fiji
New Caledonia O miles 500 0 km 160 AUSTRALIA
Figure l. Map showing the location of the Solomon Islands. Specific areas of study for this thesis are indicated by the arrows.
4 province of the lcw Georgia Group is currently the object or a separate detailed study by Professors R. J. Arculus and R.L. Stanton.
Sample Locations The locations \\'here volcanic rocks \\'ere collected and subsequently analysed as part or this study arc sho vvn i n Figure I . T he author has personally visited all of the localities described in th is thesis with the exception of those on f3ougainville. The volcanic rocks of Fauro Island in the hortland Islands group \\'ere collected from the northern tip of the remnant part of a sunken caldera (sec also fig. 5). On the island of Choiscul. the rock samples were collected while traversing along the coast as well as following rivers that drain from the itactambc and Kumboro volcanic centres. The rock samples collected from the Gold Ridge Volcanics of central Guadalcanal are mostly somewhat silicified ,·olcanic breccia. Rock samples from Bougainvillc were obtained from the collections of Dr. R. \V. Johnson or the A ustralian Geologica l Su rvey Organisation.
The major and trace clement bulk rock data for the prc-19-U, 1943-53. 1959-75 (and unkno" n age) Bagana eruptions \\'Cn~ taken from the BM R Journal of Australian Geology and Geophysics (Bullitudc. 1978). Additional data were taken from the Bougainvil l<.: Memoir (Rogerson ct al 1989). Bulk rock analytical data for samples colleclcd on Savo and Gallego (weslern Guadalcanal) were provided by Dr M G Petterson. a former senior geologist al lhc Solomon Islands Geological Survey.
Analytical wo r k Bulk rock analysis wa completed for major elements (on fused glas discs) and some trace (wilh pressed powder pellet ) u ing the automated X-ray fluorescence (XRF) spectromelcrs (PW 1400 and PW2400) in the Department of Geology (ANU). Whole rock analysis for the Bougainville volcanic rocks was previously undertaken at the Department of Geology (A. U) using the same PW 1400 spectrometer with the analytical method · of
1 orrish and Hutton ( 1969) and ~orrish and Chappell (1967,1977). Some trace element
(e.g .. REE. 1 b. Ta. Th. and U) were analysed by laser ablation, inductively-coupled plasma source mass spectometer (L A-fCP-MS) techniques at the Research School of Earlh Science · (RSES) of ANU.
5 The major geochemical composicions of the Mis and host phenocry ts were obtained by wavelength- and energy-dispersive electron microprobe analysis (EMPA) at RSES and by an ED S-equipped scanning electron microscope (SEM) at the Research School of Biological Sciences. Analyses were performed on si licate Mls that were exposed on the surface of polished thin sections.
6 1.0 Previous W ork Done
1.1 Tectonic Setting and Regional Geology
The geological setting of the Solomon Islands arc systems was first de cribcd by Coleman ( 1960 and 1970: sec Fig. 2). Coleman divided the Solomon Islands into four geological provinces: the Pacific Province (which includes Malaita and Ulawa); the Central Province (the islands of Makira, the bulk of Guadalcanal, the Florida Islands, Santa Isabel, and the bulk of Choiscul): the Volcanic Province (inc.: ludcs west Guadalcanal, Savo, the Russe l Islands, the cw Georgia Group, and the Shortlund Islands: and the Atoll Province which includes the atoll islands of Rennell. Bellona, Sikaiana and Oncong Java.
Thi province model has recently been recast using modern tcrranc modelling by Pellcrson ct al. ( 1999). This new model synthcsi cs basement and cover geology. tructurcs, geochemistry. and geochronology across the Solomon archipelago (Fig. 3).
The oldest basement within the oldest tcrrane arc Cretaceous plateau and -MORB ocean floor basalts with or without ultramafic complexes. This Cretaceous basement forms the 'keel" to the Solomon tcrranc collage (Table I i11 Pettersen et al., 1999). The development of superimposed arc-related terranes occurred in two main stages (Kroenke 1984; Coulson and Vedder 1986, Petterson ct al 1997, Table 2 and 3 of Petterson ct al 1999,): stage l arc (Eocene to Early Miocene); and stage 2 arc development from Late Miocene to the present day.
Bougainville is the large t island within the Solomon Islands (geologically- peaking) group, with an 8000-m deep submarine trench to its southwest (Fig. 2). T he island is aligned in a NW-SE direction and thus oriented similarly lO the re t of the Solomon Islands chain. Bougainvillc con isls mainly of Cainozoic volcanic and volcanic-derived sedimentary rocks together with early Miocene and Pleislocene limestones (Blake and Miezitis, 1967: Speight, 1967. i11 Bullitudc 1978).
The Solomon Island , 1cw Britain. Bougainville, and Vanualu have been commonly grouped together as the 'Greater Nlelancsian Arc' (e.g., Petterson ct al., 1999). This arc marks the collisional zone between the Australian and Pacific Plates (Fig. 3). The Solomon Islands are SLmoundcd by relatively deep ocean floor to the northca I and southwest
7 ~fo rth
'\ \
Cl Pacific Prcvince Cl Certral Prcvirce
Vcicar1cProvinces
Cl Atoll Prcv1nce _j_ Spre.:iding centre T 0 160 Active trench km
Figure 2. P.J. Coleman's province model for the Solomon I land (e.g .. Coleman 1965, 1966, 1970).
8 (Pcucrson el al 1999). The So/0111011 Block comprises a series of islands and submarine basins; ils sedimentary basins have accumulated sediments up to 4-7 km thick (Coleman 1989).
The Solomon Block is bounded by lwo trench systems, namely the North Solomon trench (presently largely inactive) lo lhe nonhcast and the 1cw Britain-San Cristobal trench to the southwest (Fig. 3). A large number of small and young ocean basins arc situated to the soulh and west of the Solomon block, and a trcnch-trcnch-tran form triple junction mark. the intersection of the Woodlark Basin spreading ridge-transform system "ith the Solomon Block.
Ocean floor spreading in the Woodlark Basin commenced at about 5 Ma. At the eastern end of the spreading ridgc-tran form system, and in particular along the Simbo Ridge, unusual high-Ti-Na basalts ha\C been erupted (Johnson ct al 1987: Crook and Taylor 199-f). The subduction of the Woodlark basin beneath the Solomon block has triggered (or in some ways has been associated with) some very important and significant geological events, including: L) the tectonic uplift of the Solomon block; 2) the production and development of picrites in the New Georgia area; 3) some leakage of calc-alkaline material occur from source regions north of the Simbo Ridge (\\C tern New Georgia Group) into the Woodlark basin: 4) an anomalously small volcanic arc -trench gap occurs (e.g., Kavachi. an active submarine volcano is situated only 30km north of the nominal suture zone (along strike with the San Cristobal Trench; Johnson and Tuni. 1987): 5) an increase in coupling between the Pncific and A ustralian plate margins (Dunkley, 1983. 1984: Crook and Taylor, 1994; and Pcllcrson cl al ., 1997).
The Cretaceous-aged Ontong Java Plateau (OJP), which is the largest extant oceanic plateau in the world (Alaska-sized), is si tuated north of rhe Solomon block. This giant plateau is estimated to be 36-42km thick (Furumoto ct al 1970: Hussong ct al 1979) and has a similar seismic st ructure to lhar of 'normal' Pacific oceanic crust. I lowever, its seismic structure has been thickened by a factor of five (Husong ct al.. 1979: Neal cl al., 1997). It was emplaced in two major eruptive episodes at 120 and 90 Ma (Mahoney ct al.. 1993).
The OJP formed as the result of a high volume of partial melting in the mantle. and a high
9 ~Otr th
Fiji ~ Rl.11nd & Bdlona ~ Ne.v Caledonk1 d1 5 0 mtft>s ? AUST~ ~: · ~ 0 km 160
N·M ORB Cre1accou; basomort plus D 1 8. 2 arcs + Sproodingcentro. diroction of spr!0\.1ding Cretaceous OJP basement no D arc oevclopmonl Cr daccous Ohgoconc plUTlc basa~ + 1rtcrbc<1cco N·MORB basemen!+ CJ stage 1 8. 2 arc ""-4-_ Tranch. inncti\.~ D Sla,:ic 1 a rc base me rt +•- stage 2 a~ A Submarine ""lcnnOC!S Urkno.vn basement. stag;: 2 arc cxpoooo
Figure 3. Tcrranc model of the Solomon Islands (after M. G. Petterson. 1999)
10 emplacement rate (M ahoney ct al., 1993; Bcrcovici and Mahoney. 199-k Tejada ct al.. 1996: Neal et al., 1997). The close geochemical similarities or sheet flow and pillow basalts (120 Ma) on Malaita and simi lar lithologies on northeastern Santa Isabel (120 nnd 90 Ma) support correlations or on-land lavas with the submarine OJP sequences . IL appears that these outcrops represent an obducted portion of the OJ P (Babbs. 1997: Petterson ct al., 1997: 1eal ct al., 1997).
The arrival of the OJP at - 10 Ma mostly marked the termination of Pacific Plate subduction beneath the Inda-Australian Plate; the anomalously thick OJP apparently choked the subduction zone triggering a reversal of subduction polarity and the intiation of Inda-Australian Plate (locally complicated of course, by the Woodlark Ridge-Transform system) subduction beneath the Pacific Plate.
The geology and the structural setting of the Solomon lslands arc complex. The islands arc developed from a number of distinct crustal units which themselves originally formed within a variety of geological environments (Petterson, 1999). As a consequence of this tectonic complexi ty, it is important before targeting any mineral exploration in the Solomon lslands, that one must understand the: l . terrane model of the Solomon Islands; 2. local tectonic complications within an overall oblique transprcssional environment; and 3. metallogcncsis complications. i) T errane Model: Ylodclling the Solomon Islands into respective tcrranes has enabled a geological framework to be established. within which the dynamic tectonic evolution and crustal accretion of the Solomon Tslands can be considered (M.G.Pcttcrson, in press). The subdivision of the Solomon Islands is based on: I . basement geology; 2. age relationships: 3. geochemistry: and 4. development of arc-dominated supracrustal sequences.
As a result of the c considerations, the Solomon Islands have been subdivided into 5 distinct terranes as follows: I. The Cretaceous-dominated basaltic basements of: the Ontong Java Plateau Terranc (OJPT) which includes the islands of Malaita, U lawa, and northeastern Santa Isabel, north of the Kia-Kaipito-K origholc fault (KKK-sec Fig. 3); 2. the N-MORB basement of the south Solomons which includes the Choiseul and Guadalcanal tcrrane (SSMT): 3. the Makira-hybrid type terranc that has accumulated through both MORB and OJP plume-related activities:
11 4. The Central Solomon Terranc (CST) of Eocene age onwards that has an arc-type, oceanic and ophiolitic character, and which includes Santa Isabel (south of the KKK); the Sh ortland fslands and Florida Groups (sec Figure 3). Basalts. basic dykes. gabbros, and ultramafic rocks dominate this group, together with more evolved andesitic-dacitic rocks; 5. The arc developments that occurred in two stages: i. The Eocene-Lower Miocene stage l arc, which created the ba cment of the Cen tral Solomon Terrnnc. and also contributed significant calc-alkalinc volcanic materials lo the South Solomon MORB Te1T;.111e: ii. The 2°11 -stage arc of Upper Miocene to Recent comprises the New Georgia Group of islands (the Tew Georgia Te1n111e, 'GT) together wiih the islands of Ru sse l. Savo. the submarine calc-alkalinc volcanoes of Coleman, Kana Keoki, and Kavachi, and the Ghizo and Simbo Ridges. ii) T ectonic style: Convergent-Transform-Composite (CTC) plate margins arc defined by Coleman ( 1991) as zones of accommodation that can reach up to l 0-> I OOkm in width having boundaries that include hi ghly oblique convergent place margins. These sorts of structures consist of a large number of discrete fault blocks that have experienced considerable rotations and translation, and are strongly affected by strike-slip, Lran spress ional, and trans-tensional tectonic forces (Pelterson in press) . The Solomon Islands have been very strongly influenced by oblique collisional tecLOnics : this fact together with the occurrence of a number of arc volcanoes (and associated magmatic centre ) makes the islands an aLtractivc target for mineral exploration. i i i) Mctallogenesis: Known mineralised areas like the Gold Ridge Volca nics and Lh c Kolou la Dioritic Complex occur within the SSMT. The Miocene-Rece nt arc volcanoes Lhal occur north of the present trench lie along ' W -SE and E-SW trending axes. They include the volcanoes from Vella Lavclla to ChoiscuL through Vangunu, through Kavachi and Nggatokae and through west Guadalcanal to Savo, all of which trend in 1 E-S\V to I NE-SSW directions. Volcanic lines arc also parallel to sub-parallel south of the trench (e.g., Kana Kcoki-Colcman seamount -Gizo Ridge) while others arc north-south trending (S imbo Ridge).
The structural and ge netic links between arc volcanism. magmatic-related arc mineralization. and fluid flow pallerns may prove usefu l as far as understanding the various metallogcne i s processes that have occurred in these areas .
12 OUA f(A'HA At' ( _...- 1A l~"' """"' · cOJ'v.;-...,.. , "'°"GI'< ....._, ..,, •Oo'ot _ .,....,.,e-o\'•
"'-£•$ rOC(H( ~ $°""°""0 1,,. -•••0"• BUKA
l.oauo, VOll:~t. ISL AND OUAt(AHAQr 10/ot..o ~ 1 or2~~:H ~ NwtftC Vd-C°"""'1.~~" U~ l O'.."°" Y06cO"l<"t M & 11 , l.A. otC".a \ol!Ot(_O"'<'-, l. t..~u~c• A • R ir .-.• VofCOl"<1 0 20 . ...
P\..<>CC"-'£ 0 l MC)C.( N( - P\.l'OC(N{ • U\dtffC' f' nt10U O W.0 • ~ ~ ffloi l •OfU l. AT( M OCCHC D ltllft'O• MOinZOt'l•O I.At( M.OCCNC P\...CX:CN( D Atc• o Co~ e roo Ml lXX.C • LAf( MOC(l l.ATE OUGOCEN€ • CArn..Y WIOC(N( ~ l< t- , ~• o l.hic:e10f'\• l AT( COCCNE • MIOCX.C ..IOC£ NC LJ g..,,,o '"°""'a''°" (.AT( COCCNE • Vo'C:Ofwe t £A R\.1' Ol.IGOC £ H( ~ ..,.,..o BOUGAINVIL LE 6 ' $ ISLAND Figure 4. Generalised geological map of Bougainville and Buka Islands after Rogerson et al. (l 989) 13 1.2 Local Geology 1.2.J 8011gai11 ville The volcanic basement of Bougainvi lie consists of the Late Eocene lo Early Oligocene Alamo Volcanic (Rogerson ct al 1989). The Alamo Volcanics comprise pillowed ba -;tit to basaltic andesite lavas'' ith interbedded \Olcaniclastic breccias. sand tone and siltstone · (Fig. 4). Unconformably overlying the Atamo Volcanics arc the Late Eocene to Middle Miocene Buka Formation, which outcrops mostly in the northern part of Bougainville and Buka Islands. Overlying the Buka Formation is the Keriakc Limestone of Late Oligocene to Miocene age. Unconformably overlying the Keriake Limestone Formation is the Middle Late Miocene Toniva Formation including the Aropa Andesite Unit. The Late Miocene-Pliocene Awara Conglomerate conformably overlies the Toniva Formation. There arc many undifferentiated volcanic intrusions from Late Miocene Pliocene age. The largest igneous intrusive complexes include the l sinai Monzonitc and the Kupci Complexes, which host the Pan guna mineralization. The Kupei Complex has other multi-phase intrusions including: I. the Panguna Andesite; 2. Kaverong quartz diorite: 3. Biuro granodiorite: and 4. the Nautango ande itc. The Bougainville Group c;onsists of a variety of Pliocene-Quaternary volcanic rocks, including: l. Emperor Range: 2. uma 1uma Volcanics: 3. Rcini Volcanics: 4. Taroka Volcanics; 5. Tore Volcanics: 6. Kunua lgnimbrites; 7. Balbi Volcanics; 8. Laluai Volcanic Complexes; 9. Takuan Volcanics; I 0. Loloru Volcanics: I .I. Billy M itchell Volcanics; and J 2. the Bagana Volcanics. These Pl iocene-Quaternary volcanic rocks consist mainly of tephras and older volcanic intrusions, mo tly outcropping in central Bougainville. These volcanic rocks have been altered lo varying degrees. The Bagana volcano is a particularly prominent and active centre, consisting of a roughly symmetrical cone made up of thick, blocky. massive to moderately vesicular lava flows with minor interspersed volcaniclastic debris (Bultitude 1976). 14 Lava flows from Bagana have been slow moving and highly viscous, and never reach the base of the volcano. Most recencly, four main eruptive episodes have been documented for Bagana: I. A pre-1943 lava flow: 2. 1943- 1953 eruptions; and 3., 1957- 1975 lava nows. The fouth group con ist of boulder, block, and bomb samples of unknown ages. The structural geology of Bougainville and Buka islands was first described by Blake ttnd Miciitis ( 1967) who documented little evidence of major faulting and no folding al all. There arc 3 sets of lineaments that have been identified which appear to control the geology and topography of the i land. These lineaments include a 320° set which defines the long axis of Bougainvillc and its volcanic cha ins; a 335-340° fault trend wh ich defines the long axis of Buka, the Parkin ·on Range, a well as the alignment of the uma Numa. Billy Mitchell and Reini Volcanoes; and finally, the 270-300° lineaments which control the Crown Prince, Deuro Ranges and the Atsilima fault. Bougainvillc appears to overlap with the Pacific and Central Province of Coleman's ( 1965, 1970) model of the Solomon Island litho tratigraphy (see Figure 2). There are wcll dcvclopcd northwesterly struclllrcs on Bougainville, which arc absent from the small-scale geological maps of the Solomon Islands except on New Georgia (Taylor, 1976). 1.2.2 Fauro Island Group The Shortland rslands comprises the Fauro, Alu and the Treasury islands groups all of which appear to be more similar geologically to Bougainville than Choiscul (Ridgway et al 1987). The Fauro group that emerged from the shelf surrounding Bougainville consists of an eroded calc-alkalinc volcano (Figs Sa. b) with an altered calc-alkaline ba altic basement (the Masamasa Volcanics). Th is ca lc-alkalinc volcano ranges in composition from basa lt, Fe-ri ch and Al-poor andesite, to andcsitc-dacite-rhyolite. The basaltic basement is exposed in the far north of the caldera. around Tanwai hill and cast of Karikc village. T he basement ha been extensively brecciated. but was originally massive and pillowed lava flows. The basa lts are dark-grey to black in colour whereas the intermediate lavas arc greenish grey. The clasts within the breccias arc rounded to sub-angular, less than Lm across. and 15 1orth A 13 ouoainville (P G) / ~ t Ovau Island Fauro caldera "orth t T~IUll:l h. -I ------P1rul. s \ '.' caldcr:i-rclatcd dioritc domc' intnision 0 ....,.... ••• l""'°'J.t'l>ll ... 111111 I 111 fll"I" ~: CJ 11~\:iJ?:~·~,:k~~ .... - 1•• 111.111·1\,.h."'11\llflh.' -'km \·1 11 .1,..."'<" . ..,_,s1l._ < .a1:p Figure 5. A Aerial view of the Fauro Ca ldera looking in the E direction; B The General Geology of the Fauro Islands 16 may have formed as a result of autobrccciation of lava flows. Vug within the brcccia arc partly filled with tarnished pyrite crystals up to lcm in length. Unconformably overlying the basement rocks is the K oria Sandstone Formation, (-1500111 in thickness). The Koria Sandstone dips away from the center of the caldera at low moderale dips. lt consist predominantly of medium-coarse grained volcanogenic sandstone together with minor conglomcrares, fine-grained sandstone. shale, and dark limestone. The volcanic clasts arc mostly of andcsite-dacite composition. Unconformably overlying these sandstones arc the high-Al basalt and pyroxene andcsitc or the Togha Pyroclastic Formation. This formation consists of andcsitic volcanic breccias, tu ff breccias, tu ff, and minor lava !lows of porphyritic andcsite and basalt. Intruding into this lithology is the high level calc-alkaline assemblage of micro-diorite (Tauna Micro dioritc), Fauro Dacite and other minor intrusives (hornblende andcsite and ryhodacite). These intrusive dykes arc up to I m wide, and their contacts with the other rocks including the basement are sharp. A roughly circular bay (horse-shoe shaped) in northern Fauro. around the Masamasa Volcanic Formation. is interpreted as a collapsed caldera of the Fauro calc-alkalinc volcano. 1.2.3 Choiseul Choiseul compri e two major and di Linet tructural units. A more deformed and faulted basement of igneous and metamorphic rocks that is unconformably overlain by a weakly tcctonised cover of sedimentary and volcunic rocks (Fig. 6) (Ridgway ct al., L987). The pre-Miocene basement includes: I . the Voza Lavas which arc basal tic (pillowed. massive and commonly brecciated) and have been extensively amphibolitiscd (Purvis and Kemp, 1975): 2. the Choiseul Schists. comprising extensive exposures of amphibolc bcaring schistosc rocks of pale to dark-green, red, purple or dark-grey colour - the schistosity locally transforms to a more gncissic foliation, especially in the east (Pudsey et al 1960): and 3. the Oaku metamicrogabbro imru ive as well as the Siruka Ultramafites. 17 'Tl (10 c: km 20 d North , I . l~O o °' gain ville 0 13'()' (') ::l ')-- . North So1. (1) • ~ · Cho1seul ------..,,. - _o'!!o,, , ; ~ -' /~ (10 I ', I)"-? (') '" oc, 'SantaIsabe ----. ,,, 0 ~ 0 (IO New ~~" () !:)) alaita Georgia~ ~ o C1 Voza Lavas Oaka Metamicrogabbro Siruka Ultramafics Choiseul ~ , () ~ b~U 00 Unconformably overlying these igneous and metamorphic basementrocks arc varic1ics of sedimentary and volcanic rocks of Upper Oligocene or Miocene age (Ridgway and Coulson ct al.. 1978). These rock formations arc listed in order of decreasing age to present day: l. the Mole Formation: 2. Mactambe Volcanics: 3. the Vaghcna Formations; 4. Kumboro volcanics; 5. Pcmba Formations: and 6. ukiki Limestone Formations. The relationship between the Voza Lava and the Choiscul Schists is not very clear and has been disputed by several authors (e.g .. Coleman 1960. 1962. 1965; cf. Ridgeway ct al.. 1987). The Oaka Meta-micro Gabbros arc the only plu1onic rocks 1hat intrude the Vora Lavas. These intrusions arc homogcnous. medium grained, cqui-granular, dark-green rocks that in some areas display glomeroporphyritic clusters of plagioclasc. The Siruka Ultramafite occurs in Eastern Choiseul 1ha1 formed a a li1hified slab or sequence of ultramafic sheets that were thrust on 1op of the Choiseul Schists and Voza Lavas at a subhorizon1al contact (Pudsey et al 1987). The calc-alkaline volcanoes of Maetambc and Kumboro consist of a variety of lithologics ( Ridgeway at cl I 978) from tu ff-breccias to breccias and breccia-conglomcrates. The breccias contain sub-angular to sub-rounded cobbles of andcsite, crystal tuffs and lithic wffs 1ha1 reach a maximum length of I .5m. The Kumboro Volcanic rock (with possibly subsidiary vents situated on Laena) arc a seq uence of andcsitc breccia, coarse lithic tuff. massive andcsice crystal turf. and tuffaceous sandstone. occurring in a limited area around the volcanic cone. Block faulting and erosion have occutTcd which exposed the underlying basemen I. The ages of 1hesc morphologically distinct calc-alkaline volcanoes arc not clear and has been dispu1ed by several authors: clearly more dating is required to solve this issue. 1. 2.4 Guadalcanal The geology of Guadalcanal has being described by H ackman (J 980) and Coulson and Vedder ( 1986). The ba emenc of Guadalcanal consi ts of two lithological types: the basalt dominated Mbirao Group and the Guadalcanal Ultraba ic Unit. This basement has been dated as Cretaceous (single K -Ar age of 92 +/-20 Ma for a gabbro from the Mbirao Volcanic Group). 19 1s9.,s·e Cape Espe1once • C~d. s.allno 01 Koloula Oiornc sulphurous sprmg LJ Gallego Lavas 6 Wo1m Sl>""ll (below 800C) t,:.)!j ~G old Rldij• Volcanu:s A Ho1 SP'•"ll (above SO•C) ~ -- -- MDJOr Fauh ••• •" Ngguraro 9 f1nahulu 17 Molou (Pu1a Ndoo Soloul (I MO.a T°'o) 10 Ch••••uOQo t32•c1 18 Namondoula 2 H0tlav4 11 Kolo0Qgulo1 19 Vatus1nahev1.t 3 Mbonehe 12 Up1>41r Suia~amo 20 Koloha•sava 4 Mboh$ahato 13 Tosi tSu10) 21 Mbolavu 5 Tas1 (Tona) 14 Uppo1 Nggounoha LOWOI Tangg111• 22 39•c 6 Upper ron1 (Tanakuo) 23 M lddle Tangg101a 7 Toni 1& Mlddlc Nggounaha 24 lango1ar• 0 10 70 )0 8 Taombo (55•C) 16 i Figure 7. Map showing rhe location of the Gold Ridge Volcanics and Gallego. 20 The cover sequence or Guadalcanal compri ses : l. the Oligocene-Miocene, highly phyric calc-alkalinc Suta Volcanics: 2. the Poha diorite, age 24.4+/_ 0.3 Ma (Chivas, 1981 ): 3. the \·olcaniclastic sediments or Oligocene-Miocene rocks of the Kavo Grc) wacke Beds. (outcropping in central-eastern Guadalcanal): 4. contemporary reef limestone formations of the Mbonche and Mbetilonga limestone Formations; 5. the Plio-Pl cistoccne Gallego Volcanics of \Vest Guadalcanal and Gold Ridge Volcanics (central Guadalcanal). Note however. that K-Ar dating shows that the Gallego Volcano is at least 6.-l+/-l .9Ma old (I lack man. 1980): 6. the Pliocene Mbokokimbo Formations of central-cast Guadalcanal: 7. the 4.5 and 1. 5 Ma (Chi vas. 198 J) Koloula Dioritic Complex (KOC); and 8. the Quaternary- Recent sediments \\'hich arc dominated by the alluvial deposits around the Guadalcanal plains as well as rhe Honiara Beds that are represented by the raised coralline reefs up to 800m (height) above sea level. Most Pleistocene-Recent deposits arc exposed around west and central-north Guadalcanal. The Gold Ri dge Volcanics (GR\/) form part of a thick, predominantly sedimentary succession ranging in age from Oligocene to Recent. The Toni Formation is a equcnces of volcaniclastics ruditcs and arenites with subsidiary pyroclast ic. extrusive and biogcnic limestone which includes: 1. the Chrikangge Grit; 2. the 1l aviha Sandstone: 3. lower Toni Conglomerates and the Gold Ridge Volcanics: 4. the M betivatu Sandstone and 5. the Upper Toni Conglomerates. Figure 7 shows a general loca lity map of Guadalcanal indicating the location of the Gold Ri dge Volcanics, Koloula Diori te Complex, and Gallego Volcanics. 1.3 Pctrologic a nd Geochemical studies 1.3.1 Pctrologic classification of the Solomon Arc The volcanic components of the islands forming the Solomon Arc system arc dominantly calc-alkaline. ranging in bulk composition from picritic basalt/basalt, andesice through dacitc to rh yolite. Ultramnfic bodies have also developed at some early stages. Calc alkaline volcanoes including other minor intrusives have intruded the basement of these islands. In the following sections, the petrographic characteristics of the various suites studied in dctai I in this thcsi • are described. 21 i) Pre-Miocene: Volcanic Basement On Fauro, che basalt arc mostly very fine-grained with an interscrtal texture including small grains of magnetite. Less common medium-grained lavas arc mincralogically similar, but arc glassier and ge nerally have more of a trachytic texture. The lavas within the Togha Pyroclastic breccias arc fcl ·par-phyric with black amphibole or green pyroxene. The tuffs are pale grey in colour and usually fairly oft. The Voza Lavas of Choiscul include a wide spectrum of basalt from those with little alteration to massive and unfoliated amphibolites. Based on their dominant textural style. the Voza Lavas have been classified into 2 groups. Group I consi ts of micro- and cryptocrystallinc basalts with granular ba altic textures. The rocks consist of micro-crystals of plagioclase needles and pyroxene granules that increase in grain sizes to dole1ite. Group 2 Voza Lavas display a progressive rep lacement or igneous features by metamorphic fabrics. The Choiseul Schist is petrographically disintinguished from amphibole-bearing Voza Lavas by its tectonic fabric. The schist is further subdivided into 2 type : I. heterobla tic schist which has a granular, heteroblastic texture where the amphibolc (mainly hornblende) displays a preferred orientation; and 2. an augcn schist consisting of fine grained ncmatoblaslic hornblende (blue-green in colour) intcrgrown with andesine. Petrographic analysis shows the Mbirao Group arc all sequences or basaltic lavas and sheets with interbcdded cherts and pelagic limestone. They also consist of dykes and. ills and larger intrusive bodies. Basalts that have been sheared arc pale bluish in colour with the pyroxene altered to ehlorite and actinolite. ii) Ultramafic Basement The Siruka ultramafitc of Choiscul has been reported a extensively scrpentinised harzburgite with fresh samples rarely to be found. Where fresh however. harzburgite consists of porphyroclastic enstatite ( l -2mm) surrounded by granular-textured olivine. Serpentine minerals rep lace both olivine and the pyroxene, and also occur in small veins. The Quadalcanal Ultraba ic Unit comprises a series of ultramafic bodies located in three linear belts (Marau Ultrabasic, Suta Ultrabasic, and Ghausava-ltina Ultrabasic), 22 predominantly harzburgitic in composition but mostly serpentiniscd. associated with anorthositc (Petterson ct al 1999). ii i) Intrusive bodies The Tauna Microdiorite outcrops mainly on Tauna Island (on Fauro) as well as the nearby mainland. It consist of mcsocratic microdiorite with megascopic c1ystals of white feldspar and dark green pyroxene. Xenoliths of dark. fine-grained, older basement-derived basalts are present The intrusive Oaka meta-micro gabbro of Choiseul is minen1logically similar to the Voza Lavas. and may have had a comparable metamorphic hi tory. The meta-micro gabbro consists of aggregate or anhedral hornblende (40%) and plagioclase (50%) plus traces or opaque minerals. The plagioc lase is mostly andesi ne but ranges from more 1 a- to more Ca- rich plagioclase. The Poha (Gu Various strongly porphyritic volcanic rocks of calc-alkalinc affinity have been collected from Bougainvillc. Fauro, Choiseul, and Gold Ridge (Fig. 8). These range in texture and compo ition from ba alt (some fine-grained pillow lava) to andcsite and dacitc. Plagioclase, pyroxene, amphibolc, and Fe-Ti oxide minerals arc the most abundant crystalline phases in these rocks. Olivi ne is present in basalts and basaltic andesitcs. Sulfide minerals arc present as trace-minerals. disseminated along fractures, joints. along veinlcts. and in vugs/porc paces within the volcanic rock (Fig. 8d). The Plio-Pleistocene calc-alkalinc rock of the Solomon Arc S!Udied here will be further divided into two group·: a) the Solomon Islands Volcanics Suites which include the calc alkalinc volcanoes of Fauro. Choiscul. Savo, West Guadalcanal and Gold Ridge: and b). the Bougainville Volcanic Groups in which the Bagana volcano is the prime focus. 23 Figure 8: Petrographic features of the volcanic rocks. a taken in plane lighr mostly dominated by plagioclasc (some strongly oscillatorily zoned), +/- clinopyroxcnc, hornblende and opaque minerals. b shows a typical medium-coarse grained andcsite with phcnocrysts ofplagioclasc, hornblende and Fe-Ti oxide. Ml arc < 10 pm in maximum dimension. Figure 8 continued, c shows a coarse-grained granodiorite where the clinopyroxene and plagiocla e have been exten ively altered. d shows disseminated ulfides that form along veinlet ,joints and vug ; the Mis in this rock are around 2-6pm in maximum dimen ion. 24 1.3.2 So lomon l slands Volcanic Suites The Fauro Dacice wa originally mapped as two units based on grain ·izc (Ridgway and Coulson, J 987). The dacitc outcrops around the sickle-shaped caldera. is well weathered. reddish brown in colour, and strongly jointed. Fresh rock samples arc lcucocratic with abundant phcnocrysls of white feldspar and quartz that have a maximun size or 3-6mrn or even up to Lem in length. Quartz phcnocryst arc absent in very fine-grained samples (micro-porphyritic). Some rock also have minor brownish-coloured biotite, and green hornblende phenocrysts. Alterations of feldspar to carbonate followed by chlorite and epidote is common. ln highly altered samples, the carbonate and chloritc sometimes form veined structures. Sulfide mineralization is widespread throughout the dacite. Pyrite is the dominant ulfide mineral that is developed along fractures, joints, and disseminated throughout the whole rock, and is abundant in weathered dacites, andesite and basalts. The calc-alkalinc volcanic rocks of Mactambe and Kumboro (~hoi cul) are highly porphyritic with phenocrysts predominantly of plagioclasc and clinopyroxcnc. Amphibolc and biotitc are abundant in a few samples, and su lfide traces are observed in some of these rocks. Feldspar phcnocrysts are the major constituent of the andcsitcs. forming cuhedral LO subhedral laths and rhombs, sometimes with sharp broken edges and occasionally showing sub-parallel alignments. Feldspar crystals range from 3- 10 mrr1 in length. arc fre hand clear with prominent concentric growth zones that are often marked by inclu ion trails. The feldspar tends to be predominantly Ca-rich (bytownitic). The Suta Volcanics of Quadalcanal arc mostly feldspar-phyric basaltic andcsirc lava flow and volcaniclasrics with white feldspar phenocrysts up to 8rnm in length (Hackman, I 980). The Gallego and GRV (also of Guadalcanal) are dominated by basaltic-andcsite to dacitic pyroclastic flows and lavas with associated diorite-granitoid intrusions. The ORV arc entirely fragmental flow breccias (Grover I 958b). The brcccia blocks arc very angular and are predominantly of altered feld parphyric basaltic andesite and ve icular basalt. The breccia ranges 1,videly in colour from blue to purple, but also includes grey, brown, orange, and green . The volcanic breccias arc well stratified, with laminated lenses of volcanic arenitc, and lapilli-tuff. The matrix of the breccias is a purplish-grey. coarse sandy turf, containing turbid plagioclasc crystals, angular fragments of basaltic glass, Fe-Ti oxides. and (occa ionally) basaltic hornblende. 25 Five samples from the GRV studied here arc highly silicified and contain significant sulfide concentrations. Most have brecciated textures. The rocks arc predominantly basalts and fcldspar-homblende-phyric andesite, and in thin section, carbonares arc observed. The feldspar phcnocrysts arc the dominant phase and arc generally in the form of euhcdral to subhedral laths, occasionally with flow-foliation fabrics. 1.3.3 Ilougainville Volcanic Groups The volcanic rocks from Bougainville have similar petrographic features to many other arc type rocks with highly porphyritic textures. Phenocrysts are dominantly plagioclase, clinopyroxene, lesser Fe-Ti oxides, and minor onhopyroxene. Amphibolc and biotitc arc generally minor but can be abundant in more fclsic samples. The Bagana rocks specifically arc very porphyritic and include glomeroporphyritic aggregates. 26 CHAPTER 2: GEOCHEMICAL STUDIES OF THE SOLOMON ISLANDS ARC SYSTEMS 2.1 Geochemistry of the Solomon Arc A series of Harker diagrams and " total alkalis-total Fe as FeO-MgO (i.e., AFM) diagrams have been plotted to disintinguish the fundamental geochemical differences between various rock types, and the basement characteristics of these areas. These diagrams can help identify the geochemical composition of various rock types that originated from different sources and were emplaced in different terranes. The approach taken here is to briefly present the results of previous (mostly Solomon Islands Geological Survey) studies, then document the geochemical compositions of the various studied centres with new data, and make some comparative comments regarding regional geochemical variations with reference in pa11icular to the New Georgia Group (e.g., Stanton, 1994). 2.1.1 Solomon Islands Arc geochemical studies i. Harker diagrams The Si02 vs. other major oxides plot (Fig. 9, data presented in Table l) shows that the older tholeiitic Masamasa Volcanic Group is chemically distinct from the younger calc-alkaline suites of the FaurQ Volcanic Group (after Ridgeway et al., 1987). The younger calc alkaline suites are assigned here to the Plio-Quatemary Fauro Series. The older tholeiitic suite has generally lower Al20 3, CaO, Na20 , K 20 and P20 5 wt% and higher FeO. MgO and Ti02 wt% than the younger calc-alkaline suites (Ridgeway et al., J 978). The Si02 vs. other major oxides plots (Fig. 10, sec Table 2) compares se lected rock types from the Solomon Islands basement in comparision to calc-alkaline rocks studied specifically here from the Faure Volcanics, Choiseul Volcanic Suite, Savo, and Gallego Volcanics. There are two distinctive litho-geochcmical groupings on Choiseul. The older Late Cretaceous Voza Lavas comprise a tholeiitic basement (dominantly basaltic composition) as described by Ridgeway et al. (1978) while the younger age (possibly Pliocene) Maetambe and Kumboro centres are calc-alkaline (andesite dominated) (Figs. 10 and J I ). 27 1.5 6 t a 5 f J 4 'Cf!. I 'J 0~ N 'J 3 0 F 0.5 ~ 2 .._.__ 19 10 ~ ... 18 b 10 g 17 0~ 8 16 s 0~ M 6 0 15 s " 4 I I I! I I I I 12 6.6 -..1t. J. .i..a.. l.o... ,1.. 6 ~ c 6 · h 5 5.5 ~ 0 5 'J 4 0~ M 'J 4.5 0 3 t 0 "' 4 if 2 ~ 3.5 3 I 0.8 2.1} • ~ I I I I It I I I I I I I IC!!! 0.7 . d 3.5 0.6 ' 3 . 0.5 . 'Cf!. 'Cf!. 2.5 'J 0.4 : 'J H 0 0.3 ; 1.5 ~ ::.'." 02 , r' 0.1 0.5 0.6 0 'l)c \llJ. ... 0.4 Y. • I 'Jf •\L.l'l !:I•'• 'Cf!. s 0.3 . e t 0 "' 02 a."' 0.1 0 45 so 55 00 65 70 75 so\',{ % 1 Figure 9. Harker diagrams for the Fauro suite (after Turner and Ridgway, 1982) 28 Ta ble l A. Analyses of Igneous Rocks from Fauro (after Turner and Ridgeway. 1982) Older Younger Suite Suite F1 F2 F3 F4 F5 F6 F7 F8 F9 F10 F11 Si02 48.68 57.85 62.95 54.87 67.17 71.2 60.89 58.51 59.59 51.45 45.75 Ti02 1.43 0.99 1.04 0.82 0.1 0.08 0.45 0.37 0.72 1.19 1.02 A1203 15 13.36 12.86 18.03 17.48 15.71 15.58 18.52 16.68 18.01 17.37 Fe203 5.48 6.83 5.08 3.13 1.51 0.79 1.41 3.01 4.17 4.39 6.49 FeO 9.19 3.81 3.63 3.7 1.07 1.26 2.8 2.17 1.68 4.06 4.67 MnO 0.36 0.14 0.3 0.2 0.05 0.03 0.75 0.14 0.11 0.16 0.24 MgO 5.76 4.7 4.99 3.37 0.1 1.34 2.98 2.24 1.79 4.17 5.81 cao 7.47 6.21 138 8.05 1.18 1.93 4.84 7.4 6.99 9.86 10.37 Na20 3.27 2.77 4.54 3.95 4.13 6.28 4.9 3.89 4.16 3.36 3.1 K20 0.48 0.3 0.24 2.25 3.56 0.7 0.8 1.64 2.21 2.02 1.46 P205 0.09 0.07 0.07 0.29 0.17 0.12 0.2 0.34 0.3 0.49 0.38 LO.I 3.61 2.04 3.54 1.32 2.52 1.38 3.56 1.47 0.79 1.15 2.56 H20 0.14 0.18 0.12 036 0.28 0.1 0.4 0.32 0.53 0.35 0.32 Total 100.96 99.25 100.74 100.34 100.32 100.92 99.56 100.02 99.72 100.66 99.54 Qz 17.76 22.67 2.34 26.17 25.62 14.58 11 .19 11 .46 c 2.87 5.2 1.41 Or 2.95 1.83 1.48 13.47 21 .57 4.14 4.96 9.87 13.29 12.05 8.98 Ab 28.5 24.27 39.67 33.92 35.86 53.45 43.39 33.49 35.78 28.67 18.41 An 25.61 23.94 6.64 25.18 4.89 8.84 18.98 28.79 20.68 28.4 30.29 Ne 4.78 Wo Di 10.03 6.18 11.46 3.87 5.17 10.31 14.51 16.92 Hy 20.49 20.17 20.68 8.47 3.48 4.96 10.73 7.19 3.01 6.72 01 5.04 2.28 13.92 Mt 4.38 3.74 3.8 3.41 2.25 1.15 2.13 2.77 3.28 3.94 3.8 llm 2.79 1.94 2.03 1.58 0.19 0.15 0.89 072 1.39 2.28 2.01 Ap 0.21 0.17 0.17 0.69 0.4 0.29 0.5 0.83 0.74 1.16 0.93 Table lB. Analyses of Basaltic rocks from va rious areas of the Solomon Islands (after Turner and Ridgway, 1982). Locations A B c D E F G H I J K L Si02 50.21 49.75 51 .21 50.89 50.41 50.25 49.53 49.09 49.29 48.47 51.4 53.93 Ti02 1.47 1.13 0.91 3.37 1.73 2.7 1.45 1.2 1.17 2.1 0.59 0.65 Al20 l 15.47 18.12 17.22 12.61 14.36 13.44 15.18 15.29 14.74 13.61 15.72 19.43 Fe20 3 3.02 2.67 2.2 2.33 6.1 5.8 3.06 4.65 2.53 3.36 2.95 3.9 FeO 11 .84 7.11 10.41 10.06 6.65 4.61 8.42 6.86 8.98 9.23 5.65 3.63 MnO 0.37 0.21 0.17 0.19 0.21 0.45 0.25 0.22 0.25 0.22 0.29 0.16 MgO 5.94 5.13 4.62 4.68 6.84 7.97 8.27 5.45 5.46 6.98 8.56 3.89 cao 7.71 10.37 9.19 10.06 11 .01 9.18 10.47 9.57 6.8 11 .15 10.85 8.92 Na20 3.37 3.31 3.44 3.98 2.36 2.95 3.15 3.08 4.05 2.86 3.06 3.49 K20 0.5 1.78 0.51 1.22 0.16 1.77 0.14 0.45 0.8 0.2 0.72 1.96 P20 s 0.09 0.44 0.1 0.61 0.16 0.9 0.1 0.17 0.13 0.18 0.15 0.38 Total 99.99 100.02 99.98 100 99.99 100.02 100.02 100.16 99.71 99.45 99.94 100.34 A Masamasa Volcanics of Pauro G Mbirao Volcanics. Guadalcanal B Togha pyroclastic, Pauro H&I Voza Lavas. Choiseul c Alu basalts J Choi. cul Schist D Mono basalt K Suta Volcanics. Guadalcanal E Older basalts, Malaita (OJP) L 13asalts and Basaltic andcsite, 'cw Georgia F Younger basalt . Malaita (OJP) 29 20 ~ 3 5 ' u 19 a e 18 3 f 17 ;ft 2 5 I *~ t 16 3 2 ~ 0 "' N ... 15 0 • < i= \.5 • 14 • ! 13 1 12 0.5 i " b 0 • 5 ! f 6 . • t 0 4 ; 5 *~ 035 .... *3 o,. 0 03 c QI •r • u. ::: 0 25 f 31 .. 2 t • J 0 ~ 0 9 - ~~ 0 1S ~ ~-----1--.a...... L...... --- • --~ g 8 c 11 f • .. • ;ft 10 ~ t *'ii ~ 6 ~ 9 0 0 t ~"' u"' 8 .L 7 ~ • 3 !-' G t • r d h 1 5 36 *3 i o ... *3 1 o .. z"' ~ • 05 H [ ...... 2 ~ • L I.. 0 • •8 49 50 51 52 53 54 48 • 9 50 51 52 53 54 S10 wt % SiO wt % i 2 Figure 10. Harker Diagrams for selected volcanic basements of different islands of the Solomon Islands. Geochemical data fo r the Masamasa volcanic, Togha pyroclastic, Alu and Mono Basalts, Old and Young Basalts of Malaita (OJP), Mbirao Volcanics of Guadalcanal, the Suta Volcanics (Stage I Arc), the cw Georgia andesite (Stage 2 arc) were obtained from Turner and Ridgway (1982) and the Choiseul volcanic basement are taken from Voza Lava, Table I, sample A45 and B89 and sample BI 9 for the Choi eul Schist of Ridgway and Coulson's geological memoir ( 1987). 30 ii. AFM Diagram T he Plio-Quaternary Fauro Series (Fauro Group. Shortland Islands) have a typical calc alkaline trend in the AFM diagram (Fig. 11). These calc-alkaline rocks may be pre Pliocene in age and formed during pre-Pliocene, NE-directed subduction. 2.1.2 Bougainville Volcanic Suites geochemical studies The Bougainvi.lle arc suite has been grouped into four (4) main classes based on a K!O vs. Si02 plot (Fig. 12a) as well as volcanic locations (Fig. l 2b) : Group l , high-K rocks of north Bougainville (includes the Balbi, Tore and Emperor ranges); Group 2 is medium-K and cons.ists mainly of the B i lly Mitchell, Bagana, and Reini volcanoes, and the Laluai volcanic complex , aJI located mostly in central and south Bougai nville; Group 3 are medium-Kand belong to the Numa Numa volcano; Group 4, the Older Tertiary Volcanics, are intrusive rocks of the Atamo Volcanics (Late Eocene-Oligocene), Aropa Andesite (Middle M iocene), the Toniva Formation (middle-late Miocene) and possibly the Awara Conglomerate (late Miocene-Pliocene). These rocks are altered (low-grade metamorphism) with chlorite, carbonate, ''iddingsite" and zeolite as secondary minerals; water and C02 are also high. The northern groups of rocks are mainly 2-pyroxene (px) basalt and andesite, with biotite phenocrysts (in felsic lavas). Amphibole is general ly absent but apatite micro-phenocrysts are common in some felsic samples. In contrast, the central and southern groups have abundant amphibole in all rocks but an absence of biotite even in felsic samples. Amphibole generally forms euhedral phenocrysts with fresh cores and typically lims that are dominated by oxidation and formation of Fe-Ti oxides. Note that high bulk rock K20 contents are manifest in the appearance of modal biotile in the northern part of Bougainville. 2.2.1 Whole rock (bulk) Geochemistry A total of 27 rock samples from Choiseul, Gold Ridge, and Fauro have been analysed for bulk rock major and trace elements. The results are listed in Tables 2 and 3. In addition to the above samples, bulk rock major and trace element analyses have been completed for rocks collected from the island of Savo (data in Tables 26 & 27 in Appendix 2) and bulk rock major and trace element analyses (data listed in Table 28 , Appendix 2) similarly for the Gallego vo lcano of Western Guadalcanal. Savo and Gallego volcanoes were selected in 31 Fe (as total FeO) C) B Fgranodiorite Fporphry H Fbasalt F Fandesite N Kumboro E Maetambe Cale-alkaline ( Savo 1(MGP) ...... 0 D Gallego ,> 0 0 0 0 0 .c:9 .6l . l? ·~ Na o + K 0 fv1g0 M Savo 2 (RLS) 2 2 Wt% Figure 11 . A-F-M diagram for selected rock suites from the Solomon Island Arc. Data projected for Fauro and Choiseul are newly obtained in this study, whi le those shown for Savo and Gallego were obtained by others. The samples from Fauro Islands are represented by the letter "F" as a prefix for each rock type e.g., Fgrano represents the Fauro granodiorite (sometimes referred to as diorite by Ridgway and Coulson ( 1989). The two volcanic centres of Kumboro and Maetambe represent the ealc-alkaline rock samples from Choiseul wh ile the samples from Savo volcano are labelled as Savo I (collected by the author as an assistant to Dr. M. Petterson) and SavoRLS (also referred to as Savo 2 in later chapters: amples co llected and analysed by Professor R.L. Stanton). 32 45 a Basalt Low-Si0 2 High-$102 Dacite andesite andesite 4 ~ G ~ ~ hrgh - K ~ 35 F I Tote l H H I r.,,.,_Babo H H ~ c,v Clo<'U>CI 0. 3 "0 G F E"'I*°' a cf ~ H H t\J 8\'ic;l\<'1 H G 25 cic I 8ag3M GG ! B 0 Re.ro ~ \ti fJ <1-l c '5 *"i 2 00-H fl .. 1 Ta\ uon ~ ON Ll 0 :.:: .i Taic>'-all.Olo 45 so 55 60 65 70 North \jBuka Island AX' Panguna mine l!,. Volcano I . International Boundary • Town ·Arawa I I o 30km Figure 12. a. shows the relationship between Si02 vs K 20 wt %. and b. the geographical locations of individual volcanoes on Bouga invillc. The dashed blue line is the international politica l boundary. The line A-B represents the boundary between the high and the low-K volcanoes (after Rogerson et al, 1989). 33 particular as examples of present-day arc volcanism in the Solomon Islands arc system. In f-i gurcs L3 to 20, graphical displays of al l these geochemical data arc shown in order to compare and contrast the vari ous analysed suites. The overall normal calc-alkalinc fractionation trends of these suite. ha ve relatively low concentrations of Al 1 0 ~ , TiO! • total Fe as F~O» MnO, M gO, and CaO, and arc high in Na10 and K! O as SiOz increases. These variations may be driven by hornblende +/ pyroxcnc-plagioclasc-bioti tc-Fe-Ti oxide fractionation. Overall, it is possible to recognise in the K 20 vs Si02 wt% plot. two major magma suites: a lower K 20 suite (< I wt%) that consists of the Gallego Volcanics plus a couple of Fauro rocks and other suites with higher K 20 ($2.5wt%), that comp1ises the volcanoes of Savo, Fauro, Kumboro and Maetambe of Choiscul. It is not obvious within these suites that a consistent relationship between K 20 wt% and distance from the nearest trench (or projection between the New Britain and San Cristobal trenches) exists. For example, there arc both medium- and high-K suites within the i\'cw Georgia Group and Bougainville (Fig. 19). Within the local pair of Gallego and Savo however. the former has lower K 20. A s a whole, the Solomon l slands- Bouga invi lie arc system straddles the alkaline-subalkaline divisor of Le Maitre et al. ( 1989), and includes trachybasah and trachyandcsite as well as basaltic andcsitc, andcsitc, dacite suites (Fig. 20). It is noticeable (Fig. l 3d) that Na?O wt% is relatively low for the Gold Ridge samples compared with the other studied suites. oting that the Gold Ridge suite is characterised by high K 20 (7- 12 wt%) compared with the other suites (sec Fig.l3h), it is possible that these compositional differences reflect some degree of a20 leaching and K 20 addition through mctasomatism. In general terms, the Ti02 (< I wt%) and MnO (<0.2 wt%) arc low while Fc 20 3 (<10 wt%) and CaO (< 15 wt%) arc high. The MgO vs SiO ~ wt% plot (Fig. l4c) shows an exponential decrease in M gO concen tration as Si02 wt% concentrations increase. M gO has been plotted against various major elements as shown in Figure 14. The concentrations of Ti02• Fc20 _, and MnO \Vt~ are positively co1Telated with M gO while Si02, Na10 and K 20 wt% arc negatively correlated with M gO. The Al ~ O , concentration seems to be independent of the abundance of M gO. Figure 16 shows the Ti01, Fe20 3, 34 Table 2. Bulk Major and Trace Element Analyses of Choiscul and Gold Ridge Choiscul Gold Ridge Y!I·.'· 71 99-:! -,, 'i102 56 5221 6121 548 53 547 71 755 (IQ 4(S 11411 6S221 1102 075 09 041 091 0.95 0')4 0.411 0 44'> 0 522 0 50' 1\1201 17 12 ,, 14 17 5 17 52 JS.U J 7 •15 12.074 , , 5•>5 15 130 14 253 h:20.1 7.78 4.3 7 '16 8.12 7 .l I 9X5 1250 l (,.11 3.\52 \lnO 0 IX 017 0.11 0 17 0.IX 015 0.001 O.QO.l () ()(IJ 0004 \lgO .l 55 J79 l.'> 34 .l 39 2 78 0.05b 0 -'05 044(1 O.• lOI'• Cao ~ l •) 24 S.• n H6 7.n 5 IJS 0 ()111 O.OI•-' O.OS9 0.062 ' a20 362 U7 4611 3.7 3.73 3.32 O.ISI () 142 012! 0 153 K20 2 I 1.711 2.37 2 17 l.91 2 I<> 10 152 x 4lQ 7.501\ l( 837 <; 0.01 001 0.()1 0.01 OOI 0()().1 0.·1~7 0. .197 o.4n 041'3 1'205 (I .lX 0.4 02 (l 37 0.43 0.53 o.oos OOIS O.OD 0.011• l"OlJl ')9.5 •n.x1 9X.04 9S.84 974 94.9 •)7 lS 96 12 .,, 31 %.17 1 race eh.• 111 cnh Oa 229.5 1654 :I02 2 JSO.<• 411.9 476.2 491.8 651> 2 547.5 690.7 Rh 4'{1. 5 28 4 5.l2 430 366 41.S lt24 7;12 (IC}' 7(,2 Sr 1136 5 Ql7 x 1501.Q X-14 3 f/.74.7 766.7 2%1 142 2 2474 J47.9 l'b S.6 S2 10.I 52 5.2 (t.11 11 3 WJ2 4M 2 105S 7 Ir 117 7 (12 ,l 100 s 131 J 140.l 1456 3'12 "5 432 -IOI\ ' b 61 :u 5.4 10.3 QX 'l5 07 07 O.~ o.~ y 22 5 IX<1 12 I H.5 303 19 5 H IU S-1 KO 13 17 (1 14 l 104 131> 16 ! 17 7 22 15 42 3.0 Cc 15.4 2 35 v "' 0 ~ ~ ~ ~ ~ ~ o M v ') :; 7 ~ 0 ,, :; co 003 0 e e ~ ~ ~ ~ r, ~ ~ 1 ; - 0 ~ .., c 0 ~ .:: c / ,.. ., ,... ,, " c 0 0 ~ 3 ;J. i:. s "'..c.~.ooog: ... ., x 0 0 : C- ·~ ~~ :J ~ :!. :;: ,(: - ,., 0 ., / - ,., .. ..,'"?000-i. ;!, 1; ? c ~ ., .. "'0 0 0 0 .. ., ~. ., .... ,., ,. ,... J ~ ~ ;; ; ~ 0 v .. 0 0 0 ..: ,, y. ,, v .. .. c ,. "' 0 .. ,, 0 .., :; = ., e ~ ,.... 0 .. " ., .... 0 ~ .,,Q Q,_ 10 2S .. e •• • 2 .,. ,, 1 s ~ ,. *~ 0 .. ,, q < I- • O OS 0 10 03 b 0 2S , s 0 2 i ... *;: *~ ... 10 0 15 0 ,.. 0 Q) c u. ::E 0 1 • • 0 05 # I 0 0 ...... 16 IS ,. c g 12 ;t ;!. 10 10 i ;: 8 0 0 ::E°' e f ()"' • : f w 0 0 ... ~ 8 3 7 d h I ,r ,1(, 2.5 , , ,. 0~ , ... , , ''("'..., : I 0~ 2 , rn ?; , Gold R1dgo , .. •...... ,•_ 1.f:tv • ta ?; d !lo •' 1 5 , 7·12 K 0 "II 'If 0 , ~µ~ ~~ · 1 N 6 .., . (I) 3 ,4 0 ... ~ N .... z ,~ Y'. , 2 . , .~" • OS ,,,,,,... ~ 0 0 40 45 50 55 60 65 70 75 40 45 50 55 60 65 70 75 $ 10 wt % 2 $ 10 wt % 2 Figure 13. I larker diagrams of elected volcanic uite from avo, Gallego, Fauro, Choiseul, and Gold Ridge. Sample from Fauro arc represented by the letter F; tho e from Choiseul by CH. The solid arrow indicates the general trend of oxide variations; the two dashed arrow in cl and h show the different trends for Gallego and Savo. 37 ,0 • 25 a ... e 2 16 ! ...... fandes•to if. ... " ~. 1 5 14 • ... ~ • *;: •• • ... 0 ... 12. .... *;: " t *;: ... 10 0 15 0 ... 0 Q> c u. ~ 0 1 0 05 0 75 c g 70 ... 6S " f 10 *;: *;: 60 r ... 0 C? (/) (.)"' 5 so ... •5.. I ... , , I) ••• If •O ...... _i. .. ,, .. 2 s 8 d h ~•- J .. ... 1 ~' 2 ' 6 ' .. .' tt 5 ' 1 5 . ' *;: ~ .', ~ ... 'l' ' ~ ' 0 )-,..., 0 ' ... ' ... •••\ " . :.: ' ' z"' ...... ' \ ' 05 ' a ,... ' I ~ ~ 0 ~~ ~ ....I. o- ~ 10 12 16 0 2 4 6 8 10 12 14 16 0 6 8 :• MgO wt% MgO wt% figure 14. Plot of MgO wt % vs other major oxide for elected bulk analyses of suites from Savo, Gallego. Fauro, Choiseul. and Gold Ridge. Abbreviations and arrows a in f igurc 13. 38 20 1 5 a e 18 • 16 • .~ ... F .andos11e ~ Gold R1dgo ~ 14 • 7-105KOwt • I 5 3 3 ...... 0 .. 12 q ...... ;;: I- ' 10 ...... 05 ... 8 • ~ '~ . 6 0 20 0 3 b ' ' 0 25 ...... 1$ ' ' .. ~ ' ~ 02 .. '\ ' . 4 3: ' ~ .... 3 ...... 10 0 IS ... 0 ' .. ~ , 0 ' ... ' c ' Cl> ' ~ • • \ 0 I u. • ' .. :::! ~ .. ' _ ..,. ,(,... ~ '• ' 0 05 .... ~ .. ~ /.~ 0 -eL...... ~ 16 r IS c 14 ~ ' 9 ' .. .. '#. ~ 10 ...... 10 I ' . .. 3:: .. 3: ' • f ' • 8 .. r 0 .. ... 0 • • (I) ' .. 6 ...... ' .... ' :::! ...... (.)"' s '.... ~ ' ' ...... ,.. ,~ ...... ~ ...... • ,. • ,. :L • 0 8 75 ;'1 d h 7 t , 70 • •· , , figure 15. Plot of K20 wt% vs major oxide for elected bulk analy c of uitc from Savo, Gallego, Fauro, Choi cul, and Gold Ridge. Abbreviation and arrow a in figure 13. 39 2 5 20 r a e 18 0 16 "$ •.. 1 5 3 14 '°'l: ...... F ·andosito 0 12 N , .... ' ... Q . .. • < I- . " 10 ~ • 05 ...... 8 . ' ~~ ... ~ • 6 0 ~- 20 03 b .. ' ' 0 25 .. r 15 ' ...... "$. ' .. 05 0 • •• _._i_• 0 0 2 6 8 10 12 0 2 6 8 10 12 Na O +K 0 w t% N a O+K 0 wt % 2 2 2 2 Figure 16. Pl ot of K20 + a20 wt% vs major oxide for selected bulk analyses of uitcs from Savo. Gallego, Fauro, Choi eul, and Gold Ridge. Abbreviations and arrows as in Figure 13. 40 100 80 a d 80 60 . E 60 . Q. Q. • .0 • Q: 40 .. ~ 20 0 0 ·20 160 0 700 b • e 14 00 600 120 0 •• 500 E 1000 Q. • E'oo Q. 800 ~· ~' Q. ~00 600 200 400 ...... 100 200 0 " 0 0 1000 1000 c f 800 800 • E .. E6oo 2: 600 Q. Q. ~ '- Cl) (.)4 00 .:oo .. , . • <1.,. ( 200 200 !' ·>~ • •• •• 0 0 40 45 50 55 60 65 70 75 40 4 5 50 SS 60 65 70 ...... ·~· SiO wt% 2 Figure 17. a to c: Abundance of selected large ion lithophile trace clement plotted against Si02 wt%. d to f: Abundances of selected ferromagnesian and ehalcophile trace clements plotted against Si02 wt%. for volcanic suites from Savo, Gallego, Fauro, Choiseul , Gold Ridge, and Bagana (pre- 1943 to unpublished, inclusive). 41 30 200 a c 2S 150 • E 20 a. • • 0 a. • ~ • N 100 • . f: :-r. : : -ic: r • 10 • ...,., 50 00 .. ·. s .. 0 0 60 12 b • d 50 10 40 8 E • E a. ... g; 30 a. 6 • QI • .0 • L (.) z • 20 r 4 • ·z. • Gol!ISRodO• 10 . . 2 ...... • s 0 0 40 45 50 55 60 65 70 75 40 45 50 55 60 65 70 SiO wt o/o 2 Figure 18. a to b: Shows abundances of elected rare earth clements plotted against Si02 wt%, and c-d: shows the abundances of two high field strength trace clements plotted against Si02 wt%. fo r volcanic suites from Savo, Gallego, Faure, Choi scul, Gold Ridge, and Bagana (prc-1943 to unpublished, inclusive). 42 12 1- T '" [ 0 ew Georgia Group Bougainville • 10 Faure • Choi ·eul • Gold Ridge 6. Gallego Savo ~ \l • 8 -·· 0 N ~ 6 ... . . ~ ~ 4 High-K 0 2 • 0 I 9j I 40 45 50 55 60 65 70 75 wt% Si0 2 Figure 19. K 0 wt% vs. SiO wt% for selected volcanic suites from the Solomon 2 2 fslands and Bougainville. Discriminant boundaries between low-, medium- and high-Kare those of Gill ( 1981). 43 12 ~ 0 New Georgia Group Bougainville • Fauro • 10 • Choiseul Gold Ridge t:,. Gallego 'V Savo trachy II a nd e~ ite 8 ~ • • dacite 4 i- •• • ,... -j 0 I I I 40 45 50 55 60 65 70 75 wt% Si0 2 figure 20. Total alkalies vs . SiO wt% fo r selected vo lcan ic suites from the Solomon 2 Islands and Bougainville. Discriminant boundaries arc those of Le Maitrc et al. ( 1989). 44 MnO, MgO and CaO wt% have negative correlations when plotted against combined alkalies (K:P + Na20 wt%). 2.2.2 Trace Element (bulk) Geochemistry The analytical results for trace elements for Choiseul and Gold Ridge are given in Table 2 and for Fauro in Table 3. The trace element data for Bougainvillc, Savo and the Gallego Volcanics are given in Tables 34-40, 41-42 and 43 respectively (in Appendix 2). The trace element geochemical relationships of each individual volcano arc described here based on: I) individual group behaviour; and 2) chondrite-nonnalizcd REE patterns and primitive mantle- normalized '·spider diagrams". 2.2.2.1 alkalies, alkaline earths, transition elements, and rare earth elements i) Large low-valency cations (alkalies. alkaline earths, and Pb) These elements have the highest abundances compared with other elements (i.e., the rare earths) of similar crystal melt (in)compatability, within the calc-alkaline rocks of Bougainville and the Solomons. Cs. Rb. Ba. Pb, and Sr increase in concentration as Si02 wt% increases (Fig. I 7a-c). Gold Ridge has high Rb (70-80ppm) and Ba but low Sr abundances. Rb and Sr concentrations are relatively high for the rocks from Choiseul, Savo, and the Fauro granodiorite (40-60 ppm and 900 to 1600 ppm respectively) while the rest have similar values (I 0-40 ppm Rb and 400-900 ppm Sr). The basalts from Fauro have the lowest abundances of these clements among the various volcanic suites analysed. ii) Ferromagnesian and chalcophile elements (Zn, Cu, Ni, Sc, V. Cr, and Ga). As is typical of most arc-related basalt-dacite suites, Sc, V, and Cr arc negatively correlated vs Si02 wt%, most probably as a result of fractional crystallisation of clinopyroxene and Fe-Ti oxides from mafic precursor melts (Fig. 17). The basaltic rocks from Fauro have higher abundances of these particular elements compared with the other selected sample suites (see Fig. 17). iii) large high-valency cations (Th. U, Zr, Nb, Hf). Zr and Si02 abundances are positively correlated (Fig. 18) while the rest of the elements in this group have no systematic correlations with Si02. The Maetambe rocks from Choiscul have the highest Zr and b 45 Gr., >Cl r' \.•fl• ,:1 80 -< Granodoonle 70 2 0 ·g 60 \ • r Figure 21. Rare earth clement abundances normalised to chondritic meteorite va lues (Rollinson, J 995) for selected rock rypes of the Fauro volcano in the Shortland Islands. -, Kumboro ma tic Jnd'l5. tu 50 r I< \1ll\bOrO•.i'1dl'S1t1• ·E --<> Kumboro-andeslle (Kagt>au) 0 ~ 40 • Q) E • l.i...ifot;1mb, .. 1r.1chy .tn,1es1t" Q) • ... 30 Goldr1d9c ·'5•·~t1 1 df'\·t c vol t:r "O 0 ~ c: (, ~ .. · l c . 0 • ' .c It R ~ 20 " ' ..... Vl GoldR id9e-s1-a ndos111c vol b r Q) ..._~ • (.) c: ...... _ • Figure 22. Rare ea rth element abundances normalised to chondritic meteorite values (Rollinson, 1995) for the calc-alkaline volcanoes of Choiseul (Kumboro and Maetambc) and sili cified volcanic breccias from the Gold Ridge volcanics (GoldRidge-si-vol.br). 46 abundances ( 150 and I Oppm respectively), while Zr in most other suites falls within a range of 50-l 50ppm. Rock samples from Gold Ridge have the lowest Zr (-<50ppm) and Nb (- I ppm). (iv) Rare-Earth Elements (REE) and Yllrium. The rare-ea11h clements (La. Ce, d) and Y show no correlation with Si02 (see Fig. 18). I Iowever. the Si02-rich rocks (granodiorites, andesite, dacites) of Fauro. Choiseul, Gallego, and a couple from Savo 2 have high values of La (14-30ppm) and Ce (30-56ppm) compared with rocks from Bagana. The tholeiitic basaltic basement rocks of Fauro and the volcanic rocks from Gold Ridge have the lowest values of La (<5ppm) and Ce ( 2.2.2.2 REE: Chondrite-Normalised patterns The combined normalized REE patterns for different volcanics of Fauro. Choiseul. Gold Ridge. and selected Bougainville volcanoes are given in Figures 2 L 22. and 23. All volcanic rocks have similar patterns and flatten off between Tb and Yb in similar fashion to many arc type volcanoes (Rogerson et al. 1989; Arculus and Johnson, 1981 ). The granodiorite intrusive from Fauro has high light REE (La 60-80ppm). followed by the andesite (La 30-50ppm) wh ile the basaltic basement has the lowest (La 20ppm). The REE values for the basalts remain relatively constant throughout while the two porphyries seem to plot in between the andesite and the basalts. The relatively silica-rich samples (granodiorite, andesite) have high l:REE abundances compared with the silica poor rocks (basalts). The light REE concentrations are higher in the samples from Choiseul than those from Gold Ridge areas (Fig. 22). There arc two well defined trends observed within the two Choiseul volcanoes. The Maetambe matic andesite and Kumboro rnafic andcsitc have 50 ppm La while the Kaghau andesite of Kumboro has much lower La (30ppm) concentration. 47 692 60 727 a ~"' -<> 728 ~ 50 • 0 Q) • Q) E 40 735 Q) • " ·.: '\ 734 "O . c: 30 • 7 ll 0 ·,...... l .c. iE',, " 7l ~ ...... e.... 717 Q) 20 ...... • 0 ' • ... c: - -~ 754 Ill r::-~-- "O ~ -• ..., 745 10 ~ - c: ------!? -ill 767 :l ~ .0 < ... ·~ 691 0 ~~ La Ce Pr N d Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu 689 60 r "'.. 711 ~ ffi b -<> 708 0 \ !.. 50 • E ~ ! 40 ' 796 ~ " ~ 804 -0 )E, c: 30 • • 800 0 ~ • •~ . / .c. ~ 798 ~ o ~ly- " '):- -J ..... _ .. 20 • 773 0 r tr • <:: ~ "· ... 805 - ' - ~ - -0"' .... 752 <:: 1 0 = -- ~ - - - -..... _ - ~ .. ' -... -ti. :l v -.... - .. - ~ . • 698 .0 ... •i ., < .; 791 0 La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Figure 23. Rare earth element abundances normalised to chondritic meteorite values (Rollinson, 1995) for selected volcanic rocks from Bougainville divided into two figures for clarity; data from Rogerson et al. ( 1989). 48 Volcanic rocks from Gold Ridge arc all low in REE abundances (La I Oppm). In this sense, the volcanics from Gold Ridge resemble the f-auro tholciitic basaltic basement (Ridgeway 1980). The REE concentrations fo r selected Bougainvillc rocks have similar geochemical characteristics to the rocks Crom Fauro. Choiseul. and Guadalcanal (Fig. 23). All the REE patterns slowly decline from La to m. and some have negative Eu anomalies. Samples 'GV# 698 and 798 (sec Fig. 23b) have somewhat unusual character[sitic:;. Sample GV# 698 has a strong negative Cc anomaly compared to the rest of the samples. probably the result of alteration and formati on of a RE carbonate. Sample 1 GV# -w; is very low in La (2ppm). probably the result of analytical error. 2.2.2.3 Primitive Mantlc-NormalisccJ "Spider·" diagrams The multi-element spider diagrams (Figs. 24 to 26) have a more hcterogenous mix of elements than the Rl ~ I ~ with a variety of peaks and troughs that reflect the different behaviour of different groups of trace clements. The more fluid-mobile. UL clements (Sr. K. Rb. Ba) are enriched relative to REE of similar melt-crystal incompatibility. comparable with other island arc type rocks. Gold Ridge samples arc exceptional (Fig. 24b). having the lowest r concentrations (<50ppm) compared with the rest of the analysed samples \\ ith high r values (>80ppm). In addition to this behaviour, recall that the Gold Ridge Volcanics have the highest K values. The abnormal abundances in the Gold Ridge suite rnay reflect the type of alteration (silicification) experienced by the samples. The less mobile HF clements (Th. Ta, 1 b, P, Zr. ll f. Ti. and Y) arc present in low abundances. Most samples are deplcti..:d relative to MORl3. In general it is known that the concentrations of the I IFS clements arc primarily controlled by the abundances in the 49 source rocks. degree of melting that occurred during the initial formation of parental melts. and the extent of fractional crystallisation that has occurred en route to the Earth·s surface. The "spider" diagrams arc generally sim ilar for the neighbouring ,·olcanocs of Savo and Gallego. but the Savo suite tends to be relatively enriched in the LIL clements (Figs. 25 and 26). In addition. there appear to be l\\'O distinct suites in the Savo sample set: one has relatively high concentrations while the other is low in LIL clements (like MORB, Fig. 24a). Given that the data were acquired by XRF, there may be problems of analytical resolution for the abundances of clements such as Th and lb in the Bagana suites (Fig. 27). Overall however. there is considerable o\·erlap in trace clement abundances for the various 20'11 century eruptive products. There is a greater degree of variability in the older eruptions with some samples containing distinctly higher Rb and Ba than the 20'h century samples, and others that have relatively flat, enriched but unfractionated relative to primitive mantle, abundances of thl: alkalies and alkaline earths. 50 C alc · 3lkallnc (Island Arc) 160 a • • Tholelo11tic (lsl;ind Arc) 140 • M ORB Fine gr<1111cd B asa lts QI 120 ... An de site + m af1c xcnolith s ~ 100 I • Granodiorit e 0 I .r:. • Mrncr::ilrzcd porphyry ~ 80 QI • \ 0. • • E 60 • • C'O (/) 40 • • ' '\ • 20 • \ • • n• 0 400 b l.latrc andcs1te + xenollths 350 ...QI 30 0 And wost of Laona • Malle andeslte ~ 250 • 0 F i ne grain andes1te .r:. ~ 200 Trachy andC'<•IC QI 0. S il1c1t1od volc;1n 1c brecc1a E 150 S 1loc1t1od volc;1n •C brccci;i C'O (/) \ • 100 • S1loc1hcd volc11111c brecco:i__, 50 0 Sr Rb T h Nb P Hf Ti Yb K Ba Ta Ce Zr Sm y Figure 24. Trace clement abundances nom1alized to "primiti c mantle" (here labelled chondritc for convenience: note however, the abundance of clements such a Kand Rb arc much lower in the primitive mantle than in chondritc ), arranged from left to right in order of incrca ing compatibiliry in the mantle. The norrnalizing values arc from Rollinson ( 1995). a selected volcanic rocks from Fauro, together with representative calc-alkalinc and tholciitic arc rocks, and average -type MORB (Rollinson, 1995). b Choi cul & Gold Ridge (bottom 4 in the legend) sui tes. 51 250 a 200 .. OJ ~- 150 c ... 0 .s::: ~ 100 c. E ra 5 0 Cl) .... • • ,• ' D ~ I'- & ' , ... 0 - - - ii ,..,,,,- - J_ -5 0 300 • ~ b 250 ..~ ';:J ~.... 200 ,' ,'i c J ~ 1 5 0 " ~ () . -OJ c. 10 0 E ra Cl) 5 0 0 .- - - 0 - 50 l Sr Rb Th Nb P Hf Ti Yb K Ba T a Ce Zr Sm y Figure 25. Trace element abundances normalized to "primitive mantle" (here labelled chondrite for convenience) arranged from left to right in order of increasing compatibility in the mantle. The normalizing values are from Rollinson ( J 995). Both parts of this Figure display data from Savo divided into a and b for clarity; data from M.G. Petterson and R.L. Stanton. 52 7 00 600 a ~ 500 ... -g 400 0 .c ~ 300 0 Q. 200 ..E (/) 1 00 u - - i 0 -- o - (.J -100 200 b 150 I 0 • ...... I "O ~ ~ 100 .c ~ 0 \ Q. 50 " \ ..E (/) Q 0 o- - -50 200 c I 150 I !!! ... i'l , ~ 100 \ I • 0 .c • 0 • • a; .. 0. 50 • E • 0 ' -50 Sr Rb Th Nb p Hf Yb K Ba Ta Ce Zr Sm y Figure 26. Trace clement abundance normalized lo "primitive mantle" (here labelled chondritc for convenience) arranged from left lo right in order of increa ing compatibility in the mantle. The normalizing value arc from Rollinson (1995). All parts of thi Figure di play data for variou (46) amples of andcsile - dacitic rock of Gallego; data from M.G. Petter on and R.L. Stanton. 53 2SO a 200 pre-194 3 #6 ... "O 1 so c: 0 .c ~ 100 QI a. .. E SO " ~ (/) u - - n 0 n - -SO 100 b 80 ·E 60 t- "O c 0 .c 40 ~ ~ 0. E 20 \ (1) l (/) n 0 -20 120 c 100 NGV#759 • NGV11690 ' 1\•1 ! 80 I -C NGV#689 ·;:;: "O • NGV#357 c 0 60 NGV#771 .c .t. ~ NGV#770 G> 40 0. E (1) (/) 20 • • . -· \ • 0 • ·20 S r Rb T h N b p Hf Ti Yb K Ba Ta C e Zr Sm y Figure 27. Trace clement abundances norma lized to "primitive mantle" (here labelled chondritc fo r convenience) arranged from left to right in order of increa ing compatibility in the mantle. The normalizing values arc from Rollinson ( 1995). All parts of this Figure are for the Bagana volcano of Bougainvillc (data source·: Bultitude et al. ( 1978) and Roger on et al. ( 1989). Li ted b abundance for ample prc-1943 #6 (upper panel) must be eIToncous. 54 CHAPTER 3: MELT INCLUS IO STUDIES 3.1 Introducti on The chemical composition of Mis can potentially directly record the evolution of the magma in which the host crystals have resided (e.g., John on ct al., 1994). The Ylls arc formed when silicate melt droplets arc trapped during the various stages of magmatic evolution. 3.2 Evolution and origin of melt inclusions The Mis trapped in magmatic rocks can optimally represent an instantaneous sample of homogcnous magma at a set stage or magmatic evolution (Roeddcr 1979, 1984). The M fs may also record variou geochemical characteristics adopted by the source magma at an earlier stage or its evolution, including volatile loads (Roedder 1992). M is can also be formed from hetcrogenous fluids, i.e .. immiscible magma( )-fluid systems (Rocdder 1991). Given that volcanic whole rocks arc genera lly a mixture of cry tats formed at varying stages or a magmatic evolution, it is also possible that MJs can preserve examples of variou ly evolved liquids that arc otherwise obscured by the presence of crystals. The composition of an evolving magma may change drastically from a low-volatile-bearing silicate melt through lo a hydrous silicate melt (possibly yielding pegmatite) to late- tage water-saturated melts and fluids (coexisting high density brine and low density vapour) which can give rise to quartz veins and perhaps the formation of ore deposits. It is possible to distinguish between two kinds of ilicatc M is depending on their origin and evolution. The first and the most common group of Mls arc primary. These inclusions occur either as isolated inclusions from the rest of the host rock or have formed along growth surfaces of host crystals. These types of M rs can occur within both the cores and rims of phenocryst phase . The second type of M rs arc secondary, normally occurring along healed fractures within the phenocrysts. fl is also well known that globule of immiscible sulfide melt can form from basaltic liquids. ft is therefore important to know at what stage in the differentiation sequence the silicate melt become saturated with respect to sulfides since some commercially valuable 55 Figure 28. a and b:Two plagioclase phenocrysts, - 0. J 1mm in length each (from Maetambe volcano, Choiseul) that have incorporated lots of transparent glass inclusions (maximum size of glass inclusion around 8-1 Op m). The Mis tend to be confined to the centre of the plagioclase together with inclusions of Fe-Ti oxides. Cracks and fractures are common within these types of plagioclase phenocryst. Figure 28 c shows clear, glassy, primary Mis (8-1 Opm) in plagioclase (O. l 5mm) that are confined to the centre of the phenocryst. Dark ci rcular and rectangular features probably are variously vapour bubbles, sulfide globules, and dark spine!; d shows characteristic types of chain like, inter-connected clear, glassy Mis that are sometimes found in these plagioclase phenocrysts. 56 Figure 29. a under cross-polarised light that shows tiny scattered, opaque filled Mls located within adjacent clinopyroxene phenocrysts. b Enlarged section of a labelled as "b" showing the Mls which all appear to be isolated and also contain bubbles and opaque minerals. 57 Figure 30 a-j . Images of plagioclase-hosted Mis taken under plane polarized light. h Clinopyroxene-hostcd Mis. All scales are in pm. 58 terrestrial deposits have formed partly because of the process of an immiscible sulfide melt (Naldrett et al, 1979, in Roedder 1979). 3.3 Rock associations and Petrographic Studies of Mis Si licate Mls are present and well developed in extrusive volcanic rocks such as volcanic tuffs, pumice or rocks that have cooled immediately after volcanic eruptions. Mls occur in some intrusive rocks but tend to be poorly developed, and have often completely devitrified and altered to an aggregate of microcrystallites. In this study, different types, sizes and characteristics of Mis are recogni sed in various phenocrysts collected from different volcanic rocks on the islands of Bougainville, Fauro, Choiseul, and Guadalcanal. Most Mis are located within plag.ioclase phenocrysts (most common mineral), Fe-Ti oxide, clinopyroxene (all except Gold Ridge samples), and quartz (all except Choiseul). The shape, form, and distribution of M is depends on: 1) the types of phenocryst in which they are trapped: 2) whether they are primary or secondary in origin; 3) the temperature of entrapment; 4) trapping mechanism; and 5) depth of formation of individual melt inc lusions. Figure 28 (a, b, c, d) shows various examples of clear, glassy, homogenous, primary Mis trapped inside plagioclase phenocrysts. The Mls are generally rounded, rectangular, diamond-like, or hexagonal in outline. The M i s are either randomly distributed (i .e., scattered) or form concentric patterns (see Fig. 28). From their shape and distribution, the Mls can be considered to be either primary or secondary (Takenouchi and lmai, 1975). The sizes of individual inclusions vary from a few microns to several millimetres. Figure 29 attempts to illustrate the general occurrence of some of the Mls located inside clinopyroxene phenocrysts in a sample from Choiseul. In this study, the Mis are observed to be mostly clear, relatively transparent, glassy, p1imary, and variably contain one or more crystalline phases . The Mis appear to be isolated as individual glass blebs (see Fig. 30, type la) or linked in chain structures (see Fig. 30, type le). The Ml s themselves sometimes have inclusions of dark spots (globules) of sulfides (with cubed, rectangular shapes) or even dark circular bubbles (Fig. 30. type 2). 59 There are few cracks developed within the crystal hosts, but these cracks do not appear to have influenced the formation, al ignment, or even the shape of the Mis. It is concluded that Mls are primary and were initially trapped on growth surfaces of their host minerals. 3.4 Different types of Mis The Mis can classified into a number of groups. on the basis of features such as: transparency, the presence of bubbles and opaque phases, and any occurrence of crystalline phases. Various types of Mis that have been identified in this study are shown in Figures 15a-j and are classified as follows: Type la. Monophase transparent glass Type lb. Di-phase transparent glass Type 1c. Chained-type structures of clear transparent glass Type ld. Elongated, long prismatic clear transparent glass Type 2. Transparent glass+/- minute opaques, crystals and bubbles Type 3. Brownish glass+/- minute opaques, crystals and bubbles Type 4. Ruptured glass+/- minute opaques and crystals Type 5 Devitrified glass+/- minute opaques and c1ystals Type 6. Opaque (including sulfide-rich)) The Mis of type I, 2, and 3 are the most common type observed, and are mainly hosted within plagioclase. Type 5 Mis are common in plutonic rocks such as the Fauro granodiorite. 3.5 Types of data recoverable from the analysis of Mis Normal Mis are defined as inclusions trapped during magmatic differentiation by fractional crystalli sation. The importance of normal Mls to petrology is that they document the line of liquid descent, compositional va1iation, and consequently sequential magmatic evolution (Roedder 1979). Previous studies show that the data which can be obtained, include: i. Trapping and closure temperature - according to Roedder (1979) this is the temperature of crystallization of the host mineral at the instant of melt entrapment, and is normally given by the temperature of homogenisation of the inclusion contents (gas. glass, and crystals) to a uniform single melt phase when 60 the sa mple is heated in the laboratory. This temperature is however, the minimum trapping temperature; ii. Constraints on maximum and minimum cooling rates - In the process of cooling, any Mis trapped in a crystal may have been affected by a range of processes. When cooling is fast, all sizes of inclusion will consist solely of glass . At intermediate rates, Mis will contain more phases and typically nucleation of a sulfide globule, a bubble, or epitaxial Fe-Ti oxide and plagioclase will occur. A much slower cooling rate maintained over a range of temperatures allows nucleation and growth of crystalline phases suc h as pyroxene, plagioclase and Fe-Ti oxides from the remaining melt (Roedder 1979); iii. Bulk composition and liquid line of descent - Mis record the composition along a liquid line of descent of the magma. The true composition of the melt trapped in an inclusion is of great potential importance to petrology. After the inclusion has been trapped, the host minerals generally continue to crystallize from the inclusion wall, and the embedding of the daughter mineral into the walls sometimes indicates that this process has taken place. The melt/crystal boundary layer and the degree to which the composition of the melt trapped in an inclusion con-esponds to that of the bulk of the melt adjacent to the crystal is sti 11 an unanswered question (Roeder L979). The major element composition at the time of entrapment is essentially determined by two factors according to Nakamura (1998). These factors are: 1) geometric relationship among the reactants and liquidus isothe1m in the phase diagrams of the system; and 2) kjnetic factors; iv. Volatile content - Mls can be rich in volatile materials such as H 20, C02, S species, Cl, and F. Analytical resu lts that yield low totals maybe due to the presence of any of these volatile compoennts. When there are secondary (later) fractures intersecting an inclusion, the open inclusion may have expelled melt and fo1med a large dark vapour bubble. 3.6 Mis resulting from melt-fluid immiscibility The history of individual Mis may be even more complex than described above. For example, some M1s may form from nonhomogeneous fluids, i.e., those formed from two or more immiscible fl uids. Roedder (1979) defines immiscibility as that combination of two or more phases that do not mix under specific conditions. The significance of Mls resulting 61 At some stages in magmatic differentiation processes, immiscible water-rich fluids evolved from magma may separate into a brine (~85-wt o/o dissolved salt) with coexisting H:P-rich vapour (Kozlowski and Karwowski 1973 in Roedder 1979). These fluids and vapour may collectively be involved in the formation of many of the porphyry Cu(+ Au) deposits of the world (Roedder, 197 lc; Nash, 1976). Figure 31 (adapted from Figure 41 of Roedder, 1979) documents 3 possible pathways of magmatic differentiation as interpreted from magmatic inclusions of melt, fluid, and vapour, showing possible relationships between crystallization and immiscibility. The first pathway is that of normal magmatic evolution, resulting from fractional crystallisation in the path from basa lt to rhyolite (granite). Immiscible, water-rich, relatively low-density fluids may evolve at any stage in this evolutionary path because of a change in composition, decrease in confining pressure, or increases in vapour pressure. Path 2 is characteristic of relatively reduced and anh ydrous, Fe-rich (tholeiitic) magma types (including lunar basalts). After crystallization of perhaps 95% of the melt, these types of magmas become enriched in iron, alkalis and silica. lmmiscible globules of high-Si melt (high-K-granite composition) coexist with a fen-o-pyroxenite mell. lmmiscible sulfide melt can also form. The last path (path 3) is that followed by relatively rare, alkalic magma types. These magmas can separate into immiscible melts of alkali-rich carbonate and silicate. 3.7: Studies of Mi s in Selected Solomon Islands Volcanic Suites 3.7.1 BougainvilJe Volcanic Suites Mi s are present in all of the volcanic rock types studied from Bougainville. The Mis are present in plagioclase, clinopyroxene, Fe-Ti oxides, and quartz phenocrysts. The Mis that are found in Fe-Ti oxides are easy to identify using the scan ning electron microscope because of the contrasts in reflective colours. Generally the sizes of the Mls range from 62 LOW-Donso sol. Phoumatotyt1c ? DENSE C02 High Dense Sot Hydrothermal BASALTIC MELT L-- ---1...J GRANITIC MELT : ' 2 ~ . ..I K-GrantticMenl .. Hydrosallne ~~~c;>us ~!SULFIDE MELTI I l·.. I L.. IFerropyrox. Met1 I i....:.;M.:.:e::..:lt ______, 3 ,..,,,Ba-s-,c--.,..At"'"ka-=t-.-r-,c..,..h""' ·!" .. I.... _D_o_ns_c_c~o_2 __ _, METTALIC' MELT Melt? Carbonate Melt H20 rich SOI. (+/·Sil, H20. Alk) Hydrothermal Hydrocarbon Fluids Fe·OXtdC Melt? ( +/· P, F. St. etc.) tmmiscibihty trend Crystal fraclionation trend Figure 3 1. Hypothetical outline of possible routes of magmatic differentiation, as interpreted from magmatic inclusion of melts, fluid, and vapour, showing the possible relationship between crystal fractionation and immiscibility [adapted from Rocdder (1979)]. 63 sub-rounded to angular morphology with abundant cracks or fractures present within both the M Js and the hosr crystal. Within the Fe-Ti oxide host phenocrysts, crystal-Mis contacts are sharp and no compositional gradient is apparent within the inclusions. In many cases, the inclusions appear to occur at the centre of the host mineral. and arc isolated (i.e., they are primary inclusions). ln a couple or samples, the Mis are not randomly distributed throughout the crystal but occur in concen tric bands, which could indicate that the inclusions were trapped by a sudden episode of crystal growth in response to some perturbation to the magmatic system (Roedder, 1984). M i s in plagioclase display a more concent1ic structure within their hosts. These M rs have generally partly or wholly crystallized. T he crystal compositions developed within the melt include small ( There i also however. evidence of cracked or fractures urrounding some M ls. According to Roedder ( l 992), there are three types of crack that can form from or within any inclu ion. These arc: 1) short, small subtle cracks which radiate from curved areas or corners of the inclusion walls; 2) large and extensive cracks which extend from the inclusion to the crystal edge: 3) Jarge and extensive cracks not directly related to the inclusions. All these cracks have fractured the inclusion glass in one way or the other and therefore must have formed after the inclusion \.vas relatively cool. Most observed cracks and fractures (within M i s) have formed after the inclusions were trapped si nce all rractures/cracks do not cut through the M i s: the M is appear to be isolated. A couple of samples show that the melt/glass of the original M i s has diffused/seeped into the surrounding fractures and cracks. Vapour bubbles within M is of the Bougainville samples arc rare and generally small (present in 3.7.2 Major Element Compositions of M is a nd Host Minerals The major clement compositions of the Ml s with respect to the host phcnocrysts were obtained by SEM and EMPA - the major element data arc tabulated in Tables 4-6. 64 Analyses were performed on Mls that are exposed on the surface of polished thin sections. It is relatively straightforward lo detect overlaps of MI glass compositions and host crystals because of the distinctive compositional characteristics of individual hosts. For example, the high CaO of clinopyroxene and plagioclase hosts, or high Ti02 of Fe-Ti oxide hosts are distinctive indicators of probe beam overlap. The Si02 concentrations of the glassy Mls fall predominantly within the range of 55 to 78 wt %. The Mfs are very rich in alkalis (up to 10 wt% K20 and 8 wt% a20) but have low MgO (usually < l wt%). Total Fe is mostly <4 wt% expressed as Fe20 3 . The Mls are generally more silicic than the Bougainville host rocks, and range to higher combined alkali contents (Fig. 32). The covariacion of a number of major element oxides of Mis hosted within Fe-Ti oxides are plotted against Si02 in Figure 33. CaO , and total Fe (and interestingly Na20) decrease with increasing silica concentration, while K 20 displays a positive correlation with silica. The variation of Al20 3 is one of initial increase and then decrease as a function of increasing silica. These trends are mostly similar to the general major oxide variation trends observed for calc-alkaline volcanic rocks (e.g., Gill, 1981), and are consistent with fractionaJ crystallisation of plagioclase-elinopyroxene-Fe-Ti oxide assemblages from mafic parental melts. The glasses are mostly daciiic-trachytic in composition, have both alkaline and sub-alkaline character (based on the classification of Irvine and Baragar, l 979). and are medium- to high-K (Fig.33d). 3.7.3 Fauro Vo lcanic Suites Mls are found in all rock types of Fauro. The i nclusions are present in Fe-Ti oxides, plagioclase, clinopyroxene and quartz phenocrysts, as is the case for Bougainville. The Mls are most abtmdant in Fe-Ti oxides (see Figs. 34 a-b) and less abundant in plagioclase and clinopyroxene crystals. The Mis range in size from < I 0-200µm in diameter, with most being between 10 and 30~tm in greatest dimension. 65 Table 4. M l trapped in clinopyroxene phcnocrysts. Bougaill\ illc - Clinopyroxcne Q\tt/4• 145.,e.,·&I ?./J·<£1t·i;:! 7./J· S102 64' 7''.I 73 5 73 ..: ~07 MO 55 ' MO iS I T102 o.i 02 02 0.1 02 o~ 02 01 02 1\1203 II o IH I 5.7 l<>.4 15.l\ l:l.1 111'.I 17.'I 10.11 fc201 ~ I I .l 1.5 o.s 23 2.1 4.7 05 lb MnO 0.1 0.1 0 .1 0.1 0.t 0 I 0 I 01 01 M~O 5.<) O. t 0.1 0.1 0.<) l.S 5.0 02 t s CnO 10 .l u I 4 1.7 3.3 3.5 11.X ()! t..2 Na20 0.9 \<) '0 2.0 3.0 n 4 3 OJ 30 K20 u 4 I 44 5.7 3.8 8.4 u 111.S IX Cr203 0.1 0.1 0.1 0.1 0.1 0.1 0 I 0.1 0.1 To1al IOO.O 100.0 1000 100.0 100.0 IOO.O IOOO 100.0 100.() Na20• K20 n 82 7.-1 77 6.8 10.7 S.9 1(1 7 4X Host crystals ()\1th• J.15"£!.'"' '4·'"f!'lr! '45·tf!.11tJ 7./.1"£>'"" 745- 66 Table 5. Mls trapped within plagioclase and quartz phenocrysts. Bougainville - Plagioclasc fe ldspar Bougainvillc - Quartz o.rnlc 79 I ·els·s6 79 t -eto-i; 7 OU5-eli;-g3 005-elx·x4 005 ·pls·s5 I!5 -pls·i; I "'£""h·_m..;..•·1'-11__ .:..:l -:..:'5_,·l("'':""'s"-t -'t:..::l..:..5·... q.;..;;1:,,_11-"-' _1;.;:1..:..5_.·q.;..;;1:.._i;3'--.;..;' -""'' 5-'.,1""1:.,s"--.s Si02 4 1.7 75.9 60.3 63.8 60.9 58.4 S102 78.0 86.9 7-U 76 3 Ti02 1..1 0.1 0.4 0.4 0. 1 0.1 Ti02 0.1 0.1 02 01 1\ 1203 13.0 IC..11 19.3 10.4 22.5 24.6 Al203 11.8 7h 10.5 10.8 FdOl 19.0 O.<> 2.3 1.6 1.5 0.8 Fc203 0.7 05 13 0.9 MnO 0.3 0.1 0.2 0.1 0.1 0.1 MnO 0.1 0.1 OJ 0.1 MgO 10.3 0 I l.S 09 0.1 0.1 ~lgO 0.3 0.1 07 O.l• C30 11.5 I.(• 3.5 2.7 4.5 7.$ C30 I.I 0.4 07 0.5 , a20 2 2 I.~ ,,.9 4.7 7 7 7.0 '1a20 2.6 1.0 3.6 2.6 K20 3.1 5.5 5.4 2.5 I.I K20 5.3 3.3 4,9 4.9 1'205 0.0 o.o 0.0 0.0 0.0 0.0 1'205 0.0 0.0 0.0 0.0 03 0.0 0.0 0.0 0.0 0.0 0.0 $03 0.0 0.0 0.0 0.0 Cr203 0. 1 0. 1 0.1 0.1 0.1 0.1 Cr203 0. 1 0. 1 3.4 3.2 Total 100.0 100.0 100.0 100.0 100.0 100.0 Tomi 10().0 100.0 100.0 100.0 Na20+K20 2.8 5.0 12.4 10.1 10.3 8.0 Na20+-K20 79 4.3 7.5 Host crystals 79 Jpls-h-8" 005-etx-hi 115-els·/JI ::E:;.;le.:.:."':.:.''':.:.:"--:..:'l:..:5.;.:·qr.:;r.::.:h.:.l _l:..:!:.;:5_,,·q:;.;r:.::.i•l::..' ____ .:.;1-:..:' 5c.:·qi..::r.::.:hc:....4 $102 4<•.1 51.2 46.3 ··,02 99.2 99 7 99.3 no2 0 I 0.1 0.1 T102 0.1 0.0 0.1 Al203 33.2 30.0 33 ..' r\1203 0. 1 0.1 0. 1 Fc203 0.8 0.7 0.7 l·c203 0.1 0.0 0.1 MnO 0.1 0. 1 0 1 MnO 0.1 0.0 0.1 MgO 0.1 0.1 0.1 MgO 0. 1 0 .0 0. 1 Cao 17.S 13.7 17.9 Cao 0.1 0.0 0.0 Na20 1 7 3.8 1.5 Na20 0.1 0.0 0.1 K20 0.1 0.3 0.1 K20 0.0 0.1 0.0 1'205 0.0 0.0 0.0 P205 0.0 0.0 0.0 03 0.0 0.0 0.0 SOJ 0.0 0.0 0.0 Cr203 0.1 0.1 0.1 Cr203 0.1 0.0 0 I rornl 100.0 100.0 100.0 Total 100.0 100.0 100.0 Na20+K20 1.8 4.0 1.5 ' '320•K20 0.2 0.1 0.2 67 ,.., e 0 e 0 0 e 0 e e 0 e o e 0 c ~ = .,. .,. ,.., 0 0 0 0 .. ... <; e g 0 = 0 0 o e .,.. 0 0 ,,, 0 0 .... e 0 rt 0 8 .... 0 0 8 0 vi (.) .. 0 0 c: 0 0 0 e co (.) 0 > ., 0 () e' 0 > c: co 0 ., 0 ; ,... C/) ::: 0 0 o e .e ::I co0 0 0 0 ° ~ 0 0 0 0 0 8 0 0 .. 0 0 e .,., e g ;:: ..:. = = e ....- 0 0 0 0 0 e .,., 0 0 0 e .e / 0 0 ..z .. e <; 0 N 0 0 0 ,..; 0 = ci 0 8 0 ....'-' ....0 >.: >.: + + ~ 9, 0 0 g ~ ~ ~ Q s ~ 9 ,., 0 0-: X; ~ ., 0 ,., 0 0 0 ('l ~, ,..; 9 8 e ., o e 0 0 ,.../ ., ,., .,., e: 0 0 0 ..... 0 ., ,... ::> "' ,., ,... .,. = 0 C' ., ., 0 0 x 0 0 0 ...... 0 ; g e o g 0 ;r. N 0 ,.. 0 c .,· ,.; ':f 8 0 .., "0 e 0 ,., 0 ,..; 0 0 .,.. ff, N N ff, .... 0 ri ~ ri "' ... 0 . 08. . '1 0 "' O c: ro 0 '"! <.> 0 0 0 > 0 ..... 0 e g ; 0 e o e 0 0 0 ,...> 0 0 ro ,., e 0 0 co ~ 8 c:i ::,.• e 0 :::i e g e ::3 0 co '! Vl ,., ., ..... 0 0 0 ::3 e ., ei 0 0 0 -"' = = ,.;'" C? 0 .... 0 e ("'i 0 "'0 0 0 ~ ci g 0 0 0 .., ,., ..... 0 v. 0 ... 0 0 ~. c "'0 o e 8 " 0 0 0 .., 0 0 0 ei e "! v. ., 0 = ..> 0 ""· e e 0 e 0 "O 0 c. c. ro ,_ 0 ., 0 ,...... ei 0 e 0 -Vl 2 ,..c :..: :..: + 8 0 0 0 8 g 0 0 "'0 -;:; ~ ~c:.o0~2No~ ,.... ('J ~ ~ ~ 0 % ~ ~ ~ % ~ u ~ -/ 20 O Whole Rock Bougainville Mis in clinopyroxene • Mis in plagioclase + Mis in quartz I Ml. in Fe-Ti oxide 15 • 0 N • • ~ • + 0 . • d" 10 • • •• z •• .,• ~ • • ~ • 0 • '· •• •: •• • 0 II ·o • .... • • 5 ' . ·-·. ••-< 0 o· • I 0 ...I.. ~~_;_I .l....._.L. .L 45 50 55 60 65 70 75 80 wt% Si0 2 Figure 32. Comparison of whole rocks and Mls hosted by va rious phcnocryst phases projected in terms of total alkalies vs SiO wt% 2 70 10 12 a d 10 8 8 ~ 6 0 *3: 3: 6 0 N 0 (1) N z ~ 2 2 mod1um·K 0 0 6 24 b e 5 22 20 0~ 18 *3: 3: 3 ... 0 0 (1) N 16 () 2 <( 14 12 0 10 20 16 c 14 15 12 alkaline *3: 0 10 *3: N ... 10 ~ 0 + 8 N Q) 0 LL N (1) 6 5 z sub-alkaline 4 0 2 50 55 60 65 70 75 80 50 55 60 65 70 75 80 S iO wt % SiO wt % 2 2 Figure 33. Major oxides of Mis plotted against sil ica for various Bougainville Volcanic rocks. The line separating the alkaline- sub alkaline fields in the total alkalies vs silica diagram (t) is taken from Irvine and Baragar ( 1981) 71 Mis in basalts, andesite and dacites tends to yield smaller size inclusion ( 'vVhich ranges from 50-SOµm and l 00-200µm but un fortunately al I t hese melts have devitrified to some extent. with a fe \;1,1except ions. The Mis trapped in Fe-Ti oxide host crysrals are sub rounded to round while others are almost squared-shaped unless the inclusions have been affected or disturbed by cracks or fractures (Figs. 34a-b). The inclusions are randomly distributed as individual grains and most of them are located within the centre of the crystal. The Mls-host crystal contacts are usually sharp, with a wel I-defined boundaries. Cracks and fractures are not common but wherever they are found, the cracks that have developed extend from the MI-crystal contact to the oxide crystal edge. These cracks and fractures are likely developed during cooling of the host rock, and are probably not related to to the origin and entrapment hi story of the Mls (see Fig. 34b). Some Mls contain vapour bubbles with dark globular structures on the inclusion walls. (Fig. 35). 3.7.4 Major Element Compositions of Mis and Host Minerals T he major element compositions of Mls trapped in various clinopyroxene, plagioclase. Fe Ti oxides, sulfide, amphibole, mica, and quartz hosts in various volcanic rocks from Fauro, are listed in Tables 7, 8. 9, 10, 11, and 12. The glasses of the Mis have wt% Si02 in the range 50-72 wt% , and tend to both higher Si02 and combined alkalies than the Fauro whole rocks (Fig. 36). The Al20 3 concentrations are 15-25 wt%. Compared w ith the inclusions in the Bougainville su ite, those of Fauro tend to be more erratic in terms of alkali content, and in particular the K 20 contents. For example, the K 20 in the Fauro Mls ranges from very low abundances (-0.2 wt%) to as high as 8 wt%. It is not clear however, despite every effort to circumvent the problem that these high values do not reflect the presence of minute feldspar/mica crystallites. Other compositional features of the Fauro Mls incl ude generally low M gO (usually <1 wt%) and Ti01 (< L wt%) concentrations, variable Cao (0.2-5.4 wt%) and total Fe as Fe20 3 (~LO wt%) . The alkalis (individually an d combined) ha ve a bi-modal compositional variation (Figs. 37a, d, and f). As a consequence, the Fauro glasses are bimodal alkaline 72 F igu re 34. Two well developed glass inclusions embedded inside an Fe-Ti oxide crystal with smooth, well-rounded edges (a) and sharp angular edges (b). Both figures (a-b) display cracks/fractures that radiate outward from the glass-crystal contact . •• •• • . .• . • · ·~_, • , . .. • • '~ , .... • Figure 35 a Shows M is trapped in a qua1iz phenocryst. All scales are in pm .. b Quartz grains of weathered, mineralised porphyry, Horn bill prospect. ote that despite the small sizes, all Mls have developed bubbles and some opaque minerals (probably magnetite or FeS) inside. 73 Tab le 7. Mis trapped within host clinopyroxene phenocrysts for various Fauro Volcanics. Fauro - Clino roxcnc O.tide 99-2-69g_2 99-2-69g3 99-2-69g.f 99-2-70,v.J 99-2-70g2 99-2-70g3 99-2-9gl 99-02-56gl 99-2-56g3a 'J9-65-g3 Si02 66.0 65.0 66.I 64.5 64.0 52.7 64.6 5 .5 64.2 58.0 Ti02 0.1 0.2 0.1 0.1 0.1 0.1 0.1 0.1 0.0 0.3 t\1203 20.8 19.6 21.6 22.0 22.6 29.4 28.3 14.3 22.2 19.-1 Fc203 0.-1 l.2 OJ 0.6 0.1 0.8 2.0 7.7 0.4 9.8 MnO 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.3 0.0 0.0 MgO 0.1 0.6 0.1 0.1 0.1 0.1 1.8 7.0 0.0 1.8 Cao I I. I 11.7 9.9 2. .3 2.9 11.7 0.1 7.0 3.9 0.6 Na20 0.2 0.2 1.0 10.2 10.0 4.9 0.5 4.8 9.0 9. K20 I. I 1.3 0.8 0.0 0.1 0.3 2.5 0.2 0.1 0.2 Cr203 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0. 1 0.1 0.0 Toial 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 'a20+K20 1.3 1.-1 1.8 10.3 10.1 5.2 3.0 5.0 9.1 10.0 Host crvst:1ls oxide 99·1-6Yh2 99-1-69/rJ 99-2-69/r4 99-2-7()h/ 99-1-70/rJ 99-1-9/r/ 01-56'1111 02-56h3b Si02 48.I 47.6 47.8 48.0 49.9 52.1 49.-1 52.5 Ti02 08 I.I 1.0 0.8 0.5 0.3 0.2 0.3 i\1203 5.7 6.1 6.0 6.1 3.8 5.1 5.9 5.2 Fc203 8.9 9.7 9.0 8.2 8.2 16.0 19.6 16.0 MnO 0.1 0.2 0.2 0.1 0.1 0.8 0.4 10.7 MgO 13.2 13.0 12.9 13.3 14.3 14.2 12.5 0.7 Cao 22.8 21.9 22.7 23.I 23.0 10.6 11.1 0.2 Na20 0.2 0.2 0.2 0.2 0.2 0.7 0.5 0.0 K20 0.0 0.0 0.0 0.0 0.0 0.2 0.3 0.0 Cr203 0.1 0.1 0.1 0.1 0.1 0.0 0.1 14.5 To1al 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Na20+K20 0.2 0.2 0.2 0.3 0.3 0.8 0.9 0.0 74 Table 8. Representati ve analyses of Mis trapped within host plagioclase phenocrysts fo r various Fauro Volcani cs. Fauro • Plagrocla~c fcl 75 Table 9. Representative analyses of M is trapped within host Fe-Ti oxide phcnocrysts for various Fauro Volcanics. Fauro ·Fe-Ti oxide 99-01-6glu 99-1-.'ig/ 99-J-56glh 99-2-.56;;.5 99-{}J./Ogl 99-J-6;;3" 99-Z-6g.5a 99-1-6/g6a 99-J-61;;7 99-01-Jg3 99-01-3g/ 99-01-5;;1 9'1-02-Vgl IJ9-IJJ.5JgJ 99-0J-53g3 Si02 69.6 653 577 623 (•1.0 72.6 726 66.1 65.-1 64.1 61.0 53.7 60.I 62.9 69.3 Ti02 0.1 00 0.1 0.1 0.2 0.5 0.1 01 0.1 0.2 0.0 119 02 0.2 0.3 Al203 17.8 18.4 23.-1 21 9 19.3 15.8 15.3 19.7 199 21.9 23.7 13 I 231 22 2 18 I ., ' Fc203 1.0 09 \9 I.I 3.7 3.6 2.6 2.2 2.2 1.0 1.7 1.6 -" 1.0 I 1 MnO 0.0 00 0.1 0.0 0.0 0.1 0.0 00 0.0 0.0 OJ> 0.0 0.1 0.1 0.0 ,\lgO 0.0 0.0 07 0.0 2.7 1.0 0.2 0.0 0.4 0.0 0.0 01 0.4 0.2 0.3 cao 0.8 00 6.6 3.5 0.9 0.1 0.7 0.6 0.7 2.9 5.0 9.1 5.1 4.7 3.9 N:'.120 8.1 1.6 7.0 4.9 5.3 0.1 7.3 11.3 11.2 9.6 8.3 0.8 7.4 S.2 6.6 K20 2.6 14.0 O.·I 6.1 6.9 5.9 I. I 0.0 00 0.5 0.3 9.4 0.9 0.2 0.4 Cr203 0.0 0.0 0.2 0.0 0.1 0.0 0.0 0.0 0.0 0.2 0.0 0.1 0.0 To1a l 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Na20 + K20 9.4 15.6 7.5 11.0 12.2 6.0 8.-1 12.0 11.0 16.0 10.1 8.6 10.2 8.3 S.-1 llos1 crvsl:lls oxule 99-01-6/r/ 99-IJJ.>/lt/ 99-0J-5(,ftJ 01-561t5a 99-01-10/r/ 99-J-61t3u 99-J.tilt.5 99.].6Jlt6 99.].()//17 99-01-21t3 99-02-3/i/ 99-02·.5/r/ t,19-112-9/r/ t)l).(J].531t1 <)t).()1-5Jlt3 Si02 0.3 1.9 03 0.8 0.5 0.1 0.3 01 0.2 0.1 0.4 0.1 09 0.2 () 2 Ti02 4 .0 7 7 5.2 4.3 5.3 +l.S 3.6 .,,_ ') l() 4.2 0.5 0.3 4.9 :n.o 23.0 ., ' Al203 I I O.R 0.6 0.3 0.2 0.3 I.I -" I 9 0.2 OJ 0.2 0.5 0.2 0.2 Fc203 94.2 ll7.l! 9-'2 932 93.7 54.3 9-13 91.6 94.2 94.5 98.-1 99.1 92.8 75 . .i 75.3 ~1 n0 0.0 0.5 0.0 0.-1 0.1 0.-1 03 Cl. I 0 0 0.6 0.5 0 I 0.0 0.6 06 MgO 0.0 07 0.1 01 0.0 0.0 00 0.1 01 0.0 0.0 Cl.I 0.2 0.1 01 Cao 0.0 00 0.5 0.1 0.3 01 Cl. I 0.1 0.3 0.1 0.0 04 02 0.2 Na20 0.2 0.9 0.1 0.2 0.0 0.1 0.1 OJ 0.3 0 2 0.3 0.1 () 2 0 I 0.1 K20 0.0 00 00 0.0 0.0 0.0 00 0-0 QO 00 0.0 00 00 Cr203 0.2 03 0 I 02 0.1 0.0 0.2 0.0 0.3 0.1 0.9 0.0 0 2 01 0.3 To1al 100.0 100.0 100.0 100.0 100.0 100.0 IOOO 1000 I 00.0 I 00.0 I 00.0 100.0 100.0 100.0 100.0 Na20+ K20 0.19 0.70 o.1g 0.21 0.05 0.09 0.05 OJ3 0.28 0.18 0.14 0 20 0.27 0.12 0.12 76 Table 10. Representative analyses of Mis rrapped within host sulfide phenocrysts for various Fauro Volcanics. Fauro - Sullide Ox id« 99-Ul-5/gl 99-1-5/gla 99-1-5/glh 99-1-5/gJ 99-2-51g4 99-l-5/g5a 9Y-l-18gl 99-2-Jl?gJ 99-l-18g4 99-l-18g5 Y9-l-18gf> 99-l -/Xg7 99-2-ll>g.\ Si02 <•5.4 54 7 66 3 65.S 65.4 65.6 52.0 67.6 64.5 65.0 53.9 52 2 59 4 Ti02 0 0 0.2 -0.9 0.0 0.0 0 0 0.0 0.0 0. I 0.0 0.0 0.0 0 I Al'.!03 21.4 :B.'I 20.3 20.5 15.5 20.0 110 19 7 21.5 21.4 15.8 17.4 19.5 Fc203 0 5 7 2 2.7 1.2 6.3 I 3 6 0 0 9 0 9 0.9 4.4 5 2 ~ 2 MnO 0.0 0.2 0 3 0.0 0.1 0.1 0 I 0.0 0.0 0 0 0.2 0 I 0 0 MgO 0.0 2.4 0.2 0.0 0.1 0.0 0.7 0 () 0.5 0.0 0.1 0.0 0.0 Cao 1.6 1.1 1.5 1.0 0.7 0.8 2. 7 0 7 3.0 2.5 0.0 2.1 D Na20 10.5 5.6 9 9 10.9 8.9 10.2 7.7 10.7 9.4 9.7 0.9 8.2 8.8 K20 0.5 4.1 0.0 0.1 0.1 u 0.6 0.2 0.2 0.1 12.9 0.2 0.2 P205 0 0 0.2 0.0 -0.1 0.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 S0 3 0 I 0.4 04 0.7 2.5 0.6 12.1 0.5 0.3 0.4 12.0 14.8 6.4 Cr20J 0.1 0 I 0.0 0.1 0.0 0.U 0.0 0.0 0.0 0.0 0.0 0.0 0 I To1al 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 IOO.O 100.0 Na20 + K20 11 0 9.7 9.7 11.0 9.0 11.5 8 "l 10.9 9.5 9.7 13.8 8.4 9.1 Most C r v~ta l Ond,, "'Jfiw 99-IJ2-5 I h I 99-2-Jlhltl 9Y-l-51'12b 99-2-51 '13 99-1-51 h4 99-2-51'15 99-2-18'12 99-1-18'13 99-1-18/i.J 99-1-18'15 99-2-ISM 99-l-llU17 IJ9-2-18'18 Si02 36 0.2 0.2 0.4 0.1 0.2 I)] 0.3 QZ Ql Ql Ql Ql Ti02 00 0.0 0.0 0.1 0.0 00 0.0 00 0.0 0.0 0.0 0 0 0.0 Al203 I I 0.1 0 I 0.2 00 01 0.1 0.1 01 01 QI 01 01 Fc203 0.0 47.4 47 4 46.8 60.0 60.0 ·16.5 46.5 46.5 46.8 46.ll 47 7 46.9 \lnO 0.0 ()() 0.0 0.1 0.2 01 0 2 0.2 0 I 0.1 0 I 00 0 I 1\lgO 0.3 00 0.0 02 0.0 0.1 0.0 0.0 0.1 0.1 0.1 00 0 I Cao 02 00 0.0 0.0 0.0 0.0 0.0 00 0.1 0.0 0.0 0 I 0 0 Na20 07 01 () l 0.1 0.1 0.1 01 01 0.1 0.0 00 0.0 ()() K20 02 00 0.0 0.0 0.0 00 0.0 00 0.0 0.0 0.0 0.0 0.0 P205 01) 00 00 0.0 0.0 0.0 o.u 0.0 0.0 0.0 0.0 0.0 0 0 s 94.0 52.4 52.4 52.3 39.8 39.9 53.0 53.0 53.1 52 6 52.6 52.4 52. 7 Cr203 ().J 00 0.0 0.0 0.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Total 1000 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 IOO.O Na20 ... K20 0.9 0.1 0.1 0.1 0.1 0.1 0.2 0.2 0.1 0.0 0.0 0.0 00 77 Table I 1. Representative analyses of Mis trapped within host amphibole, mica and quartz phenoc1ysts for various Fauro Volcanics. Fauro - am~hibolc Fauro - mica Fauro - guartz oxidet 99-2-69gJ 99-2-SgJ 99-2-69g5 99-2-9g_2 99-02-6g_4 99-02-20g2 99-02-618. 7 99-2-22g2b 99-2-22g4b 99-2-1Jg2a99-2-12g2 99-2-27g2 Si02 58.0 40.6 58.9 61.0 62.7 86.8 88.3 61.6 66.4 58.I 60.4 63.7 Ti02 0.1 2.4 0.1 0.1 0.1 0.2 0.1 0.0 0.0 0.1 0.1 0.1 Al203 29.4 14.3 32.0 27.9 25 .6 4.6 7.3 22.7 19.2 26.2 22.7 19.1 ., ., Fc203 .) ..) 12.3 1.8 2.0 1.7 3.0 0.1 1.5 1.9 0.1 0.4 0.0 MnO 0.1 0.1 O.l 0. I 0.1 0.1 0.1 0.0 0.0 0.1 0.1 0.0 MgO 3.5 14.6 3.8 4.3 0.1 0.1 0.1 0.0 0.0 0.3 0.1 0.0 Cao 2.2 12.I 1.8 1.9 0.7 5.1 2.2 4.1 0.7 6.2 2.0 0.3 Na20 0.3 2.7 0.2 0.3 0.1 0.1 1.7 9.4 11.0 8.4 0.1 1.3 K20 3.0 1.0 l.4 2.4 8.8 0.1 0.l 0.1 0.1 0.4 14.0 15.4 Cr203 0.1 0.1 0.1 0.1 0.1 0.1 0. 1 0.1 0.0 0.1 0.1 0.0 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Na20 + K20 3.3 3.6 l.5 2.7 8.9 0.2 1.8 9.5 11.1 8.9 14.1 16.7 Host crvstals oxide 99-2-69'1 I 99-2-8h I 99-2-69'15 65-3-m 99-02-6'14 99-02-201t2 99-02-61It7 99-2-22'12b 99-2-22'14b 99-2-27'12 Si02 41.1 40.4 43.0 32.3 99.3 99.4 99.4 98.6 99.2 99.8 Ti02 1.9 2.5 0.1 0.0 0.1 0.1 0.1 0.0 0.1 0.1 Al203 14.3 14.3 19.9 21.7 0.1 0.1 0.1 0.6 0.3 0.0 Fc203 12.3 12.4 17.8 26.3 0.1 0.1 0.1 0.6 0.2 0.1 MnO 0.1 0.1 0.5 0.7 0.1 0.1 0.1 0.0 0.0 0.0 MgO 14.5 14.5 13.4 18.0 0.1 0.1 0.1 0.0 0.0 0.0 Cao 12.5 12.1 0.6 0.1 0.0 0.1 0.0 0.0 0.0 0.0 Na20 2.1 2.7 0.6 0.5 0.1 0.1 0.1 0.2 0.1 0.0 K20 1.2 1.0 4.1 0.4 0.0 0.0 0.0 0.0 0.0 0.0 Cr203 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.0 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 I00.0 100.0 ~., Na20 + K20 .) ..) 3.7 4.7 0.9 0.2 0.2 0.1 0.2 0.1 0.0 78 Table 12. Representati ve analyses of matrix glasses for vari ous Fauro Volcanics. Fauro - Matrix oxide 99-02-56g I 99-2-5 I hlh 99-2-5 lg_5c 99-02-4g_2 99-02-4g_3 99-2-2 7g I 99-2-27g_3 99-2-27g_4 99-2-27g_5 99-02-9g2 Si02 58.5 66.5 66.2 64.7 63.9 67.9 68.0 68.0 67.8 63.4 Ti02 0.1 0.0 0.1 0. 1 0.0 0.1 0.0 0.0 0.1 0.2 Al203 14.3 20.4 20.6 18.5 21.8 20.I 20.2 20.0 20.1 20.7 Fe203 7.7 0.5 0.5 0.4 0.9 0.0 0.0 0.2 0.1 1.6 MnO 0.3 0.1 0.0 0.0 0.0 0.1 0.1 0.1 0.1 0.1 MgO 7.0 0.0 0.0 0. 1 0.0 0.1 0.0 0.1 0.1 0.4 Cao 7.0 1.0 1.5 0.1 2.7 0.2 0.2 0.1 0.0 3.5 Na20 4.8 10.8 10.6 1.0 9.6 11.7 11.6 11.7 12.0 9.0 K20 0.2 0. 1 0.1 14.9 l.O 0.0 0.1 0. 1 0.0 0.3 P20 5 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.7 S0 3 0.0 0.4 0.4 0.1 0.0 0.0 0.0 0.0 0.0 0.1 Cr20 3 0. 1 0. 1 0.0 0. 1 0.0 0. 1 0.0 0.1 0.1 0.1 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Na20 + K20 5.0 11.0 10.7 16.0 10.6 11.8 11.6 11.7 12.0 9.4 79 20 T r-r-r- -r-r-r ...,... r-'"" Fauro Whole Rock M is in clinopyroxene • M is in plagiocla e • MI. in amph-mica-quanz M is in r e-Ti oxide • &'.\ /\ Matrix glass 1 5 • • 0 • ("I ~ • ?A + • 0 • • • ~~p. · 1 0 • z«f' •• • • /\ e • ~ • • • 3 • •• • • •• • 0 • • (f) 0 0 •0 o • • 0 0 5 0 • D. 0 0 0 0 • • • • • 0 • • 0 L -l_ -1 45 50 55 60 65 70 75 80 wt% Si0 2 Figure 36. Comparison of whole rocks, MI ho tcd by various phenocryst pha cs, and matrix glasses projected in term of total alkalies v SiO wt% 2 80 14 •,• t n CP " 20 a '.II n pf.19 d 12 l'.' in QU 15 10 X Ml 1n amp :.!! • 0 8 A \'lnFLT 10 .... 3: ·,/ .... 6 • M l ~ ~ hd• • 0 ...... 5 z"' • / 0 0 • -5 14 35 b e 12 f 30 10 25 8 ;:: 3: 20 6 .... 0 0 .. 15 (.)"' 4 2 10 .... 0 .... 5 x ·2 0 20 20 c 15 alkaline 15 • 10 ;:: • 0 ...... 10 • >£.-.M sub-alkaline 0 ~ N 5 + ·~ · ·!f: .... Q) 0 ...... l.L...... z"' 5 0 • .5 0 40 50 60 70 80 90 40 50 60 70 80 90 SiO wt o/o SiO wt% 2 2 Figure 37. Major oxide concentrations of Mls and matrix glasses from fauro plotted against silica concentrations; Mis are hosted by clinopyroxene, plagioclase, Fe-Ti oxides, sulfides, amphibole, mica and quartz. Analytical data are shown in Tables 7, 8, 9, I 0, 11 , and 12 respectively. 81 and sub-alkaline (dacitic-trachytic in composition)(Fig. 37f). In general terms, there is a large range of analytical scatter both for whole rock as well as Mis compositions for Fauro, but the overlap of matrix glasses and compositions of Mis is noteworthy. 3.7.5 Choiseul Volcanic Suites Petrographic studies of Mis were can-i ed out on a number of samp les col lected from the calc-alkaline volcanoes of Maetambe (East Choiseul) and Kumboro (Southeast Choiseul). i. Maetambe The Mis in the Maetambe suite are abundant in plagioclase phenocrysts, less abundant in clinopyroxene and amphibole and do not appear to be present in qua1tz. A medium-coarse rock (possibly pegmatite fragment) samp led in the Kagono River (near Voza vi llage) has a remarkable abundance and variety of MTs. Both the plagioclase and clinopyroxene in this sample are riddled with many glass inclusions (see Fig. 30a-j and Fig. 38a-d). In addition, the andesites of Maetambe contain more preserved Mis than any other types of andesite collected for this study (e.g., Fig. 22). The shapes of the Mls vary from rectangular and square forms to more in-egular shapes with sharp edges. Well-developed Mis have rounded and smooth edges. The crystal-melt contacts appear to be sharp with no compositional gradients. In most cases, Mrs occur in concentric zones within plagioclase hosts (see Fig. 34b). Cracks or fractures are not common. However, some Mrs show evidence of fracturing and cracking, but these seem to be unrelated to the original entrapment process. Vapour, bubbles, or opaque blebs are common in a range of different M i s, and appear as dark, circular spherules near the edges of the Mis-crystal contacts. In some case the nuclei have developed together with growth of Fe-Ti oxides and su lfide crystals. The degree of devitrification is limited due to the rapid cooling of the magma that generated these inclusions. This is confirmed by the size of the Mls and its crystal development within the inclusion. Some inclusions are quenched, sh runken, or ruptured either along concentric zones or in isolation (Fig. 30j). 82 Table 13. Representat ive analyses of M fs trapped within Fe-Ti oxide phenocrysts for the Choiseul Volcanics. Fe-Ti ox idc MO 71 <• 628 66.6 l>I> 7 74!'1 7J 7 O.l\ 00 02 O• 0.1 03 01 07 04 0 I 03 \l~O\ 20.:- 16.0 l'J.2 IX.0 lt\O 156 12 4 117 111 1 1 1.4 2.7 14.5 OJ 2.1 I.'> 14 \lnO 0.0 00 0.1 0.0 0.1 0.0 OS O.~ 0 I 02 0.0 \ 3 0.1 0 I 0.1 02 2.0 OS ·O I 0 I 00 00 2.0 2.-1 2.0 1.7 04 J.O 1.2 2.0 01 00 0.3 7.1{ 4.9 7.1 6.5 \,J 3.~ ~.3 23 4.4 KlO 3.0 1.') 4.-l 5.1 3.2 74 4 I 7 7 5 4 s.s 0.0 0.0 0.0 0.0 0.1 0.0 0.4 0 4 0.0 00 0.0 low! 1000 100.0 1000 100.0 100.0 100.0 1000 100.0 100.() 100.0 100.0 ' .120 K20 !U 10<) ll.7 11.5 11.(> II 2 'H too I-l ost crvs tals 11-0.1 hi u.oJ 11.1 lJ.oJ 1:x :!J.IJJ M 11-031111 l /.OJ /1/3 15-0! Iii .'J·Ol h.1 lS·lll "5 l5-tJ9 hi 0.1 o.~ 0.1 0.3 0 ..\ 02 03 0 I 0 I 02 7.0 1.6 3.J 2.5 40 3 I 09 \120\ 2.6 2.7 3.\ 11 I 2 10 I 2 X7 t\ 9J I Q()() 914 925 90 \ 92 9 •H.I 93 x \lnO o.•> 0.6 09 o.x 0.6 OS oc. 10 09 Ill I.ti 1.0 1.4 11 1.6 OS ()<) OS l.t\ 00 0 I 0.0 0.1 0.0 00 00 0.1 0.0 00 02 0.1 0 I 0.2 0.1 0 I 02 02 01 0.3 KlO 0.0 0.0 0.0 0.0 0.1 00 0.0 0.0 0 I 0 I 0.0 00 0.0 0.0 0.0 00 0.0 0.0 0.0 0.0 Tue.ti 100.0 IOOO 100.0 100.0 100.0 100.0 100.0 1000 100.0 100.0 0 l'I 0.11 0.15 0.20 0.IS 0.17 0.22 0 17 0 II 0 34 3 Table 14. Representative analyses of matrix glasses within various Choiseul Volcanics. Choiscul - Matrix oruli' ]/.(13 i;l 21 -UJ /:5 :!J.(1Ji;6 2J.03i;7 11-1151:1.M 11·05i;2-! 25-0ls.Sme.Js.7 15-0.Ys.S·I 15-IJSs_S-1 15-0.~s.9 15 - 09,~] ]5.IJ9-s.3 25-09-i;.t .?5-IJ9-i;5 Si02 <>6.5 70.0 72.0 68.8 65.0 64.8 55.1 7D 77.4 7S.O 7-l.8 ?R.6 73.-l 73.S Ti02 0.2 0.2 0.3 0.2 0. 1 0. 1 o.•> 0.3 0.3 0.3 0.2 0.5 0.2 0.2 i\1203 17.6 lfi.7 16.0 17 2 2!.li 20.9 21.7 12.1> 12.5 12.3 15. I 11.6 14.4 15.1 FcWJ 3.2 1.9 l.S 1.0 1.6 1.7 9.0 1.5 LS 1.3 18 04 11 0.(1 FcO \lnO 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.1 0.0 ~l gO 0.5 0.2 0.0 0.0 0.2 Oh 2.1 0.4 0.4 0.3 0.1 0.8 0.0 0.1 CaO l.S 2.0 2.4 11 3.6 1.5 4.3 I 2 1.2 I.I oc. 0.5 0.C> O,<) Na20 5.5 S.7 5.8 S.3 7.6 10.5 3.3 1.9 1.8 1.6 2.5 3.9 3.(i -l .4 K20 4.8 3.2 1.9 6.6 0.4 0.0 3.7 .\ .9 4.9 5.1 .) 7 -l .3 6.5 5.1 1'205 0. 1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 00 0.0 02 0.0 0.0 0.0 S03 0. 1 0 I 0. 1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.2 1.2 0.1 0.0 C'r203 0.0 0.1 0.0 0.0 0.0 00 00 o.o 0.0 0.0 ,. 0.0 0.0 0.1 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 t\a20 • K20 10.J 8.9 7 7 11.S s.o 10.5 7.0 6.8 6.7 6.7 7.2 N.3 10. 1 9.5 84 Table 15. Representative analyses of M l s trapped within elinopyroxcne and amphibole phenoerysts for various Choiseul Volcanics. Choiscul - Clinopyroxcnc _C_h_o1_·sc_u_l_-_A_n_.1p_h_ib_o_lc______t>ticlt• .'1-UJ 119 !l-11.1 11 111 .'1-0.1 8 1 1,.01.J s1 "'""' JJ-04 s;J .'J-{}-1 8.s.1 11.()-1 0.J-1 lj.111 01 1.1.fl! 0.J 15-0.\ i;I .'5-081;1-1 15-0Si;J-! !5-0\ 111 .'.I ~\ o"" "' !.1-09-i;tS S102 M! (>q 50.'I l Fc203 2 I> 1.0 17.•> 09 h:201 0.8 I 0 1.0 0.9 I 0 IJ 11 1.0 0.Cl 12 (> 1.2 MnO 0.0 01 02 0.1 \lnO 0.1 0.1 0.1 00 00 OJ 02 0.1 0.1 0 I 0 I MgO 2..1 01 45 02 :>.lsO 0.3 0.1 0.1 0.1 0 I OJ 0.2 0 I 0 I 6.2 01 C30 2.5 OC) 171> 09 Cao 6.oi 1.7 I 7 02 0.2 U! l.S 2.0 1.5 50 0 1 l\a20 5.4 5.~ 0. 1 2 l Na20 7.9 3.2 2 9 4 (i ox 4.4 )6 u .u 111 2' K20 3.1 7X 0.2 H K20 0.7 4.4 4.4 ,, 6 !SJ 4 •) 4.4 3.S 4.9 1~ 5 5 Cr20:1 0.7 0.0 01 0 I Cr20.1 0.1 0.1 0.1 0.1 ·O I 0.3 0.2 0.0 0.0 0.0 00 ., 0131 100 0 1000 IOO.o 100.0 ro1al 100.0 100.0 100.0 1000 100.0 1000 100.0 100.0 100.0 100.0 1()0,0 N320•K20 N 5 117 (U (1 I N320 • K20 8.6 1.5 7.3 II .I 16.1 93 100 85 'I 2 s.o u: Host crystals Host crvstals (l'(I(/(_~ !l-0.1119 1/-11.l-h/{) 11-115"' fi-1-.llr! Ol.1tk JJ-04/:J !1-0Jfi././ !l-IJ./lt-1-! 2.1-0!h! !5-11!/i.J 2.1.().~ltJ !J-OS/i!-1 :!5-0Sh1-1 !.I-I/Sid ! .l·IJ.\lrtiv1 .'.1.IJ'lltfl $102 50 I 4•17 56 I )0.7 S102 4 1.'1 39.7 JQ ,7 40.11 41.5 40-l 40-1 40.4 40.4 42 x .lJJ T102 Al20J 31 ! .•) 2~ \1203 l·U 15.S 15 5 12 s 10 ~ 111 13 I 13.1 13.1 122 \(I Fc:?03 110 11"'.2 00 n<> l'c!OJ 17 J 18.5 IS 5 17.0 IS 6 178 17 s 17li 17.\ 1<1.'1 20 2 \lnO Cl s 0 .3 CIO OS \lnO 0.4 0.3 0.1 01 0.1 0 .3 0 .1 0 .3 OJ 0.4 2.1 \lgO 11 2 12 l 12.S 11 .1 105 10.S 10 5 10.5 11 S 10 s Cao :?:?.S 2:? .i :?2 4 11 2 Cao IU1 II S llli II 6 11.~ II 9 11 9 II 9 11 .9 12.0 11.2 '\a20 0 .4 OS 00 0.0 -.:;120 :?.2 23 2 .1 2J :! .2 24 24 2.4 2 -I 2 :! 22 K20 0.0 0() ()O o.n K!O I. I I 4 I 4 I. I I 3 14 1.4 1.4 1.-1 011 1.0 Cr20J 0.1 0.0 o.i 0.0 Cr203 0 .0 00 00 00 00 00 0 .0 0.0 0.0 oo 110 Tot;il 100.0 100.0 IUOO 100.0 To13I 100.0 100.0 1000 1000 100.0 1110.0 100.0 100.0 1000 1000 l 85 Ta ble 16. Representative analyses of Mls trapped within plagioclase phenocrysts for the Choiseul Volcanics. Choiscul - Plagioclasc fel~Qar otidc 11-04 i;I 11-0.J s;l 15-0t s' e5 .'5-01115 µ5 15-01 s7 v7 15-07 sl µ13 15-07 J;6" 25-07 gS" 15-07 ;;9" 15-07 sJO e4 25-07 J;ll "15-07s!J pll 15-011 S"' 15-08 s5 15-08 glO 6-01-J xi Si02 64.3 59.6 59.7 74.S 63.6 72.5 73.9 70.0 62.1 78.8 37.9 77.3 68.1 70.9 58.9 779 Ti02 0. 1 0.1 0.1 0.0 M 0.2 0.0 0.0 0.5 0.3 1.7 0.3 OJ 0.3 0.1 05 Al203 19.1 23.2 23.3 14.7 176 l(\.f> 16.0 17.4 17.5 11.2 15.4 12.2 19.0 17.2 24.4 13.0 Fc20J 2 0 0.7 4.4 2.7 7.0 12 0.7 1.2 11 .4 1.9 20.7 1.2 1.3 1.3 0.8 1.0 /\lnO 0.0 0.1 0.0 0.0 0.0 00 0.0 0.0 0.0 0.0 0.4 0.0 0.1 0.0 0.1 0.1 MgO 2.2 0.4 0.2 0.3 0.0 0. 1 0.1 0.0 0.8 0.3 8.2 0.0 0.1 0.0 0.1 0.1 CnO 2.3 6.2 5.9 I.') 2.3 1.9 2.3 2.9 2.6 0.6 12.4 0.7 1.9 1.3 83 1.7 Na20 5.4 (>.3 4.0 2.6 2.7 39 3.8 4.7 2.9 l.S 2.3 3.3 3.<> 3.-1 4.7 2.2 K20 4.6 3.4 2.4 3.4 5.8 3.6 3.3 3.8 '.!.I 5.2 1.0 4.9 5.S 5.6 2.7 3.6 Cr203 0.0 0.1 0.0 0.0 0.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 \la20+K20 10.0 9.8 6.4 6.0 8.5 7.5 7.1 $.5 5.0 6.9 3.3 8.2 9.4 9,0 7.4 5.:1 Host crystal oxult· ]!.Q.f.hJ 11-04-h] 25-0lrif11[13 l5-0I "5 15-01117 25-07 It] 15-07 "6 15-07 h/O 25-07 !tJJ 15-08 h4 25-0S '15 25-0S !tJO Si02 55. 1 54.2 57.6 5-1.1 54.S 57 9 57.9 47.5 52.0 58.3 57.1 55.1 T102 0.0 0.0 0.0 0.0 0.0 00 0.0 0.0 0.0 0.0 0.0 0.0 1\ 1203 28.5 29.0 26.7 29.1 28.8 26.5 26.S 33.(i 30.7 21i.0 2<>.7 2lU t'e203 0.1 0.2 0.3 OA 0.3 0.3 0.3 0.4 0.2 0.4 0.4 0.0 MnO 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 .0 0.0 MgO 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Cao 10.1 11.3 s.o 10.7 10.3 7.6 7.(, 16.0 12.4 ~ . I 9.0 10~ Na20 S.4 5.1 7.0 5.4 5.7 72 7.2 2.-1 4.6 6.7 6.3 5.4 K20 0.2 02 0.4 O.:l 0.3 0.5 0.5 0.1 0.2 0.6 0.5 0.-1 Cr203 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 JOO 0 100.0 100.0 Na20' K20 5.6 5.3 7.4 5.7 6.0 7.6 7.6 2.5 ~ . $ 7.3 6N 5.8 86 a -< +,f ! .. ~-.....a::::.,._ .- • .. • • · ~ •I Figure 38. Mis hosted by various crystals within a coarse-grained, pegmatitic[samplc # 99-06-03-0 I). a,b and care Mis trapped in clinopyroxcnc, and d arc Mis hosted by a plagioclase phcnocryst. All scales arc in µm. • •• Figure 39. Mis trapped in a plagioclase host. a & b arc inclusions trapped in a coarse-grained andesite (Kao River, Mactambe). All scales are in pm. lote: the laminar arrangement of Mis in b which were likely formed along a healed crack or staged growth surface. 87 ii. Kumboro As for the Maetambe samples, the MJs in the Kumboro volcanic suite are most prevalent in plagioclase hosts and less abundant in clinopyroxene and amphibole. An exception is a sample from Kaghau on the east flank of Kumboro where the Mls occur abundantly in amphibole as well as plagioclase and clinopyroxene hosts (Fig. 40a-c). Within amphibole hosts (e.g., sample 99-5-21-04), the Mis are individually isolated but distributed throughout the whole crystal. These Mis are primary in origin and range in length from 2-8µ.m. Some Mls are ruptured and/or pa11ially crystallized. Some very small scale inclusions are present along fractures/cracks within the host phenocrysts. These Mls likely represent fracture-fill and are classified as secondary inclusions. Some other inclusions present in am phi bole are also very small ( <0. l -2µm) and contain opaque spherules. The plagioclase-hosted Mls are very small scale structures (<2-5µm) and are partially quenched or crysta.llized. The Mis occur within crystals as circular trails or patches. Pyroxene- (both clinopyroxene and othopyroxene) hosted Mls tend to be very small in size (<2-6µm) and occur as individual or isolated melts (e.g., in sample 99-05-21-05) (see Fig. 40). The Mls typically have two phases: vapour/bubble + glass and some of these inclusions are probably multiphase silicate-Mis containing very small gas rich bubbles which are filled with dark spherules. 3.7.6 Major Element Compositions of Mis and Host Minerals The compositions of the glasses are those of mostly sub-alkaline, aluminous, felsic, and a minority of alkali-rich magmas, ranging from andesite to rhyolite in bulk composition. The analytical data for Mls and their respective host phenocryst minerals are listed in tables 13, L4. 15 and 16. Various major elements oxides are plotted against Si01 in Figure 4 1. There is some degree of scatter in these plots, an overall similarity to the Mis compositions from Bougainville, no indication of a group with very low alkali contents as is the case for Faure, nor any consistency in terms of host phenocryst type as to compositional characteristics. There do appear however, to be at least two groupings (with both having a negative correlation) of 88 Figure 40. Mis and various c1ystalline phases (including Fe-Ti oxides) trapped in phenocrysts of hornblende (a), biotite (b), and clinopyroxcne (c). Al l scales are in pm. 89 12 20 a d 10 '.' I p 1a 15 ( Ml In Fo·TI 8 ~ 0 10 3: 6 0 N (1) 5 z 0 2 0 .5 20 25 b e 15 20 10 3: 3: 15 0 (1) 0 5 N (.) 10 0 .5 5 25 20 c 20 15 alkaline 0~ 15 3: 0 N 10 0 "' ~ N 10 + Q) 0 lL. N (1) z 5 5 0 0 30 40 50 60 70 80 90 30 40 50 60 70 80 90 SiO wt% 2 SiO wt % 2 Figure 41 . Plots of major oxide concentrations vs si lica of Mis and matrix glasses in volcanic rocks from Choiseul. Most of the data were obtained from crystals of clinopyroxenc, plagioclase, amphibole, and Fe-Ti oxides. 90 20 r-r- r I 0 Whole Rock Choiseul r Mis in cpx-arnph 1 ~ • Mls in plagioclasc i I • Mls in Fe-Ti oxide Matrix glass 1 15 ~- 0 ~ · • + • • • 0 • · 10 • z~ •• • • • .• ' (;)':: ••• ~ • • • • ii • 0 •• • • • 0 Bo • • 5 0 • 0 45 50 55 60 65 70 75 80 wt% Si0 2 Figure 42. Comparison of whole rocks, Mls hosted by various phenocryst phases, and matrix glasses projected in terms of total alkalies vs SiO wt% 2 91 a20 vs Si02 wt% (Fig. 4 La). Similarly, negative correlations with bulk SiO:? are observed for CaO, total Fe as F~0 3 , and A l20 3. On the other hand, K10 is positively correlated with wt% Si0.2. Other key geochemical parameters include low concentrations of MgO, Ti02, and P20 5 (<2 wt%). As a group, the bulk compositions of the Mls from Choiseul are more Si02-rich than the host whole-rocks, but there is some overl ap between compositions of the Mls and matrix glasses (Fig. 42). 3.7.7 Gold Ridge (Guadalcanal) Volca nics Rock samples from the Gold Ridge area of Guadalcanal are highly silicified and altered. Good quality Mis that can be analysed are located mainly in crystals of sulfide, plagioclase, and quartz. Inclusions are broadly distributed within the sulfide, quartz, and plagioclase. The Mis within the Gold Ridge samples ranges in size from 2 - 30µm in diameter, but most are between 5 and LOµm in maximum dimension. The sulfide phenocrysts are usually subhedral-euhedral in shape. and arc rectangular or cubed shape, ranging in size from 100 to 450µm. Sulfides have formed closed to vein lets of quartz, and in vugs, and are sometimes display preferred orientation fabrics parallel to vein lets and vugs. Within the su lfide hosts, the Mis have a sub rounded-angular shape and are greyish white-dark in colour when observed as electron back scattered images with the SEM. In general, the Mis that occur at the centre of crystals are too small to be analysed (< l~nn ). Two types of Mls are encountered: one high in Na20 wt% and the other extremely low in NaiO but rich in K 20 wt%. These compositions do not resemble any normal magma type, and most likely reflect K-metasomatism. 3.7.8 Maj or Element Compositions of Mis and Host Minerals Geochemical data deri vcd from EMPA of both the M l s and host crystals are shown in tables 17 and 18 and presented graphically in Figure 43. The glasses have Si02 concentrations in the range 60-75 wt%, Al20 3 from !1-25 wt%, and as mentioned above, are very rich in alkalis (up to 16 wt% K 20 and 10 wt% Na20 ). The MgO is low (usually <0.5 wt%), total Fe as Fe20 3 is around 1-2.5 wt%, while TiO ~, MnO, and Cr20 3 are all very low (<0.15 wt%). The Mls are mostly dacitic in terms of Si02 content but with high Al 20 3 and very high K ~ O probably resulting from metasomatic additions. 92 Table 17. Representati ve analyses of Mis trapped within plagioclase and quartz phenocrysts of the Gold Ridge Volcanics. Gold Ridge - Plagioclasc feldspar ua11z S102 63.1 75.2 M .I (>.1.3 M.5 64.7 63 ') M .3 64.S 64.0 (,(),4 r102 0 I 0. 1 0 .1 0.1 0. 1 0.1 0. 1 0.1 O. l 0.1 5.0 Al203 1711 12.3 lR.4 18. l 20.5 18.2 17.9 lR.5 fc203 I) I 0.2 01 0.1 1 6 0.0 0. 1 0.1 Ol 0.0 0.3 MnO 0. 1 0 .1 0. 1 0.1 ·0.1 O.l 0. 1 0.1 00 01 0 I MgO 0 I 0.1 O.l 0.1 o.o O.l 0.1 O.l 0.0 0. 1 01 CaO 0.2 0.1 0.1 0.5 1.6 0.2 0.2 0.1 0 I 02 1).2 ~a20 0.1 0. 1 01 0.1 10.7 02 0,1 0 1 0.3 02 1U KlO 18.'I 11.$ 16. Tomi 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.C) 100.0 1000 100.0 Na20• K20 19.0 11 .9 17,0 l i.O 10.'l 16 1) 17.2 16.7 17.3 15.(\ Host crystals ox1 Si02 63.7 63.9 64.3 Q')(i Q'J 5 99 4 99..1 9'l.4 99.7 '>')A T102 0.0 0.1 0. 1 0.0 0.0 0.0 0.1 0. 1 O. l 0.0 0.0 Al203 18 0 IS 7 18.3 1$.4 0.1 0.3 0.1 0 .1 0.1 o~ l·c203 0.0 0 l 0.1 0.0 02 0.0 0.1 0.1 0.1 0.(1 0.0 1-l nO 0.0 0.1 0.1 0.0 0. 1 0.0 0.1 O.l 0.0 00 00 l\tgO 0.0 0.1 0.1 0.0 0 .1 0.0 0.1 0.1 0.0 0.0 0.0 cao o. ~ 0.1 0.1 0. 1 0.0 0.1 0.0 0.0 00 0.0 0.0 Na20 o.o 0.1 0.2 1) 0 0.0 0.1 0 l 0.1 0.0 0.0 0.0 K20 17.0 16.9 1i .O 17.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 P205 00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 S03 IJ.O 00 0 C) 0.0 0.1 0.0 0.0 0.0 0.5 0.0 0.0 Cr203 o.o 0. 1 0. 1 0.0 0.0 0.0 0.1 0.1 00 0.0 0.0 To1al 1000 100.0 100.0 100.0 10<).0 100.0 100.0 100.0 100.0 100.0 1000 Na20+K20 17.0 17. l 17.2 17. 1 0.0 0.1 0.2 0.2 0.0 0.0 0.0 93 T able 18. Representative analyses of Mis trapped within ul fide and Matrix glasse of the Gold Ridge Volcanics. Gold R1d12c - Sulfides Matrix otult w.: lfJi;I 9'1.!-'!Si;! Y'l-1-77s..t 99.;. --i;5 99-1· --!;la 9'1-!- --!;lh 99-!.;-I> I ,,t1-!~i ..~.Jh 'i9· ..'··'no~ 'i10~ Ml 7 5•1' 1>2 'I 63 6 6n 62 ,, 62' M.' M2 1102 01 00 0.1 01 0.1 0.1 0.1 0 I 01 t\120J 17.7 111 11 JI( 3 200 llU 17 7 17 l\ ll\ ~ 174 1·<20' n 2 'i I; 19 0.2 1.7 20 114 0.0 \lnO 00 01 I) I 00 0.0 02 00 00 Ill \lg() 00 00 0 I I) I 00 OJ 0.5 00 01 CnO ()() 0.0 0.0 2.1 0.0 0.0 00 o.o 1).1 Na20 04 01 03 6.7 0.3 OJ 0. ~ OA 0.1 K20 I.' I 1 l!) lld 48 11>.5 15 ,, 15 'I 11'1 lh' r20s 0.0 00 o.o 0.0 0.0 011 O.Q 00 00 'i <) N.' ()'/ 07 00 \ 'i O.S O' 0.0 "°'Cr!<>' 0.0 00 00 0.1 00 0.1 00 0 I 0 I r,>1al 1000 IOOI) 100.0 1000 100.0 1000 llXHJ 1000 1000 \ ;i20 K20 I~:! 166 II ' 16.l\ ir.o 16' If• 4 '" '" 5 Host crystals otul,· Y'I-! -611/ yy.;. -Mo.' tJv.;.;;/:.t "9-!- --1:5 St02 02 0 I 0.3 OA T102 Ill 00 00 0.1 \1203 0.0 Oil 00 0.1 lc20J Wl 'b.I> 46 !( 46b \lnO IJO 04 00 0.1 \lgO 1.2 01• 02 0.0 C:.10 0.0 01 0.0 0.0 \ n20 0.4 114 01 0.0 K20 00 Oil 0 I 0.0 P205 0.0 o.o 0.11 o.o s f\0.0 h.2 ~ 52 4 52.5 Cr20.' no 02 02 0.0 To1al 100.0 100.0 1000 1110.0 \ n20 • K20 04 0.4 03 0.1 94 22 a 20 c 20 15 at 18 ~ ~ ... ~ ~ 10 0 .. 16 0 :;( :.:: .. 14 5 12 0 20 b d 2 5 18 ~ alkaline ~ l: ;.: . 16 l: I 5 0 ... :.:: .. 0 .. \ + Q) 14 u. 0 .. $10 wl % $10 wt % 2 7 Figure 43. Major oxide compositions of Mls and matrix glas cs plotted against silica. 95 I I I 20 [ I I Gold Ridge 0 Whole Rock rr- Mls in plagiocla e 18 • Mis in quartz Mls in sulfide • Matrix glass , • • 16 .. I•·• ...... •.. .. .• ...... "• 0 N ~ ..... + 14 0 ~- • • zd" ~ 12 ~ 3 • ~ • 0 lO L 0 0 8 0 ..f 6 1--1. ..__~ 45 50 55 60 65 70 75 80 wt% Si0 2 Figure 44. Comparison of whole rocks, Mls hosted by various phcnocryst phases, and matrix glasses projected in terms of tota l alkalies vs SiO wt% 2 96 There arc some simi laritics between the compositions of the Gold Ridge Mls and those of Bougainvillc, Fauro, and Choiseul. However, unlike these other localities, the Mis of Gold Ridge have distinctively high K 20 concentrations (I 5- l 9-wt%) as seen in Figure 43, that arc much grctcr than the whole rocks (Fig. 44). A s a consequence of these elevated K20 concentrations, the compositions of the Mis of Gold Ridge all plot within the alkaline field of the TAS diagram (Fig. 43d). Note that the Mis obtained from the calc-alkaline volcanics of Fauro (Fig. 37f) and Choiseul (Fig. 41) project mostly within the sub-alkaline field of the TAS diagram. 3.8 Compa rison of major element compositions of the Mis and host crystals fo r Ilougainville, Fauro, Choiseul, and Gold Ridge The major clement compositions obtained for M is and host crystals (phcnocrysts) from the various respective volcanoes (normalised to 100%) arc discussed in this section, based on simple I larker diagrams. The original data for M Is found in various phcnocrysts in different volcanic centres are given in Tables 19 to 3 1 (sec Appendix 2). The purpose of showing glass and host compositions together is so that contamination/crystallisation vectors of the former by the latter may be more clearly recognised. Some general features are apparent from Figure 45, which shows inclusion compositions coupled with their clinopyroxene hosts. The majority of the M is arc relatively rich in Al 2 0 ~ . and there is generally a negative correlation of Al20 3 with Si02 when Si02 exceeds - 55 wt%. Similarly, MgO, total Fe as Fc20 3 , CaO and Na20 arc all negatively correlated with SiOz. l n contrast, K 20 is positi vely correlated with Si0 2, whereas MnO and Ti02 do not show any consistent variations/correlations. lt is noteworthy that distinct offsets between the host compositional populations and Mls exist for Al10 3, Na20, and MgO. This fcalurc lends some support to the hypothesis that the M l s' compositions, as screened during the course of this study. really may represent magma compositions. cvcnhelcss. judging from the extremes in individual oxide concentrations (e.g., K 20), it is likely that some of the. e analyses arc not those of glass alone but arc likely contaminated by crystallites. Inclusion compositions and plagioclase hosts are displayed in Figure 46. To a first approximation. the compositional variation of some of the oxides (Al20 3 in particular) of the Mls mimics that of the plagioclase hosts. l n other cases however. and when considered together with the clinopyroxene host data (Fig. 45), the compositional va riation of the Mls define a vector that is primarily controlled by the combination of plagioclasc and 97 30 1 2 a • e 25 • • 0.8 • * 20 • Fhoit.I • 06 • • 15 De .... 0 " O< o .. I- ~ 10 0 2 :...______--v- 0 • 0 ·O 2 25 b 20 • 10 • 15 .... 10 0 o.," c: ..... :i! 0 • .5 ., 20 25 c 9 15 20 10 15 0 0 ('O 10 :i!"' <.) 0 • • 5 20 12 d 10 h 15 8 10 6 5 0 • 0 • ... .5 2 30 40 50 60 10 80 90 30 Figure 45. Plot of major oxides vs. Si02 fo r the Mis (open symbols) found in host clinopyroxene (fill ed symbols) in various vol ca nic rocks from Bougainvillc (B), Fauro (F) and Choiseul (CH). A rrO\.VS indicate schematically the compositional trends fo r the Mfs. 98 35 B ' 10 a • l • e Fglass e 30 lI • Fhos1 ' ..,. ' ••"'' 6 I 25 3 *J:; N • • 0 "' 0 N 20 < • • I- 2 t 15 ea11ClkU ;; . I> 0 ' • • • 10 · 2 ' u 0 5 f b 20 0 -! 't t • 0 IS t 0 3 " *i ' *3: ' ., 10 ~ 0 2 l 0 0 N c G> ::!: 0 \ • H f WE!fi!W'_I> IL s t • .... ! 0 [ 0 r • ... rm. - • • ...... ·S. l .... .&...... _ ~~ ·O 1 _._ ...... _.._ ...... __ - "'--~ 12 25 c g 10 0 20 8 15 *J:; 6 *J:; 0 0 C> 10 ~ (.)"' 2 0 • ·2 '- ...... l ...... _ 0 20 12 d h 10 15 8 10 r *3 6 *:: 0 0 N N I> ::£ z u "' 'l 0 ... ,.... ,... Ci> 0 • • .5 ·2 ...... ~~ 30 40 50 60 70 80 30 •O 50 60 70 80 SiO wt % S iO wt o/o 2 1 Figure 46. Plot of major oxides vs. silica fo r M Is (open symbols) found in host plagioclase (tilled symbols) for various volcanic rocks from Bougainvillc (B), Fauro (F) and Choiscul (CH), and in selected Gold Ridge (G R) rnincrals. Arrows indicate schematically trends of compositional variations. 99 35 r a e 30 f e • 25 • Fglass 20 • F ~os t ... 15 .& r.. H" > 0 N 0 < I- • • • 100, b 80 • 60 .• 1 5 40 l 0 f c 20 ::; - ~..,...,..., . 0 ~ • • ..!] ·· ..J ·~ f • ·20 ·O 5 15 c I g 10 • 5 0 (.)"' oO • • • . 5 12 r ?.O r d h 10 • 15 ~ 8 • f 6 f 10 • • 0 . N < f z"' r :•. • • ·2 0 20 60 80 100 S iO wt % $ 10 WI % 2 2 Figure 47. Plot of major oxides vs silica fo r Mis (open symbols) found in Fe-Ti oxides (filled symbols) for various volcanic rocks from Bougainville (B), Fauro (F) and Choiseul (CH). Arrows are schematic fits (by eye) through the compositional data. ote however, that some of these are inconsistent with "normal" frac ti onation trends (e.g., relatively invariant alka lies vs silica). 100 30 6 25 e 20 • Fglau 15 • Fhos1 ) 1 I 10 • 5 0 • • 0 • --'-----l>ll .5 ·I 35 O.• 3 b 03 • 25 02 ... 1 5 0 1 ~=----1>1> lt1 0 ON c: ~ u.. • ~ 05 ·O 1 0 • ·0.5 ·O 2 I < g I 2 • c 6 5 08 06 0 0 0 . () 2 ~"' "' ~ 1 ~ 02 • 0 0 ·O 2 ·1 20 12 ,' d 10 , 15 , . ' . 10 *3: 6 I '-1\. 0 " z"' • 0 • 0 ~ c...... 5 ·2 0 20 60 80 100 0 20 40 60 80 100 $ 10 I'll % S iO wt % 2 : Figure 48. Plot of 1najor ox ides vs. silica for Mls (open symbols) hosted by quartz (fi lled symbols) for various vo lcan ic rocks from Bougainville (B), Fauro (F) and Gold Ridge (GR). Arrows are schematic fits (by eye) to the compositional data. However, these do not conform to "normal" fractionation trends, and probably result from other processes. IOI e 0 8 20 ~ ....: * 06 18 -; .. 0 . 18 q ... 0 2 0 12 ·O 2 25 0 . b 20 0 3 0 2 15 0 1 10 0 c 0 5 ::E ·O I 0 ·O 2 .5 ·O 3 6 c g 0 0 ::;:"' (.)"' 0 [ I> 0 ·2 14 20 d h 12 15 10 10 8 / 6 0 c I> 2 0 .5 55 60 65 70 75 80 55 60 65 70 75 80 Si O wt % S10 wt % , 2 Figure 49. Plot of major oxides vs. silica for matrix glasses (open symbols) for various volcanic rocks from Fauro (F), Choiseul (CH) and Gold Ridge (GR). Arrows are schematic fits (by eye) to the compositional data. flowever, not all of these conform to "normal" fractionation trends, and probably result from other processes. 102 clinopyroxene fractionation. For example, the increase and then decrease in Al.!0 3 content with increase in Si02 of the Mls probably results from the combination of plagioclase and clinopyroxene fractional crystallisation of evolving magmas, that are trapped at differing stages .in the inclusion suite. The total Fe content of a number of the low- Si02 Mis exceeds that of clinopyroxene, so that fractional crystallisation of this phase alone is insufficient to account for the overall negative correlation of Fe (as Fe:P 3 ) vs. Si02 observed for Mls. It is likely that the additional fractionation of an Fe-Ti oxide is al l that is required. Accordingly, in Figure 47 , inclusion compositions and host Fe-Ti oxides are displayed. While fractionation of Fe-Ti oxide phases of course is incapable of accounting for covariation of oxides such as Al20 3• they are clearly implicated in the covariation of total Fe with SiO ~ . Evidence of modification of compositions of Mls by the host phase can be discerned from Figure 48, which displays inclusion compositions and their quartz hosts. For example, the fact that there is a linear vector of Al20 3 compositions towards the 100% Si02 point is indicative of either EMPA beam overlap, or continued crystal lisation of inclusion host wall. A similar argument can be advanced for the observed covariation of K 20 with SiO?. (Fi g. 48h) in that the curvilinear trend points towards the 0% K 20 -l00% Si02 point. Mls in amphibole and mica are of the mafic end of the total spread on inclusion glasses (Fig. 50). The compositional spread of the Mls show no distinctive vector trends away from the mineral hosts. 103 Famp_glass 30 • Famp_host a .. c ~ 25 .. 20 .... 0 0 15 ~ "' ::E"' 10 5 0 L"------~- 8 16 I • d 7 b 14 6 i 12 \ 10 t o ... 8 • z"' .6 . 2 0 .___ 0 • 30 • O 50 60 70 80 30 • O 50 60 70 60 S i0 w t % S iO wt % 2 2 Figure 50. Plot of major oxides vs. silica for Mis (open symbols) hosted by arnphibole and mica (fi lled symbols) for the Fauro (F) and Choiseul (CH) Volcanics. Arrows are schematic fits (by eye) to the compositional data. 104 C HAPTER 4: DISC SSION, CO 'C L SIONS, ANO F RTHER ST DIES -t I The texture aod nature of the :vtls and their fractionation trends The M IS arc preserved in many djfferent volcanic rock types and occur either as isolated inclusions or in clusters with preferred orientations within host crystals. The M l s range from tiny inclusions(< lµm) to a maximum of several hundreds of µm. Most of the Mis appear to be primary in origin being homogcnous, clear and transparent (type I) with minor development of vapour bubbles or daughter crystal s. However, in some Mls (regardless of sizes), bubbles appear as dark globules on the inclusion walls. Some of the M is have minute inclusions ( 4.2 Geochemical compositions of the Mis and matrix glasses The geochemical compositions of M l s analysed in clinopyroxcnc hosts show that the trapped Bougainville melts were highly felsic (66-80 wt% Si02); Mls in Fauro and Choiscul on the other hand preserve both fclsic (60-66 wt% Si02) and mafic compositions (<55 wt % SiO,). The compositions of M l s hosted within plagioclasc have a wide range. The Choiseul Mis seem to represent the whole range of melts from a basaltic composition (<53 wt% Si02) to highly felsic (>70 wt% Si02), whereas the Bougainvillc-Fauro-Gold Ridge Mls arc moslly quite felsic (-60 wt% Si02). The Na20 wt% in Mis from Fauro resembles the matrix glass compositions and may have been co-genetic. Mis hosted by Fe -Ti oxides are mostly felsic in composition ( ~60 % wt Si02) as shown in Figure 47. The volcanic glass compositions of Fauro are high in alkalis compared to those of Bougainvillc and Choiscul. The quartz-hosted Mls for all volcanic suites have highly fclsic compositions. The K 20 and a20 wt% are high for the Fauro and Gold Ridge Volcanics (15 wt%) and (6-8 wt%) respectively, and low for Bougainville (<5 wt% K:P) and (2-4 wt% a20). 105 Overall, the matrix glasses range from basaltic to highly felsic in composition, overlapping with the M1s' compositional range. Matrix glasses from Bougainville and Fauro are generally highly felsic (70 wt% Si02) like the Mls hosted by plagioclase. The matrix glass composition has high K 20 wt% for Fauro and Gold Ridge compared with Choiseul. An overview of the trends in compositional ranges of the Mls reveals a possible dichotomy at high Si02 (e.g., Figs. 32, 36, 42). Recognising that the Solomons and Bougainville include volcanic suites that are alkalic by the Le Maitre et al. (1989) definition, it is interesting that some of the populations of Mls trend towards highly alkaline and even peralkaline compositions, while others trend towards subalkaline compositions. ft is not clear that there is any association between the development of peralkalinity with mineralised systems in the Solomons-Bougainvil le region, but the study of Mis clearly reveals that highly evolved magmas have been developed, even if they have not been recognised as eruptive units. These alkaline melts may have significantly different transporting capacities of base and noble metals. 4.3 Whole rock geochemical co mpositions Given the generally incompatible behaviour of K within island arc suites experiencing fractional crystal lisation of spinel-olivine-pyroxene-plagioclase assemblages, genetical ly related rock types typically define linear trends of K 20 that are positively con"elated with increasing Si02 . These relationships are observed in plots of Si02 wt% vs K:P wt% in Figure 13h. The covariations of Ti02, MnO, MgO, Fe20 3 and CaO with Si02 (Fig. 13a-f) also show good correlations; it can be argued that the various rock types from a given volcano suite are probably related, primarily through the fractional crystallisation of pyroxene-plagioclase-Fe-Ti oxide assemblages, as argued generally for andesite suites by Gill (1981). For many arcs , the primary magmas are reckoned to be basalts with high MgO contents (>6% MgO), 250-300 ppm Ni and 500-600 ppm Cr (Wilson (1989)). However, it is rare thac such compositions are found. In rhe majority of the suites studied here, and with the speci fie exception of the comparator New Georgia Group, the predominant rock types are andesitic-dacitic. 106 4.4 Trace element geochemical compositions While major oxides can reveal the importance of fractional crystallizalion as the most important process in the formation of evolved magmas, trace element abundance systematics can help clarify other petrogenetic detai ls. For example, the Ni contents for all rocks are low which suggest they are not primary magmas derived from an olivine-rich upper mantle source. Arc type rocks are enriched in the incompatible elements of low ionic potential (e.g., Sr, K, Rb, and Ba; also known as large ion lithophile elements of L fLE) as discussed in Arculus and Johnson (1981), and have low abundances of elements with high ionic potential (e.g., Ta, Nb, Ce, P, Zr, Hf, Sm, Ti, Y, Yb, aJso known as high field strength elements or HFSE). The standard interpretation of these features is that the LILEs have been enriched with respect to the HFSEs through preferential mobilisation in a subducted slab-derived fluid phase (Wilson 1989). The REE patterns (Figs 22 and 23) cover a wide range from light-REE depleted through flat to strongly light-REE enriched relative to chonddtes. These differences are co1Telatcd with the KzO content for each magma series as illustrated in Figure 24. The basement tholeiires have light REE-depleted patterns while the calc-alkaline basalts are light-REE enriched. The potential role of fluids, and their interaction with the various elements are important to understand in these types of arc te1nne when one has to consider the 01igins of base metal deposits. 4.5 Conclusions A number of preliminary conclusions concerning the comparative geochemistry of these selected Solomon Islands and Bougainville volcanic suites can be made: l. The slrnctural and genetic links between arc volcanism, locations of arc volcanoes, magmatically-related mineralization, crustal fluid flow patterns, and above all the geochemical characteristics of specific rock suites are all important factors for understanding and evaluating the various metallogenetic processes that occun-ed in thi s region. 2. The Gold Ridge rocks have very high K-concentrations, most likely the result of mineralising K -metasomatism. 107 3. Bulk rock analysis shows that the ferrmomagnesian oxides and lime (i .e. , Mg0- Fe20 3-Ti02-Mn0-Ca0) generally decrease while K 20-Na20 increase as the Si02 contents of the rocks increase. A l20 3 seems to be relatively constant in most cases. 4. Some Gallego, Fauro, and Choiseul volcanic sources have high K 20-Na:P whereas other Gallego and Kurnboro volcanic suites are relatively depleted in aJkalis. 5. The REE patterns are similar to many other arc sui tes world-wide. In addition, the potentially fluid-mobile LILEs (Sr, K, Rb, and B a) have greater relative degrees of enrichment than those that are regarded as fluid-immobile elements together with the HFSEs (e.g., Th, Ta, Nb, Zr, Hf, Ti, Y and Yb). 6. The systematics of the REE abundances for the Au-related Gold Ridge Volcanics are similar lo those of the Fauro calc-alkaline basement; it is possible that this may indicate s.imilar mineralisation characteristics on Fauro. 7. Mls are either isolated or occur in preferentially aligned clusters, ranging in size from 8. M l s range from basaltic (45 wt% Si02) through felsic (60 wt% Si02) to highly silicic in compostion(>70 wt% Si02) . Geochemical compositions of glass in the rock matrices resemble those of glasses trapped in crystalline hosts. 4.6 Further Studies l. It will be interesting to conduct laser ab lation analysis of some of the Mls and to compare these resu lts to bulk rock trace element data. 2. Radiogenic isotopic studies should be pursued in the future to augment the geochemical data obtained so far, and to help refine the petrogenetic models especially those that recognise distinctions between older basements and younger volcanic suites in Bougainville, Faure, Choiseul, and Guadalcanal. 3. For any future projects in the Solomon Islands, a key requirement is an improved temporal framework; geochronology is critica.lly needed both for the younger calc alkaline volcanic rocks as well as the basement suites. 4. Finally, a combinations of geochemical efforts (Mls, bul k major, trace, isotope and dating) will enhance and improve our geological knowledge, highlight the important and significant geochemical signatures, and thus help us understand the evolving tectonic framework of the Solomon Islands. This is a prerequisite for understanding the nature of mineral development in the region. 108 REFERENCES Arculu , J .R., and Johnson, W.R. I 981. Island-arc magma source: a geochemical assessment of the roles of slab-derived components and crustal contamination. Geochemical Journal; 15, 109-133. Ayres, D.E., Burns, M.S .. and Smith, J.W., 1981. A sulphur isotope reconnaissance study of the porphyry copper deposits at Panguna and Mt Fubilan, Papua lew Guinea. Pacific Geology 15, 37-50 Babbs , T.L., 1997. Geochemical and Petrological Investigations of the deeper portions of the Ontong Java Plateau: Malaita, Solomon Islands. PhD Thesis, University of Leicester. Baldwin, J.T., 1977. Discovery, evaluation and development, Panguna porphy1y coppers deposit, Bougainville Island, Papua New Guinea. James Cook University of No11h Queensland MSc. thesis (unpubl.) Baldwin. J.T.. Swain H.D., and Glark. G.H .. 1978. Geology and grade distribution of the Panguna porphyry copper deposit. Bougainville. Papua e\ Guinea. Economic Geology 73. 690-702. Bercovici. D., and Mahoney. J.J., 1994. Double flood basa lts and fluid plume head separation at the 660 km discontinuity. Science 266, 1367-1369. Beyer ,L.A., 1986. Gravity anomalies of the Solomon Islands between I 56 and 161 E longitude, in Vedder, J.G., Pound, K.S., and Boundy, S.Q. (eds), Geology and offshore resources of the Pacific island arcs - central and western Solomon Islands, 123- 134. Circum-Paci fie for Energy and Minera l Resources Earth Science Series 4. Circum - Pacific Council for Energy and Mineral Resources, Houston. Blake, D.H., 1968. Post-Miocene volcanoes on Bougainville Island, TerTitory of Papua New Guinea. Bulletin Volcanologique 32, 121 -138. Blake, O.H., and Miezitis. Y., I967. The Geology of Bougainville and Buka I slands, New Guinea, Australia, Bureau of Mineral Resources Bulletin 93. Branch, C.D .. l 965b. Bougainville volcanoes, Mount Bagana and Lake Loloru. Australia, Bureau of Mineral Resources Record 1965/65. Bultitude, R.J ., 1976. Erupti ve history of Bagana volcano, Papua New Guinea, betwee n 1882-1975, in Johnson, R.W., (ed.), Volcanism in Australasia, 317-336. Elsevier. Amsterdam. 109 Bultitude, R.J. , Johnson, R.W., 1976. Eruptive history of Bagana volcano, Papua 1ew Guinea, between 1882-1975. in Johnson, R.W., (ed.). Volcanism in Australasia, 3 17-336. Elsevier, Amsterdam. Bullitude. R.J. , Johnson, R.W., and Chappel. B.W., 1978. Andesites of Bagana volcano, Papua New Guinea: chemical stratigraphy and a reference andesitc composition. Bureau of Mineral Resou rces Journal of Australian Geology and Geophysics 3, 28 1-295. Bultitude, R.J ., 1979. Bagan a volcano, Bougainville i sland: geology, petrology and summary of erupti ve history between 1875 and 1975. Papua New Guinea Geological Survey Memoir 6. Bultitude, R.J. , and Cooke, R.J.S. , 198 1. Note on activity from Bagana volcano from 1975 lo 1980 in Johnson, R.W. (ed.), and Cooke Ravi an Volume of Vulcanological Papers, 243-248. Papua New Guinea Geological Survey Memoir 10. Chase, T.E., Seekins, B.A., and Lund, K.E., 1986. Submarine topography of the Solomon Sea region, in Vedder, J.G., Pound, K.S., and Boundy, S.Q. (eds), Geology and offshore resources of Pacific Island-arcs central and western Solomon Islands, 21- 3 1. Circu m Pacific Council for Energy and Mineral Resources Earth Sciences Series 4. Circum Pacific Council for Energy and Mineral Resources, Houston. Cl1ivas, A.R. , 198 l. Geochemical evidences for magmatic fluids in porphyry copper mineralization, Part J. M afic silicates from the Koloula Igneous Complex. Conlrib. Mineral. Petrol. 78, 389-403. Chi vas, A.R. , 1978. Porphyry Copper Mineralization at the Koloula Igneous Complex, Guadalcanal, Solomon Islands. Economic Geology, 73, 645-677. Clark, G.H., in press. The Pangun a porphyry copper/gold deposit, Papua New Guinea, in Hughes, F.E., (ed. ), Geology and mineral depos its of Australia and Papua New Guinea. Australasian Institute of Mining and Metallurgy, Melbourne. Coleman, P.J., 1960. An introduction to the geology of the island of Choiseul in the Western Solomons 1957. British Solomon Islands Geological Record. 1957-1958. 16-26. Coleman, P.J ., 1962. An outline of the geology of Choiseul, British Solomon Islands. Journal of Geological Society of Australia. 8, 36-58. Coleman, P.J ., 1965. Stratigraphy and structural notes on the British Solomon Islands with reference to the first geological map, 1962. Report o. 29, British Solomon Islands Geological Record Vol.2, 1959- 1962, pp. 17-33. Coleman, P.J ., 1970. Geology of the Solomon and New Hebrides Islands as part of the Melanesian Re-entrant, SW Pacific. Pacific Science 24, 289-3 14. 110 Coleman, P.J., L989. Petroleum potential of Solomon Islands, SW Pacific. Ministry of Natural Resources, Honiara, Solomon Islands, Published by Bureau of Mineral Resources , Canberra, 24 pp. Coleman, P.J., 1991. Dynamic strike-slip fault systems with respect to the Solomon Islands, and their effect on mineral potential. M arine Geology. 98. 167- 176. Coleman, P.J., and Kroenke, L.W., 198 l. Subduction without volcanism in the Solomon Island Arc. Geomarineletters J, 129- 134. Cooper. P.A., and Taylor, B., 1984. The spatial distribution of earthquakes, focal mechanisms and subducted lithosphere in the Solomon I slands. In Taylor, B .• Exon, N.F., (eds.), Marine Geology, Geophysics, and Geochemistry of the Wood lark Basin - Solomon Islands. Circum-Pacific Counci I for Energy and Mineral Resources, Earth Science Series Vol. 7, pp 67-88. Cooper, P., and Taylor, B., 1987 . Seismo-tectonics of New Guinea: a model for arc reversal following arc-continental collision. Tectonics 6, 53-67. Coulson, F.l., and Vedder, J.G., 1986. Geology of the central and western Solomon I slands, in Vedder, J.G., Pound, K.S., and Boundy, S.Q. (eds), Geology and offshore resources of Pacific Island-arcs central and western Solomon Islands, 89- 108. Circum Pacific Council for Energy and Mineral Resources Earth Sciences Series 4. Circum Pacific Council for Energy and Mineral Resources, Houston. Crook, K.A.W., and Taylor, B .. 1994. Structural and Quaternary tectonic history of the Woodlark triple junccion region, Solomon Islands. Marine Geophysics Resource 16, 65-89. Davies, H.L., Honza, E., Tiffin. D.L., Lock, J., Okuda, U., Murakami, F., and Kisimolo, K. , 1987. Regional selling and the structure of the western Solomon Sea. Geo Marine Letters 7, 153-160. Davies, H.L., Symonds, P.A., and Ripper, I.D., l984. Structure and Evolution of the southern Solomon Sea region. Bureau of Mineral Resources Journal Australian Geology and Geophysics 9, 49-68. Dunkley, P.N., J 983. Volcanism and the evolution of the ensimatic Solomon Islands Arc. 225-24 1 in Arc volcanism: Physics and Tectonics. Shimozuru, D and Yokohama. I (editors). (Tokyo Terra Scientific Publishing.) Dunbar, .W. and K yle, R.P., ( 1992). Lack of volatile gradient in the T aupe Plinian ignimbrite transition: Evidences from melt inclusion analysis. American Mineralogist, 78, 612-618. Dunbar, . W. and Hervig, R.L., (1993). Petrogenesis and volatile stratigraphy of the Bishop Tuff: evidence from M lsanalysis. Journal of Geophysical Research 97, 15129-15150. 111 Eastoe, C.J ., l 978a. A fluid inclusion study of the Panguna porphyry copper deposit. Bougainville, Papua New Guinea. Economic Geology 73, 721-748. Eastoe, C.J., 1978b. Formation of Panguna porphyry copper deposit, Bougainville. University of Tasmania PhD thesis (unpubl.). Eastoe, C.J. , 1982. Physics and chemistry of the hydrothermal system al the Panguna porphyry copper deposit, Bougainville, Papua Jew Guinea. Economic Geology 78, 201-213. Eastoe, CJ., 1983. Sulfur isoLope data and the nature of the hydrothermal systems at the Panguna and Frieda porphyry copper deposits, Papua 1 ew Guinea. Economic Geology 77, 127-153. Eastoe, C.J., and Eadington, P.J., 1986. High temperature fluid inclusions and the role of the biotite granodiorite in mineralisation at the Panguna porphyry copper deposit, Bougainville, Papua New Guinea. Economic Geology 81 , 478-483. Frezotti, M.L. 1990. Magmatic immiscibility and fluid phase evolution in the M. Genis Jeucogranite (S.E. Sardjnia, Italy) (abstr.) PACROFl, Pan-American Conference on Research on Fluid Inclusions, III Symp, 20-22 May 1990, 37. Fisher, N.H., 1936. Geological report, Kupei goldfield, Bougainville, Territory of New Guinea. Papua New Guinea Geological Survey Archive File 36/00J (unpubl.). Fisher, N.H., 1957. Catalogue of the active volcanoes of the world including solfara fields, Pait V , Melanesia. International Volcano logical Association, Naples. Ford, J.H. , 1976. A geochemical and stable isotope study of the Panguna porphyry copper deposit, Bougainville. University of Queensland PhD thesis (unpubl.). Ford, JJ-L 1978. A chemical study of alteration at the Panguna porphyry copper deposit, Bougainville. Papua New Guinea. Economic Geology 73, 703-720. Fountain, R.J , 1972. Geological relationship in the Panguna porphyry copper deposit, Bougainville Island. New Guinea. Economic Geology 67, 1049-1064. Fountain, R.J. , L973. The Panguna porphyry deposit - a genetic study. Univer·sity of Sydney PhD thesis (unpubl.). Grover, J.C., 1958b. Two coast to coast walks across Choiseul 1954. 118-120 in The Solomon Islands-geological exploration and RESEARCH 1953-56. Geological Survey B1itish Solomon I slands Memoir, No. 2. Hackman, B.D., 1980. The geology of Guadalcanal, Solomon Islands. Overseas Memoir of the Institute of Geological Sciences, London, UK. Overseas Memoir No. 6. 112 Henly, R.W., 1985. The geothermal frame work for epithermal deposit, in Berger, R.B., and Bethke, P.M., (eds), Geology and geochemistry of epithermal system, 1-24. Reviews in Economic Geology 2. Society of Economic Geologists, Chelsea. Honza, E., Davies, H.L., Keene, J.B., and Tiffin, D.L., 1987. Plate boundaries and evolution of the Solomon Sea region. Geo-Marine Letters 7, 161-168. Hussong, D.M., Wiperman, L.K., and Kroenke, L.W., 1979. The crustal structure of the Ontong Java and Mahiniki oceanic plateaus. Journal of Geophysical Resource, 84, 6003-6010. Hyba, D.0., Be11hke, P.M., Heald, P., and Foley, N.K., 1985. Geologic, Mineralogic, and geochemical characteri stics of volcanic hosted epithermal precious - metal deposits, in Berger, B.R., and Bechke, P.M, (eds), Geology and geochemistry of epithermal systems, 129-167. Reviews in Economic Geology 2. Society of Economic Geologists, Chelsea. Johnson, R.W., 1977. Distribution and major element chemistry of late Cainozoic volcanoes at the southern margin of the Bismarck Sea, Papua New Guinea. Australia, Bureau of Mineral Resources Repo11 188. Johnson, R.W., Jaques, A.L., Langmuir, C.H., Perfit, M.R .. Staudigel, H., Dunkley, P.N., Chappel, B.W., Taylor, S.R., Baekisapa, M., 1987. Ridge subduction and fore-arc volcanism: petrology and geochemistry of rocks dredge from the Western Solomon Arc and Woodlark Basin. In Taylor B., Exon, N.F., (Eds), Marine Geology. Geophysics, and Geochemistry of the Woodlark Basin-Solomon Islands. Circum-Pacific Council for Energy and Mineral Resources. Earth Science Series. 7, 155-226. Distribution and major element chemistry of late Cainozoic volcanoes at the southern margin of the Bismarck Sea, Papua New Guinea. Australia, Bureau of Mineral Resources Report 188. Kroenke, L. W ., 1972. Geology of lhe Ontong Java Plateau. Hawaii Institute of Geophysics Report HIG-72-5. Kroenke, L.W., 1984. Cenozoic tectonic development of the Southwest Pacific. UN ESCAP Committee foe Co-ordination of Joint Prospecting for Mineral Resources in South Pacific Offshore Areas Technical Bulletin 6. Kroenke, L.W., Resig, J.M., and Cooper, P.A., 1986. Tectonics of the South-eastern Solomon Islands: formation of the Malaita anticlinorium, in Vedder, J.G., Pound, K.S., and Boundy, S.Q. (eds.), Geology and Offshore resources of Pacific island arcs-central and western Solomon Islands, 109-116. Circum-Pacific Council for Energy and Mineral Resources Earth Science Series 4. Circum Pacific Council for Energy and Mineral Resources, Houston. 113 Kroenke, L.W., 1995. A morphotectonic interpretation of SOPACMAPS 1: 500 000 charts. Cencral Solomon Islands to southern Tuvalu, SOPAC Tech. Report. 220, SOPAC, Fiji. Laudon, T.S., 1968. Land gravity survey of the Solomon and Bismarck Islands, in Knopoff, L., Drake, C.L., and Han, P.J. (eds.), The crust and upper mantle of the Pacific area, 279-285. American Geophysical Union, Geophysical Monograph 12. Lindley, D., J988. Early Cainozoic stratigraphy and structure of the Gazelle Peninsula, east New Britain: an example of extensional tectonics in the New Britain arc-trench complex. Australian Journal of Earth Sciences 35, 231-244. Lock, J., Davies, H.L., Tiffin, D.L., Murakami, F., and Kisimoto, K., 1987. The Trobriand subduction system in the western Solomon Sea. Geo-Marine Letters 7, 129-134. Luyendyk, B.P., MacDonald, K.C., and Bryan, W.D., 1973. Rifting History of the Woodlark Basin in the southwest Pacific. Geological Society of America Bulletin 84, 1125-1133. Mahoney, J.J., Storey, M., Duncan, R.A., Spencer, K.J., and Pringle, M., 1993. Geochemistry and geochronology of the Ontong Java Plateau. ln: Pringle, M., Sager, M., Sliter, W. , and Stein, S., (eds.). The Mesozoic Pacific. Am. Geophysics. Union, Geophysics. Mongr. 77, 233-261. Marlow, M.S., Dadisman, S.V., and Exon, N.F., (eds.) 1988. Geology and Offshore resources of Pacific island arcs - New Ireland and Manus region, Papua New Guinea. Circum-Pacific Council for Energy and Mineral Resources Earth Science Series 9. Circum-Pacific Council fo r Energy and Mineral Resources, Houston. McKee, C., Johnson R.W., and Patia, H., 1988. Introduction to volcanic hazards on Bougainville Island. Geological Survey of Papua New Guinea Report 88/23. Miura, S., Shinohara, M., Takahashi, E., Araki, A., Taira, K., Suyehiro, K., Coffin, M.F., Shipley. T., Mann, P. , 1996. Researchers investigate fate of ocean plateaus at subduction zones. Eos 77, 282-283. Nakamura, M., and Shi maki ta, S., 1998. Dissolution ong1n and syn-entrapment compositional change of melt inclusion in plagioclase. Earth and planetary Science Letters 161 , 119-133. Nash, T.J. 1976. Fluid-inclusion petrology - data from porphyry copper deposits and appli cation to exploration. U.S. Geological Survey Prof. Paper 907 D, 16p. Neal, C.R., Mahoney J.J., Duncan R.A., Jain, J.C., Petterson M.G., 1997. The origin, evolution and ultimate fate of the Ontong Java Plateau, SW Pacific: evidence from exposed plateau basement on Malaita, Solomon Islands. In Mahoney, J.J., Coffin, M., (eds), Large Igneous Provinces. Am. Gcophys. Union, Geophys. Monogram. 100, 183-216. 114 Norrish, K., and Chappel, B.W., 1967. X-ray fluorescence specLrography, in Zussman, J. (ed.), Ph ysical methods in determinative mineralogy, 161 -214. Academic Press, London. orrish, K., and Hutcon, J.T., 1969. An accurate x-ray spectrographic method for the analysis of a wide range of geological samples. Geochimica et Cosmochimica Acta 33, 43 1-454. Norrish, K. , and Chappel, B.W., 1977. X-ray fluorescence spectrography. in Zussman, J. (ed.), Physical methods in determinative mineralogy, 201-272. Academic Press, London. Packman, G.H., and Andrews, J.E., 1973. Results of Leg 30 and the geologic history of the southwest Pacific arc and marginal sea complex, in Andrews, E.J., Packman, G.H., et al. Initial reports of the Deep Sea Drilling Project 30, 691-705. Page, R.W. , and McDougall , I. , 1972. Geochronology of the Panguna porphyry copper deposit, Bougainville Island, Papua New Guinea. Economic Geology 67, 1065- 1074. Petterson, M.G. , and Bi liki, N., 1995. A volcano-morphological map of the Gallego volcanic field, western Guadalcanal, Solomon Islands. Water and Mineral Resources Di vision, Ministry of Energy, Water and Mineral Resources. Honiara, Solomon Islands, Technical Report TR4/94, 15pp. Petterson, M.G., Neal, C.R. , Mahoney, J.J., Kroenke, L.W., Saunders A.O. , Babb's, T., Duncan, R.A., Tolia, D., and McGrail , B.A., 1997. Structure and Deformation of north and central Malaita, Solomon Islands: tectonic implications for the Ontong Java Plateau-Solomon Arc colli sion and for the fate of ocean plateaus. Tectonophysics 283, 1-33. Petterson, M.G., Talia, D., Papabatu, A. , Toba, T., and Qopoto, C., J 998. A volcanic hazard of Savo volcano, Solomon Islands, SW Pacific. ln Taylor, P.W, (Ed), Volcanic Hazards and Emergency Management in the SW Pacific. Tech. Bull. 11, SOPAC, Suva, Fiji (in press). Petterson, M.G., Babbs, T., Neal, C.R., Mahoney, J.J., Saunders A.D., Duncan, R.A., Tolia, 0., Magu, R., Qopoto, C., Mahoa, Hand Natogga, D. 1999. Geological-tectonic framework of Solomon Islands, SW Pacific: crustal accretion and grovvth within an intra-oceanic setting. Tectonophysics, International Journal of Geotectonics and the geology and physics of the interior of the earth , 30 1, 35-60. Petterson, M.G., Coleman, P.J., Talia, D., Magu, R., and Mahoa, H. Application of Terrane Modelling of Solomon Islands, SW Pacific, to Mettalogenesis and Mineral Exploration in Composite Arc-Ocean Floor Ten-ane Collages. In press 11 5 Pudsey-Dawson, P.A. 1960. Notes on geological reconnaissance on Choiseul 1957. British Solomon Islands Geological Record 1957-1958, 14-15. Purvis, J.G., and Kemp. G.J. 1975. Final reports on reconnaissance permits R33, R35 and R36 Choiseul. Unpublished report Con:tinc Riotinto Australia Exploration Ply Ltd (CRA). (Open-file, Geology Division, Ministry of Natural Resources, Honiara, Solomon Islands). Ridgway, J., and Coulson, F.l.E.. and 9 others. 1987. The geology of Choiseul and the Shortland's IsJands, Solomon Islands. Overseas Memoir of the In sti tute of Geological Sciences, London, UK. Overseas Memoir No. 8. Rees, B. V.L., 1969. The Geology of Panguna deposit, Bougainville, Terri tory of Papua New Guinea. Un iversity of Adelaide B.Sc (Hons) thesis (unpubl.). Robertson Research Australi a Limited and Flower Doery Buchan Limited, 1985. Petroleum potential of the Cape Vogel, Bougain ville and New Ireland Basins, Papua New Guinea. Papua New Guinea Geological Survey, Port Moresby. Roedder, E. 197lc. Natural and laboratory crystallization of lunar glasses from Apollo 11. Mineral. Soc. Japan, Spec. Paper 1, 5-12 (Proc. IMA-IAGOD Meeting '70, IMA Vol.). (303, 536). Roedder, E. 1979. Origins and significance of magmatic inclusions. Bulletin Mineral, 102, 487-510. Roedder, E. 1984. Fluid Inclusions: Reviews in mineralogy. Mineralogical Society of of America, vol. 12. Roedder, E. 1992. Fluid inclusion evidence for immiscibility in magmatic differentiation. Geochimica et Cosmi chimica Acta, 56, 5-20. Rogerson, R.J., Hilyard, D.B., Finlayson, E.J., Johnson, R.W., McKee, C.O., Chappel, B.W., Price, R.C., Nion, S.T.S., Joseph, L., Sumaiang, R. , Duguman, J. , and Patia, H., 1989. The Geology and Mineral Resources of Bougainville and Buka Islands, Papua New Guinea. Rollinson, H.R. 1995. Using geochemical data: evaluation, presentation, and interpretation. Geochemistry Series. Rose, J.C., and Woolard, G.P., and Malahoff, A., 1968. Marine gravity and magnetic studies in the Solomon Islands, in Knopoff, L., Drake, C.L., and Harl, P.J., (eds), The crust and upper mantle of the Pacific area, 379-410. Ame1ican Geophysical Union, Geophysical Monograph 12. Rudnick, R.L., and Jackson, I., 1995. Measured and calculated elastic wave speed in partial equilibrated mafic granulite xenoliths: implications for the properties of an under plated lower continental crust. J. Geophys. Res. 100910, 2ll-218. l 16 Scotc, R.M., Heyligers, P.B., McAlpine, J.R. , Saunders J.C., and Speight, J.G., 1967. Lands of Bougainville and Buka Islands, Territory of Papua New Guinea. Land Research Series 20. Australia, Commonwealth Scientific and Industrial Research Organization, Melbourne. Sopacmaps, 1994. Results of swath mapping surveys in Melanesia. Report numbers 192 (Central Solomon Trough), L94 (Malaita), and 192 (Melanesian Arc Gap). SOPAC, Suva, Fiji. Stanton, R.L. 1994. Ore Elements in arc lavas. Oxford monographs on Geology and Geophysics, No.29. Stewart, W.D., Francis, G., and Pederson, S.L., 1986. Hydrocarbon potential of the Bougainville and southeastern ew Ireland Basins, Papua New Guinea. Papua New Guinea Geological Survey Report 86/11. Stewart, W.D., and Sandy, J.M., 1986. Cainozoic stratigraphy and structure of New Ireland and Djaul Island, New Ireland Province, Papua New Guinea. Papua ew Guinea Geological Survey Report 86112. Stewan, W.D., and Sandy, J.M., 1988. Geology of New Ireland and Djaul Islands, north eastern Papua New Guinea, in Marlow, M.S. , Dadjsman, S.V., and Exon, N.F. (eds), Geology and Offshore resources of Pacific island arcs-New Ireland and Manus region, Papua New Guinea, 13-31. Circum-Pacific Council for Energy and Mineral Resources Earth Science Series 9. Circum Pacific Council for Energy and Mineral Resources, Houston. Tachenouchi, S., and Imai , H. 1975. Glass and fluid inclusion studies in acidic igneous rocks from some mining areas in Japan. Econ. Geol. USA, 70, 750-769. Taylor, G.R., Olangi, R., Iko, S., and Pule, N. 1974. 1:50,000 geological maps of the West Florida Islands. Publications of the Ministry of Nawral Resources, Honiara, Solomon Islands. Taylor, G.R. , 1976. Styles of mineralization in the Solomon Islands - A review. UN ESCAP Committee for Co-ordination of Joint Prospecting for Mineral Resources in South Pacific Offshore Areas Technical Bulletin 2, 83-91. Taylor, B. , and Exon, N.F. , 1987. An in vesrigalion of ridge subducti on in the Woodlark Solomon 's region: Introduction and overview, in Taylor, B., and Exon, .F. , (eds), Marine geology, geophysics, and geochemistry of the Woodlark Basin - Solomon Islands, J-24, Circum Pacific Council for Energy and Mineral Resources, Houston. Taylor, S.R., Capp, A.C., and Graham, A., 1969. Trace element abundances in andesites. Contributions to Mineralogy and Petrology 23, 1-26. I 17 Tejada, M.L.G., Mahoney, J.J., Duncan, R. A., Hawkins, M.P., 1996. Age and geochemistry of basement and alkali rocks of Malaita and Santa Isabel, southern margin of the On- tong Java Plateau. J. Petrol. 37, 361-394. Thompson, J.E., 1962. The Pumkuma copper-gold prospect, Bougainville Island, Territory of Papua New Guinea. Australia, Bureau of Mineral Resources Record J 962/39. Turner, C.C., and Ridgway, J. 1982. Tholeeitic, calc-alkaline and (?) alkaline igneous rocks of the Shortland Islands, Solomon Islands. Tectonophysics, 87, 127-146. Vedder, J.G., Bruns, T.R., and Cooper, A.K., 1987. Geologic framework of Queen Emma Basin, Eastern Papua New Guinea, in Vedder, J.G., Bruns, T.R., and Colwell, J.B. (eds), Aspects of regional geology, Solomon Islands and Papua New Guinea joint cruise report, R/V S.P. Lee, 1984. United States Geological Survey, Branch of Pacific Marine Geology, Manlo Park, California (unpubl. ). Vedder, J.G., Pound, K.S., and Boundy, S.Q., (eds), 1986. Geology and Offshore resources of Pacific island arcs-central and western Solomon Islands. Circum-Pacific Council for Energy and Mineral Resources Earth Science Series 4. Circum Pacific Council for Energy and Mineral Resources, Houston. Vedder, J.G., and Coulson, F.I., 1986. Regional offshore geology of the Solomon Islands, in Vedder, J.G., Pound, K.S., and Boundy, S.Q., (eds), Geology and offshore resources of Pacific island arcs-central and western Solomon lslands, 59-87. Circum-Pacific Council for Energy and Mineral Resources Earth Science Series 4. Circum Pacific Council for Energy and Mineral Resources, Houston Weisse!, J.K., Taylor, B., and Karner, G.D., 1982. The opening of the Woodlark Basin, subduction of the Woodlark spreading system, and the evolution of northern Melanesia since mid-Pliocene time. Tectonophysics 87, 253-277. Wilson, M. 1995. Igneous Petrogenesis: A global tectonic approach. Department of Earth Sciences, University of Leeds. Wulff-Pedersen, E., Neumann , E-R, Vannuci, R., Bottazzi, P., and Ottolini, L. 1999. Silicic melts produced by reaction between peridotite and infiltrating basaltic mells: ion probe data on glasses and minerals in veined xenoliths from La Palma, Canary Islands. Contrib. Mineral Petrol, 137, 59-82. 118 APPENDIX 1 Sample preparations Rock samples collected from the field were prepared for geochemical analysis in the laboratory. Fresh rock samples were crushed and ground in tungsten carbide mil ls to for whole rock geochemical analysis. For electron microprobe analysis, polished thin sections 60 by 30 mm in size, were prepared, polished and carbon coated. The Mis were analysed using both the wavelength dispersive spectrometer (WDS)-equipped electron microprobe (EMP) al the Research School of Eanh Sc ience (RSES), and the energy dispersive equipped SEM at Research School of Biological Science (RSBS). The SEM was used to determine Both the he major chemical composition of the Mis and its host crystal using the JEOL 6400 scanning electron microprobe machine available at the RSBS , Australian National University. Bulk Major and Trace Element analysis: XRF and ICP-MS The Fauro, Choiseul, and Gold Ridge suites were analysed using X-ray fluorescence (XRF) techniques available at the Department of Geology (ANU) while trace elements (including the REE) were analysed using the Laser Ablation JCP-MS techniques al RSES. The whole rock analysis for the Bougainville volcanic rocks had previously been undertaken at the Department of Geology, Australian National University (ANU) in Canberra as well as at the Department of Geology, La Trobe University (Melbourne) using the analytical methods of Norrish and Hutton (1969) and Norrish and Chappell (1967.1977). 119 APPENDIX 2 Major and trace element analyses of volcanic rocks from Bougainville, Savo, and Gallego Table 19. Bougainvillc (Bagana) Bulk Major and Trace Element Analyses Volcnno B ~I BM Bagan a Bagana Lava Flow NGVNo. 689 357 771 770 789 SI02 6347 496 55.25 55 49 5583 Ti02 0 48 023 0.8 08 078 Al203 1583 17 17 17.39 17 4 1 17 67 Fo203 271 1.44 4 51 4 5 4 79 Foo 2 16 3.81 3.7 3 72 3.45 MnO 011 0 13 0. 17 0 17 0 18 MgO 201 10.02 3.4 3.31 3 4 1 cao 58 15.1 816 8.04 804 Na20 4 09 1.6 4 16 4 18 4.23 K20 184 035 1 54 1.55 1.55 P205 021 0.03 035 0.35 037 s <002 001 <002 <0.02 <002 H20+ 085 024 013 02 004 H20· 0 11 0 14 0.07 0.08 0.06 C02 0 13 0 1 006 009 002 rost 0 17 0 13 0 18 0. 19 021 o:s Total 99 97 1001 99 87 10008 10063 Trnco Elomonts Ba 320 60 245 245 2•13 Rb 37 5 22 22.5 25 Sr 730 191 780 790 841 Pb 5 3 5 4 6 Th 1 5 1 1 5 3 u 0.5 0.5 <0.5 <1 Zr 86 16 74 77 Nb 5 <1 4.5 4.5 6 y 13 7 18 18 20 La 9 2 9 9 10 Cc 19 5 n 2 1 17 Sc 14 33 19 18 v 94 117 If>:! 154 179 Cr 15 329 4 2 1 Ni 10 140 5 6 2 Cu 16 20 52 49 46 Zn 53 37 67 69 80 Ga 15 11 5 17 17 19 120 Table 20. Bougainville (Bagana) Bulk major and Trace clement (memoir) NGV No. 759 690 689 357 771 770 Sl02 6167 6267 6347 496 5525 5549 TI02 051 0 51 0 48 0 23 08 08 AJ203 1651 1589 1583 17 17 17 39 17 41 Fo203 22 248 2 71 144 .: 51 45 FeO 261 2 58 216 381 37 37' MnO 0 12 011 0 11 0 13 0 17 0 17 MgO 2 15 215 201 1002 3 ~ 3 31 cao 5 66 586 58 151 816 8 04 N320 4 08 398 4 09 16 4 16 4 18 K 20 183 195 184 035 154 1 55 P205 021 022 0 21 003 035 035 s •002 <002 <002 001 <002 <002 H20+ 191 1 58 085 024 0 13 02 H20· 0 47 012 0 11 014 0.07 0.08 C02 0 16 009 0 13 0 1 006 009 rost 0 17 0 17 0 17 013 018 019 o=s 10026 10036 9997 100 1 9987 100.08 Tr3co Elomont Ba 325 350 320 60 245 245 Rb 36 35.5 37 5 22 22 5 Sr 715 715 730 191 780 790 Pb 5 5 5 3 5 4 Th ?5 2 15 1 15 u 05 <05 05 05 <05 Zr 87 82 86 16 74 77 Nb 55 45 5 <1 4 .5 ~ 5 y 13 12 13 18 18 l a 10 11 9 2 9 9 Co 20 20 19 5 22 21 Sc 13 11 14 33 19 18 v 9-l 84 g.; 117 153 154 Cr 11; 12 15 329 4 2 NI 8 8 10 140 5 6 Cu 18 9 16 20 52 49 Zn 55 48 53 37 67 69 Ga 15 14 5 15 ,, 5 17 17 121 Table 2 1. Geochemical compositions of the Pre-1943 eruptions of the Bagana volcano al •2 •3 "-I M• 11- ll 'I "10 S102 5599 562 566 566 ~h .7 51> 77 56l->3 ~6Q 5109 57<1 T102 0 7~ 073 08 0.73 0 ?J O.M 0.69 0.1•7 fl.7 0.6~ \120.\ 1686 17 3 17 SS 17.65 IX 17 7 17.M PY 17.72 IS I ld03 393 sos 542 5.36 435 4 <>5 5 07 St.~ 4 ,9 395 h·O 389 :>8 2 36 2.36 3.IS 2N 2.41 J <) 24•> 2.9 Mn() 0.17 0 16 018 0.17 0 1(• 017 0.18 0 lb 0 .17 0. 11> M~O 4 1 38 2 77 2.76 27 2 61! 2.7 2.7~ 2<•' 2.SS (°;\() 807 8 1 7 45 7.5 7.4 7 42 7.53 25.S 7.4• 7.3 \ ;120 4 1 3.83 36 27 u 3.XS 4..12 4 2-1 ' 15 41~ .'95 K20 134 14 163 1.63 I 11(1 l.M I 59 161 1111 1 7 rim 0.3 0.31 033 0.3\ 0 \S OJ2 031 tJ \5 0 .31 0 ..'4 s 002 0.01 001 0.01 001 002 001 001 O.G2 0.02 H20• 031 0 (I.I 024 0 31 05 0..11> 0.31 0.0<1 1109 0 21 H20· o.os 0()(· 0 1 0.05 0 .1 II .II 0.1 017 0.11 0 .1 1 CO:> 0.05 005 005 0.05 . 005 1105 0.01 005 oos 005 re~I 02 0.2 019 ll.19 0 19 0 .1•> 019 01•1 0 .19 019 o ·~ 001 001 tJOI O.GI TO!al '>9 .~8 •I•) XI '''' •>s 100 t>~uu, <)') 89 99.82 '14.'J '1'> .11 9'l.81 K201Na20 0.35 0 39 0. 'l< 0.38 04\ MgO/FoOI 0.55 0 52 Mg value 1 "5 Mg V(ltuo 2 54 Trace Element frpm) Rb 20.5 21 2(, ~ 26 2<.S 211 s z5_5 25 2q 27.5 Ba .?~O 2'5 275 280 ~75 2ll0 2' 0 270 2"~ 2~0 Pl> 5 .j J J 7 s 6 4 b Sr ~2; 735 l\10 SIO .,so MIO 810 ~00 -·15 X2"' Ki Rb 545 555 510 520 520 SIS 515 ;\S szs 515 K/Sa 50.5 J•) 5 J9 48 5 50 4\ 5 47 5'1 s 4!( 5 505 K/Sr 15.5 If> 111.5 16 5 175 17 16.5 1/1 5 17 17 Rb Sr 0.0285 0.02115 0 0325 0.032 0.0\J 0-0'.1 0.0315 00.'15 00,12 0.0.135 L.1 10 II II 10 II 10 10 JI II JI __, Co 24 24 24 25 211 :?5 ,, 25 25 23 Nd 10 II 12 11 10 II 0 10 IU 10 y 15 It> II• II> 15 IS I<• It• Iii 16 Ir l\2 lll '~ 3' Sl' S5 XI> s~ \1 l<9 4 .j 4 .j 84 .. .. 4 ., 'In" 7<> 72 h'1 M 71 1>'1 M 71> ,.. , ·~,_ .,,_ ·- ('u 4' 21 21 ~" 23 ~o Z4 24 \1 2lo. ll< 4 .; .j 3 I ~ .,, IX 17 14 13 p 12 l.l 14 11 II \ ISO 1'12 141 132 I.IQ 143 1)4 155 144 123 ( r 110 51> 5 ,, s (I.a I ~ l'I IX" 5 l'l' IX' S 11/ s l'I l'I l'I 5 20 122 Table 22. Geochemical compositions of the 1943 - 1953 lava flows from Bagana volcano Ill 112 t/3 U4 ttS lib rl7 #X <;102 52.94 5336 53.53 53.l>S 53.69 53.7 53.7 53 .76 T102 0.88 089 087 O.•>I 09 0.96 o.~l( 0.89 1\1203 16 45 1661 16 67 17.55 17.3 17.8 17.I 17.2 rc203 519 4 48 4.59 4.06 4.34 4. 1 4.15 4.n lcO 4.45 .: 99 4 82 5.0'! 4.94 5.1 4 95 .1 9.1 MnO 0 18 0 18 018 0.19 0.19 0 IS 0.17 019 MgO 4 63 4.32 4 31 >.85 3.S~ 3.S5 4.25 .l.82 C30 8 77 8 5'l 8.5 8.;\9 8.41 XJ :os X.37 Na20 3.8 395 3.87 J.88 3.% 3.6 ~(> 4.02 K20 1.56 158 1 59 1.5\) 1.61 1,6 1.61 I.<>! 1'205 0.35 0.36 0 36 0.32 0.35 035 0.36 0.34 s OJJI 001 001 0.02 O.o2 0.01 0.02 O.lll H20• 0.31 0.37 0.36 0.27 0.22 0,02 0 17 027 H20· 0.o7 0.0•) 0.06 0.05 0.04 o.os 0.o7 0.07 C02 0.05 005 0. 1 0.05 0.05 ... 005 0.05 0.05 rest 021 0.21 o.n 0.21 0.21 0.21 0.21 02 100.0R 100.11 ')<).ll4 o•s 0.01 0.01 0.01 Total 9'1.!(5 99.99 10004 100.08 100.11 Trace Elemen1 (ppm) ,,. Rb 24 5 25 25 .j 23.5 22 .5 25 23 Ba 245 245 245 230 230 225 245 HO Pb s 5 4 7 5 4 5 Sr XIO S35 ~20 ~00 1110 790 825 790 KlRb 530 525 530 575 570 5QO 535 580 KlBa 53 53.5 54 575 58 59 54.5 sx KISr lC> 15.5 1(> 16.5 16.S 17 16 17 Rb/Sr 0.03 O.oJ 0.0305 0.029 o.oi•> 0.02S5 0.0305 0029 La 10 II 12 to 11 10 10 10 Ce 24 23 25 24 23 26 24 24 Nd 12 II II 13 11 II II II y 1<1 16 16 17 19 IS 16 19 Zr 72 71 73 75 7(> 76 74 75 1' b 3 4 4 4 4 4 4 Zn 73 75 76 80 77 82 75 79 Cu 65 65 81 (.~ <>S 72 64 (,() '11 14 12 12 7 <· 9 II x Sc 22 20 22 I\) 19 19 21 21 v 227 215 225 215 20<• 233 225 216 Cr 31 25 25 8 7 'I ·"r R (j3 IX 20 20 1'1 l'l.S 205 IS.5 I<) 123 Table 23. Geochemical compo it ions of the 1959 I 975 lava flows from Bagana volcano •I "' r:3 •4 ~s #(• S102 S5.4 5S 5 $4 91 552 553 5' ' 1102 082 081 082 o.ss ox 0 Ill J\ll 124 Table 24. Geochemical composition of the unknown age of lava now from Bagana volcano '>102 5345 5393 54 3 54.X '5 O<> 'C1.511 56 f) 1102 0 91 087 085 0 X3 0 X2 0 72 0.73 ,\1201 17 6 17 45 17 6 17.S 17 52 IHI IX I I c20l 4 37 45 39 4 15 4 41 3 55 3.65 ' · ~ l cO 507 4 49 4.8 41 4 1' l 9S 35 :IS MnO 0 19 019 0 17 0 17 0 17 0.1 11 O.I X 0.16 016 :-11:0 3 9 1 3 73 3 85 3 65 l 4S .U? 274 27 2.<.5 (".10 84 8 26 85 X05 x 2 7 'IX 7 sx 7 45 7 3 Nn20 395 3.97 3 75 J . ~ 1 7 4.02 431 4 4.05 K20 1 58 16 161 I ('5 I 5 I 51 1.59 l.M 1.7 1'20' 0 '4 O.H 036 0.15 () l4 0 14 0 3:1 0.32 0 '4 s 001 0.01 002 0.01 001 ()01 0.01 0.01 001 H20• 0 24 031 002 <01 006 0 l'I 0.51 0.19 0.15 H20· oos 0.06 006 0.()(1 OOfl OOCI oox 0 13 (I 15 C02 003 0.04 0.05 00' OO' 01 005 005 005 IU$1 0.2 0.2 021 0 21 0.2 01'1 0 l•I o~ 0.2 'IY.)i 10005 0 s O.ot 001 To:.11 99'!9 100.M 1000-l Tr3eo Elemcnl 125 Table 25. Bougainvillc: Trace Element Analyses by Instrumental Neutron Activation NGVNO. 692 727 728 694 703 735 734 721 723 717 754 745 767 691 689 711 708 780 Lii 13 15 14 19 16 19 18 i4 15 12 19 8 9 10 11 9 9 11 Co 27 32 28 39 33 38 37 30 32 25 41 19 19 21 21 20 20 26 Nd IS 18 15 19 17 19 18 16 17 13 20 12 11 11 12 12 12 16 Sm 36 42 38 34 37 42 38 ~ 37 32 42 33 2 Table 25 continued NGVNO. 793 796 804 800 798 773 805 752 698 791 La 11 10 11 11 1 12 12 5 20 10 Co 27 2•1 24 ?S 25 26 21 12 21 22 Nd 17 13 14 14 14 14 12 8 24 13 Sm 38 32 31 3 31 3 2.5 24 s 1 33 Eu 133 122 1 1 097 096 074 082 165 1.1 Tb 06 06 05 OS 04 04 04 05 07 06 Ho 08 07 07 07 06 05 03 06 09 06 Yb 24 2 .1 2 19 19 19 12 2 3 .1 25 Lu 0 36 032 03 029 029 028 0 19 034 047 04 Sc 19 21 10 13 10 9 9 39 12 25 Cr 2 50 2 9 8 4 14 116 56 10 Sb 126 Table 26. Savo I 13ulk Major and Trace Element Analyses Element SV8 SV27 SV6A SV28 SVl8 SV19 SV29 510 SVI SV7B SV20 57532 57542 53 SV25 SV17 SV7A 5102 40 62 42 69 43 76 47 19 502 5097 51 54 52 43 52 58 52.08 5224 53 27 54 46 5618 5985 6147 61 75 Al20 3 7 83 12 75 10 73 16 45 1632 1661 17 47 1638 15 41 17 61 17 2 17 8 1763 1881 17 59 1862 18 s.: T102 103 1 558 0328 0815 0 738 o.n5 087 0 72 065 0758 0754 0 78 074 049 0528 0404 044 Fo203 17 96 16 95 1125 896 1006 9 a.: 989 80b 904 9().1 9 56 8 88 806 685 5 75 4 44 4 58 Mno 0 188 0 183 0233 0165 0.135 0 132 0 17 0 21 016 0151 0 190 01 016 0 ..14 0 106 0095 0101 MgO 14.75 10 18 1563 84 4 7 4.77 4 84 383 691 36 343 323 4 31 345 306 13 2.16 C30 14 65 109 1396 128 1061 10 78 9 II 891 10 11 79 9 79 9 24 8 76 831 542 43 512 Nn20 146 2 73 249 321 357 348 385 3 58 346 4 22 4 04 4 06 4()..: 4 52 529 615 562 K20 0317 0 548 0 369 0526 1439 1492 113 138 1 29 1 423 1591 1 22 149 I 09 1857 2087 1 732 P205 002 009 003 006 0 .19 021 02 02 0 17 024 02 0 22 023 0 17 0 16 012 02 L.0.1 059 4 61 0 1.7 072 098 TOTAL 988 986 988 986 98 991 9966 10032 9917 976 99 1005 100 59 10101 997 99 1 1003 Zr 303 502 25 39 716 712 55 52 49 127.8 776 55 68 53 113 1364 82 y 11 4 206 128 213 19 20 1 16 13 14 11 4 20 2 14 12 9 13 1 9 22 1 Nb 16 25 1 2 1 s 31 3.7 1 1 1 5.7 32 2 2 2 44 59 31 Cr 417 277 8611 1884 1339 1569 24 33 2 14 44.4 634 33 43 16 302 25 5 361 v 5609 603 2 1993 352 6 413 3 482 7 168 368 247 114 6 2786 408 335 271 153 3 955 3059 Rb 26 56 44 44 11 8 163 15 16 18 28.5 161 11 22 16 316 253 17 7 Sr 1726 373 2 1379 3889 8552 8574 634 586 696 11595 737 8 601 835 742 7998 1451 6693 Ba 601 144 2 395 1064 5304 4939 251 371 322 743 328 1 363 513 369 5516 794 5 3331 Nf 187 4 352 170 416 33 7 33 13 37 53 226 166 44 47 33 156 102 202 Cu 15 7 661 194 813 119 7 1089 119 201 111 20 I 95 1 227 127 153 33 3 198 104 I Zn 805 963 952 651 865 92 7 75 152 68 586 688 133 151 142 39 494 74 8 Ga 14 7 158 ,, 2 18 17,9 207 21 7 162 21 5 205 203 Sc 706 719 609 558 33 383 25 18 44 309 6 54 212 Co 11::> 75 3 65 4 358 30.6 32 I 32 40 9 1 24 2 97 61 24 1 Th 4 5 2 u 0 2 <3 2 1 1 Pb 2 2 7 3 9 6 Co 124 139 01 149 402 36 8 22 35 5 129 227 33.7 Nd 68 47 06 4 8 17 9 163 7 1 129 8 88 17 9 L~ 48 66 08 3 18 3 163 83 1::> 3 38 72 114 127 Table 27. Savo 2 Bulk Major and Trace Element Analyses (RLS data) Sample# RLS31 RLS32 RLS33 RLS34 RLS35 RLS36 RLS37 RLS38 RLS39 RLS40 RLS41 RLS42 RLS43 RLS4•1 RLS.15 RLS46 RLS47 RLS48 Si02 64 17 5542 5333 51 07 53 22 5301 54 83 581 51.73 539 65 36 S4 49 52 79 60 18 628 6288 5347 5508 TI02 0 32 067 0.72 067 085 078 0 75 055 084 074 027 0 76 087 0 53 0 39 043 0 75 0 71 Al203 17 58 17 53 17 84 14 58 17 45 17 55 17 36 17 91 18 13 17 78 17 85 17.31 17 06 17 77 1807 17 58 1849 176 Fo203 295 531 4 01 4 37 4 63 652 4 55 526 5 37 48 142 47 637 358 369 2 78 7 29 ~ 86 FcO 0().1 195 331 4 25 4 08 1.91 331 059 34 301 103 3 .12 281 149 102 084 263 MnO 007 0 14 0 17 0 15 015 0 14 0 14 011 016 015 007 0,15 0 .15 0 I 009 008 016 0 .14 MgO I 27 364 348 7 41 4 43 396 4 II 288 4 25 383 106 4 23 4 .74 205 156 2 .1s 292 391 Cao 3 77 7 88 932 1093 4 08 869 841 595 961 896 327 852 895 62 4 43 381 869 8 36 Na20 603 4 22 4 27 308 382 4 4 23 5 398 387 64 4 25 384 4~ 59 tl21 4 37 4 14 K20 187 146 161 12 129 148 1 47 168 142 148 205 1 46 129 195 211 2 35 1 71 142 P20S 0 13 02 024 0 17 02 02 02 022 023 021 0 I 021 018 018 0 18 0 23 0 25 019 s <002 <002 <002 <002 002 <002 <002 <002 <0 .02 <002 <0.02 <002 <0.02 <002 <002 <002 <0.02 002 H20+ 0.9-1 0 73 044 093 063 0.66 037 07 037 064 042 033 057 04 0 39 034 0 38 041 H20· 0.33 063 03 073 0 19 069 0 .15 045 0 19 021 012 0 12 0 14 0 17 0 16 008 022 0 17 C02 0 24 0 12 086 04 011 0 11 0 12 018 0. 13 012 0 11 01 0 14 019 0 12 0 12 015 0 11 Rost 0 26 0 22 0 21 025 022 021 021 026 023 022 027 0 22 02 0 22 03 0 33 022 022 O=S 9986 001 Total 9997 100 12 100.11 10019 10035 9991 10021 9985 10004 9992 998 9997 1001 9995 100 19 10039 9991 9995 Tr.lee Elements Ba 640 320 300 300 240 285 300 525 270 300 640 310 245 410 670 745 320 305 Rb J9 225 20 5 15 5 17 5 18 5 215 285 155 30 385 21 17 5 30 365 37 20 25 Sr II~ 751 739 639 657 701 726 1040 723 753 1220 738 633 806 1430 1570 765 771 Pb 10 5 5 5 4 5 5 9 5 5 10 5 4 7 12 16 6 5 Zr 138 85 64 61 74 70 84 133 70 77 129 83 73 106 141 137 87 81 Nb 4 1 s 1 1 1 1 2 3 1 15 3 1 5 15 25 35 3 15 y 7 15 16 13 17 19 16 15 17 16 6 16 16 14 9 9 17 16 La 8 8 8 8 7 9 7 11 6 7 7 8 7 10 9 12 7 7 Co 16 18 16 15 17 15 16 23 18 18 18 17 16 20 17 25 17 19 Sc 5 18 16 27 2•1 22 20 II 19 20 4 20 28 12 6 7 17 19 v 43 217 225 253 286 243 226 131 267 256 52 223 273 189 57 86 182 220 Cr II 39 16 234 21 19 35 20 24 20 13 39 26 17 14 17 18 44 Mn Co 7 23 23 38 27 24 24 18 28 25 5 22 27 14 9 11 2~ 27 NI 7 17 11 56 16 II 16 :>s 13 11 5 17 15 8 8 11 7 18 Cu 10 97 110 130 138 82 90 S4 173 ~ 23 103 87 19 14 29 197 80 Zn 52 67 73 75 81 69 66 81 76 74 46 66 n 52 55 53 7~ 70 Ga 22 5 205 20 165 205 20 20 21 5 20 5 20 22 5 20 21 20 5 23 22 5 21 21 128 Table 28. Gallego (West Guadalcanal) Bulk Major and Trace Element Analyses $3mplc r RLSI RLS2 RLS3 RL$4 RLS6 RLS7 RLS8 RLS9 RLSll RLS12 RLSl3 RLSl4 RLS16 Si02 6317 5887 61 18 5803 51 69 5086 5086 53 75 6433 s.:aa S4 17 558 5178 TI02 0 • 6 046 044 047 084 083 085 081 034 088 069 068 07 AJ203 1807 1812 178 1853 14 68 14 72 1998 1858 16 57 1791 1845 1867 1899 Fc203 35 333 5 11 58 4 68 553 54 527 171 559 3 3\ 4 28 805 FcO 085 2 .16 029 4 76 39 326 2 64 19 193 4 45 223 03\ MnO 0 .11 0 11 01 0 II 0 .16 0 17 0 15 015 007 016 013 0 13 0 15 M90 06 291 2 53 316 6 37 6 35 3 44 38 2 68 343 44 337 5" cao 501 7 ?9 603 7 27 1092 1088 993 921 531 868 869 808 863 Na20 4 82 4.18 4 21 4 , 1 268 268 353 33 ,. 73 368 349 369 313 K20 11 103 I 18 092 126 125 116 054 \IS 055 0 71 075 038 P205 0 17 0 .13 013 0 13 031 03 0 19 015 0 I 02 0 15 017 0 12 s <002 <002 <002 <002 <002 <002 <002 <002 <002 <002 <002 <0.02 <002 H20+ 067 0 71 0 76 063 0 .73 066 073 098 058 I 07 0 .73 089 181 H20· 075 022 032 021 04 2 048 042 068 0 1 063 03 098 068 C0 2 014 009 0 11 0 11 024 006 007 008 0 I 008 012 0.08 005 Res I 0 16 0 17 0 .16 0 17 025 025 0 18 0 14 017 016 016 0 14 0 16 10001 Ots 001 Total 9958 99 78 10006 9994 100 99 75 100.15 10008 99 8-1 9983 9995 9994 10008 Traco Elements Ba 335 285 370 300 140 150 220 115 330 160 150 155 175 Rb 205 155 19 9 45 435 24 5 8 195 10 12 75 Sr 667 601 577 593 676 663 550 397 599 468 453 424 442 Pb 6 5 4 4 5 5 4 2 4 3 3 4 3 Zr 84 65 75 64 94 92 74 83 67 109 70 92 73 Nb 2 I 5 2 1 5 25 25 25 15 15 35 2 15 y 10 II 12 11 24 2• 19 22 8 23 19 20 17 u 8 4 6 4 22 23 18 5 3 9 7 6 6 Co 16 13 13 10 45 47 23 15 9 26 20 17 15 Sc 9 17 15 17 39 37 22 24 14 20 33 17 27 v 69 142 68 142 253 258 236 189 89 157 191 151 131 Cr 7 24 22 26 235 245 12 28 100 80 33 17 92 Mn Co 11 18 19 22 38 40 2;; 26 15 26 33 19 33 NI 6 9 10 12 40 38 8 16 19 19 2\ 12 34 Cu 28 60 36 58 144 102 96 53 25 18 78 48 63 Zn 56 53 50 56 80 79 74 84 43 72 72 68 78 Go 18 5 195 18 18 5 \65 165 21 19 16 19 5 17 19 17 5 129