Metamorphic sole of the Papuan Ultramafic Belt ophiolite : field mapping, petrochemistry, geochemistry, geochronology and associated reconnaissance experimental studies on boninite genesis

Abstract

The Papuan Ultramafic Belt (PUB) ophiolite comprises former oceanic crust and upper mantle that was emplaced onto continental crust in Papua New Guinea (PNG) in a zone of convergence between the Pacific and Australian plates. The metamorphic sole beneath the ophiolite is best exposed in the Musa-Kumusi divide and comprises a 40 to 300 m thick body of greenschists, amphibolites, granulites, and gabbroic rocks which are separated from the overlying ophiolite by a Transition Zone of banded ultramafic rocks. Sampling and thin section examinations of the metamorphic rocks have established a sequence from greenschist and albite + epidote amphibolite facies through amphibolite to hornblende granulite and hornblende + olivine granulites close to the ultramafic contact. There is also a strong compositional contrast within these metamorphic facies, expressed by Mg# of olivine and co-existing phases, and by decreasing silica and alkalis. The compositional gradient within the metamorphic sole is irregular in that layering conformable with the attitude of the metamorphic foliation provides 'scatter' in bulk compositions and variability in mineral phase proportions. Similarly, there is a compositional gradient within the ultramafic sheet in which extremely refractory harzburgites distant from the metamorphic sole contact, become more variable in mineralogy and relatively enriched in Fe/Mg, Al/Cr and with low but significant Ca, Al, Na, Ti contents proximal to the contact. The strongly banded rocks in the Transition Zone have an appearance approaching dunite or harzburgite mylonite. These are interlayered with lherzolite and pyroxenites with minor saussuritized plagioclase and/or pargasitic hornblende. The brown hornblende amphibolites of the metamorphic sole recrystallized at approximately 4 kbars, 800°C, whereas the granulites recrystallised at -3 kbars, 900-1000°C. The gabbros, Transition Zone rocks and the harzburgite the above the metamorphic sole show a cooling history from higher temperature to approximately 1000°C. Major and trace element compositions indicate that the amphibolites, granulites, and gabbros are LREE-enriched within plate tholeiitic basalts (WPB) that are similar or transitional to the mafic rocks of the Emo Metamorphics. The observed rocks were subject to a complex sequence of chemical modifications that occurred both prior to and during the metamorphism related to emplacement of PUB ophiolite. Trace element compositions and trace element partial melting models suggest that the bulk of the granulites within the sole were formed by up to 10% melt extraction, and the more depleted granulites and gabbros are residues formed by up to 50% melting. Melt extraction alone cannot explain the high MgO, Cr, and Ni bulk compositions of the mafic granulites and gabbros. The range of major element compositions observed within the metamorphic sole can be accounted for by a simple variant of the melt-residue model, whereby a range of starting compositions produced from probable mechanical mixtures of typical sole metabasites and PUB harzburgite compositions, is subject to partial melting and melt extraction. Geochronological studies on the metamorphic sole, using amphiboles from the granulites and amphibolites, yield measured K-Ar ages, average 40 Ar-39 Ar direct total fusion ages, and plateau ages. Five of the six 40Ar-39Ar plateau ages, derived from age spectra, lie between 58.6 ± 0.2 and 57.8 ± 0.2 Ma with an overall mean age of 58.3 ± 0.4 Ma. The large spread in measured K-Ar and 40 Ar-39 Ar total fusion ages is thought to be caused by the presence of variable amounts of excess argon. The mean plateau age for five samples of 58.3 ± 0.4 Ma is interpreted to mark the time of cooling of the metamorphic sole following peak metamorphism. It is suggested that the development of the metamorphic sole and emplacement of the PUB ophiolite onto the PNG crust occurred in a relatively short time interval in the Paleocene. The age of the metamorphic sole formation coincides with the age of eruption of Dabi Volcanics and associated boninitic lavas at 58.9 ± 1.1 Ma (Walker and McDougall, 1982). The age relations suggest that sole formation and eruption of the Cape Vogel boninites were associated with emplacement of the PUB ophiolite wedge and its interaction with the proto lith of the Emo Metamorphics in a subduction zone environment. Melts and fluids from slab metamorphism and melting migrated into the wedge environment causing melting and possible genesis of boninitic liquids. The composition of the melt derived from the Emo Metabasite slab in the basalt/eclogite melting experiment at 1050°C, 2.0 GPa is that of a rhyodacite. This melting scenario was explored at 2.0 GPa establishing a large temperature interval (at least 850-11 00°C) over which the rhyodacite/siliceous melt co-exists with gar + cpx + rutile + plagioclase (to around 1050°C). The residue is mafic to ultrabasic in composition, enriched in CaO, MgO, FeO and Ti02, and strongly depleted in K20 and SiOz and depleted in Na20 and Alz03• At lower pressures and high temperatures, this bulk composition would crystallise to olivine + clinopyroxene + hypersthene + plagioclase + ilmenite if it is anhydrous, or to a hornblende-rich composition in the presence of 1-2% H20. Loss of a rhyodacitic melt fraction from the chosen 'Emo Metamorphics' composition produces an olivine-normative and silica-depleted composition that matches the hornblende granulites/hornblende gabbros at the base of the ophiolite. However, the Mg# of these residues, at -50 % melting, is too low in comparison with values of-70-80 in the high grade rocks near the ultramafic contact. The model tested was for extraction of 25% of rhyodacite melt. Higher degrees of melting (probably in several stages) would be required to produce residues as Mg-rich as Mg#75. However, initial melts eliminate quartz/coesite, potassic phases and most of the Na-component originally in pyroxenes or plagioclase. The residue is refractory in terms of solidus temperature and further melting is difficult without fluxing by 'water-rich fluids. In the fluxed-melting study of refractory mantle wedge, the olivine-control lines through the Maaloe and Aoki peridotite composition show that liquids with harzburgite residues match the plotting position of high-Ca, low-Ca Types I and II and some Type III boninites, which on this basis, may be derived from this peridotite. Projection from '01' shows that high-Ca and Type III low-Ca boninites lie above the harzburgite residue control line. The ratios between Na, Ca and Al of the rhyodacite flux superimposed on the melt products, and suggests that high-Ca boninites and Type III boninites could be derived from refractory lherzolites by H20-fluid fluxing alone, with little need for addition of any major and minor element by the fluxing components. The experimental techniques applied in this study can also be utilised in the study of sole formation and related processes in other similar ophiolites throughout the world, particularly those with associated boninitic lavas. The Emo Metamorphics Derived Model (and its variant the Retrogressed Eclogite Model) discussed in Chapter 8, suggests that the metamorphic sole formed when the proto lith of the Emo Metamorphics and base of the PUB ophiolite came into contact at a subduction zone or at deeper parts of the Owen Stanley Fault. Water, fluids and melts moved into the actively deforming contact zone, and ssociated secondary faulting, thinning and stacking of proto lith of the Emo Metamorphics and ophiolitic rocks formed the metamorphic sole and Transition Zone rocks. The mineralogical banding textures observed within the metamorphic sole and Transition Zone represent complex individual recrystallisation events during ophiolite emplacement. Further work on process of mechanical mixing and re-equilibration of the mixtures to form a different bulk composition needs to be done to better understand the complex process of metamorphic sole formation. Show more