John Tarney

Geochemical characteristics of basaltic volcanism within back-arc basins

Summary Back-arc basins are formed by extensional processes similar to those occurring at mid-ocean ridges. However, whereas the magmas erupted along the major ocean ridges are predominantly LIL element-, Ta- and Nb-depleted N-type MORB, many back-arc basins are floored by basalts transitional between N-type MORB and island arc or even calc-alkaline basalts (viz. enrichment of LIL elements (K, Rb, Ba, Th) relative to HFS elements (Nb, Ta, Zr, Hf, Ti)). On a broad scale, it is possible to relate basalt composition, tectonic setting of the basin, and maturity of the adjacent subduction zone. Thus, the Parece Vela Basin, formed during the earliest stages of the Mariana subduction system, is floored by basalts indistinguishable from N-type MORB, whereas the later Mariana Trough is erupting N-type MORB and basalts with calc-alkaline characteristics, commonly in close spatial proximity. The calc-alkaline component is best developed in narrow, ensialic basins such as Bransfield Strait, where the extension is adjacent to mature, continent-based magmatic arcs. This range of compositions, from N-type MORB to calc-alkaline basalt, can be satisfactorily explained only by invoking chemical variations in the composition of the mantle material supplying the back-arc basin crust. Two major processes may be suggested: (i) selective contamination of the mantle wedge by LIL-enriched hydrous fluids, perhaps together with sediments, derived from the descending, dehydrating oceanic lithosphere; and (ii) repeated melt (and incompatible element) extraction during basalt genesis. The former process will enrich the mantle source of back-arc basalts with LIL elements; the latter will deplete the source in all incompatible elements, but the net effect of both processes is to increase the LIL/HFS element ratio of the source regions. Consequently, as the subduction zone matures, the LIL/HFS element ratio of successive back-arc basalts will be expected to increase, from initial N-type MORB ‘background’ values, to ratios more typical of island-arc basalts. The model has implications for mantle dynamics in back-arc regions, because transfer of material from the subducted slab may destabilize the overlying mantle, potentially leading to diapiric uprise when tectonic conditions permit extension.

Subduction of pelagic sediments: implications for the origin of Ce-anomalous basalts from the Mariana Islands

Attempts to assess the significance of subducted sediments in the genesis of island arc magmas have been limited by the lack of trace element data on pelagic oozes. Accordingly, we have analysed a series of pelagic clays and nannofossil oozes from the Nazca Plate for REE and other trace elements. A calculated average––Pacific Authigenic Weighted Mean Sediment (PAWMS)––exhibits light REE-enrichment (Lan/Ybn∼4.5), high contents of Ba and Sr, but low abundances of Rb, Nb and Ta. Most significant, however, is the occurrence of large, negative Ce anomalies (Ce/Ce*~0.2). We have attempted to model the contribution of PAWMS-type material to the source of the magmas of the Mariana Island Arc, an intra-oceanic arc far removed from the effect of continent-derived detritus. Only small amounts of pelagic sediment, between 0.3 and 0.5% of the source, are required to develop the small negative Ce anomalies, high Ba/La ratios, and high LIL/HFS element ratios characteristic of these and other arc lavas. However, a small fluid contribution from the dehydrating subducted oceanic crust is required to produce the high Rb/Ba observed in several Mariana suites. The ternary mixing between sediment, mantle host and dehydrating oceanic crust also produces very low Nb and Ta abundances in the arc lava source. However, the very high abundance of Sr (>1000ppm) and the estimated high 87Sr/86Sr ratio (~0.7150) of PAWMS, results in a model 87Sr/86Sr ratio of 0.7070. This is far higher than the measured ratios in the Mariana arc lavas (0.7033–0.7040) and may suggest that the subducted sediment has a lower Sr content (<200 ppm) or a lower 87Sr/86Sr ratio, or that the carbonate-rich component is not involved in source contamination. Volumetrically it would appear that much of the sediment approaching the Mariana trench (~45 km3 Ma−1 per km of arc) may be recycled into the deeper mantle. This will have the effect of introducing high Ba, Sr, Th, 87Sr/86Sr, 208Pb/204Pb and 207Pb/204Pb material into the mantle. However, incorporation of such material cannot alone satisfactorily account for the trace element and isotope chemistry of ocean islands; oceanic sediments have LIL/Ta, Nb ratios far too high to produce the trace element characteristics of most intraplate magmas.

Trace element geochemistry of orogenic igneous rocks and crustal growth models

Some of the more important constraints on crustal growth mechanisms are reviewed in relation to the trace element composition of orogenic igneous rocks, with specific reference to granitoids, and in relation to tectonothermal models of subduction zones. The wide range of magma types associated with subduction zones is difficult to explain by a single magma-generating mechanism. This variation can be accommodated if some lavas ( are tholeiites and boninites) are formed through wet melting of upwelling asthenosphere at the initiation of oceanic subduction, whereas others are generated from the mantle wedge during the more mature stage of ares, with either hornblende or phlogopite dominated assemblages, where induced convection in the mantle wedge enables source replenishment. Yet other products are generated through melting of the slab (±subducted sediment), but only where the subducting ocean crust is very young and/or warm. The range and variation of trace-element patterns displayed by basic lavas is also prevalent in more siliceous compositions, implying that the mantle component in many granitoids may be much higher than is commonly assumed. As well as the traditional I-type, S-type and A-type granites with generally low-Ba and low-Sr characteristics, there is another important group of granitoids that has high-Ba and high-Sr concentrations, and that consistently has about 10 trace-element characteristics that are complementary in nature to the first groups. This type of granitoid is common in the Arehaean, rare in the Proterozoic, but is surprisingly abundant amongst post-Cretaceous to Recent orogenic volcanic and plutonic rocks. Trace element patterns may be inherited from mineralogically-controlled element fractionation occurring deep in the lithosphere; major element characteristics are more dependent on mineral-fluid control during magma generation.

Fluid Influence on the Trace Element Compositions of Subduction Zone Magmas

Subduction zones represent major sites of chemical fractionation within the Earth. Element pairs which behave coherently during normal mantle melting may become strongly decoupled from one another during the slab dehydration processes and during hydrous melting conditions in the slab and in the mantle wedge. This results in the large ion lithophile elements (e.g. K, Rb, Th, U, Ba) and the light rare earth elements being transferred from the slab to the mantle wedge, and being concentrated within the mantle wedge by hydrous fluids, stabilized in hydrous phases such as hornblende and phlogopite, from where they are eventually extracted as magmas and contribute to growth of the continental crust. High-field strength elements (e.g. Nb, Ta, Ti, P, Zr) are insoluble in hydrous fluids and relatively insoluble in hydrous melts, and remain in the subducted slab and the adjacent parts of the mantle which are dragged down and contribute to the source for ocean island basalts. The required element fractionations result from interaction between specific mineral phases (hornblende, phlogopite, rutile, sphene, etc.) and hydrous fluids. In present day subduction magmatism the mantle wedge contributes dominantly to the chemical budget, and there is a requirement for significant convection to maintain the element flux. In the Precambrian, melting of subducted ocean crust may have been easier, providing an enhanced slab contribution to continental growth.

Lewisian gneiss geochemistry and Archaean crustal development models

Abstract The geochemistry of Lewisian amphibolite-facies gneisses from northwest Scotland is described with particular reference to the rare earth elements (REE) and compared with the geochemistry of Lewisian granulite-facies gneisses. The results show that there are no significant differences between “Laxfordian” amphibolite-facies and “Scourian” granulite-facies gneisses in terms of REE and other immobile trace elements (at equivalent silica levels), although the mobile radioactive heat-producing elements, K, Rb, Th, U, are significantly lower in the granulites. In both types the basic gneisses have moderately fractionated REE patterns while the intermediate and acid gneisses have strongly fractionated REE patterns with low heavy REE abundances and decreasing levels of total REE with increasing SiO 2 . The most silicic gneisses develop large positive europium anomalies. These gross chemical similarities between gneisses from intermediate (amphibolite-facies) and lower (granulite-facies) crustal levels constrain models for the evolution of the Archaean crust. The depletion of K, Rb, Cs, Th and U in granulites, but not other incompatible trace elements cannot be explained by magmatic processes. The positive Eu anomaly in the more siliceous gneisses of both facies is a function of the primary processes of crustal generation and not secondary processes such as intracrustal melting or fractional crystallisation. Fractionation of radioactive heat-producing elements from other trace elements is a result of granulite-facies metamorphism with these elements being removed by an active fluid phase. The apparent lack of partial melting in lower crustal granulites suggests a model for Archaean crustal growth largely through underplating by primary tonalitic magmas.

Dynamic melting in plume heads: the formation of Gorgona komatiites and basalts

Abstract The small Pacific island of Gorgona, off the coast of Colombia, is well known for its spectacular spinifex-textured komatiites. These high-Mg liquids, which have been linked to a late Cretaceous deep mantle plume, are part of a volcanic series with a wide range of trace-element compositions, from moderately enriched basalts ( La/SmN ∼ 1.5) to extremely depleted ultramafic tuffs and picrites ( La/SmN ∼ 0.2). Neither fractional crystallization, nor partial melting of a homogeneous mantle source, can account for this large variation: the source must have been chemically heterogeneous. Low 143Nd/144Nd in the more enriched basalts indicates some initial source heterogeneity but most of the variation in magma compositions is believed to result from dynamic melting during the ascent of a plume. Modelling of major- and trace-element compositions suggests that ultramafic magmas formed at ∼ 60–100 km depth, and that the melt extraction that gave rise to their depleted sources started at still greater depths. The ultra-depleted lavas represent magmas derived directly from the hottest, most depleted parts of the plume; the more abundant moderately depleted basalts are interpreted as the products of pooling of liquids from throughout the melting region.

A new plate tectonic model of the Caribbean; implications from a geochemical reconnaissance of Cuban Mesozoic volcanic rocks

Accreted terranes, comprising a wide variety of Jurassic and Cretaceous igneous and sedimentary rocks, are an important and conspicuous feature of Cuban geology. Although the Mesozoic igneous rocks are generally poorly exposed and badly altered, we have collected and geochemically analyzed 25 samples that place new constraints on plate tectonic models of the Caribbean region. From our recognizance sampling, six main lava types have been identified within the Mesozoic igneous rocks of Cuba: rift basalts, oceanic tholeiites, backarc basin lavas, boninites, island arc tholeiites (IAT), and calc-alkaline lavas. We suggest that the rift-related basalts may have formed during the development of the proto-Caribbean, as the Yucatan block rifted away from northern South America in Jurassic–Early Cretaceous time. The Early Cretaceous oceanic tholeiites have flat rare earth element patterns, and are compositionally similar to Pacific mantle plume–derived oceanic plateaus of similar age. The Early Cretaceous arc-related rocks are either backarc basalts, boninites, or relatively trace element–depleted IAT lavas. A limited amount of geochemical and field evidence hints that two parallel arc systems existed in the western proto-Caribbean area in Early Cretaceous time. This leads us to speculate that in the proto-Caribbean at this time there was a western arc with a northeast-dipping subduction zone erupting IAT lavas (with Farallon plate being consumed), and a more eastern boninitic arc with a southwest-dipping subduction zone (with proto-Caribbean plate being consumed). This latter arc was relatively short lived and after being aborted was mostly eroded away. The Cretaceous primitive (IAT) arc survived and, later in Cretaceous time, as this arc system moved into the widening gap between North and South Americas, calc-alkaline lavas began to be erupted. The evidence suggests that the change from IAT to calc-alkaline lavas was gradual and not abrupt. These new data, although limited, provide geochemical constraints on the tectonic development of the northern part of the Caribbean plate. In consequence, we present a new plate tectonic model for this area of the Caribbean.

The petrogenesis of Gorgona komatiites, picrites and basalts: new field, petrographic and geochemical constraints

Gorgona Island, Colombia is remarkable not only because it contains the only Phanerozoic komatiites, but also because it has mafic to ultramafic lavas with a wide range of compositions, from moderately enriched to extremely depleted (relative to Bulk Earth). The komatiite flows are, in many respects similar to Archaean komatiites; they formed from MgO-rich (18%) liquids and have upper spinifex zones and lower cumulate zones. The cumulate zones of Archaean komatiites contain many solid grains, in contrast more than 90% of the olivine in the Gorgona cumulates is highly skeletal. This combined with the fact that the Gorgona cumulate zones are thinner than those in Archaean komatiites, suggests that the komatiite magma became strongly superheated en route to the surface. The komatiites have trace element contents intermediate between those of the basalts and the ultramafic tuffs. Some basalts have isotope compositions indicative of long-term enrichment in incompatible elements, whereas other basalts and ultramafic volcanics have isotopic signatures that imply corresponding depletion. It is apparent that the plume source region of the Gorgona magmas was markedly heterogeneous, with at least two source components contributing to the observed variation in composition. This heterogeneity may have resulted from the incorporation of different components into the plume source, or it may be the result of complex melting and melt extraction processes during the ascent of a heterogeneous plume. Despite earlier suggestions that there may have been a significant age gap between depleted komatiite and basalt flows and the enriched basal& new 4oAr-39Ar dating of basalts and gabbros are more consistent with all being generated at 87 Ma during formation of the Caribbean/Colombian plateau, possibly at the Galapagos hotspot.

Role of subducted sediment in the genesis of ocean-island basalts: Geochemical evidence from South Atlantic Ocean islands

The South Atlantic Ocean islands of Ascension, Bouvet, St. Helena, Gough, and Tristan da Cunha display considerable inter-island (and to a variable extent intra-island) heterogeneity in ratios of highly incompatible trace elements. Basaltic and hawaiitic lavas from Ascension, Bouvet, and St. Helena have consistent trace-lenient ratios (e.g., La/Nb, Ba/Nb, Ba/La, Ba/Th, Rb/Th). In contrast, Tristan da Cunha and Gough (and Walvis Ridge) lavas are depleted in Nb and enriched in Ba relative to other highly incompatible trace elements as compared to the other islands. The trace-element and Pb isotopic geochemistry of these lavas is explicable by contamination of the ocean-island basalt source that gave rise to Ascension, Bouvet, and St. Helena lavas by variable, but small (about 1%), amounts of ancient (1.5–2.0 Ga) pelagic sediment.

Modification of an oceanic plateau, Aruba, Dutch Caribbean: Implications for the generation of continental crust

The generation of the continental crust may be connected to mantle plume activity. However, the nature of this link, and the processes involved, are not well constrained. An obstacle to understanding relationships between plume-related mafic material and associated silicic rocks is that later tectonic movements are liable to obscure the original relationships, particularly in ancient greenstone belts. Studies of younger analogous regions may help to clarify these relationships. On the island of Aruba in the southern Caribbean, a sequence of partly deformed mafic volcanic rocks intruded by a predominantly tonalitic batholith is exposed. The mafic lavas show geochemical and isotopic affinities with other basaltic, picritic and komatiitic rocks that crop out elsewhere in the Caribbean - these are well documented as belonging to an 88-91 Ma plume-related oceanic plateau, which is allochthonous with respect to the Americas, and is thought to have been formed in the Pacific region. The ~85 to ~82 Ma tonalitic rocks share some geochemical characteristics (high Sr and Ba, low Nb and Y) with Archaean tonalite-trondhjemite-granodiorite (TTG) suites. Field relationships suggest that deformation of the plateau sequence, possibly related to collision with a subduction zone, was synchronous with intrusion of the Aruba batholith. New incremental heating 40Ar/39Ar dates, combined with existing palaeontological evidence, show that cooling of the batholith occurred shortly after eruption of the plateau basalt sequence. Sr-Nd isotopic data for both rock suites are uniform (87Sr/86Sri≈0.7035, eNdi≈+7), whereas Pb isotopes are more variable (Plateau sequence: 206Pb/204Pb=18.6-19.1, 207Pb/204Pb=15.54-15.60, 208Pb/204Pb=38.3-38.75; Aruba batholith: 206Pb/204Pb=18.4-18.9, 207Pb/204Pb=15.51-15.56, 208Pb/204Pb=38.0-38.5). This suggests that there has been a minor sedimentary input into the source region of the batholith. However, the limited time interval between basaltic and tonalitic magmatism makes a normal subduction-related origin for the tonalites improbable. Instead, models involving derivation of tonalite from partial melting of the plateau sequence, or alternatively, genesis in an unusual subduction zone environment, are investigated.