James B. Gill

Rare earth elements and the island arc tholeiitic series

Abstract It is chemically inappropriate to call many of the rocks in island arcs calc-alkaline and we suggest they be known as the “island arc tholeiitic series”. They differ from calc-alkaline rocks by having a lower silica mode, more iron-enrichment, higher Na2O/K2O ratios, less K and associated trace elements, K/Rb∼ 1000,Th/U∼ 1–2, and chondritic REE patterns with La/Yb= 1–2. When erupting concurrently, rocks with these characteristics occur shallower earthquake foci on the trenchward side of island arcs than do the calc-alkaline. They and not rocks of the traditional calc-alkaline series are the most dominant in many western Pacific and Atlantic Island arcs and represent the earliest stages in arc evolution. Although sharing some tholeiitic features, they differ from normal tholeiitic series by having a higher percentage of intermediate and acid members and too little normative olivine, for example, to have been in equilibrium with peridotite. In such distinctive features as REE, Th, and U contents and La/Yb, Th/U, and K/Rb ratios, they are more like mid-ocean ridge tholeiites than any other terrestrial rocks, but they differ from them in silica mode, alkali content and isotopic composition, and MgO, FeO, Ni, and Cr contents. Some of their spatial, temporal, and chemical characteristics may be explained by mixing of the partial melting products of resorbed lithosphere and overlying upper mantle. We report REE data for 19 samples, chosen to highlight the different patterns found in modern and ancient island arcs.

Isotopic and paleomagnetic constraints on the Mesozoic tectonic evolution of south China

In order to better constrain the paleogeographic evolution of south China we measured Sm-Nd and Rb-Sr isotopic compositions for 23 Mesozoic granites that crop out throughout the area. Tightly grouped neodymium depleted mantle model ages (1.4 ± 0.3 Ga) suggest the region is underlain by relatively homogeneous Proterozoic crust and fail to define crustal provinces. Neither the isotopic nor geologic data suggest that a Mesozoic suture exists. However, granites possessing anomalously high Sm (>8 ppm) and Nd (>45 ppm) concentrations, relatively high initial epsilon neodymium (−4 to −8), and high but variable initial 87Sr/86Sr (0.759 to 0.713) form a northeast trending zone that coincides with two prominent Mesozoic basins. Southeast of the zone lie the majority of Mesozoic intrusives and Upper Triassic to Lower Cretaceous extensional basins found in south China. Mesozoic paleomagnetic poles are well clustered northwest of the zone. Pre-Cretaceous poles southeast of it are discordant with respect to those from the northwest. The only recognized tectonostratigraphic terrane in south China lies southeast of the zone. The terrane is bordered by a northeast trending sinistral fault that was active in the Mesozoic. Other faults in south China have similar attitudes, ages, and sense of shear. Together, the observations suggest that the Mesozoic tectonic regime in south China consisted of strike-slip activity plus concomitant rifting as terranes or fragments of similar crust were transported north along sinistral faults. The zone, defined by the granites enriched in Nd and Sm, demarcates displaced terranes to the southeast from relatively stable land to the northwest.

Geochemistry of Viti Levu, Fiji, and its evolution as an island arc

Available geochemical and field data suggest that three different periods of igneous activity have occurred on Viti Levu. Rocks of the first period are “island arc tholeiites”; of the second, a calc-alkaline series; and of the third, shoshonites. Each period has a geochemical distinctiveness. The temporal sequence also corresponds to a spatial zonation analogous to that observed in many currently active island arcs. K2O and the larger trace elements increase south to north and with time, whereas iron enrichment and Na2O/K2O ratios decrease. Initial Sr 87/86 ratios average about 0.7041 in rocks of all periods. If Viti Levu magmas were genetically related to underthrust oceanic crust, as is argued for comparable ones of modern island arcs, then the island must have drifted or been rotated to its present position within the last 5–10 m.y.

Effects of partial melting on the uranium decay series

Distinction between possible partial melting processes using stable incompatible trace elements depends largely on the mass fraction of liquid produced. At low degrees of melting (less than 5%), both incompatible element concentrations and interelement fractionations are greater for batch, fractional and accumulated fractional melting than for dynamic or continuous melting. At higher degrees of melting, uncertainty in initial source composition and small fractionations between incompatible elements make it unlikely that the most realistic melting models, batch and dynamic, could be discriminated. However, it is possible for these melting processes to be distinguished by the magnitude of radioactive disequilibria that is produced within the 238U decay series. Even for durations of partial melting much longer than the half-lives of 230Th and 226Ra, both230Th-238U and 226Ra-230Th disequilibria can occur because of differences in the bulk solid/liquid partition coefficients for these elements. The magnitude of these disequilibria depends on melting process. The (230Th)(232Th) of a dynamic melt can differ significantly from that of a batch melt, and will not necessarily indicate the original (238U)(232Th) of the source. If MORB is formed by a dynamic melting process, then the Th/U of the upper mantle near ridge axes may be higher than that calculated from the (230Th)(232Th) of the basalts.

Th isotope and U-series studies of subduction-related volcanic rocks

The activities of 6–7 radionuclides measured by alpha spectrometry are reported for 35 samples of recent volcanic rocks from 24 volcanoes in 9 arcs. These include continental margins (Cascades, Alaska, Costa Rica), island arcs (the Sunda, Banda, and Sangihe arcs of Indonesia, Marianas, Japan, Aleutians, Antilles, Tonga), and a backarc basin (Lau Basin). Enrichment of 238U over 230Th is more common (over 25 of the samples) and greater (up to 59%) in these subduction-related volcanics than in those from other tectonic environments. 230Th232Th ratios also extend to higher values, both in absolute numbers and relative to other isotope ratios. Enrichment of 210Po and 226Ra over 230Th is widespread and is more common and greater in island arcs than continental margins. The level of Po or Ra enrichment (50–300% in island arcs, 10–50% at continental margins) is similar to that in ocean island and ridge basalts and decreases with differentiation. The differences in Th-U systematics between subduction-related and other volcanic rocks are attributed to variations in the process of melt extraction, changes in bulk partition coefficients within the mantle wedge, or preferential addition of U from subducted lithosphere. Ra enrichments are attributed to partial melting processes which are similar to those at ocean ridges. Smaller excesses at continental margins are attributed either to slower ascent or to differences in the process of melt formation beneath continents.

Composition and age of Lau Basin and Ridge volcanic rocks: Implications for evolution of an interarc basin and remnant arc

The Tonga-Kermadec Ridge, Lau Basin, and Lau-Colville Ridge are, respectively, a frontal arc, interarc basin, and remnant arc at the Australian-Pacific plate boundary. Basement rocks of the Lau-Colville Ridge (Lau Volcanics) are 9- to 6-m.y.-old basaltic andesites to dacites with 55 to 66 percent SiO 2 , K 60 = 1.0 to 1.5 percent, little Fe enrichment, Sr 87 /Sr 86 = 0.7030 to 0.7034, and enrichment in light rare-earth elements. Westward increases in K, Rb, Th, and U suggest that subducted lithosphere was un-derthrust from the east. Variations in rock composition are consistent qualitatively with derivation from basalt by low-pressure crystal-liquid fractionation involving removal of phenocryst phases: plagioclase + clinopyroxene + orthopyroxene + magnetite. Volcaniclastic turbidites of the same age and derived from western sources are found in Tonga. These andesitic vol-canogenic recks are overlain on both ridges by Pliocene limestones, which are capped on the Lau-Colville Ridge by 3.9- to 3.5-m.y.-old olivine + hypersthene normative tholeiites (Korombasanga Volcanics) having minor 56 to 60 percent SiO 2 andesitic differentiates. Lau Basin basalts are transitional between ocean-floor and island-arc tholeiites, sharing with the latter their higher Rb, Ba, light rare-earth element, and Sr 87 contents and lower Ti, Zr, and Hf contents. These data support Karig9s idea that the Lau and Tonga Ridges represent a once-united island arc now dismembered by rifting, which has formed the intervening Lau Basin. This rifting began about 5 m.y. B.P. The change in volcanism on the Lau-Colville Ridge reflects its removal from a subduction site.

Mantle Melting and Basalt Extraction by Equilibrium Porous Flow

The chemical composition of mid-ocean ridge basalt, the most prevalent magma type on the planet, reflects the melts continuous reequilibration with the surrounding mantle during porous flow. Models of basalt extraction that account for the observed uranium-series disequilibria on the Juan de Fuca ridge constrain both the abundance of melt beneath ridges (0.1 to 0.2 percent) and the style of mantle melting. Unlike models that incorporate near-fractional melts (dynamic melting), mixing of equilibrium porous flow melts derived from heterogeneous source materials quantitatively explains the uranium-series observations.

Volcanism in the Sumisu Rift, I. Major element, volatile, and stable isotope geochemistry

A bimodal volcanic suite with KAr ages of 0.05–1.40 Ma was collected from the Sumisu Rift using alvin. These rocks are contemporaneous with island arc tholeiite lavas of the Izu-Ogasawara arc 20 km to the east, and provide a present day example of volcanism associated with arc rifting and back-arc basin initiation. Major element geochemistry of the basalts is most similar to that of basalts found in other, more mature back-arc basins, which indicates that back-arc basins need not begin their magmatic evolution with lavas bearing strong arc signatures. Volatile concentrations distinguish Sumisu Rift basalts from island arc basalts and MORB. H_2O contents, which are at least four times greater than in MORB, suppress plagioclase crystallization. This suppression results in a more mafic fractionating assemblage, which prevents Al_2O_3 depletion and delays the initiation of Fe_2O_3_((tot)) and TiO_2 enrichment. However, unlike arc basalts,Fe^(3+)/ΣFe ratios are only slightly higher than in MORB and are insufficient to cause magnetite saturation early enough to suppress Fe_2O_3_(tot) and TiO_2 enrichment. Thus, major element trends are more similar to those of MORB than arcs. H_2O, CO_2 and S are undersaturated relative to pure phase solubility curves, indicating exsolution of an H_2O-rich mixed gas phase. High H_2O/S, high δD, and low (MORB-like) δ^(34)S ratios are considered primary and distinctive of the back-arc basin setting.

Compositional controls on the partitioning of U, Th, Ba, Pb, Sr and Zr between clinopyroxene and haplobasaltic melts: implications for uranium series disequilibria in basalts

The partitioning of U, Th, Pb, Sr, Zr and Ba between coexisting chromian diopsides and haplobasaltic liquids at oxygen fugacities between the iron-wustite buffer and air at 1285°C has been characterized using secondary ion mass spectrometry. The partition coefficients for Th, U and Zr show a strong dependence on the Al and Na content of the clinopyroxene. A good correlation between IVAl and DTh exists for all recent Th partitioning studies, providing a simple explanation for the two order of magnitude variation in DTh observed in this and previous studies [1,2]. Because mantle clinopyroxenes generally have greater than 5 wt% Al2O3, we suggest that the relevant partition coefficients for U and Th are between 0.01 and 0.02. While variations in Al and Na in clinopyroxene affect the absolute value of the Th and U partition coefficients, they have no effect on their ratio, DThDU. Our results reinforce the inference that equilibrium partitioning of U and Th between clinopyroxene and melt cannot explain the observed 230Th excesses in basalts. Indeed, under the oxygen fugacities relevant to MORB petrogenesis, clinopyroxene has little ability to fractionate U from Th (DThDU < 2), implying that chemical disequilibrium between melt and wall rock during transport is not required to preserve 230Th excesses generated in the garnet stability field. If the Ba partition coefficient serves as an analog for Ra and the partition coefficient of U5+ serves as an analog for Pa5+, then 226Ra and 231Pa excesses can be generated by clinopyroxene-melt partitioning. Using compositionally dependent partition coefficients, a melting model is used to show that equilibrium porous flow can explain variations in uranium series activities from the East Pacific Rise by varying the depth of melting.

Volcanism in the Sumisu rift. II, Subduction and non-subduction related components

Abstract A bimodal suite of volcanic rocks collected from the Sumisu Rift by alvin provide present day examples of the first magmatic products of arc rifting during the initiation of back-arc spreading. The trace element and isotopic composition of these rocks, which are contemporaneous with island arc tholeiite lavas of the Izu-Ogasawara arc 20 km to the east, differ from those of arc rocks and N-MORB in their relative incorporation of both subduction-related and non-subduction-related components. Subduction-related components, i.e., those that distinguish volcanic arc basalts from N-MORB, are less pronounced in rift lavas than in arc lavas. Alkali and alkaline earth to high field strength element and REE ratios as well as 87 Sr/ 86 Sr are intermediate between those of N-MORB and Izu arc lavas and indicate that Sumisu Rift basalts are similar to BABB erupted in other, more mature back-arc basins. These results show that back-arc basins may begin their magmatic evolution with BABB rather than more arc-like lavas. Evidence of non-subduction related components remains after the effects of subduction related components are removed or accounted for. Compared to the arc, higher HFSE and REE concentrations, contrasting REE patterns, and ⩽e Nd in the rift reflect derivation of rift lavas from more enriched components. Although SR basalt resembles E-MORB in many trace element ratios, it is referred to as BABB because low concentrations of Nb are similar to those in volcanic arcs andH 2 O/REE andH 2 O/K 2 O exceed those of E-MORB. Differences in HREE pattern ande Nd require that the E-MORB characteristics result from source heterogeneities and not lower degrees of melting. Enriched mantle beneath the rift may reflect enriched blobs entrained in a more depleted matrix, or injection of new, more enriched mantle. High 208 Pb/ 204 Pb and moderate 207 Pb/ 204 Pb ratios with respect to Pacific MORB also reflect ancient mantle enrichment. Trends on Pb isotopic diagrams and ⩽e Nd in the rift than in the arc indicate that recent sediment recycling is not an important process in the generation of these back-arc lavas.