Trace Element Systematics Research Articles

Evidence for magma mixing and a heterogeneous mantle on the West Valley Segment of the Juan de Fuca Ridge

We studied basaltic rocks of the northern end of the Juan de Fuca Ridge to compare the volcanology, geochemistry, and petrogenesis of spatially related rocks of an active spreading center (West Valley and West Ridge), the Sovanco Fracture Zone, the Middle Ridge, and the adjacent Endeavour and Northern Symmetrical ridges. Rocks were analyzed for mineral chemistry and major, trace, and rare earth elements (REE). Crystallization temperatures and equilibrium phase compositions were calculated. The complex zoning and disequilibrium compositions of plagioclase and olivine megacrysts and phenocrysts from all areas except West Ridge, Sovanco Fracture Zone (7123‐7), and Endeavour Ridge indicate mixing in shallow level magma chambers. The rocks range from primitive (up to Mg # = 68) in the West Valley south and West Ridge areas to evolved ferrobasalts (Mg # as low as 52) at West Valley central, West Valley north, and Northern Symmetrical Ridge. The rocks from the West Valley axis span the range of Mg #. Rocks from West Valley are transitional to enriched (e.g., [La/Ce]n= 0.89–1.06), as are rocks from Endeavour Ridge (e.g., [La/Ce]n = 0.88–1.06). Rocks from West Valley wall, Middle Ridge, Sovanco Fracture Zone, and Northern Symmetrical Ridge are light REE depleted (e.g., [La/Ge]n = 0.76–0.88), and the most depleted rocks are from West Ridge (e.g., [La/Ce]n = 0.62). The chemically heterogeneous character of the lavas exists both within site and within a single dredge haul (West Valley north and Endeavour Ridge rocks) and between sites, indicating that the mantle source is heterogeneous on a variety of scales. The range in incompatible element ratios can be modelled as binary mixing between an enriched source and a depleted source. Further, trace element systematics require variations in the degree of partial melting. Lava chemistry and ridge morphology are consistent with southward propagation of spreading in West Valley to form an overlap with the Endeavour Segment. West Valley (West Valley north, central and south) is dominated by extension over volcanism and is thus thermally and magmatically depressed. Rocks from the West Valley Segment span a broader compositional range than ridge systems to the south, reflecting the heterogeneity of the source and the lack of robust volcanism to homogenize the chemistry. Zero age rocks from West Valley are similarly enriched in [La/Ce]n as zero age rocks from the Endeavour and Explorer ridges. Rocks from the West Valley wall, the Middle Ridge, and the older Sovanco Fracture Zone rocks are depleted similar to rocks of the Juan de Fuca Ridge south of the Cobb Offset. This difference suggests that enrichment of the mantle source north of the Cobb Offset has been a recent phenomena.

Genesis of collision volcanism in Eastern Anatolia, Turkey

Late Cenozoic volcanism in Eastern Anatolia extends in a broad SW - NE trending belt across the Arabia - Eurasia collision zone, from the Arabian foreland basin in the southwest and to the Kars Plateau and Lesser Caucasus in the northeast. Foreland volcanism is dominated by basaltic shield and fissure eruptions of transitional tholeiitic - alkaline composition. Volcanism on the thickened crust north of the Bitlis Thrust Zone varies from the mildly alkaline volcano, Nemrut, and older Mus volcanics in the south, through the transitional calc-alkaline/alkaline volcanoes Bingöl and Süphan and the alkaline volcano Tendürek to the calc-alkaline volcano Ararat and older Kars plateau volcanics in the north. Isotope (Sr, Nd) and trace element systematics indicate that the lavas from the foreland were derived from the mantle lithosphere of the Arabian continent which had been enriched by small volumes of asthenospheric melts over a period of time; and that lavas from the alkaline volcanic area around Mus, and the volcanoes Nemrut and Tendürek north of the Bitlis Thrust Zone were derived from a lithospheric source of similar composition, either from the same, underthrust, Arabian continent or from the Bitlis Massif microcontinent. By contrast, the transitional lavas from Bingöl and Süphan and the calc-alkaline lavas from Ararat and Kars were derived from lithosphere carrying a distinct subduction signature inherited from pre-collision subduction events. Positive correlations between 87Sr/ 86Sr and SiO 2 and Rb/Nb and SiO 2 in the alkaline and transitional lavas suggest that combined assimilation and fractional crystallization was an important process within at least part of the thickened crust of the collision zone. Trace element covariation diagrams such as Y-Rb indicate the importance of hornblende crystallization at depth (and orthopyroxene at shallow levels) within the calc-alkaline provinces, in contrast to the consistently anhydrous crystallization sequences of the alkaline lavas. Trace element diagrams, based on the covariation of compatible and incompatible elements, point to moderate - low degrees of partial melting with residual clinopyroxene throughout, and residual garnet in the foreland province. Consideration of mineral stabilities, mantle solidi and geothermal gradients before and after collision suggest that lithospheric thickening should both increase the thickness of metasomatized lithosphere and depress the metasomatized zone to greater depths, probably beneath the amphibole and dolomite breakdown curves. Perturbation of the thickened lithosphere by delamination of the thermal boundary layer, perhaps coupled with local stretching associated with pull-apart basins on strike-slip fault systems, is then sufficient to generate melt, the composition of that melt being largely dependent on the enrichment history of the lithosphere in question. In Eastern Anatolia, volcanism appears to have started between about 8 and 6 Ma ago, some 5 Ma after the start of rapid uplift of the East Anatolian Plateau.

Petrology, geochemistry, and age of the Spurr volcanic complex, eastern Aleutian arc

The Spurr volcanic complex (SVC) is a calc-alkaline, medium-K, sequence of andesites erupted over the last 250000 years by the eastern-most currently active volcanic center in the Aleutian arc. The ancestral Mt. Spurr was built mostly of andesites of uniform composition (58%–60% SiO2), although andesite production was episodically interrupted by the introduction of new batches of more mafic magma. Near the end of the Pleistocene the ancestral Mt. Spurr underwent avalanche caldera formation, resulting in the production of a volcanic debris avalanche with overlying ashflows. Immediately afterward, a large dome (the present Mt. Spurr) formed in the caldera. Both the ash flows and dome are made of acid andesite more silicic (60%–63% SiO2) than any analyzed lavas from the ancestral Mt. Spurr, yet contain olivine and amphibole xenocrysts derived from more mafic magma. The mafic magma (53%–57% SiO2) erupted during and after dome emplacement from a separate vent only 3 km away. Hybrid block-and-ash flows and lavas were also produced. The vents for the silicic and mafic lavas are in the center and in the breach of the 5-by-6-km horseshoe-shaped caldera, respectively, and are less than 4 km apart. Late Holocene eruptive activity is restricted to Crater Peak, and magmas continue to be relatively mafic. SVC lavas are plag ±ol+cpx±opx+mt bearing. All postcaldera units contain small amounts of high-Al2O3, high-alkali amphibole, and proto-Crater Peak and Crater Peak lavas contain abundant pyroxenite and anorthosite clots presumably derived from an immediately preexisting magma chamber. Ranges of mineral chemistries within individual samples are often nearly as large as ranges of mineral chemistries throughout the SVC suite, suggesting that magma mixing is common. Elevated Sr, Pb, and O isotope ratios and trace-element systematics incompatible with fractional crystallization suggest that a significant amount of continental crust from the upper plate has been assimilated by SVC magmas during their evolution.

A combined chemical and Pb-Sr-Nd isotope study of the Azores and Cape Verde hot-spots: the geodynamic implications

Summary The mafic lavas of the Azores and Cape Verde Islands are of a highly varied composition, reflecting a complex history of magma genesis and a variety of source compositions. The lavas of the Cape Verde Islands are characteristically highly silica undersaturated, with alkali-rich ankaramites, larnite-normative melilitites and carbonatites. In contrast, the lavas of the Azores vary from strongly nepheline- to hypersthene-normative types. Isotopic ratios and trace elements also show considerable variation, consistent with derivation from multicomponent mantle sources. Three distinct groupings of lava compositions can be seen in Pb-Sr-Nd isotope space, and each is characterized by its own trace-element enrichment. (i) The majority of the lavas from the Azores and northern Cape Verdes Islands have isotope and trace-element systematics similar to basalts recovered from elevated segments of the mid-Atlantic Ridge ( eg during Deep Sea Drilling Project Leg 82) (Ba/La <12, 206 Pb/ 204 Pb=19.3–20.0). (ii) The islands of Sao Miguel and Faial in the Azores are characterized by radiogenic strontium and lead isotope ratios and marked fractionation of the most incompatible trace elements ( eg 87 Sr/ 86 Sr=0.70522, 206 Pb/ 204 Pb=19.88, 207 Pb/ 204 Pb=15.75, Ba/Nb=12, La/Nb=0.84). (iii) In contrast, the southern Cape Verde Islands have relatively unradiogenic Pb-Sr-Nd isotope ratios ( eg 87 Sr/ 86 Sr=0.70393, 206 Pb/ 204 Pb=18.74, 207 Pb/ 204 Pb=15.53, Ba/La=22, Ba/Nb=15, La/Nb=0.67). Mixing relationships between isotope and trace-element ratios indicate the involvement of at least four, and possibly five, chemically distinct source regions in the petrogenesis of mafic lavas. It is shown that these different source regions represent combinations of the depleted upper mantle (mid-ocean ridge basalt source), recycled oceanic lithosphere and two components of the subcontinental lithospheric mantle. No direct unequivocal evidence is found for a fifth component, namely undepleted primitive mantle reservoir, although helium isotope data suggest influxes of material and heat from the lower mantle. Similarly, it is suggested that continental crust has no direct contribution to the petrogenesis of these ocean island basalts and its role in the mantle is limited to the production/modification of the subcontinental lithosphere above subduction zones.

Crustal sources involved in continental arc magmatism: A case study of volcan Mocho-Choshuenco, southern Chile

Volcan Mocho-Choshuenco, a Pleistocene-Holocene basaltic andesite to dacite shield volcano located at lat 40°S in the Southern Volcanic Zone (SVZ) of the Andes, is built on the thinnest SVZ continental crust (≈30 km), which consists largely of Paleozoic sedimentary and metamorphic rocks and Mesozoic plutons. Sr, Pb, Nd, and O isotopic ratios of Mocho-Choshuenco lavas ( 87 Sr/ 86 Sr = 0.70399 to 0.70422; 206 Pb/ 204 Pb =18.57 to 18.61; 207 Pb/ 204 Pb = 15.60 to 15.63; 208 Pb/ 204 Pb = 38.45 to 38.57; 143 Nd/ 144 Nd = 0.512787 to 0.512881; δ 18 O = +5.8‰ to +6.8‰ Standard Mean Ocean Water) are similar to those of other SVZ centers and exhibit no correlation with indices of differentiation. However, incompatible trace-element systematics indicate that evolved lavas have been contaminated by crustal plutonic rocks by 5% to 15%. Rb/Ba, K/Ba, Rb/La, and K/La ratios increase with increasing SiO 2 content as a result of assimilation of orogenic calc-alkalic plutonic rocks enriched in Rb and K relative to Ba and La. Xenoliths of plutonic rocks recovered from Mocho-Choshuenco lavas are slightly more radiogenic in 87 Sr/ 86 Sr, less radiogenic in 143 Nd/ 144 Nd, and comparable in Pb isotopic ratios, and they have variable 18 O contents ( 87 Sr/ 86 Sr = 0.70482 to 0.70557; 206 Pb/ 204 Pb = 18.60 to 18.64; 207 Pb/ 204 Pb = 15.60 to 15.62; 208 Pb/ 204 Pb = 38.53 to 38.56; 143 Nd/ 144 Nd = 0.512686; δ 18 O = +3.3‰ to +5.5‰ SMOW). Contamination of Mocho-Choshuenco parental magmas by crustal melts with small isotopic contrast imposed little variation on isotopic heterogeneities existing in the parental compositions. These heterogeneities probably were produced during magma-crust interaction at the crust-mantle boundary or in the lower crust.

A model for the simulation of combined major and trace element liquid lines of descent

Variations of major and trace element partition coefficients as functions of temperature and composition have severely limited the range of composition for quantitative modeling of the differentiation of natural magmatic systems. Application of a two-lattice melt model greatly reduces compositional dependence for trace elements. This allows the temperature dependence of isobaric trace element partitioning to be evaluated independently, thus making it possible to model an expanded range of conditions and compositions. The first goal of this investigation was to construct a model to simulate major and trace element systematics of natural differentiation processes, by combining trace element partitioning with a major element phase equilibrium model ( Nielsen, 1985b). The second goal was the application of this model to the simulation of magma chamber recharge, assimilation, eruption and fractional crystallization. These simulations indicate that the proportions of minerals fractionating from an open system are not necessarily the same as those in a closed system. Recharge or assimilation will drive a magma composition off the cotectic along which it is evolving. However, fractionation will drive the liquid back towards the cotectic. In a recharged magma chamber, as the recharge flux increases, the fractionating mineral proportions will be increasingly biased towards those phases which appear first from the primitive magma. The bulk partition coefficient for an open system can be considerably different (5–10× or more), for some elements, compared to a closed system. Paired assimilation and fractional crystallization are often cited as the mechanisms responsible for the origin of high Al basalts and andesites. However, the addition of an Al-rich assimilant into a fractionating basaltic magma will increase the proportion of plagioclase to the extent that Al in the derived liquids will be effectively buffered, or even decrease. The assimilation of forsteritic olivine by reaction ( Keleman and Ghiorso, 1986), increases the proportions of pyroxene and olivine, thus reducing the Fe-enrichment and Al-depletion characteristic of low pressure basalt fractionation.

Comparative Re [sbnd]Os, Sm [sbnd]Nd and Rb [sbnd]Sr isotope and trace element systematics for Archean komatiite flows from Munro Township, Abitibi Belt, Ontario

A series of komatiite flows from Munro Township, Ontario, have Re and Os concentrations that range from 0.545 to 1.27 ppb and 1.00 to 2.20 ppb, respectively. The 187Re/ 186Os in the suite range from 14.05 to 52.37 and the 187Os/ 186Os range from 1.49 to 3.22, giving a crystallization age of2726 ± 93Ma and an of initial 187Os/ 186Os of0.873 ± 0.035. This age is in good agreement with the crystallization ages that have been reported for these rocks using other isotopic systems. The Re Os isochron indicates that the system was resistant to the low grade metamorphism that altered the rocks. The initial Os isotopic composition is consistent with the derivation of the melts from a mantle source(s) that originated with a chrondritic ( 187Os/ 186Os) i and evolved with an approximately chondriticRe/Os. Sm Nd data for the same suite cluster about anε Nd value at 2.72 Ga of +2 to +3. Rb Sr isotope data indicate significant mobilization of these elements by the greenschist-grade metamorphism that has altered the rocks. Comparisons of Re and Os concentrations with those of MgO, FeO, Ni, Cr and S suggests that Re was incompatible ( D < 0.01) with the mantle residues and also with the predominantly olivine assemblages that crystallized from the komatiite melts. A lack of correlation between Os and MgO indicates that Os concentrations were not controlled by the presence of olivine in the mantle residue or olivine crystallization within the flows. Os was, however, moderately compatible with the mantle residues ( D > 1). Os concentrations were likely controlled by the presence or precipitation of trace metallic or sulfide phases both during melting and crystallization, and not by the modally predominant silicate minerals.

Crustal Influences in the Petrogenesis of the Naivasha Basalt--Comendite Complex: Combined Trace Element and Sr-Nd-Pb Isotope Constraints

The bimodal Naivasha complex (central Kenya) comprises 2 suites of transitional basalts and 7 chemostratigraphic groups of comendites. The early basalt series (EBS) predates the Group 1 comendites with the later series (LBS) erupted between Groups 5 and 6. Basalts from both suites are notable for their relatively radiogenic 207Pb/204Pb isotope ratios which are higher than in the majority of ocean island basalt (OIB, Zindler & Hart, 1986), and 87Sr/86Sr ratios more radiogenic than basalts from northern Kenya. Both basalt suites exhibit systematic trace element and isotopic variations which appear related to greater assimilation of Proterozoic amphibolite facies crust by the chemically more evolved rock types. Their mantle source regions show evidence of residual plagioclase and have a 'Dupal'-like OIB trace element and Pb-Sr-Nd isotope signature (Hart, 1984). A contribution from the sub-continental lithosphere is proposed in basalt genesis.The seven comendite groups have distinct trace element and isotope systematics. Hydration of comendite glass causes significant changes in Sr and Pb isotope ratios. In terms of their Sr-Nd isotope relationships the unaltered comendites could be derived from the basalts by an assimilation-fractional crystallization (AFC) process dominated by the fractional crystallization of feldspars. However, the Pb systematics clearly demonstrate that the basalts and comendites are not part of a cogenetic suite. Chemical variations within individual comendite groups are predominantly the result of fractional crystallization of the observed phenocryst assemblages (i.e. alkali feldspar dominated) and minor crustal interaction. The majority of the chemical and isotopic differences between Groups 1-7 cannot be explained by fractional crystallization and appear to represent crustal melts derived from close to the interface between Pan African basement and the overlying Miocene-Holocene volcanoclastic rocks, at approximately 6 km depth (KRISP working group, 1987).Halogens play a fundamental role in the petrogenesis of the comendites (Cl+F<1·7 per cent) permitting small degree melts of low viscosity to be extracted from the crust and causing the breakdown of minor phases e.g, zircon. These factors explain the extreme enrichment of certain incompatible trace elements (Zr<2500, Nb<700) in the comendites and coupled with the retention of zircon in the source of the halogen poor comendites (Group 1<0·6 per cent Cl+F) result in notable fractionation among the HFSE (Zr/Nb 1·5-5·5). Halogens may be concentrated in the source region from the surrounding crust by the presently active hydrothermal system. Each of the chemostratigraphic comendite groups is chemically distinct, implying that partial melting of the heterogeneous crust is on a limited scale and that no extensive magma chambers exist beneath Naivasha.

Geochemistry and origin of some unusually oxidized alkaline rocks from Kauai, Hawaii

We report major-element and trace-element (transition metal, REE, Ba, Rb, Sr, Th) analyses and Sr, Nd and O isotopic ratios from a poorly exposed plug that may represent the remnant of a feeder pipe to the post-erosional Koloa Volcanic Series on Kauai, Hawaii. The plug is zoned with respect to oxidation state and is composed of ijolite with a relatively low oxidation state (near NNO), that apparently forms a central core, and a highly oxidized (>MH) border facies consisting of abundant mantle-derived xenoliths within a highly vesicular composite matrix of alkali picrite and alkali gabbro, both of which have exceptionally high Fe3+/Fe2+ atomic ratios (∼9.0).The isotopic evidence demonstrates that the xenoliths are indeed unrelated to their host lithologies yet allows a genetic relationship between the ijolite, alkali picrite and alkali gabbro. The trace-element systematics of the alkali picrite strongly suggest that it represents a primary melt derived by 5–7% partial melting of a recently metasomatized garnet-lherzolite source very similar in trace-element characteristics to that postulated for the Honolulu Volcanic Series lavas by Clague and Frey (1982). Apart from their disparate oxidation states, the ijolite and alkali gabbro are compositionally identical. Least-squares and fractional crystallization calculations suggest an origin for these lithologies by 50% fractionation of equal parts olivine and clinopyroxene from a parental melt like the alkali picrite.We contend that had the contrasting oxidation states been established at the time of fractionation, the alkali gabbro and ijolite would have evolved into very distinct compositions due to the effect of fO2 on mineral/melt distribution coefficients and the order in which phases crystallize upon cooling. We conclude, therefore, that the mechanism by which the alkali gabbro developed its high Fe2+/Fe2+ ratio operated late in its history, after fractionation, and that the same mechanism probably operated on the similarly oxidized alkali picrite. We believe that the unusually high oxidation states of the alkali gabbro and alkali picrite are most likely the result of atmospheric contamination of the margins of the crystallizing plug. The position of the ijolite in the core of the plug apparently isolated it from this process.

Trace element determination in lunar rocks — quantitative pixe analyses on thick samples

Information about the formation processes and geochemical coherence can be obtained by determination of the partitioning of trace elements in coexisting minerals. Systematic trace element analysis among coexisting mineral phases requires the measurement of the elemental concentrations in a small volume, which can be obtained by applying a proton microprobe. Different opaque minerals, like armalcolite (Fe, Mg) Ti 2O 5, ilmenite FeTiO 3, ulvöspinel Fe 2TiO 4 and chromite FeCr 2O 4 (of the spinel solid solution series Fe 2TiO 4FeCr 2O 4) in samples of three Apollo missions were investigated. Comparative analyses of the mineral pair armalcolite/ilmenite in two Apollo 17 basalts (sample 74242,19), which differ in their crystallization sequences, show that the partitioning of Zr among the two mineral phases is different in the two basalt types. Coexisting ilmenite/ulvöspinel were studied in the Apollo 15 basalts (samples 15065,93; 15555,42). The trace elements Zr and Nb were found to partition in favour of ilmenite. The investigated mineral pairs of chromite/ulvöspinel in Apollo 12 (samples 12051,60; 12051,62) and Apollo 15 basalts are zoned in their major element contents. The trace element Zr detected by proton microprobe analyses also shows zoning behaviour. Primarily, crystallized chromite does not incorporate Zr, whereas the Zr content in the coexisting ulvöspinel rises with increasing Ti content.

Case studies on the origin of basalt

This paper presents a method for the systematic trace element modelling of a cogenetic suite of lavas. It is based on the geochemical inversion technique of Allegre and coworkers and utilizes the variations in the trace element concentrations of the lavas to calculate initial concentrations and source mineralogy. We reduce this inversion to a simple, step-by-step procedure: (1) correcting for fractional crystallization; (2) testing the inferred primary melt compositions for consistency with a model of equilibrium partial melts (with constant partition coefficients) formed from identical sources; (3) estimating the proportions of mineral phases entering the melt; (4) computing concentrations and bulk partition coefficients in the initial source relative to the concentration of a common reference element; (5) estimating relative mineral abundances in the source. Except for the fractionation correction, the calculations are done element by element using a direct analytic solution. For the purpose of comparison we apply this method to the same set of data used by Minster and Allegre (1978), a suite of lavas from Grenada (lesser Antilles) originally analyzed by Shimizu and Arculus (1975). The results of both methods agree well for the source abundances of the light REE, whereas the heavy REE abundances are shown to be poorly constrained by the data. Both methods require residual clinopyroxene and garnet in the source, but the ratio of these minerals is not well constrained. We are unable to reproduce the shape of D0 pattern (=bulk partition coefficients of the initial source) given by Minster and Allegre. The reason for this cannot be evaluated without repeating their calculations in detail. The set of data from Grenada is useful for comparison of the methods only, because it is now known from isotopic data that the samples are not truly cogenetic. Possibly better suited sets of samples for petrogenetic modelling are presented in parts II and III of this series.

Isotopic geochemistry (O, Sr, Pb) of the Golda Zuelva and Mboutou anorogenic complexes, North Cameroun: mantle origin with evidence for crustal contamination

The annular (6–8 km diameter) Golda Zuelva and Mboutou anorogenic complexes of North Cameroun are composed of a suite of alkaline plutonic rocks ranging from olivine gabbro to amphibole and biotite granite. For the Mboutou complex there are two overlapping centres. In the Golda Zuelva complex the plutonic rocks are associated with a later hawaiite to rhyolite volcanic suite. A Rb/Sr whole rock isochron gives an age of 66±3 Ma for the Golda Zuelva granites, with initial 87Sr/ 86Sr ratio of 0.7020, and demonstrates that plutonism and volcanism were essentially contemporaneous and probably cogenetic. For Golda Zuelva and the north Mboutou centre 18O/ 16O (5.6–6.2), 87Sr/ 86Sr (0.7030–0.7045) and Pb isotopic ratios ( 207Pb/ 204Pb: 15.60–15.64) support a mantle origin for the initial magmas. Unlike Sr isotopes, the O isotopic ratios of the granitic end members at Golda Zuelva (∼7.5) indicate crustal contamination. Post-magmatic alteration was not significant. For the younger south Mboutou centre the O-, Sr- and Pb-isotopic data indicate more extensive magma-crust interaction and in a different (higher level?) crustal environment with δ 18O granite=3.3‰, 87Sr/ 86Sr ratios up to 0.706 and Pb isotopic ratios more markedly displaced from the oceanic volcanic field. The low- 18O granites probably record, at least in part, a magmatic process with subsequent minor post-magmatic alteration effects. The major and trace element systematics between the north and south Mboutou centres are directly comparable. The evolution of the magmas were dominated by fractional crystallisation and progressive crustal contamination processes.