Nd and Sr isotopic crustal contamination patterns in an Archaean meta-basic dyke from northern Labrador

Samples from the core to the margin of a 20 cm wide meta-dolerite dyke are sequentially enriched in K, Rb, Sr, and the light REE's. Rb-Sr and Sm-Nd compositional and isotopic profiles in the dyke are interpreted to be the result of selective contamination with components of country rock derivation, rather than the result of simple bulk mixing. ${}^{87}\text{Rb}{}^{86}\text{Sr}$ ratios are higher at the edge of the dyke than at its centre, although they are somewhat irregular, due probably to the effects of subsequent alteration. This profile and one shown by unsupported ⁸⁷Sr are both consistent with contamination of the dyke by a fluid phase derived by the breakdown of biotite. Common Sr shows a parallel, albeit weaker, contamination profile which is interpreted to reflect the contribution of a Sr-bearing phase such as plagioclase. ${}^{147}\text{Sm}{}^{144}$Nd ratios and ¹⁴⁴Nd concentrations increase and decrease respectively from the margin to the core of the dyke. In addition, the margin of the dyke is significantly less radiogenic than the interior. This contrasts with the relatively radiogenic character of an adjacent pegmatite vein. As this sample does not lie on an anticipated contamination profile between the Uivak gneisses and the dyke it is concluded that the REE contamination of the dyke occurred by the addition of a REE-enriched fluid phase which gained access to the dyke by flow along the dyke-pegmatite interface. If it is assumed that both the Rb-Sr and Sm-Nd contamination profiles are the result of diffusion limited processes, then the observations of scale made in this paper suggest that the rate of diffusion of Nd is an order-of-magnitude slower than that for Sr. In view of the scale and nature of these profiles, ages obtained from isotopic data for such mafic dykes must be interpreted with some care. Nevertheless, in spite of these limitations the ?_Nd values for the least contaminated specimens provide a clear indication that the Saglek dykes were derived from a depleted mantle source with ?_Nd? +2.     

Chemical and isotopic homogeneity of a 400 km long basic dyke in central West Greenland

Major and trace element chemistry, and Sr and Pb-isotope data are presented for a 400 km long and ca. 100 m wide basic dyke in the Rinkian mobile belt of West Greenland. The dyke is relatively alkaline in composition (Na₂O+K₂O:4.0–5.5%) but silica saturated (SiO₂:47–49%). Its age is determined as 1,645±35 Ma (Rb-Sr whole-rock isochron, initial ⁸⁷Sr/⁸⁶Sr ratio 0.7028±0.0001) and the Pb-isotope data indicate a source with model ₁= 8.00±0.02. Although the dyke cuts very diverse Archaean and Proterozoic country rocks, it displays a high degree of chemical and isotopic homogeneity along its entire outcrop and has probably not suffered any significant crustal contamination. The dyke shows a distinct chemical zonation, and variations in composition across the dyke at individual localities are often greater than variations between chill samples along the dyke. Chemical homogeneity of the dyke is consistent with two models: (1) rapid lateral emplacement from a localised intrusive centre, or (2) derivation of the magma from a deep, homogeneous mantle source of great extent. Both models pose severe problems, and the evidence does not permit a reliable choice between them.     

Use of Sr isotopes as a tool to decipher the soil weathering processes in a tropical river catchment,southwestern India

River water composition (major ion and ⁸⁷Sr/⁸⁶Sr ratio) was monitored on a monthly basis over a period of three years from a mountainous river (Nethravati River) of southwestern India. The total dissolved solid (TDS) concentration is relatively low (46 mg L⁻¹) with silica being the dominant contributor. The basin is characterised by lower dissolved Sr concentration (avg. 150 nmol L⁻¹), with radiogenic ⁸⁷Sr/⁸⁶Sr isotopic ratios (avg. 0.72041 at outlet). The composition of Sr and ⁸⁷Sr/⁸⁶Sr and their correlation with silicate derived cations in the river basin reveal that their dominant source is from the radiogenic silicate rock minerals. Their composition in the stream is controlled by a combination of physical and chemical weathering occurring in the basin. The molar ratio of SiO₂/Ca and ⁸⁷Sr/⁸⁶Sr isotopic ratio show strong seasonal variation in the river water, i.e., low SiO₂/Ca ratio with radiogenic isotopes during non-monsoon and higher SiO₂/Ca with less radiogenic isotopes during monsoon season. Whereas, the seasonal variation of Rb/Sr ratio in the stream water is not significant suggesting that change in the mineral phase being involved in the weathering reaction could be unlikely for the observed molar SiO₂/Ca and ⁸⁷Sr/⁸⁶Sr isotope variation in river water. Therefore, the shift in the stream water chemical composition could be attributed to contribution of ground water which is in contact with the bedrock (weathering front) during non-monsoon and weathering of secondary soil minerals in the regolith layer during monsoon. The secondary soil mineral weathering leads to limited silicate cation and enhanced silica fluxes in the Nethravati river basin.     

Multiple fluid components of salt diapirs and salt dome cap rocks, Gulf Coast, U.S.A.

Salt diapirs contain a few percent of anhydrite that accumulated as residue to form anhydrite cap rocks during salt dissolutions. Reported ⁸⁷Sr/⁸⁶Sr ratios of these salt-hosted and cap rock anhydrites in the Gulf Coast, U.S.A., indicate their derivation from Middle Jurassic seawater. However, a much wider range of ⁸⁷Sr/⁸⁶Sr ratios, incorporating a highly radiogenic component in addition to the Middle Jurassic component, has been found in several Gulf Coast salt domes. This wide range of ⁸⁷Sr/⁸⁶Sr ratios of anhydrite within the salt stocks records Sr contributions from both marine water and formation water that has equilibrated with siliciclastics. During cap rock formation this anhydrite either recrystallized in the presence of, or was cemented by, a low-Sr fluid with a Late Cretaceous seawter-type Sr isotope ratio or simply lost Sr during recrystallization. Later, the cap rock was invaded by warm saline brines with high Sr isotope ratios from which barite and metal sulfides were precipitated. Subsequently, low-salinity water hydrated part of the anhydrite bringing to six the total number of fluids that interacted througout the history of salt dome and cap rock growth. The progenitor of these salt diapirs, the Louann Formation, is generally thought to have formed from marine water evaporated to halite and, rarely, higher evaporite facies. Salt domes in the East Texas, North Louisiana, and Mississippi Salt Basins have ⁸⁷Sr/⁸⁶Sr and δ³⁴S values that corroborate a Mid-Jurassic age for the mother salt. However, salt domes in the Houston and Rio Grande Embayments of the Gulf Coast Basin have ⁸⁷Sr/⁸⁶Sr ration ranging to values higher than both Middle Jurassic seawater and all Rb-free marine Phanerozoic rocks. These anomalous ⁸⁷Sr/⁸⁶Sr ratios are probably derived from radiogenic Sr-bearing fluids that equilibrated with siliciclastic rocks and invaded the salt either prior to, or during, diapirism. Potential sources of the radiogenic ⁸⁷Sr component include clay and/or feldspar (located either in older units beneath the Louann Formation or younger units flanking the salt diapirs) and K-salts within the Louann evaporites. Because partial Sr exchange in anhydrite had to take place in a fluid medium, admittance of radiogenic ⁸⁷Sr-bearing fluids into the salt may have led to diapirism by lowering the shear strength of the crystalline salt. The slight number of anomalous ⁸⁷Sr/⁸⁶Sr values in the interior basins indicates that anomalous values are related to areally discrete structural or stratigraphic controls that affected only the Gulf Coast Basin.     

Anomalous unradiogenic 87Sr//86Sr ratios in ultrahigh‐pressure crustal carbonates – evidence for fluid infiltration during deep subduction?

Marbles from Changpu (Dabie Shan, eastern China), subducted to 4.4 GPa, have ⁸⁷Sr/⁸⁶Sr values < 0.7040. These low ⁸⁷Sr/⁸⁶Sr values, which would imply a sedimentation age > 2 Ga if considered as primary signature, reflect fluid–rock interaction with a fluid from a low‐⁸⁷Sr/⁸⁶Sr source. The introduction of low‐⁸⁷Sr/⁸⁶Sr was paralleled by introduction of Mg and loss of Si, K and Na in such a way that carbonates from the purest marbles have the least evolved Sr isotopic composition. Introduction of Mg is also indicated by the distribution of calcite and dolomite. Calcite forms inclusions in garnet, whereas dolomite is restricted to the matrix. These chemical changes, inferred from the mineralogy, in combination with textural evidence require a mobile metamorphic fluid. P–T–X constraints for fluid generation and for permeability increase related to mineral reactions and phase transitions suggest that the marbles acquired their anomalous Sr‐isotopic composition during subduction below 60 km. The marbles with the least radiogenic Sr isotopic composition demonstrate that crustal rocks may lose their isotopic fingerprint during deep subduction.     

Strontium and oxygen isotope decoupling in the Hercynian Trois Seigneurs Massif,Pyrenees: evidence for fluid circulation in a brittle regime

Systematic shifts of oxygen isotopic compositions in the higher grade parts of the high temperature-low pressure Hercynian metamorphic sequence, exposed in the Trois Seigneurs Massif, have previously been explained as a result of an influx of surface-derived water during the prograde part of the metamorphic cycle. It has been suggested that this caused a regional lowering of ⁸⁷Sr/⁸⁶Sr in the metamorphic sequence. Mapping of strontium isotopic compositions across a 15 m meta-carbonate horizon in the higher grade pelite-psammite sequence shows that strontium isotopic compositions were homogenised over length scales of metres or less during the Hercynian metamorphism, which brought the carbonate and pelite-psammite to oxygen isotopic equilibrium with a common fluid. Comparison of model pre-Hercynian ⁸⁷Sr/⁸⁶Sr profiles across the carbonate (based on a depositional/diagenetic age of 450 Ma and initial ⁸⁷Sr/⁸⁶Sr ratio of 0.7086 given by 10 m length scale averaging) with the post-Hercynian ⁸⁷Sr/⁸⁶Sr profile (calculated from analysed ⁸⁷Sr/⁸⁶Sr and Rb/Sr compositions) implies strontium isotopic diffusion distances of ca. 0.4 m in the carbonate and ca. 7 m in the pelite-psammite. The limited Sr-isotopic diffusion distance of 0.4–0.7 m within the carbonate is compatible with pervasive oxygen-isotopic exchange over distances restricted to 4–15 m if fluid strontium concentrations were between 4 and 50 ppm. The strontium isotopic transport distances are not compatible with pervasive oxygen isotopic alteration over the observed 5 km regional scale. Either the flow was perfectly layer-parallel or, more probably, the regional-scale alteration of oxygen took place by fluid circulation in the brittle regime early in, or prior to, the Hercynian metamorphic event. Flow along cracks with incomplete diffusive exchange between fluid and wall rock would allow greater decoupling of oxygen and strontium isotopic transport than pervasive advective transport with local fluid-solid equilibrium.     

Transport of Pb and Sr in leaky aquifers of the Bufa del Diente contact metamorphic aureole, north-east Mexico

The Sr and Pb isotopes from the 31.6 ± 0.3 Ma (2σ) old Diente del Bufa alkali syenite, northeastern Mexico, and marbles of its contact aureole were used to trace the sources and the mobility of these metals during hydrothermal activity. Chert layers form aquifers within the marbles. The marbles represent aquitards. During fluid-wallrock reaction, the chert layers developed wollastonite rims. Early wollastonite rims have Sr and Pb isotopic compositions similar to those of their immediate host marbles, which indicates that the isotopic composition of Sr and Pb is initially buffered by the marble. Later wollastonite and other replacement minerals rimming the aquifer have Sr and Pb isotopic compositions that carry with time increasingly larger contributions from the high-salinity magmatic brine. The Sr and Pb contributions from the alkali syenite can be traced isotopically for more than 90 m away from the contact of the intrusion. In contrast, Sr and Pb originating from the alkali syenite are traceable within the marbles only for 3 to 5 cm from the aquifer-marble boundary. This distance is comparable to the spatial distribution of isotopic alterations of C and O implying that Sr and Pb were transported into the marbles through a fluid phase. The isotopic variation of Sr, Pb, C, and O across the aquifer-marble profiles reflects infiltration as a transport mechanism rather than diffusion. Because Sr and Pb are minor components in both the infiltrating fluid and the rock and because their concentrations are strongly affected by the distribution coefficients among the solid phases present, there is little correlation between the isotopic compositions of the trace elements Sr and Pb and those of C and O, which are major components in fluid and rock. Very thin meta-argillite rinds at the outer margin of the aquifer represent residual material after the dissolution of calcite. They are distinctly enriched in Rb, Sr, and U. The Rb and Sr are to some extent residual from the original limestone mineralogy, whereas U is dominantly derived from the magmatic fluid and leaked from the aquifer with the escaping immiscible CO₂-rich H₂O-CO₂ fluid that was produced by decarbonation. The ²³⁸U/²⁰⁴Pb values ranging from 100 to 250 and distinctly lowered Th/U in the meta-argillite rims (1) demonstrate that U was transported with the magmatic fluid along the aquifer and (2) imply that during unmixing of the highly saline magmatic fluid U fractionated into the CO₂-rich H₂O-CO₂ fluid from which it precipitated selectively in the meta-argillite band across the aquifer. Radioautographs demonstrate that the upper meta-argillite rim has 20 to 40 times more U than the lower rim, which implies that 20 to 40 times more CO₂-rich H₂O-CO₂ fluid has left through the upper aquifer contact. Received: 30 September 1997 / Accepted: 15 December 1997     

湘中锡矿山锑矿床的Sr同位素地球化学

对湘中锡矿山锑矿床围岩灰岩、硅化灰岩、煌斑岩和脉石矿物进行了系统的Sr同位素研究。结果表明，矿区围岩发生了隐性蚀变，灰岩中Sr亏损，而^87Sr/^86Sr高于同时代的海相碳酸盐，这种隐性蚀变很可能是水／岩反应所致。矿体附近的硅化灰岩中Sr更加亏损，而^87Sr/^86Sr明显增加。成矿期方解石的^87Sr/^86Sr较高，成矿体系中变化的W／R比造成了方解石中^87Sr/^86Sr值的明显波动。成矿流体为一富放射成因^87Sr的溶液。成矿流体来自或流经基底地层，流体中的Sr由基底碎屑岩提供，矿质Sb也可能主要来自富Sb的元古宇基底。水／岩反应的理论模拟显示，锡矿山成矿流体中的Sr约为3．0μg/g，^87Sr/^86Sr为０．７１７；蚀变－成矿体系为一开放体系，矿石的沉淀机制主要为水／岩反应，成矿体系中W／R 比较高。     

REE and Sr isotope characteristics of carbonate within the Cu–Co mineralized sedimentary sequence of the Nchanga Mine,Zambian Copperbelt

Carbonate, largely in the form of dolomite, is found throughout the host rocks and ores of the Nchanga mine of the Zambian Copperbelt. Dolomite samples from the hanging wall of the mineralization show low concentrations of rare-earth elements (REE) and roof-shaped, upward convex, shale-normalized REE patterns, with positive Eu*_SN anomalies (1.54 and 1.39) and marginally negative Ce anomalies (Ce*_SN 0.98,0.93). In contrast, dolomite samples associated with copper and cobalt mineralization show a significant rotation of the REE profile, with HREE enrichment, and La/Lu_SN ratios <1 (0.06–0.42). These samples also tend to show variable but predominantly negative Eu*_SN and positive cerium anomalies and an upwardly concave MREE distribution (Gd-Er). Malachite samples from the Lower Orebody show roof-tile-normalized REE patterns with negative europium anomalies (Eu*_SN 0.65–0.80) and negative cerium anomalies (Ce*_SN 0.86–0.9). The carbonate ⁸⁷Sr/⁸⁶Sr signature correlates with the associated REE values. The uppermost dolomite samples show Neoproterozoic seawater-like ⁸⁷Sr/⁸⁶Sr ratios ranging from 0.7111 to 0.7116, whereas carbonate from Cu–Co mineralized samples show relatively low concentrations of strontium and more radiogenic ⁸⁷Sr/⁸⁶Sr, ranging between 0.7136–0.7469. The malachite samples show low concentrations of strontium, but give a highly radiogenic ⁸⁷Sr/⁸⁶Sr of 0.7735, the most radiogenic ⁸⁷Sr/⁸⁶Sr ratio. These new data suggest that the origin and timing of carbonate precipitation at Nchanga is reflected in the REE and Sr isotope chemistry. The upper dolomite samples show a modified, but essentially seawater-like signature, whereas the rotation of the REE profile, the MREE enrichment, the development of a negative Eu*_SN anomaly and more radiogenic ⁸⁷Sr/⁸⁶Sr suggests the dolomite in the Cu–Co mineralized samples precipitated from basinal brines which had undergone significant fluid–rock interaction. Petrographic, REE, and ⁸⁷Sr/⁸⁶Sr data for malachite are consistent with the original sulfide Lower Orebody being subject to a later oxidizing event.     

Strontium isotope composition of runoff from a glaciated carbonate terrain

The relationship between subglacial chemical weathering processes and the Sr isotope composition of runoff from Robertson Glacier, Alberta, Canada, is investigated. This glacier rests on predominantly carbonate bedrock of Upper Devonian age, but silicate minerals are also present. The provenance of solute in meltwaters is found to vary systematically with solute concentration and, by inference, subglacial water residence time. In dilute waters, the principal process of solute acquisition is calcite dissolution fueled by protons derived from the dissolution of CO₂ and subsequent dissociation of carbonic acid. At higher solute concentrations, dolomite dissolution coupled to sulfide oxidation is more important. Sr concentration is found to increase with total solute concentration in two separate meltwater streams draining from the glacier, but ⁸⁷Sr/⁸⁶Sr only increases in the eastern melt stream. Carbonate and K-feldspar sources are shown to dominate the Sr content of the western stream, irrespective of concentration. They also dominate the Sr content of the eastern stream at low and intermediate concentrations, but at higher concentrations, muscovite (with high ⁸⁷Sr/⁸⁶Sr) is also an important Sr source. This reflects the outcrop of muscovite-bearing lithologies in the catchment of the eastern stream and an increase in the rate of weathering of K-silicates relative to that of carbonates as more concentrated solutions approach saturation with respect to carbonates. Nonstoichiometric release of ⁸⁷Sr/⁸⁶Sr and preferential release of Sr over K from freshly ground K-silicate surfaces may also occur. This may help to explain the radiogenic nature of runoff from distributed subglacial drainage systems, which are characterized by long water:rock contact times and water flow through environments in which crushing and grinding of bedrock are active processes.Although the exchangeable Sr in tills has higher ⁸⁷Sr/⁸⁶Sr than local carbonate bedrock, only the more concentrated meltwaters from the eastern stream display similarly high values. The most dilute waters, which probably transport the bulk of the dissolved Sr flux from the glacier, have ⁸⁷Sr/⁸⁶Sr characteristic of local carbonate bedrock. Thus, the results suggest that although enhanced weathering of silicate minerals containing radiogenic Sr (such as muscovite) does occur in glaciated carbonate terrains, it is unlikely to contribute to any enhanced flux of radiogenic Sr from glaciated continental surfaces to the oceans.     

Effects of fluids on the interaction of granites with limestones: The Notch Peak stock,Utah

In this paper we consider the mechanisms by which the mineralogy and composition of the margins of the Notch Peak granitic stock, Utah, were affected by calcareous argillite country rocks. The contact zone of the granite relative to the unaffected granite is enriched in: K₂O from about 4 to 10 wt.%, Rb from 250 to 510 ppm, Sr from 150 to 790 ppm and Ba from 250 to 2160 ppm. Locally, some of the intrusive rocks at the contact are nearly devoid of quartz and can be classified as syenites. The initial ⁸⁷Sr/⁸⁶Sr ratios range from 0.7069 in the unaffected rocks to 0.7100 in the potassium-enriched samples, approaching the values of the calc-silicate country rocks of about 0.7110.Calculations show that the characteristics of the contact zone near the top of the stock are the result of a number of sequential processes. CO₂-rich fluids derived from the heated calcareous argillites apparently caused a shift in the phase boundaries in the magma, enhancing accumulation of K-feldspar and high-Ca augite at the expense of other phases. The accumulation resulted in the high Ba and Sr concentrations in some samples. However, the high K₂O and Rb concentrations and magmatic ¹⁸O values indicate infiltration of magmatic fluid emanating from the solidifying lower portions of the pluton subsequent to solidification of the cap. The minimum fluid-rock ratios of 4.6 by mass, calculated on the basis of K₂O and Rb concentration shifts, indicate that a substantial amount of the fluid was channeled through this contact zone. The desilication of the rocks forming the syenitic samples at the contact apparently occurred when a chemical potential gradient of silica between the granite and wall-rocks was established as quartz was consumed in the wall-rocks during calc-silicate reactions. The infiltrating magmatic fluid probably acted as a medium for transport of silica across the contact and perhaps exchange of Sr between the country rocks and the intrusion where up to 30% of strontium in the granitic and syenitic samples from the contact zone was derived from the calc-silicates. The syenitic rocks cannot be explained by desilication reactions involving assimilation of the calc-silicates by the granite magma. The results of this study show that fluids interacting with the country rocks need to be considered to explain the effects of country rocks on the composition of the margins of granitic intrusions.     

Strontium isotopic evidence on the chemical evolution of pore waters in the Milk River Aquifer,Alberta, Canada

Strontium isotope ratios were measured on 13 rock, 18 leachate and 28 pore-water samples from the Milk River aquifer, the confining argillaceous formations, and the glacial till mantling the recharge area. Strontium isotope ratios (⁸⁷Sr/⁸⁶Sr) of pore waters from the aquifer, confining units, and the glacial till ranged from 0.7069 to 0.7082. The ⁸⁷Sr/⁸⁶Sr ratios in aquifer pore waters decrease with increasing distance from the aquifer recharge area, and this is interpreted to be the result of mixing and water–rock interaction within the aquifer.The solute composition of the recharging groundwater is modified by the local lithology, causing distinct geochemical patterns along different flow paths within the aquifer. Whole-rock ⁸⁷Sr/⁸⁶Sr ratios indicate that the shales and till are generally more radiogenic than the aquifer sandstone. The authigenic carbonate cements and rock-forming minerals comprising the major lithologic units had little apparent influence on the pore-water Sr chemistry. Carbonate cement leachates from the till and the aquifer sandstone are more radiogenic than those from the confining shale formations. Feldspar separates from the aquifer sandstone have relatively radiogenic Sr isotope ratios, whereas bentonites from the Milk River and Colorado Shale Formations have whole-rock and leachate Sr isotope ratios that are relatively unradiogenic. Ratios of most Milk River aquifer pore waters are lower than those of any leachates or whole rocks analyzed, except the bentonites.The ⁸⁷Sr/⁸⁶Sr ratios of exchangeable Sr in the bentonites are similar to ratios found in the more evolved pore waters. Simple rock–water interaction models calculated for the whole-rock, leachate, and exchangeable-ion/pore-water pairs indicate that ion exchange with bentonite clays within the Milk River and Colorado Shale Formations appears to influence the isotopic evolution of the pore-water Sr in each of these units.     

A strontium isotopic study of formation waters from the Illinois basin, U.S.A.

The isotopic composition of Sr has been measured in 73 formation-water samples from Paleozoic strata in the Illinois basin; ⁸⁷Sr/⁸⁶Sr ratios range from 0.7079 to 0.7108. With the exception of four samples, the waters are more radiogenic than corresponding Paleozoic sea-water values. The relatively narrow range of slightly elevated ⁸⁷Sr/⁸⁶Sr rations is uniformly distributed in waters throughout the stratigraphic column and in Silurian waters across the basin. Isotopic analyses of core samples from reservoir rocks show an absence of water-rock Sr isotopic equilibration. Basin lithology and analyses of detrital rock units indicate that clay minerals in shales and in quartz sandstone matrices represent the only significant source of radiogenic Sr for the waters. Silurian and Devonian water show a two-component mixing relation which suggests that they comprise a single hydrogeological system that evolved when radiogenic water from New Albany shales entered Silurian-Devonian carbonate rocks and mixed with marine interstitial water. Regional migration of the waters and associated petroleum within the Silurian-Devonian strata, proposed in other studies, is consistent with the Sr isotopic data. Under favorable circumstances subsurface waters are capable of retaining a Sr isotopic recor of their evolution.     

Compositional gradients in large reservoirs of silicic magma as evidenced by ignimbrites versus Taylor Creek Rhyolite lava domes

The Taylor Creek Rhyolite of southwest New Mexico consists of 20 lava domes and flows that were emplaced during a period of a few thousand years or less in late Oligocene time. Including genetically associated pyroclastic deposits, which are about as voluminous as the lava domes and flows, the Taylor Creek Rhyolite represents roughly 100 km³ of magma erupted from vents distributed throughout an area of several hundred square kilometers. Major-element composition is metaluminous to weakly peraluminous high-silica rhyolite and is nearly constant throughout the lava field. The magma reservoir for the Taylor Creek Rhyolite was vertically zoned in trace elements, ⁸⁷Sr/⁸⁶Sr, and phenocryst abundance and size. Mean trace-element concentrations, ranges in concentrations, and element-pair correlations are similar to many subalkaline silicic ignimbrites. However, the polarity of the zonation was opposite that in reservoirs for ignimbrites, for most constituents. For example, compared to the Bishop Tuff, only ⁸⁷Sr/⁸⁶Sr and Sc increased upward in both reservoirs. Quite likely, a dominant but nonerupted volume of the magma reservoir for the Taylor Creek Rhyolite was zoned like that for the Bishop Tuff, whereas an erupted, few-hundred-meter-thick cap on the magma body was variably contaminated by roof rocks whose contribution to this part of the magma system moderated relatively extreme trace-element concentrations of uncontaminated Taylor Creek Rhyolite but did not change the sense of correlation for most element pairs. The contaminant probably was a Precambrian rock of broadly granitic composition and with very high ⁸⁷Sr/⁸⁶Sr. Although examples apparently are not yet reported in the literature, evidence for a similar thin contaminated cap on reservoirs for large-volume silicic ignimbrites may exist in the bottom few meters of ignimbrites or perhaps only in the pumice fallout that normally immediately precedes ignimbrite emplacement. ⁸⁷Sr/⁸⁶Sr in sanidine phenocrysts of the Taylor Creek Rhyolite is higher than that of their host whole rocks. Covariation of this isotope ratio with sanidine abundance and size indicates positive correlations for all three features with decreasing distance to the roof of the magma reservoir. The sanidine probably is more radiogenic than host whole rock because growing phenocrysts partly incorporated Sr from the first partial melt of roof rocks, which contained the highly radiogenic Sr of Precambrian biotite ± hornblende, whereas diffusion was too slow for sanidine to incorporate much of the Sr from subsequently produced less radiogenic partial melt of roof rocks, before eruption quenched the magma system. Disequilibrium between feldspar phenocrysts and host groundmass is fairly common for ignimbrites, and a process of contamination similar to that for the Taylor Creek Rhyolite may help explain some of these situations.     

Isotopic evidence (Sr/Sr, δLi) for alteration of the oceanic crust at deep-rooted mud volcanoes in the Gulf of Cadiz, NE Atlantic Ocean

The chemical and isotopic composition of pore fluids is presented for five deep-rooted mud volcanoes aligned on a transect across the Gulf of Cadiz continental margin at water depths between 350 and 3860 m. Generally decreasing interstitial Li concentrations and ⁸⁷Sr/⁸⁶Sr ratios with increasing distance from shore are attributed to systematically changing fluid sources across the continental margin. Although highest Li concentrations at the near-shore mud volcanoes coincide with high salinities derived from dissolution of halite and late-stage evaporites, clayey, terrigenous sediments are identified as the ultimate Li source to all pore fluids investigated. Light δ⁷Li values, partly close to those of hydrothermal vent fluids (δ⁷Li: +11.9‰), indicate that Li has been mobilized during high-temperature fluid/sediment or fluid/rock interactions in the deep sub-surface. Intense leaching of terrigenous clay has led to radiogenic ⁸⁷Sr/⁸⁶Sr ratios (∼0.7106) in pore fluids of the near-shore mud volcanoes. In contrast, non-radiogenic ⁸⁷Sr/⁸⁶Sr ratios (∼0.7075) at the distal locations are attributed to admixing of a basement-derived fluid component, carrying an isotopic signature from interaction with the basaltic crust. This inference is substantiated by temperature constraints from Li isotope equilibrium calculations suggesting exchange processes at particularly high temperatures (>200 °C) for the least radiogenic pore fluids of the most distal location.Advective pore fluids in the off-shore reaches of the Gulf of Cadiz are influenced by successive exchange processes with both oceanic crust and terrigenous, fine-grained sediments, resulting in a chemical and isotopic signature similar to that of fluids in near-shore ridge flank hydrothermal systems. This suggests that deep-rooted mud volcanoes in the Gulf of Cadiz represent a fluid pathway intermediate between mid-ocean ridge hydrothermal vent and shallow, marginal cold seep. Due to the thicker sediment coverage and slower fluid advection rates, the overall geochemical signature is shifted towards the sediment-diagenetic signal compared to ridge flank hydrothermal environments.     

Fluid-rock interaction during progressive migration of carbonatitic fluids, derived from small-scale trace element and Sr, Pb isotope distribution in hydrothermal fluorite

Associated with the Cretaceous Okorusu carbonatite complex (Namibia) is a hydrothermal fluorite mineralization hosted in Pan-African country rock marbles, which resulted from fluid-rock reaction between the marbles and orthomagmatic, carbonatitic fluids expelled from the carbonatite. Yellow fluorite I was deposited in veins up to 5 cm away from the wallrock contact, followed by purple and colorless fluorite II, smoky quartz and barite, a Mn-rich crust on early calcite, and pure calcite. This clear-cut sequence of mineral growth allows an investigation into fluid-rock interaction processes between the marble and the migrating carbonatitic fluid, and element fractionation patterns between the fluid and subsequent hydrothermal precipitates.Fluorite I shows a progressive change in color from dark yellow to colorless with purple laminations over time of deposition. Subsequent fluorite I precipitates show an increase in Ca, and a continuous decrease in F, Sr, REE, Y, Th, U and Pb contents. The ratios (Eu/Eu*)_cn, Th/Pb and U/Pb increase whereas Y/Ho, Th/U and (La/Yb)_cn decrease. The Sr-isotopic composition remains constant at ⁸⁷Sr/⁸⁶Sr = 0.70456-0.70459, but with varying, highly radiogenic Pb (²⁰⁶Pb/²⁰⁴Pb = 32-190, ²³⁸U/²⁰⁴Pb = 7-63). Fluorite II has ⁸⁷Sr/⁸⁶Sr = 0.70454-0.70459, ²⁰⁶Pb/²⁰⁴Pb = 18.349, and ²⁰⁷Pb/²⁰⁴Pb = 15.600, and a chemical composition similar to youngest fluorite I. The Mn-rich crust on early calcite accumulated REE, Ba, Pb, Zr, Cs, Th and U, developing into pure calcite with a prominent negative Ce anomaly and successively more radiogenic Sr. The calculated degrees of fluid-rock interaction, f = weight fraction of fluid/(fluid + marble), decrease from fluorite I and most fluorite II (f = 0.5) to calcite (f = 0.2-0.3) and hydrothermal quartz (f ? 0.1). A crush-leach experiment for fluid inclusions in the hydrothermal quartz yielded a Rb-Sr isochron age of 103 ± 12 Ma. Crush-leach analysis for the carbonatitic fluid trapped in the wallrock yielded a trend from the fluid leachate to the host quartz (²⁰⁶Pb/²⁰⁴Pb = 18.224 and 18.602, ²⁰⁷Pb/²⁰⁴Pb = 15.616 and 15.636, respectively) extending from carbonatite towards crustal rocks.Calculated trace element distribution coefficients fluorite/fluid are below unity throughout, and increase from La to Yb. Elements largely excluded from fluorite (Ba, Pb, LREE relative to HREE) were incorporated later into the Mn-rich crust on calcite. The trace element patterns of the hydrothermal minerals are related to changing aCO₂ and aF⁻ in the fluid during continued fluid-marble reaction. A predominance of carbonate over fluoride complexing in the fluid as reactions proceeded controlled the Y/Ho, Th/U and REE patterns in the fluid and the crystallizing phases. Deviations from these trends indicate discontinuous processes of fluid-rock reaction.     

Trace elements (REE) and isotopes (O,C, Sr) to characterize the metasomatic fluid sources: evidence from the skarn deposit (Fe,W, Cu) of Traversella (Ivrea,Italy)

The skarn complex of Traversella was formed at the expense of various rock types (calcic hornfels, gneiss, dolomitic marble) occurring in the contact aureole of the dioritic intrusion of Traversella (30±5 Ma). Application of phase equilibria has fixed the temperature of the primary stage of skarn formation between 550° C to 625° C. Similar applications indicate a larger range of temperature (525° C to 300° C) for the secondary stage. The different types of skarn (primary stage) are enriched in REE relative to the corresponding precursor rock (T.R.=126 ppm (protolith) to 228 ppm (inner zone) for the skarn on gneisses; T.R.=14 ppm to 71 ppm for the skarn on calcic hornfelses; T.R.=12 ppm to 200 ppm for the skarn on dolomitic marbles), but all the inner zones of these different types of skarn show a similar REE distribution with a slight LREE fractionation and no Eu anomaly. It is inferred that the primary metasomatic fluid has a parallel REE pattern. The oxygen isotope composition of water in equilibrium with the early stage of skarn at T=600° C ranges from 8.3 per mil to 8.9 per mil. At the beginning of the first hydroxylation stage (secondary stage), the fluid σ ¹⁸O remains in the range observed in the primary stage but within it, there is a sharp decrease from 8.0 per mil to 5.0 per mil. During the sulphidation stage, the fluid σ ¹⁸O decreases more gradually from 5.0 per mil to 3.0 per mil. The I _Sr of the early skarn silicates ranges from the values observed in the dolomitic marbles (0.70874 to 0.70971) to the I _Sr of the intrusion (0.70947 to 0.71064). During the secondary stage, there is a progressive increase of the minerals I _Sr up to 0.71372. The REE pattern of the primary metasomatic fluid does not put any precise constraint on the primary fluid source. On the other hand, both stable and radiogenic isotopes suggest that the early high-temperature metasomatic fluid was isotopically equilibrated with the dioritic intrusion. This implies that this early fluid is either exsolved from the crystallizing intrusion or a metamorphic water previously equilibrated with the intrusion. During the secondary stage, the replacement of the early anhydrous phases by hydrated parageneses is accompanied by the mixing with meteoric fluid as indicated by stable (σ ¹⁸O) and radiogenic (⁸⁷Sr/⁸⁶Sr) isotopes.     

Patterns of Nd and Sr isotopic ratios produced by magmatic and post-magmatic processes in the Shiant Isles Main Sill, Scotland

The Shiant Isles Main Sill of the British Tertiary Igneous Province is a classic example of a differentiated, alkaline basic sill. Four separate intrusions, each emplaced internally in rapid succession, form a 165-m-thick sill hosted by Lower Jurassic sedimentary rocks. Extensive Nd and Sr isotopic studies were conducted on samples from a vertical section through the sill where the relationships of samples to one another are well defined. The results illuminate patterns of modification of isotopic ratios and clarify the petrogenesis (magma sources, crustal contamination), magmatic processes (bulk mixing, interstitial liquid mixing), and post-magmatic alteration (hydrothermal effects on Sr and Nd). Overall, the whole-rock initial ⁸⁷Sr/⁸⁶Sr ratios range from ∼0.7037 to 0.7061 while initial ¹⁴³Nd/¹⁴⁴Nd ratios vary from ∼0.51243 to 0.51286 (ɛ_Nd∼−0.7 to +5.7) – values that contrast markedly with those of the country rock. Acid leaching (HCl) of the whole-rock samples that removes analcime indicates that most of the scatter in the ⁸⁷Sr/⁸⁶Sr is caused by the ubiquitous sub-solidus, aqueous alteration during which more-radiogenic Sr was introduced into the sill, especially along the margins, and also reveals magmatic isotopic ratios. In contrast, Nd was immobile during fluid interaction so that the sill ¹⁴³Nd/¹⁴⁴Nd ratios were not affected, even <1 m from the country-rock contact. Using leached rock values, ⁸⁷Sr/⁸⁶Sr and ¹⁴³Nd/¹⁴⁴Nd ratios are inversely correlated from magmatic processes. Magmas with two distinct isotopic compositions were involved: a more primitive one with ¹⁴³Nd/¹⁴⁴Nd ∼0.51285 and ⁸⁷Sr/⁸⁶Sr ∼0.7035 that produced the first two intrusions and a more evolved one (with 0.51252 and 0.7048) that produced the third intrusion. Mixing of the two magmas was very limited, restricted to near contacts between units, and apparently occurred by interstitial melt migration. The more evolved crinanitic magma was probably produced from a batch of the more primitive picritic melt by a small degree of crustal contamination and crystal fractionation during a short crustal residence prior to ascent and emplacement. Received: 20 December 1999 / Accepted: 5 May 2000     

The strontium isotopic budget of Himalayan rivers in Nepal and Bangladesh

Himalayan rivers have very unusual Sr characteristics and their budget cannot be achieved by simple mixing between silicate and carbonate even if carbonates are radiogenic. We present Sr, O, and C isotopic data from river and rain water, bedload, and bedrock samples for the western and central Nepal Himalaya and Bangladesh, including the monsoon season. Central Himalayan rivers receive Sr from several sources: carbonate and clastic Tethyan sediments, High Himalayan Crystalline (HHC) gneisses and granitoids with minor marbles, carbonates and metasediments of the Lesser Himalaya (LH), and Miocene-Recent foreland basin sediment from the Siwaliks group and the modern flood plain. In the Tethyan Himalaya rivers have dissolved [Sr] ≈ 6 μmol/l and ⁸⁷Sr/⁸⁶Sr ≈ 0.717, with a large contribution from moderately radiogenic carbonate. Rivers draining HHC gneisses are very dilute with [Sr] ≈ 0.2 μmol/l and ⁸⁷Sr/⁸⁶Sr ≈ 0.74. Lesser Himalayan streams also have low [Sr] ≈ 0.4 μmol/l and are highly radiogenic (⁸⁷Sr/⁸⁶Sr ≥ 0.78). Highly radiogenic carbonates of the LH do not contribute significantly to the Sr budget because they are sparse and have very low [Sr]. In large rivers exiting the Himalaya, Sr systematics can be modeled as a mixture between Tethyan rivers, where slightly radiogenic carbonates (mean ⁸⁷Sr/⁸⁶Sr ≈ 0.715) are the main source of Sr, and Lesser Himalaya waters, where extremely radiogenic silicates (>0.8) are the main source of Sr. HHC waters are less important because of their low [Sr]. Rivers draining the Siwaliks foreland basin sediments have [Sr] ≈ 4 μmol/l and ⁸⁷Sr/⁸⁶Sr ≈ 0.725. Weathering of silicates in the Siwaliks and the flood plain results in a probably significant radiogenic (0.72–0.74) input to the Ganges and Brahmaputra (G-B), but quantification of this flux is limited by uncertainties in the hydrologic budget. The G-B in Bangladesh show strong seasonal variability with low [Sr] and high ⁸⁷Sr/⁸⁶Sr during the monsoon. Sr in the Brahmaputra ranges from 0.9 μmol/l and 0.722 in March to 0.3 μmol/l and 0.741 in August. We estimate the seasonally weighted flux from the G-B to be 6.5 × 10⁸ mol/yr with ⁸⁷Sr/⁸⁶Sr = 0.7295.     

A Rb-Sr geochronologic study of the pegmatites of the Middletown area,Connecticut

The pegmatites of Eastern Connecticut have a mineralogy consistent with a magmatic origin yet occur in a non-igneous environment. Various theories of genesis have been investigated by the Rb-Sr geochronologic method.Rb-Sr measurements on early stage pegmatite minerals indicate an age of 258±1 m.y. with initial Sr⁸⁷/Sr⁸⁶=0.734±0.0096. Previously reported K-Ar and U, Th-Pb ages for pegmatite minerals are 249±8 m.y. and 260±3 m.y. respectively. Rb-Sr whole rock data for the host rocks vary between 285±10 m.y. and 472±15 m.y. in age and between 0.705±002 and 0.7167±0.0016 in initial Sr⁸⁷/Sr⁸⁶. A direct genetic relationship between the pegmatites and their host rocks is thus precluded. In addition, whole rock samples of the Brimfield schist taken at variable distances from the Strickland Quarry pegmatite have remained essentially closed systems with respect to Rb and Sr and thus an in situ origin for this pegmatite is unlikely. Mixing of pegmatite and country rock systems has occurred only locally, and isotopic studies of these mixed rocks yield a date of 231±4 m.y. with initial Sr⁸⁷/Sr⁸⁶=0.7188±0.004, an age not inconsistent with previously reported K-Ar and Rb-Sr mineral dates on host rock minerals (approximately 220 to 240 m.y.).Late stage cleavelandites are anomalously enriched in radiogenic Sr-87, the source of which was most probably other zones within the crystallizing pegmatite. This is indicated by analyses of pegmatite whole rocks which show both enrichment and depletion of radiogenic Sr-87 in local systems. The conclusion is drawn that there was widespread movement of radiogenic Sr-87 within each pegmatite system, but that pegmatite-host rock reactions were minimal.