Search

To understand the characteristics of Ca isotope composition and fractionation in silicate-dominated Quaternary aquifer system, hydrochemical and isotope studies (87Sr/86Sr, 13CDIC and 44/40Ca) were conducted on groundwater, sediment and rock samples from the Datong basin, China. Along the groundwater flow path from the basin margin to the center, groundwater hydrochemical type evolves from Ca-HCO3 to Na-HCO3/Na-Cl type, which results from aluminosilicate hydrolysis, vertical mixing, cation exchange between CaX2 and NaX, and calcite/dolomite precipitation. These processes cause the decrease in groundwater Ca concentration and the associated modest fractionation of groundwater Ca isotopes along the flowpath. The groundwater δ44/40Ca value varies from −0.11 to 0.49‰. The elevated δ44/40Ca ratios in shallow groundwater are attributed to vertical mixing involving addition of irrigation water, which had the average δ44/40Ca ratio of 0.595‰. Chemical weathering of silicate minerals and carbonate generates depleted δ44/40Ca signatures in groundwater from Heng Mountain (east area) and Huanghua Uplift (west area), respectively. Along the groundwater flow path from Heng Mountain to central area of east area, cation exchange between CaX2 and NaX on clay mineral results in the enrichment of heavier Ca isotope in groundwater. All groundwater samples are oversaturated with respect to calcite and dolomite. The groundwater environment rich in organic matter promotes the precipitation of carbonate minerals via the biodegradation of organic carbon, thereby further promoting the elevation of groundwater δ44/40Ca ratios.

This paper presents an initial evaluation and concept description of an approach to CO2 storage where the reservoir rocks are volcanic terrains that have been built up from the seafloor and consist of several kilometers of stacked lava, pyroclastic, and volcano-sedimentary rocks, and where CO2 could be injected in large quantities in the supercritical or liquid state. These coastal “Saline Volcanic Basins” (SVB) have massive volume, heterogeneous internal structure, and relatively low temperatures. SVB's occur in island arcs and so-called hot spots such as Hawaii and Iceland. In both settings, the volcanic formations are exceedingly thick, border the ocean, are below sea level, and are saturated with seawater at storage depths. Many SVB reservoirs are accessible with onshore wells within which intercalated high and low-permeability layers and low temperatures can keep supercritical CO2 in a relatively high-density state and promote solution and capillary trapping, in addition to mineralization. Some regions of the subsurface may be at low enough temperature to allow for CO2 hydrate formation as an additional trapping mechanism. Our initial assessment of storage potential focuses on the northeast portion of the island of Hawaii, where there is direct information about the subsurface volcanic stratigraphy and hydrology based on observations made during two scientific drilling projects that penetrated to 3.5 km with continuous coring. Initial analysis, including simulations of a 50 Megaton injection of supercritical CO2, suggests that this region could be effective for permanent storage, and potentially for gigatons of CO2.

Elucidating geodynamic processes at depth relies on a correct interpretation of petrological and geochemical features in magmatic records. In southern Tibet, both potassic volcanic rocks and adakitic intrusions exhibit high Sr/Y and La/Yb, and low Y and Yb concentrations. But these two rock types have contrasting temporal–spatial distributions and isotopic variations. Here we present a systematic study on the postcollisional potassic and adakitic rocks in order to investigate their petrogenetic links with the coeval mantle-derived ultrapotassic rocks and shed light on the potential input from underthrusted Indian continental crust. We found that adakitic intrusions with higher K2O/Na2O tend to display lower Y and higher SiO2, suggesting that the mantle-derived ultrapotassic melts, showing relatively high Y and Yb concentrations, only played a minor role in adakitic magmatism. Therefore, the unradiogenic 143Nd/144Nd and the dramatic decrease of zircon eHf(t) values since ~ 35 Ma shown by postcollisional adakites should be interpreted as reflecting the crustal input from Indian plate. Unlike adakitic intrusions in southern Lhasa subterrane, potassic volcanic rocks share similar spatial distributions with ultrapotassic rocks, and their isotopic discrepancy is diminishing with volcanic activity becomes younger and migrates eastward. Evidence from whole-rock Pb and zircon Hf isotopes further indicates that potassic volcanic rocks are more likely to originate from partial melting of the overthickened and isotopically heterogeneous Lhasa terrane crust rather than the underthrusted Indian continental crust. The elevated Rb/Sr and varying Sr/CaO in potassic volcanic rocks provide an argument for sanidine + plagioclase + clinopyroxene as the major fractionating phases during magmatic differentiation. These findings not only highlight the significance of potassic and adakitic rocks in providing constraints on the geodynamic processes beneath southern Tibet, but also imply that special caution is needed if we attempt to probe into the nature of mantle lithosphere using isotopic tracers of the Tibetan ultrapotassic rocks.

High iodine concentrations in groundwater have caused serious health problems to the local residents in the Datong basin, northern China. To determine the impact of water-sediment interaction and irrigation practices on iodine mobilization in aquifers, isotope ( 2 H, 18 O and 87 Sr/ 86 Sr) and hydrogeochemical studies were conducted. The results show that groundwater iodine concentrations vary from 14.4 to 2180 μg/L, and high iodine groundwater (>150 μg/L) mainly occurs in the central area of the Datong basin. Sediment iodine content is between 87 Sr/ 86 Sr values and groundwater chemistry suggest that aluminosilicate hydrolysis is the dominant process controlling hydrochemical evolution along groundwater flowpath, and the degradation of TOC/iodine-rich sediment mediated by microbes potentially triggers the iodine release from the sediment into groundwater in the discharge area. The vertical stratification of groundwater 18 O and 2 H isotope reflects the occurrence of a vertical mixing process driven by periodic surface irrigation. The vertical mixing could change the redox potential of shallow groundwater from sub-reducing to oxidizing condition, thereby affecting the iodine mobilization in shallow groundwater. It is postulated that the extra introduction of organic matter and O 2 /NO 3 /SO 4 could accelerate the microbial activity due to the supplement of high ranking electron acceptors and promote the iodine release from the sediment into shallow groundwater.

International audience; Kinetic calcium isotope effects can be used as growth-rate proxies for volcanic and subvolcanic minerals. Here, we analyze Ca isotopic compositions in experimental and natural samples and confirm that large kinetic effects (>2‰) can occur during magmatic plagioclase crystallization. Experiments confirm theoretical predictions that disequilibrium isotope effects depend mainly on the rates for crystal growth relative to liquid phase Ca diffusivity (R/D). Plagioclase phenocrysts from the 1915 Mount Lassen rhyodacite eruption, the ∼650-y-old Deadman Creek Dome eruption, and several mafic subvolcanic orbicules and plagioclase comb layers from Northern California have disequilibrium Ca isotopic compositions that suggest rapid crystal growth rates (>1 cm/y to 15 cm/y). The Ca isotope results, combined with complementary crystal-size distribution analyses, suggest that magmatic rejuvenation (and eruption) events, as reflected in crystal growth times, can be as short as ∼10−3 y. Although mafic systems are predicted to have shorter magmatic rejuvenation periods, we find similarly short timescales in both mafic and silicic systems. These results are consistent with a growing body of evidence suggesting that dominantly crystalline volcanic magma reservoirs can be rapidly reactivated by the injection of fresh magma prior to eruption. By focusing on a common mineral such as plagioclase, this approach can be applied across all major magmatic compositions, suggesting that Ca isotopes can be used as a tool for investigating the dynamics and timing of volcanic eruptions.

Water-rock interactions in mid-ocean ridge hydrothermal systems are a critical part of Earth system evolution. Extensive insights have been developed from vent fluid chemistry and laboratory experiments, but these leave unanswered many questions about the temporal evolution and spatial structure of the hydrothermal systems that can only be addressed with reactive transport simulations. Other issues are the effects of changing spreading rates and seawater chemistry through Earth history. We are addressing this problem using the Toughreact code, starting with 2D static (no seafloor spreading) simulations of the near-axis region where most of the interaction occurs. The simulations use a dual-permeability grid to represent fractured rocks, and also have a formulation for Sr isotope exchange. Vent fluid Ca, Mg, SO4, and Na concentrations and Sr isotopes can be used as a guide to fluid chemical evolution. Initial simulations reproduce modern vent fluid chemistry even with maximum temperature only at 380°C, and suggest that fluids need not be in equilibrium with the rocks at any point in the system. Model fluids continue to evolve chemically even in the upflow zone prior to venting. The effects of different seawater chemical composition, as proposed for the Cretaceous, for example, can be captured with charge-balance models.

Volcanogenic sandstones are typically rich in volcanic rock fragments that can provide reactive minerals for CO 2 mineralization in a scenario of CO 2 sequestration. To quantitatively evaluate the extent and time scale of CO 2 mineralization in potential reservoir formations, we characterized example high‐porosity volcanic sand from the San Joaquin Valley of southern Central California. Samples were collected from the volcanic‐rich members of the Etchegoin Formation in the San Joaquin Basin near Coalinga, California. Thin sections made from these samples were examined under petrographic microscope to identify mineral compositions, and then scanning electron microscopy (SEM) and energy dispersive x‐ray spectroscopy (EDX) were used to gather quantitative information on mineral abundances, distribution, and reactive surface areas. Porosity and permeability were also measured using core plugs made from outcrop samples. Results show that the Etchegoin volcanic sandstone has a high percentage (10–15%) of reactive minerals (pyroxenes, Fe‐Ti spinel and clays), and high reactive surface areas at about 1 m-super-2/kg. Reactive transport modeling is conducted and shows that these reactive minerals could mineralize 92% of injected and capillary‐trapped CO 2 within 1000 years of injection. Possible effects of heterogeneity on CO 2 injection and mineralization are also studied using the reactive transport code TOUGHREACT. Vertical heterogeneity of mineralogy and hydrology increases both CO 2 injectivity and mineralization. Available volcanogenic sandstones worldwide like the Etchegoin sandstone are summarized from the literature, with formations from the western USA that have CO 2 sequestration potential emphasized.© 2015 Society of Chemical Industry and John Wiley & Sons, Ltd

Variations in the Mg, Ca, Sr, and SO4 concentrations of paleoseawater can affect the chemical exchange between seawater and oceanic basalt in hydrothermal systems at midocean ridges (MOR). We present a model for evaluating the nature and magnitude of these previously unappreciated effects, using available estimates of paleoseawater composition over Phanerozoic time as inputs and 87Sr/86Sr of ophiolite epidosites and epidote-quartz veins as constraints. The results suggest that modern hydrothermal fluids are not typical due to low Ca and Sr relative to Mg and SO4 in modern seawater. At other times during the last 500 million years, particularly during the Cretaceous and Ordovician, hydrothermal fluids had more seawater-derived Sr and Ca, a prediction that is supported by Sr isotope data. The predicted 87Sr/86Sr of vent fluids varies cyclically in concert with ocean chemistry, with some values much higher than the modern value of ∼0.7037. The seawater chemistry effects can be expressed in terms of the transfer efficiency of basaltic Ca and Sr to seawater in hydrothermal systems, which varies by a factor of ∼1.6 over the Phanerozoic, with minima when seawater Mg and SO4 are low. This effect provides a modest negative feedback on seawater composition and 87Sr/86Sr changes. For the mid-Cretaceous, the low 87Sr/86Sr of seawater requires either exceptionally large amounts of low-temperature exchange with oceanic crust or that the weathering flux of continentally derived Sr was especially small. The model also has implications for MOR hydrothermal systems in the Precambrian, when low-seawater SO4 could help explain low seawater 87Sr/86Sr.

We proposed to use volcanogenic sandstones for CO2 sequestration. Such sandstones with a relatively high percentage of volcanic rock fragments (VRF) could be a promising target for CO2 sequestration in that they have a sufficient percentage of reactive minerals to allow substantial mineralization of injected scCO2, which provides the most secure form of CO2 storage, but can also be porous and permeable enough to allow injection at acceptable rates. The limitation in using volcanogenic sandstones as CO2 reservoir rocks is that porosity and permeability tend to decrease with increase of volcanic rock fragments (VRF) and with the length and complexity of the diagenetic history. Decreased porosity limits the rate at which CO2 can be injected. We evaluated these tradeoffs to assess the feasibility of using volcanogenic sandstone to achieve highly secure CO2 storage. Using relationships between VRF percent, porosity and permeability from available geological data, the reactive transport code TOUGHREACT was used to estimate the rate and extent of CO2 mineralization over 1000 years, and the trade-off between higher reactivity and lower porosity and permeability. For the parameter set that we believe is defensible, the optimal VRF percent for the largest amount of mineralized CO2 is around 10-20%. The results show that as much as 80% CO2 mineralization could occur in 1000 years and still allow sufficient injectivity so that ca. 1 megaton of CO2 can be injected per year per well. The key to estimating how much CO2 can be injected and mineralized is the relationship between permeability (or injectivity), reactive mineral content, mineral dissolution rates and reactive surface areas. We have sampled examples of volcanogenic sandstones from the Miocene Etchegoin Formation, central California to examine these parameters. Characterizations of these samples by SEM, XRF and XRD show that they are very rich in reactive minerals with around 32% plagioclase, 10% orthopyroxene, 2% clinopyroxene, and 1% ilmenite. Porosities range from 10% to 20%, and permeabilities range from 10 mD to 1000 mD. Batch experiments are also conducted to measure dissolution rates of one of the common minerals in volcanic sandstones, i.e. chlorite. Results show that the released Mg from chlorite in such sandstones could provide enough cations to potentially mineralize about half the stored CO2 in the pore space within 1000 years. The last chapter presents an extension of the TOUGHREACT code to include carbon 13 isotope into reactive transport modeling. 13 C isotope proves to be a great tracer for monitering subsurface flow of CO2.

It is the purpose of this study to demonstrate that the molecular scale reaction mechanisms describing calcite precipitation and calcium isotope fractionations under highly controlled laboratory conditions also reproduce field scale measurements of δ44Ca in groundwater systems. We present data collected from an aquifer during active carbonate mineral precipitation and develop a reactive transport model capturing the observed chemical and isotopic variations. Carbonate mineral precipitation and associated fluid δ44Ca data were measured in multiple clogged well bores during organic carbon amended biogenic reduction of a uranium contaminated aquifer in western Colorado, USA. Secondary mineral formation induced by carbonate alkalinity generated during the biostimulation process lead to substantial permeability reduction in multiple electron-donor injection wells at the field site. These conditions resulted in removal of aqueous calcium from a background concentration of 6 mM to

Obtaining quantitative information about the timescales associated with sediment transport, storage, and deposition in continental settings is important but challenging. The uranium-series comminution age method potentially provides a universal approach for direct dating of Quaternary detrital sediments, and can also provide estimates of the sediment transport and storage timescales. (The word "comminution" means "to reduce to powder," reflecting the start of the comminution age clock as reduction of lithic parent material below a critical grain size threshold of ~50μm.) To test the comminution age method as a means to date continental sediments, we applied the method to drill-core samples of the glacially-derived Kings River Fan alluvial deposits in central California. Sediments from the 45m core have independently-estimated depositional ages of up to ~800ka, based on paleomagnetism and correlations to nearby dated sediments. We characterized sequentially-leached core samples (both bulk sediment and grain size separates) for U, Nd, and Sr isotopes, grain size, surface texture, and mineralogy. In accordance with the comminution age model, where 234U is partially lost from small sediment grains due to alpha recoil, we found that ( 234U/ 238U) activity ratios generally decrease with age, depth, and specific surface area, with depletions of up to 9% relative to radioactive equilibrium. The resulting calculated comminution ages are reasonable, although they do not exactly match age estimates from previous studies and also depend on assumptions about 234U loss rates. The results indicate that the method may be a significant addition to the sparse set of available tools for dating detrital continental sediments, following further refinement. Improving the accuracy of the method requires more advanced models or measurements for both the recoil loss factor f α and weathering effects. We discuss several independent methods for obtaining f α on individual samples that may be useful for future studies. © 2010 Elsevier B.V.

We use uranium (U) isotope ratios to detect and quantify the extent of natural U reduction in groundwater across a roll front redox gradient. Our study was conducted at the Smith Ranch-Highland in situ recovery (ISR) U mine in eastern Wyoming, USA, where economic U deposits occur in the Paleocene Fort Union formation. To evaluate the fate of aqueous U in and adjacent to the ore body, we investigated the chemical composition and isotope ratios of groundwater samples from the roll-front type ore body and surrounding monitoring wells of a previously mined area. The (238)U/(235)U of groundwater varies by approximately 3‰ and is correlated with U concentrations. Fluid samples down-gradient of the ore zone are the most depleted in (238)U and have the lowest U concentrations. Activity ratios of (234)U/(238)U are ∼5.5 up-gradient of the ore zone, ∼1.0 in the ore zone, and between 2.3 and 3.7 in the down-gradient monitoring wells. High-precision measurements of (234)U/(238)U and (238)U/(235)U allow for development of a conceptual model that evaluates both the migration of U from the ore body and the extent of natural attenuation due to reduction. We find that the premining migration of U down-gradient of the delineated ore body is minimal along eight transects due to reduction in or adjacent to the ore body, whereas two other transects show little or no sign of reduction in the down-gradient region. These results suggest that characterization of U isotopic ratios at the mine planning stage, in conjunction with routine geochemical analyses, can be used to identify where more or less postmining remediation will be necessary.

Late Quaternary, post-shield lavas from the Mauna Kea and Kohala volcanoes on the Big Island of Hawaii have been dated using the {sup 40}Ar/{sup 39}Ar and U-Th/He methods. The objective of the study is to compare the recently demonstrated U-Th/He age method, which uses basaltic olivine phenocrysts, with {sup 40}Ar/{sup 39}Ar ages measured on groundmass from the same samples. As a corollary, the age data also increase the precision of the chronology of volcanism on the Big Island. For the U-Th/He ages, U, Th and He concentrations and isotopes were measured to account for U-series disequilibrium and initial He. Single analyses U-Th/He ages for Hamakua lavas from Mauna Kea are 87 {+-} 40 ka to 119 {+-} 23 ka (2{sigma} uncertainties), which are in general equal to or younger than {sup 40}Ar/{sup 39}Ar ages. Basalt from the Polulu sequence on Kohala gives a U-Th/He age of 354 {+-} 54 ka and a {sup 40}Ar/{sup 39}Ar age of 450 {+-} 40 ka. All of the U-Th/He ages, and all but one spurious {sup 40}Ar/{sup 39}Ar ages conform to the previously proposed stratigraphy and published {sup 14}C and K-Ar ages. The ages also compare favorably to U-Th whole rock-olivine ages calculated from {sup 238}U - {sup 230}Th disequilibria. The U-Th/He and {sup 40}Ar/{sup 39}Ar results agree best where there is a relatively large amount of radiogenic {sup 40}Ar (>10%), and where the {sup 40}Ar/{sup 36}Ar intercept calculated from the Ar isochron diagram is close to the atmospheric value. In two cases, it is not clear why U-Th/He and {sup 40}Ar/{sup 39}Ar ages do not agree within uncertainty. U-Th/He and {sup 40}Ar/{sup 39}Ar results diverge the most on a low-K transitional tholeiitic basalt with abundant olivine. For the most alkalic basalts with negligible olivine phenocrysts, U-Th/He ages were unattainable while {sup 40}Ar/{sup 39}Ar results provide good precision even on ages as low as 19 {+-} 4 ka. Hence, the strengths and weaknesses of the U-Th/He and {sup 40}Ar/{sup 39}Ar methods are complimentary for basalts with ages of order 100-500 ka.

Amphitheater-headed canyons have been used as diagnostic indicators of erosion by groundwater seepage, which has important implications for landscape evolution on Earth and astrobiology on Mars. Of perhaps any canyon studied, Box Canyon, Idaho, most strongly meets the proposed morphologic criteria for groundwater sapping because it is incised into a basaltic plain with no drainage network upstream, and approximately 10 cubic meters per second of seepage emanates from its vertical headwall. However, sediment transport constraints, 4 He and 14 C dates, plunge pools, and scoured rock indicate that a megaflood (greater than 220 cubic meters per second) carved the canyon about 45,000 years ago. These results add to a growing recognition of Quaternary catastrophic flooding in the American northwest, and may imply that similar features on Mars also formed by floods rather than seepage erosion.

This paper presents new U–Th/He crystallization and 3 He exposure ages on olivine for a suite of basalt samples from the Snake River Plain, Idaho. The results provide the first demonstration that U–Th/He geochronology can be successfully applied to Late Quaternary basalts using measurements of olivine phenocrysts as opposed to U-rich minerals. The approach employs abrasion to remove the outer rinds of olivine crystals to avoid problems with 4 He implantation from the groundmass, and a specially designed extraction system that allows U, Th and He concentrations and isotopic compositions to be measured on the same olivine grains. We demonstrate the viability of the U–Th/He technique on basalts with ages of 373 ka and 3400 ka previously dated by Ar–Ar, and show that concurrent measurements of U–Th– 4 He ages and 3 He cosmogenic exposure ages improve the accuracy of both methods. The combinedU–Th– 4 He– 3 HetechniqueisappliedtoaseriesofbouldersandflowsfromBoxCanyonnearHagerman,Idaho,toevaluate the ages of the basalts and to use exposure ages to make inferences about the age and mode offormation of the canyon. The U–Th/He ages of the Box Canyon basalts are between 86±12 ka and 130±12 ka (1-sigma errors) and suggest at least two periods of eruption. The ages are similar to a ca. 95 ka Ar–Ar age previously assigned, and distinguish the Box Canyon basalts from nearby ca. 395 ka basalt. Exposure ages of boulders within the canyon at the mouth, head and middle of Box Canyon are 21±1, 19±3 and 48±3 ka; scoured bedrock at the canyon head has an exposure age of 45±5 ka. These data indicate that the canyon was carved before the

Both chemical weathering rates and fluid flow are difficult to measure in natural systems. However, these parameters are critical for understanding the hydrochemical evolution of aquifers, predicting the fate and transport of contaminants, and for water resources/water quality considerations. 87 Sr/ 86 Sr and ( 234 U/ 238 U) activity ratios are sensitive indicators of water–rock interaction, and thus provide a means of quantifying both flow and reactivity. The 87 Sr/ 86 Sr values in ground waters are controlled by the ratio of the dissolution rate to the flow rate. Similarly, the ( 234 U/ 238 U) ratio of natural ground waters is a balance between the flow rate and the dissolution of solids, and α-recoil loss of 234 U from the solids. By coupling these two isotope systems it is possible to constrain both the long-term (ca. 100’s to 1000’s of years) flow rate and bulk dissolution rate along the flow path. Previous estimates of the ratio of the dissolution rate to the infiltration flux from Sr isotopes ( 87 Sr/ 86 Sr) are combined with a model for ( 234 U/ 238 U) to constrain the infiltration flux and dissolution rate for a 70-m deep vadose zone core from Hanford, Washington. The coupled model for both ( 234 U/ 238 U) ratios and the 87 Sr/ 86 Sr data suggests an infiltration flux of 5 ± 2 mm/yr, and bulk silicate dissolution rates between 10 −15.7 and 10 −16.5 mol/m 2 /s. The process of α-recoil enrichment, while primarily responsible for the observed variation in ( 234 U/ 238 U) of natural systems, is difficult to quantify. However, the rate of this process in natural systems affects the interpretation of most U-series data. Models for quantifying the α-recoil loss fraction based on geometric predictions, surface area constraints, and chemical methods are also presented. The agreement between the chemical and theoretical methods, such as direct measurement of ( 234 U/ 238 U) of the small grain size fraction and geometric calculations for that size fraction, is quite good.

High precision uranium isotope measurements of marine clastic sediments are used to measure the transport and storage time of sediment from source to site of deposition. The approach is demonstrated on fine-grained, late Pleistocene deep-sea sediments from Ocean Drilling Program Site 984A on the Bjorn Drift in the North Atlantic. The sediments are siliciclastic with up to 30% carbonate, and dated by δ18O of benthic foraminifera. Nd and Sr isotopes indicate that provenance has oscillated between a proximal source during the last three interglacial periods – volcanic rocks from Iceland – and a distal continental source during glacial periods. An unexpected finding is that the 234U/238U ratios of the silicate portion of the sediment, isolated by leaching with hydrochloric acid, are significantly less than the secular equilibrium value and show large and systematic variations that are correlated with glacial cycles and sediment provenance. The 234U depletions are inferred to be due to α-recoil loss of 234Th, and are used to calculate “comminution ages” of the sediment — the time elapsed between the generation of the small (≤ 50 μm) sediment grains in the source areas by comminution of bedrock, and the time of deposition on the seafloor. Transport times, the difference between comminution ages and depositional ages, vary from less than 10 ky to about 300 to 400 ky for the Site 984A sediments. Long transport times may reflect prior storage in soils, on continental shelves, or elsewhere on the seafloor. Transport time may also be a measure of bottom current strength. During the most recent interglacial periods the detritus from distal continental sources is diluted with sediment from Iceland that is rapidly transported to the site of deposition. The comminution age approach could be used to date Quaternary non-marine sediments, soils, and atmospheric dust, and may be enhanced by concomitant measurement of 226Ra/230Th, 230Th/234U, and cosmogenic nuclides.

New Sr and Nd isotopic data are combined with available information on the composition and petrology of lavas and the thermal and seismic structure of the underlying crust to develop a detailed model for the deep structure and magmatic processes of the Clear Lake volcanic system in northern California. The isotopic data require a two-stage model for magmatic evolution. In stage I, basaltic magma (eNd = + 6 to + 8; 87Sr / 86Sr = 0.703 to 0.7035; SiO2 ≤ 50%) is fed from the mantle into the lower and middle crust and evolves through combined crustal assimilation and fractional crystallization to basaltic andesite (eNd = + 5 to + 0.4; 87Sr / 86Sr = 0.70328 to 0.70485; SiO2 ≈ 55% to 57%). In stage II, the basaltic andesite magmas are transported upward and are either erupted at the surface or stored in shallow magma chambers where they evolve by fractional crystallization to form dacitic and rhyolitic magmas (SiO2 ≈ 65% to 70%). High-silica rhyolites (SiO2 ≈ 75%; high 87Sr / 86Sr) show evidence that further crustal assimilation can occur where upper crustal temperatures are elevated. Calculated densities of Clear Lake lavas indicate that basalt should pond at a depth of 12–18 km where seismic data show a pronounced density boundary within the crust. Thermodynamic models of assimilation require that mid-crustal temperatures are at least 600–800 °C to allow for enough assimilation to explain the isotopic data. Both surface heat flow and thermobarometry of crustal xenoliths in andesites are consistent with these inferred high temperatures. The Clear Lake volcanic system provides an opportunity to cross-calibrate petrological, geochemical and geophysical approaches. The results confirm that magma supply, magma buoyancy, and crustal temperatures control magmatic evolution. A temporal trend of increasing eNd over the past 2 million years suggests that magma supply in the Clear Lake volcanic field has been increasing and is still high. This is consistent with high heat flow in the area and high 3He / 4He in thermal gasses. The current pause in eruptive activity could represent a period of silicic magma accumulation in the crust. Future eruptions in the Clear Lake area may therefore be silicic and larger in volume than those over the past 100 ky.

Strontium isotope ratios (87Sr/86Sr) measured in groundwater samples from 273 wells in the Pasco Basin unconfined aquifer below the US Department Of Energy Hanford Site show large and systematic variations that provide constraints on groundwater recharge, weathering rates (for saturated zone dissolution) of the aquifer host rocks, communication between unconfined and deeper confined aquifers, and vadose zone–groundwater interaction. The impact of millions of cubic meters of wastewater discharged to the vadose zone (103–105 times higher than ambient drainage) shows up strikingly on maps of groundwater 87Sr/86Sr. Extensive access through the many groundwater monitoring wells at the site allows for an unprecedented opportunity to evaluate the strontium geochemistry of a major aquifer, hosted primarily in unconsolidated sediments, and relate it to both long term properties and recent disturbances. Groundwater 87Sr/86Sr increases systematically from 0.707 to 0.712 from west to east across the Hanford Site, in the general direction of groundwater flow, as a result of addition of Sr from the weathering of aquifer sediments and from diffuse drainage through the Pasco Basin vadose zone. The lower 87Sr/86Sr groundwater reflects recharge waters that have acquired Sr from Columbia River Basalts. Based on a steady-state model of Sr reactive transport and drainage, there is an average natural drainage flux of 0–1.4 mm/yr near the western margin of the Hanford Site, and ambient drainage may be up to 30 mm/yr in the center of the site assuming an average bulk rock weathering rate of 10−7.5 g/(g/yr). The weathering rates cannot be higher than those assumed by more than a factor of 2–3, in order to match Sr isotope data.

A mathematical model is presented that describes the effects of pore fluid aqueous diffusion and reaction rate on the isotopic exchange between fluids and rocks in reactive geo-hydrological systems where flow is primarily through fractures. The model describes a simple system with parallel equidistant fractures, and chemical transport in the matrix slabs between fractures by aqueous diffusion through a stagnant pore fluid. The solid matrix exchanges isotopes with pore fluid by solution–precipitation at a rate characterized by a time constant, R (yr −1 ), which is an adjustable parameter. The effects of reaction on the isotopes of a particular element in the fracture fluid are shown to depend on the ratio of the diffusive reaction length for that element ( L ) to the fracture spacing ( b ). The reaction length depends on the solid–fluid exchange rate within the matrix, the partitioning of the element between the matrix pore fluid and the matrix solid phase, the porosity and density of the matrix, and the aqueous diffusivity. For L / b L / b relative to a standard porous flow (single porosity) model. For L / b > 1, the parallel fracture model is no different from a porous flow model. If isotopic data are available for two or more elements with different L values, it may be possible to use the model with appropriate isotopic measurements to estimate the spacing of the primary fluid-carrying fractures in natural fluid–rock systems. Examples are given using Sr and O isotopic data from mid-ocean ridge (MOR) hydrothermal vent fluids and Sr isotopes in groundwater aquifers hosted by fractured basalt. The available data for MOR systems are consistent with average fracture spacing of 1–4 m. The groundwater data suggest larger effective fracture spacing, in the range 50–500 m. In general, for fractured rock systems, the effects of fracture–matrix diffusive exchange must be considered when comparing isotopic exchange effects for different elements, as well as for estimating water age using radioactive and cosmogenic isotopes.

We agree with George et al. that high 226 Ra/ 230 Th and correlated Ba/Th, 238 U/ 230 Th, and other elemental and isotopic ratios are the result of fluid addition to the wedge. The Feineman and DePaolo [Earth Planet. Sci. Lett. 215 (2003) 339–355] manuscript addresses more specifically the pathway followed by the fluid en route to the surface. In particular, we address whether the 226 Ra/ 230 Th data directly yield the total time available for the fluid to migrate from the point of origin in the slab to the melting region in the wedge, and ultimately to the surface with the melt. All of the processes involved in generating arc volcanism would of necessity have to take place very rapidly within the constraints proposed by George et al. The alternative is that the upward fluid movement is hindered by the fluid flow regime and solid mantle flow, allowing more time for reaction with the mantle wedge. There is evidence in the U-series data for more than one time scale, and our model helps to explain how this could come about. The extent to which grainscale 226 Ra/ 230 Th disequilibrium affects the inferred melt transport time is yet another issue, and this could be achieved either with the small-degree melt mechanism we discussed or by additional effects associated with melt and fluid migration at later stages in the melt generation process. We agree that the final melt transport event is likely to happen quickly. One of the most puzzling features of volcanic arcs is that the volcanic front emerges considerably behind the presumed location of primary water release from the subducting slab (depth to slab is ∼120 km beneath the volcanic front, as opposed to the predicted ∼80 km). This offset between the expected site of fluid release and the location of the volcanic front can be attributed to coupled transport of the fluid—in the form of hydrous minerals, such as amphibole and/or phlogopite in the wedge [J. Geophys. Res. 97 (1992) 2037–2070] or redirection of the melt due to flow and stresses in the mantle matrix as the magma slowly percolates upward [J. Geol. 95 (1987) 285–307]. In either case, the fluid and/or melt is required to ascend slowly, at least for some part of its journey, to explain the location of the volcanic front. Furthermore, 238 U/ 230 Th disequilibria in arc lavas seem to suggest a relatively long interval for fluid addition (∼30–60 ka) [Science 292 (2001) 1363–1366; Earth Planet. Sci. Lett. 179 (2000) 581–593; Geochim. Cosmochim. Acta 61 (1997) 4855–4884], which is difficult to reconcile with the 226 Ra/ 230 Th disequilibrium observed in the lavas. For these reasons, we sought to find a means whereby apparently short-lived isotopic disequilibria could be maintained during complex fluid and melt migration in the mantle wedge. Below, we address a few of the main issues presented by George et al. [Earth. Planet. Sci. Lett. this issue] in their comment.

Waste material buried in drums in the shallow subsurface at the Radioactive Waste Management Facility (RWMC) of the Idaho National Engineering and Environmental Laboratory (INEEL) contained significant amounts of organic compounds including lubricating oils and chlorinated solvents. CO{sub 2} concentrations in pore gas samples from monitoring wells in the vicinity of the disposal pits are 3 to 5 times higher than the concentrations in nearby background wells. The stable carbon isotope ratios ({delta}{sup 13}C values) of CO{sub 2} from the disposal pits averaged 2.4. less than CO{sub 2} from the background wells, indicating that the elevated CO{sub 2} concentrations around the pits were derived from source materials with {delta}{sup 13}C values in the range of -24{per_thousand} to -29{per_thousand}. These {delta}{sup 13}C values are typical of lubricating oils, but higher than most solvents. The radiocarbon ({sup 14}C) contents of CO{sub 2} across most of the site were significantly elevated above modern concentrations due to reactor blocks buried in a subsurface vault at the site. However, several samples collected from the high-CO{sub 2} zone on the far side of the RWMC from the reactor blocks had very low {sup 14}C contents (less than 0.13 times modern), confirming production from lubricating oils manufactured from fossil hydrocarbons. The magnitude of the CO{sub 2} anomaly observed at the site is consistent with intrinsic biodegradation rates on the order of 0.5 to 3.0 metric tons of carbon per year.

The U–Th/He system can potentially be used for dating volcanic rocks with ages as young as a few thousand years and as old as several million years, thus providing a valuable supplement to radiocarbon and K–Ar dating. Garnet phenocrysts from the 79 AD eruption of Mt. Vesuvius were dated to evaluate the accuracy with which the necessary measurements and corrections can be made. The determined age, corrected for diffusive loss of He, alpha ejection, and initial U-series disequilibrium, is 1885±188 yr which compares well with the known age of 1923 yr. U and Th concentrations were measured by isotope dilution on different aliquots than were used for He concentration measurements. Step-wise degassing yielded an Arrhenius relationship for He diffusion in garnet with an activation energy of 91.31±5.76 kJ/mol and ln D0/a2=−2.00±0.56. The uniformity of U and Th concentrations in garnet was checked by ion microprobe analysis. The 234U/238U and 230Th/238U activity ratios were measured by MC-ICPMS. The results suggest that with proper analysis and corrections, the U–Th/He method can be used to date young volcanic minerals with useful precision and accuracy, and may therefore be valuable for dating volcanic rocks that have low K or are otherwise difficult to date accurately with Ar–Ar or radiocarbon.

The relatively low rates of magma production in island arcs and continental extensional settings require that the volume of silicic magma involved in large catastrophic caldera-forming (CCF) eruptions must accumulate over periods of 105 to 106 years. We address the question of why buoyant and otherwise eruptible high-silica magma should accumulate for long times in shallow chambers rather than erupt more continuously as magma is supplied from greater depths. Our hypothesis is that the viscoelastic behavior of magma chamber wall rocks may prevent an accumulation of overpressure sufficient to generate rhyolite dikes that can propagate to the surface and cause an eruption. The critical overpressure required for eruption is based on the model of Rubin (1995a). An approximate analytical model is used to evaluate the controls on magma overpressure for a continuously or episodically replenished spherical magma chamber contained in wall rocks with a Maxwell viscoelastic rheology. The governing parameters are the long-term magma supply, the magma chamber volume, and the effective viscosity of the wall rocks. The long-term magma supply, a parameter that is not typically incorporated into dike formation models, can be constrained from observations and melt generation models. For effective wall-rock viscosities in the range 1018 to 1020 Pa s–1, dynamical regimes are identified that lead to the suppression of dikes capable of propagating to the surface. Frequent small eruptions that relieve magma chamber overpressure are favored when the chamber volume is small relative to the magma supply and when the wall rocks are cool. Magma storage, leading to conditions suitable for a CCF eruption, is favored for larger magma chambers (>102 km3) with warm wall rocks that have a low effective viscosity. Magma storage is further enhanced by regional tectonic extension, high magma crystal contents, and if the effective wall-rock viscosity is lowered by microfracturing, fluid infiltration, or metamorphic reactions. The long-term magma supply rate and chamber volume are important controls on eruption frequency for all magma chamber sizes. The model can explain certain aspects of the frequency, volume, and spatial distribution of small-volume silicic eruptions in caldera systems, and helps account for the large size of granitic plutons, their association with extensional settings and high thermal gradients, and the fact that they usually post-date associated volcanic deposits.

Interpretation of U-series disequilibria in midocean ridge basalts is highly dependent on the bulk partition coefficients for U and Th and therefore the mineralogy of the mantle source. Distinguishing between the effect of melting processes and variable source compositions on measured disequilibria (238U-230Th-226Ra and 235U-231Pa) requires measurement of the radiogenic isotopes Hf, Nd, Sr, and Pb. Here, we report measurements of 238U-230Th-226Ra and 235U-231Pa disequilibria; Hf, Nd, Sr, and Pb isotopic; and major and trace element compositions for a suite of 20 young midocean ridge basalts from the East Pacific Rise axis between 9°28′ and 9°52′N. All of the samples were collected within the axial summit trough using the submersible Alvin. The geological setting and observational data collected during sampling operations indicate that all the rocks are likely to have been erupted from 1991 to 1992 or within a few decades of that time. In these samples, 230Th excesses and 226Ra excesses are variable and inversely correlated. Because the eruption ages of the samples are much less than the half-life of 226Ra, this inverse correlation between 230Th and 226Ra excesses can be considered a primary feature of these lavas. For the lava suite analyzed in this study, 226Ra and 230Th excesses also vary with lava composition: 226Ra excesses are negatively correlated with Na8 and La/Yb and positively correlated with Mg#. Conversely, 230Th excesses are positively correlated with Na8 and La/Yb and negatively correlated with Mg#. Th/U, 230Th/232Th, and 230Th excesses are also variable and correlated to one another. 231Pa excesses are large but relatively constant and independent of Mg#, La/Yb, Th/U, and Na8. The isotope ratios 143Nd/144Nd, 176Hf/177Hf, 87Sr/86Sr, and 208Pb/206Pb are constant within analytical uncertainty, indicating that they were derived from a common source. The source is homogeneous with respect to parent/daughter ratios Lu/Hf, Sm/Nd, Rb/Sr, and Th/U; therefore, the measured variations of Th/U, 230Th, and 226Ra excesses and major and trace element compositions in these samples are best explained by polybaric melting of a homogeneous source, not by mixing of compositionally distinct sources.

Preferential flow paths are expected in many groundwater systems and must be located because they can greatly affect contaminant transport. The fundamental characteristics of radiogenic isotope ratios in chemically evolving waters make them highly effective as preferential flow path indicators. These ratios tend to be more easily interpreted than solute-concentration data because their response to water-rock interaction is less complex. We demonstrate this approach with groundwater {sup 87}Sr/{sup 86}Sr ratios in the Snake River Plain aquifer within and near the Idaho National Engineering and Environmental Laboratory. These data reveal slow-flow zones as lower {sup 87}Sr/{sup 86}Sr areas created by prolonged interaction with the host basalts and a relatively fast flowing zone as a high {sup 87}Sr/{sup 86}Sr area.

The chemical compositions, ages, and Nd and Sr isotopic compositions of Miocene basaltic volcanic rocks from the southwestern Basin and Range Province were measured. The data presented, and other data obtained from the literature show that temporal variations in the Nd and Sr isotopic compositions are typically correlated with the timing of local crustal extension. The isotopic variations cannot be accounted for by crustal contamination, and are best interpreted as indicating the involvement of subcontinental lithospheric mantle as a major source of basalt magma before and during the early stages of extension, giving way to asthenospheric sources as extension proceeds. It is inferred that the basalt isotopic and age data provide information about the amount and timing of changes of the overall lithosphere thickness during extension. If this inference is valid, the inferred lithosphere thickness vs. time histories can be compared to the amount and timing of upper crustal extension to evaluate competing models for lithospheric strain during continental extension. The depth of the lithosphere–asthenospshere boundary is inferred primarily from the isotopic data, but can be refined by accounting for differences in the depth of origin of the basalts as suggested by differences in the degree of silica saturation (nepheline normative vs. hypersthene normative). The transition between silica saturated and undersaturated magmas is dependent on many factors, but for the conditions likely to apply to the Basin and Range Miocene volcanism, it corresponds to a depth of origin of 50±10 km. The data presented show strong support for the inferred mantle geochemical stratigraphy, and the inferred modern depths to the lithosphere–asthenosphere boundary are consistent with available seismic refraction and structural data. In the highly extended regions near the eastern and western margins of the southwestern Basin and Range, temporal changes in basalt isotopic compositions indicate that the subjacent lithospheric mantle thinned contemporaneously with the upper crust, as would be predicted for pure shear. These areas are also characterized by relatively high (although still low) basalt extrusion rates. In contrast, upper crustal extension in the Death Valley area was not accompanied by significant thinning of the subjacent lithospheric mantle, which implies that the location of lithospheric thinning was offset from the location of upper crustal extension as in a simple shear model. The total amount of lithospheric thinning is generally less than would be expected from the magnitude of upper crustal extension and is more evenly distributed; this suggests that there is also a component of distributed shear.

We expand a model for isotope ratios in groundwater-rock systems to consider the ratio of 14C to dissolved inorganic carbon, or “percent modern carbon.” The model couples one-dimensional advection and dispersion with water-rock interaction and radioactive decay. We use a Fickian formulation to model longitudinal macrodispersion and present a formula for the groundwater velocity error introduced when dispersion is ignored, as a function of dispersivity and decrease in percent modern carbon. For example, if the longitudinal dispersivity were one tenth the length of observation and percent modern carbon decreased tenfold along a flow path, the error would be 23%. We also present a method for inversion of 87Sr/86Sr and 14C data to obtain carbonate dissolution rates and dissolution-corrected water velocities and apply this technique using published data from the Lincolnshire aquifer, England. The results are similar to those obtained using the δ13C method.

We have examined the Sr, Nd, and Pb isotopic compositions of Mauna Kea lavas recovered by the first drilling phase of the Hawaii Scientific Drilling Project. These lavas, which range in age from ∼200 to 400 ka, provide a detailed record of chemical and isotopic changes in basalt composition during the shield/postshield transition and extend our record of Mauna Kea volcanism to a late-shield period roughly equivalent to the last ∼100 ka of Mauna Loa activity. Stratigraphic variations in isotopic composition reveal a gradual shift over time toward a more depleted source composition (e.g., higher 143Nd/144Nd, lower 87Sr/86Sr, and lower 3He/4He). This gradual evolution is in sharp contrast with the abrupt appearance of alkalic lavas at ∼240 ka recorded by the upper 50 m of Mauna Kea lavas from the core. Intercalated tholeiitic and alkalic lavas from the uppermost Mauna Kea section are isotopically indistinguishable. Combined with major element evidence (e.g., decreasing SiO2 and increasing FeO) that the depth of melt segregation increased during the transition from tholeiitic to alkalic volcanism, the isotopic similarity of tholeiitic and alkalic lavas argues against significant lithosphere involvement during melt generation. Instead, the depleted isotopic signatures found in late shield-stage lavas are best explained by increasing the proportion of melt generated from a depleted upper mantle component entrained and heated by the rising central plume. Direct comparison of Mauna Kea and Mauna Loa lavas erupted at equivalent stages in these volcanoes' life cycles reveals persistent chemical and isotopic differences independent of the temporal evolution of each volcano. The oldest lavas recovered from the drillcore are similar to modern Kilauea lavas, but are distinct from Mauna Loa lavas. Mauna Kea lavas have higher 143Nd/144Nd and 206Pb/204Pb and lower 87Sr/86Sr. Higher concentrations of incompatible trace elements in primary magmas, lower SiO2, and higher FeO also indicate that Mauna Kea lavas formed through smaller degrees of partial melting at greater depth than Mauna Loa lavas. These chemical and isotopic differences are consistently found between volcanoes along the western “Loa” and eastern “Kea” trends and reflect large-scale variations in source composition and melting environment. We propose a simple model of a radially zoned plume centered beneath the Loa trend. Loa trend lavas generated from the hot plume axis reflect high degrees of partial melting from a source containing a mixture of enriched plume-source material and entrained lower mantle. Kea trend lavas, in contrast, are generated from the cooler, peripheral portions of the plume, record lower degrees of partial melting, and tap a source containing a greater proportion of depleted upper mantle.

We have investigated the isotopic compositions of Os, Sr, Nd, and Pb in a suite of primitive Mauna Loa lavas from the upper 280 m of the Hawaii Scientific Drilling Project pilot core drilled near Hilo, Hawaii. These lavas were probably erupted from Mauna Loa's northeast rift. Correlations between Os (hosted by olivine) and other isotopes indicate that olivine crystals in these flows are closely related in space and time to the enclosing lava, despite the presence of deformation features and Fe/Mg disequilibrium in some olivines. The temporal isotopic evolution of the lavas matches published data for basalts from Mauna Loa's southwest rift, indicating that the two rifts (as well as the summit) share a common magma feed zone which is distinct from that of Kilauea. The composition of the lowermost HSDP Mauna Loa sample shows some isotopic similarities to modern Kilauea compositions and in this respect compares well with published data on submarine lavas from Mauna Loa's southwest rift. The good correlations among the isotopic tracers of compatible (Os) and incompatible (Sr, Nd, Pb) elements indicate that a depleted upper mantle component is very minor or nonexistent in Mauna Loa lavas. The Os isotope results definitively rule out equilibrium porous flow as a means of melt transport through the lithosphere. The isotopic variations in shield-stage lavas are most consistent with partial melting of two distinct sources within the Hawaiian plume followed by partial mixing and rapid transport of melts through the oceanic lithosphere. The passage of the Pacific lithosphere over a heterogeneous Hawaiian plume can account for the systematic differences in the compositions of volcanoes from the “Kea” and “Loa” trends as well as the geochemical evolution of individual shields on the island of Hawaii.

Helium concentrations and isotopic ratios have been measured in a suite of basaltic olivines from the Hawaii Scientific Drilling Project (HSDP) core at Hilo, Hawaii, which allows a characterization of the temporal helium isotopic evolution of Mauna Kea volcano. Typically more than 85% of the helium within the olivines is released by crushing, which demonstrates that helium is dominantly contained within the melt inclusions, and strongly suggests that helium behaves as an incompatible element during silicate melting. Lavas from the base of the Mauna Loa section of the core (240–270 m depth) have 3He/4He ratios between 13.9 and 15.8 times atmospheric (Ra). These values, when combined with Sr and Nd isotopic data, correspond to those previously found in subaerial Mauna Loa lavas that are greater than 30 ka in age, place a minimum age of 30 ka for the top of the Mauna Kea section, and are consistent with radiocarbon chronology. The upper part of the Mauna Kea lava section (290–620 m) has 3He/4He ratios of 7.0 to 6.8 Ra, similar to most mid-ocean ridge basalts and is thought to represent the normal asthenospheric mantle signature. Between 620 and 670 m depth there is a transition to higher values, with 3He/4He ratios between 10.2 and 12.5 Ra at depths greater than 670 m, suggesting a greater contribution of undegassed plume material to the older Mauna Kea tholeiites. The correlations between He, Nd, and Pb isotope ratios demonstrate that the helium isotopic variations are related to mixing of source materials rather than shallow contamination effects (i.e., via addition of radiogenic 4He). The transition from high to low 3He/4He ratios occurs within the tholeiitic shield of Mauna Kea. The highest 3He/4He ratios in the Mauna Kea lavas (up to 12.5 Ra) are significantly lower than the values found in the shield tholeiites of the adjacent volcanoes (Loihi Seamount, Mauna Loa, and Kilauea); this relates to a smaller hot spot contribution to the Mauna Kea lavas. The temporal helium isotopic evolution of Hawaiian shield volcanos (with highest 3He/4He ratios found in the oldest lavas) and the inter-volcano variations (with the Mauna Kea shield having the lowest 3He/4He ratios on the island) are attributed to movement of the Pacific plate over a radially zoned hot spot.

Intraplate or "hot spot" volcanic island chains, exemplified by Hawaii, play an important role in plate tectonic theory as reference points for absolute plate motions, but the origin of these volcanoes is not explained by the plate tectonic paradigm [Engebretson et al., 1985; Molnar and Stock, 1987; Morgan, 1971, 1981, 1983; Wilson, 1963]. The most widely held view is that these chains of volcanoes form from magma generated by decompression melting of localized, buoyant upwellings in the mantle [Ribe and Christensen, 1994; Richards et al., 1988; Sleep, 1990; Watson and McKenzie, 1991] . These upwellings, or "plumes," are believed to originate at boundary layers in the mantle (e.g., at the core-mantle boundary or near the boundary at-670 km between the upper and lower mantle), and the cause of the buoyancy may be both compositional and thermal [Campbell and Griffiths, 1990; Griffiths, 1986; Richards et al., 1988; Watson and McKenzie, 1991]. Mantle plumes are responsible for about 10% of the Earth's heat loss and constitute an important mechanism for cycling mass from the deep mantle to the Earth's surface. Studies of the chemical and isotopic compositions of lavas from intraplate volcanoes, especially ocean island volcanoes, have contributed significantly to our knowledge of magma genesis in the mantle [Carmichael et al., 1974; Macdonald et al., 1983] and the compositional heterogeneity of the mantle [Allègre et al., 1983; Hart, 1988; Hart et al., 1986; Kurz et al., 1983]. Of particular importance is the identification of distinct compositional end members in the mantle, the origin and distribution of which provide insight into the long-term differentiation of the mantle-crust system, the recycling of oceanic crust and continental sediment into the mantle, and the history of the lithosphere [Allègre et al., 1995; Farley et al., 1992; Hart, 1988; Hofmann and White, 1982; McKenzie and O'Nions, 1983; Weaver, 1991; Zindler and Hart, 1986].

We describe a method of uranium-thorium-lead (UThPb) isotopic age dating for Quaternary rocks. The approach uses an instrumental mass discrimination correction for lead isotope ratios, which allows small enrichments of radiogenic 206Ph and 208Ph to be detected at the level of 0.001%. Igneous rocks hosting minerals with a range in 238 U 204 Pb values of 100 can be dated with uncertainties of approximately ±15–20 kyr. A Quaternary rhyolite dated at 1.19 Ma by KAr yields a 238U206Ph age of 1.03 ± 0.10 Ma. A Holocene dacite (9.5 ka) has uniform 206 Pb 207 Pb to within 0.0015% in groundmass phases, but 1 mm plagioclase phenocrysts have lower 206 Pb 207 Pb by 0.105 ± 0.002% indicating contamination of the magma after plagioclase crystallization. High precision 206 Pb 207 Ph ratios may be a useful new tool for petrogenetic studies.

The reconstruction of Laurentia, Australia, and Antarctica into a Proterozoic supercontinent is evaluated by analyzing the fit of Precambrian provinces defined by isotopic and geochronologic mapping. The analysis is complicated by allochthonous segments of the Antarctic and eastern Australian margins. Removal of the allochthonous provinces produces a closer fit of the continents; there is a match of Early Proterozoic basement between southwestern Laurentia and the only exposure of craton known from the paleo-Pacific margin of Antarctica. In addition, western Laurentia is brought closer to the Australian Gawler block, consistent with provenance interpretations of the Belt Supergroup. Removal of the allochthonous provinces by right-lateral translation relative to the Antarctic craton margin places them in a pre-750 Ma position where they could be southwestward extensions of the Yavapai-Mazatzal and Grenville provinces of southern Laurentia. This modified reconstruction leads to a prediction of extensive Archean basement in Antarctica between the South Pole and Victoria Land, a prediction partly borne out by Archean rocks in the Miller Range of the Transantarctic Mountains; it also predicts the presence of 1.4Ga rapakivi granites in the Transantarctic Mountains basement. This configuration implies assembly of the Australia-Antarctica Gondwana margin by terrane accretion following, or accompanied by, left-lateral translation. This is compatible with a tectonic regime of clockwise rotation of Laurentia relative to Australia and Antarctica after rifting. Thus, the proposed supercontinent, with some modifications, has potential for explaining several aspects of the pattern of Precambrian provinces in the three continents.

Strontium isotopic compositions of carbonate mollusk shells from estuarine sediments cored beneath San Francisco Bay are used to derive a record of mean annual salinity and average freshwater inflow to the estuary for intervals during the past 4300 yrs. The large difference in the 87Sr/86Sr ratio between seawater (0.7092) and the average freshwater entering the estuary (0.7065) produces a correlation between 87Sr/86Sr and salinity in bay waters that can be detected with high-precision measurements. Paleosalinity is inferred from the 87Sr/86Sr ratio measured in fossil carbonate shells of bivalves preserved in the sediment. Because salinity in San Francisco Bay is primarily controlled by the freshwater inflow from the Sacramento and San Joaquin rivers, the paleosalinity record can be converted to a paleodischarge record using a transfer function derived from historical data. Salinity data are referenced to modern values estimated for the estuary after correcting for the lower discharge caused by diversion since 1850 A.D. Data are presented from Richardson Bay, located near the estuary mouth (modern mear arrival salinity (MAS) of ca. 24‰), and from San Pablo Bay, located about 30 km upstream (modern MAS of ca. 18–20‰). For Richardson Bay, the data cover the periods 0–600 yrs B.P. and 2200–4300 yrs B.P., based on radiocarbon dating. For San Pablo Bay, the periods 0–130 yrs B.P. and 2500–2800 yrs B.P. are represented. The Richardson Bay data indicate extended periods of low salinity relative to modern, indicating higher freshwater inflow, at ca. 80, 220, 310, 440–500, 3100–3400 and 4300 yrs ago. The data indicate high salinity relative to modern, hence lower freshwater inflow, at about 40, 140–210, 270, 2100 and 3450–3700 yrs B.P. For San Pablo Bay, where salinity resolution is better, salinity was lower than modern during the period at ca. 50 and 2550–2650 yrs B.P., and higher than modern at ca 90–110, 155 and 2510–2530 yrs B.P. The data suggest that mean annual discharge to San Francisco Bay typically varies between values that are below the modern value (600 m3/s) and values 2 to 3 times as high; the average paleosalinity is typically substantially lower than the modern values. The 4300 yr average freshwater inflow is estimated to be ≥ 1200 ± 200 m3/s. The record indicates that quasi-cyclic variations in mean annual discharge occur naturally with a period of about 200 yrs or less.

SILICA-rich lavas, erupted at island-arc or continental volcanoes, are often produced by a complex process involving the assimilation of crust into a crystallizing, mantle-derived basaltic magma1. The different strontium, neodymium and oxygen isotopic compositions of mantle-derived magmas and continental crust provide a powerful method for tracing the different contributions to continental silicic magmas, and for understanding the parameters controlling the composition and volume of erupted magma1–4. In the large rhyolite eruptive centres of the western United States, the largest-volume, explosive rhyolite eruptions have more mantle-like Nd isotope ratios than other silicic lavas from the same centre2–4, a relationship that has been interpreted as reflecting increased influx of mantle-derived basaltic magma to a crustal magma chamber before large-volume eruptions1. Here we report isotope data for lavas from Unzen volcano, which suggest a similar relationship: the Nd isotope composition is more mantle-like in three larger-volume dacite eruptions (>0.1 km3) than in one small-volume (0.02 km3) eruption. We accordingly suggest that, in small-volume systems like Unzen, where the timescales for magma-chamber evolution are of the order of decades, isotope data such as those presented here might be used in volcanic hazard evaluation.

During the last two decades, expanding industrial activity in east Asia has led to increased production of airborne pollutants that can be transported to North America. Previous efforts to detect this trans-Pacific pollution have relied upon remote sensing and remote sample locations. We tested whether Pb isotope ratios in airborne particles can be used to directly evaluate the Asian contribution to airborne particles of anthropogenic origin in western North America, using a time series of samples from a pair of sites upwind and downwind of the San Francisco Bay Area. Our results for airborne Pb at these sites indicate a median value of 29% Asian origin, based on mixing relations between distinct regional sample groups. This trans-Pacific Pb is present in small quantities but serves as a tracer for airborne particles within the growing Asian industrial plume. We then applied this analysis to archived samples from urban sites in central California. Taken together, our results suggest that the analysis of Pb isotopes can reveal the distribution of airborne particles affected by Asian industrial pollution at urban sites in northern California. Under suitable circumstances, this analysis can improve understanding of the global transport of pollution, independent of transport models.

[1] Sr-Nd-Pb-Hf isotopic compositions of postshield lavas from two pairs of Hawaiian volcanoes, Mauna Kea and Kohala (Kea trend) and Hualalai and Mahukona (Loa trend), allow for identification of smallscale (tens of kilometers) heterogeneities in the Hawaiian mantle plume and provide constraints on their distribution. The postshield lavas range from transitional/alkalic basalt to trachyte and are enriched in incompatible trace elements (e.g., LaN/YbN = 6.0–16.2). These lavas are characterized by a limited range of Sr-Nd-Hf isotopic compositions ( 87 Sr/ 86 Sr = 0.70343–0.70365, 143 Nd/ 144 Nd = 0.51292–0.51301, and 176 Hf/ 177 Hf = 0.28311–0.28314) and have distinct Pb isotopic compositions ( 206 Pb/ 204 Pb = 17.89–18.44, 207 Pb/ 204 Pb = 15.44–15.49, and 208 Pb/ 204 Pb = 37.68–38.01) that correspond to their respective Kea or Loa side of the Pb-Pb isotopic boundary. Mauna Kea lavas show a systematic shift to less radiogenic Pb isotopic compositions from the shield to postshield stage and they trend to low 87 Sr/ 86 Sr toward, but not as extreme as, compositions characteristic of rejuvenated stage lavas. Hualalai postshield lavas lie distinctly above the Hf-Nd Hawaiian array and have much lower Pb isotopic ratios than shield lavas, including some of the least radiogenic values (e.g., 206 Pb/ 204 Pb = 17.89–18.01) of recent Hawaiian volcanoes. In contrast, comparison of Kohala with the adjacent Mahukona volcano shows that these older postshield lavas become more radiogenic in Pb during the late stages of volcanism. The isotope systematics of the postshield lavas cannot be explained by mixing between Hawaiian plume end-members (e.g., Kea, Koolau, and Loihi) or by assimilation of Pacific lithosphere and are consistent with the presence of ancient recycled lower oceanic crust (±sediments) in their source. More than one depleted component is sampled by the postshield lavas and these components are long-lived features of the Hawaiian plume that are present in both the Kea and Loa source regions. The depleted components in the postshield lavas, particularly as sampled at Hualalai, are different from the much more homogeneous component present in rejuvenated lavas. The geochemistry of the postshield lavas provides evidence for a bilateral symmetry to the plume where the compositional boundary between the Kea and Loa sources is complex and vertical components of heterogeneity are significant.

We use a high-resolution seawater Sr isotopic evolution curve for the last 100 m.y. in conjunction with modern riverine Sr flux measurements, and also geologic, tectonic and geochronological data, to make the case for a close relationship between seawater Sr isotopic composition and the India Asia continental collision. Using a simple seawater Sr budget model we begin by showing that the Sr flux associated with alteration of seafloor basalts is too small and does not have the right time evolution to account for much of the seawater Sr isotopic curve of the last 100 m.y. The flux of dissolved Sr carried by rivers originating in the Himalaya-Tibet region on the other hand is presently a significant fraction of the global Sr budget. We calculate how this riverine flux would have had to change with time in order to match the observed seawater Sr isotopic curve and find that the riverine flux remains relatively constant prior to the collision of India with Asia but then increases very significantly after collision. We note that the period of most rapid change in seawater Sr isotopic ratio, from 20 Ma to 15 Ma, is also a period of exceptionally high erosion in parts of the Himalayas and the Tibetan Plateau. As further evidence that Sr derived from the collision of India with Asia plays a major role in the Sr isotopic evolution of seawater we show that the total amount erosion of the Himalaya-Tibetan Plateau since collision, which we calculate separately, represents a total amount of Sr that is very nearly the same as the cumulative amount required by the Sr isotopic change of seawater since collision. The relationship between erosion and riverine Sr flux allows us to use the Sr isotopic evolution of seawater to reconstruct a history of erosion since collision, and we find that the erosion rate accelerates with time since collision, with the present having the largest rate. When we apply the Sr budget model to the entire Phanerozoic using a new compilation of deformed continental area versus time we find that we can account for the large-scale structure of the seawater Sr isotopic curve, but fail to reproduce several local maxima and minima, especially in the period 100-300 Ma. The present high S7Sr/S6Sr of seawater and similar highs in the Devonian and Cambrian do correlate with extensive deformation on the continents.

Geochemical studies of Paleozoic granites in northern Victoria Land (NVL) and the central Transantarctic Mountains (CTM) are integrated with structural and lithologie studies into a model for the tectonic development of the Antarctic-south-east Australian Gondwana margin in late Precambrian-middle Paleozoic time. Isotopic data (Nd, Sr, O) are used to define crustal basement age boundaries and subduction polarity and thus help identify suspect terranes. In the CTM, three basement age-blocks are recognized: The outboard two provinces are inferred to be allochthonous, with accretion to the Gondwana margin occurring between ≈ 750 and ≈ 500 Ma. This tectonic scenario is inferred to apply also to NVL and southeastern Australia (SEA). In the CTM and NVL, chemical and isotopic polarity of ca. 500 Ma granitoids indicate a cratonward-dipping subduction zone. In NVL, terranes outboard of the belt of ca. 500 Ma granites contain Devonian (ca. 400 Ma) I-type granites with chemical and isotopic polarity opposed to that of the ca. 500 Ma granites. In the furthest outboard position, S-type granites emplaced at ca. 600 and 525 Ma occur with ca. 400 Ma granites and indicate the presence of early Proterozoic basement. This NVL terrane collage is allochthonous; it was assembled, intruded by Devonian granites, and then tectonically accreted to the Gondwana margin between latest Devonian and Permian time. Terrane correlations with SEA imply that much of the latter was accreted to the Gondwana margin after Devonian time. Other fragments of this allochthonous continental collage are probably in Marie Byrd Land (West Antarctica) and New Zealand (including the Campbell Plateau).

Ocean island volcanoes and large igneous provinces, both of which represent mantle plume volcanism, are attractive drilling targets because of compelling scientific issues, the layered structure of lava accumulations, and the demonstrated capability to retrieve a high percentage of core in many environments. Although the programs supporting on-land scientific drilling — ICDP, and in the U.S., NSF-Continental Dynamics — are nominally about continental drilling, the primary dichotomy is between on-land drilling and ship-based offshore drilling (IODP). Mantle plume volcanic rocks are accessible to on-land drilling in many oceanic settings as well as on continents. Drilling and coring ordered, datable sequences of lavas can address questions about the mechanisms of magma generation and transport, the geochemical and geophysical structure of mantle plumes and their relationship to the structure and composition of the deep mantle, the structural evolution of large volcanoes, the subsidence history of volcanic piles, the flexural properties of the lithosphere, and the hydrology, alteration history and geobiology of subsurface volcanic environments.

[1] Sr, Nd, and Os isotopic measurements were made on 110 Mauna Kea lava and hyaloclastite samples from the drillcore retrieved from the second phase of the Hawaii Scientific Drilling Project (HSDP-2). The samples come from depths of 255 to 3098 meters below sea level, span an age range from 200 to about 550-600 kyr, and represent an ordered record of the lava output from Mauna Kea volcano as it drifted a distance of about 40 km over the magma-producing region of the Hawaiian hot spot. The deepest (oldest) samples represent the time when Mauna Kea was closest to the center of the melting region of the Hawaiian plume. The Sr and Os isotopic ratios in HSDP-2 lavas show only subtle isotopic shifts over the ∼400 kyr history represented by the core. Neodymium isotopes (ɛNd values) increase systematically with decreasing age from an average value of nearly +6.5 to an average value of +7.5. This small change corresponds to subtle shifts in 87Sr/86Sr and 187Os/188Os isotope ratios, with small shifts of ɛHf, a large shift in 208Pb/204Pb and 208Pb/207Pb values, and with a very large shift in He isotope ratios from R/RA values of about 7–8 to values as high as 25. When Mauna Kea was closest to the plume core, the magma source did not have primitive characteristics for Nd, Sr, Pb, Hf, and Os isotopes but did have variable amounts of “primitive” helium. The systematic shifts in Nd, Hf, Pb, and He isotopes are consistent with radial isotopic zoning within the melting region of the plume. The melting region constitutes only the innermost, highest-temperature part of the thermally anomalous plume mantle. The different ranges of values observed for each isotopic system, and comparison of Mauna Kea lavas with those of Mauna Loa, suggest that the axial region of the plume, which has a radius of ∼20 km, is a mixture of recycled subducted components and primitive lower mantle materials, recently combined during the formational stages of the plume at the base of the mantle. The proportions of recycled and primitive components are not constant, and this requires there be longitudinal (vertical) heterogeneity within the core of the plume. The remainder of the plume, outside this plume “core zone,” is less heterogeneous but distinct from upper mantle as represented by mid-ocean ridge basalt (MORB). The plume structure may provide a detailed view of mantle isotopic composition near the core-mantle boundary.

In the mid-1990s, a groundwater plume of uranium (U) was detected in monitoring wells in the B-BX-BY Waste Management Area at the Hanford Site in Washington. This area has been used since the late 1940s to store high-level radioactive waste and other products of U fuel-rod processing. Using multiple-collector ICP source magnetic sector mass spectrometry, high-precision uranium isotopic analyses were conducted of samples of vadose zone contamination and of groundwater. The isotope ratios 236U/238U, 234U/238U, and 238U/235U are used to distinguish contaminant sources. On the basis of the isotopic data, the source of the groundwater contamination appears to be related to a 1951 overflow event at tank BX-102 that spilled high-level U waste into the vadose zone. The U isotopic variation of the groundwater plume is a result of mixing between contaminant U from this spill and natural background U. Vadose zone U contamination at tank B-110 likely predates the recorded tank leak and can be ruled out as a significant source of groundwater contamination, based on the U isotopic composition. The locus of vadose zone contamination is displaced from the initial locus of groundwater contamination, indicating that lateral migration in the vadose zone was at least 8 times greater than vertical migration. The time evolution of the groundwater plume suggests an average U migration rate of approximately 0.7-0.8 m/day showing slight retardation relative to a groundwater flow of approximately 1 m/day.

[1] We report measurements of U-series disequilibria, Sr, Nd, Hf, and Pb isotopic compositions and major and trace element abundances in a suite of well-located, off-axis MORBs that span the East Pacific Rise (EPR) ridge crest from 9°48′–52′N and across it for ∼4 km on either side. The geological context of the samples are well constrained as they were collected by submersible in an area that has been extensively imaged by remote sensing techniques. Sr, Nd, Hf and 208Pb/206Pb isotopic compositions of the off-axis N-MORB are identical to the axial lavas from this same region, suggesting that their sources are similar and that melting processes are the dominant influence in establishing the U-Th-Ra disequilibria and trace element fractionations. A majority of off-axis samples have U-Th and Th-Ra disequilibria that are larger, and model ages that are younger, than would be predicted from their off-axis distance and the time-integrated spreading rate. There are, however, a few off-axis samples with U-Th ages that are consistent with their spreading rate ages. It is likely that these samples erupted within or close to the axial summit trough (AST) and aged at a rate proportional to the spreading rate. The anomalously young ages determined for most of the off-axis lavas suggest that volcanic construction along this region is occurring over a zone that is wider (at least 4 km) than the AST (10s to 100s of meters). The combined observational, chemical and isotopic data support a model for the 9°0′N area that includes a significant component of crustal accumulation resulting from lavas that breach the AST and flow down the flanks of the EPR ridge crest. However, these data also require a minor component of off-axis eruptions that occur on distinct pillow mounds and ridges. This suggests that MOR construction involves several volcanic and tectonic processes acting in concert to form a complex patchwork of lava ages and compositions along, and across, this fast spreading ridge crest.

[1] New He isotopic data from the HSDP pilot hole core, lava accumulation rate models, and data from the literature are used to develop a 200,000 year isotopic record for the lava erupted from the Mauna Loa volcano. This record, coupled with an analogous record from Mauna Kea from the Hawaii Scientific Drilling Project (HSDP) pilot hole project and other literature data from the GEOROC database, are used to construct a “map” of lava isotopic compositions for the island of Hawaii. The isotopic map is converted to a map of the He and Nd isotopic compositions of melts from the mantle plume, which can be compared with a published melt supply map derived from geodynamic modeling. The resulting map of the plume indicates that values of helium 3He/4He > 20 Ra are confined to the core of the plume (radius ≈ 20–25 km) and correspond to potential temperatures >1565°C, suggesting the He isotopic signal is derived from deep in the mantle. The 3He/4He map has closed contours down to 10 Ra; the contours are teardrop-shaped and elongated in the general direction of plate motion. The closed contours indicate that most of the plume He signal is lost during the early stages of melting, which is consistent with helium behaving as a strongly incompatible element (KHe ≤ 0.001). The eNd contours (and by inference the contours for Sr, Pb, Hf, and Os) do not all close on the scale of the island of Hawaii but instead partially follow material flow lines within the plume beneath the lithosphere. The plume signal for Nd extends circa 100 km in the direction of plate motion, which is consistent with the moderately incompatible behavior of Nd (KNd ≈ 0.02). Downstream from the plume core epicenter, plume Nd occurs with asthenospheric He; this could be mistaken for an additional plume component, whereas it may be only a manifestation of differing incompatibility. Data from Mauna Loa suggest the presence of a low-3He/4He plume component that has low eNd and high 87Sr/86Sr. The plume map suggests that this component may be a blob (circa 20 km scale), located between Mauna Loa and Hualalai and separated from the main plume core by a zone of more asthenosphere-like material. The HSDP data preclude a proposed model where this material represents a ring of entrained material from the lower mantle. The orientation of the elongation of contours on the plume He and Nd isotope maps (∼N45°W) does not match the modern plate motion as measured by GPS (N65°W) nor does it match the trend of the ridge axis between Maui and Loihi (N30°W). The geochemical evidence, as well as the locations and growth histories of the Hawaiian volcanoes, suggest that the plume, as well as the Pacific plate, has been moving at a velocity of several centimeters per year over the past 1 to 2 million years.

Hawaiian volcanoes are the most comprehensively studied on Earth. Nevertheless, most of the eruptive history of each one is inaccessible because it is buried by younger lava flows or is exposed only below sea level. For those parts of Hawaiian volcanoes above sea level, erosion typically exposes only a few hundred meters of buried lavas (out of a total thickness of up to 10 km or more).Available samples of submarine lavas extend the time intervals of individual volcanoes that can be studied. However, the histories of individual Hawaiian volcanoes during most of their ~1-million-year passages across the zone of melt production are largely unknown.

The shapes of typical Hawaiian volcanoes are simply parameterized, and a relationship is derived for the dependence of lava accumulation rates on volcano volume and volumetric growth rate. The dependence of lava accumulation rate on time is derived by estimating the eruption rate of a volcano as it traverses the Hawaiian plume, with the eruption rate determined from a specified radial dependence of magma generation in the plume and assuming that a volcano captures melt from a circular area centered on the volcano summit. The timescale of volcano growth is t = 2 R/ν_plate where R is the radius of the melting zone of the (circular) plume and νplate is the velocity of the Pacific plate. The growth progress of a volcano can be described by a dimensionless time t′ = tν_plate/2R, where t′ = 0 is chosen to be the start of volcano growth and t′ = 1 approximates the end of “shield” growth. Using a melt generation rate for the whole plume of 0.2 km^(3)/yr, a plume diameter of 50 km, and a plate velocity of 10 cm/yr, we calculate that the lifetime of a typical volcano is 1000 kyr. For a volcano that traverses the axis of the plume, the “standard” dimensions are a volume of 57,000 km^3, a summit thickness of 18 km, a summit elevation of 3.6 km, and a basal radius of 60 km. The volcano first breaches the sea surface at t′ ≈ 0.22 when it has attained only 5% of its eventual volume; 80% of the volume accumulates between t′ = 0.3 and t′ = 0.7. Typical lava accumulation rates start out over 50 m/kyr in the earliest stages of growth from the seafloor, and level out at ∼35 m/kyr from t′ ≈ 0.05 until t′ = 0.4. From t′ = 0.4 to t′ = 0.9, the submarine lava accumulation rates decrease almost linearly from 35 m/kyr to ∼0; subaerial accumulation rates are about 30% lower. The lava accumulation rate is a good indicator of volcano age. A volcano that passes over the plume at a distance 0.4R off to the side of the plume axis is predicted to have a volume of about 60% of the standard volcano, a lifetime about 8% shorter, and lava accumulation rates about 15–20% smaller. The depth-age data for Mauna Kea lavas cored by the Hawaii Scientific Drilling Project are a good fit to the model parameters used, given that Mauna Kea appears to have crossed the plume about 15–20 km off-axis. The lifetime of Mauna Kea is estimated to be 920 kyr. Mauna Loa is predicted to be at a stage corresponding to t′ ≈ 0.8, Kilauea is at t′ ≈ 0.6, and Loihi is at t′ ≈0.16. The model also allows the subsurface structure of the volcanoes (the interfaces between lavas from different volcanoes) to be modeled. Radial geochemical structure in the plume may be blurred in the lavas because the volcanoes capture magma from a sizeable cross-sectional area of the plume; this inference is qualitatively born out by available isotopic data. The model predicts that new Hawaiian volcanoes are typically initiated on the seafloor near the base of the next older volcano but generally off the older volcano's flank.

Many volcanoes are associated with subduction zones or mid-ocean ridges, but other areas of unusually high volcanism (or "hotspots") do not have such a direct connection to plate tectonic processes. DePaolo and Manga argue in their Perspective that some hotspots are caused by narrow upwelling plumes originating in the lower mantle, while others are caused by broad, hot upwellings or "superswells." Yet others are not obviously associated with either deep plumes or superswells. Direct evidence for mantle plumes will require seismic imaging at higher resolution than is currently accessible. In a related Perspective, Foulger and Natland take the opposing view that hotspot volcanism is a by-product of plate tectonics and does not require deep plumes.

Temporal and spatial variations in the Nd isotopic compositions of Tertiary caldera-forming rhyolite tuffs, and Cretaceous and Tertiary granites of the western U.S.A. are used as a basis for a model that accounts for the observed proportions of crustal versus mantle contributions to silicic magmas in terms of two parameters: the ambient crustal temperature and the rate of supply of basaltic magma from the mantle. The crustal contribution to silicic igneous rocks is measured in terms of the Neodymium Crustal Index (NCI). The relationships between crustal temperature, basalt supply and NCI are quantified using a model of a magma chamber undergoing continuous recharge, wall-rock assimilation and fractional crystallisation