Deep thermohaline Circulation in the low-latitude Atlantic during the Last Glacial

Present-day low-latitude eastern and western Atlantic basins are geochemically distinct below the sill depth of the Mid-Atlantic Ridge. While Antarctic Bottom Water (AABW) circulates freely in the western Atlantic, flow into the eastern Atlantic is restricted below 4 km which results in filling the abyssal depths of this basin with water of geochemical similarity to nutrient depleted North Atlantic Deep Water. Using carbon isotopes and Cd/Ca ratios in benthic foraminifera we reconstruct the geochemistry of these basins during the last glacial maximum. Results indicate that deep eastern and western Atlantic basins became geochemically identical during the last glacial. This was achieved by shoaling of the upper surface of AABW above the sill depth of the Mid-Atlantic Ridge, which allowed bottom waters in both basins to be filled with the same water mass. Although AABW became the dominant water mass in the deep eastern Atlantic basin during the glacial, Holocene-glacial delta13C-PO4 shifts in this basin are in Redfield proportions, unlike the disproportionate Holocene-glacial delta13C-PO4 shifts observed in the Southern Ocean. By examining the composition of deep and intermediate waters throughout the Atlantic, we show that this effect was induced by a change in gradient of the delta13C-PO4 deepwater mixing line during glacial times. Evidence from high-latitude planktonic data suggests that the change in gradient of the deepwater mixing line was brought about through a significant reduction in the thermodynamic effect on Southern Ocean surface waters. By using coupled delta13C-PO4 data to constrain the composition of end member water masses in the glacial Atlantic, we conclude that deep waters in the low-latitude glacial Atlantic were composed of a mixture of northern and southern source waters in a ratio of 1:3.

Geochemistry of recent sediments from the Indian Ocean

During the International Indian Ocean Expedition (1964/65) sediment cores were taken on six profiles off the western coast of the Indian Subcontinent. These profiles run approximately perpendicular to the coast, from the deep-sea over the continental slope to the continental shelf. Additional samples and cores were taken in a dense pattern in front of the delta of the Indus River. This pattern of sampling covered not only marine sediments, but also river and beach sediments in Pakistan. The marine samples were obtained with piston, gravity and box corers and by a Van Veen grab sampler. The longest piston core is about 5 meters long. 1. Distribution of the elements on the sediment surface The area of maximal carbonate values (aprox. 80-100% CaCO3) essentially coincides with the continental shelf. The highest Sr values were observed largely within this area, but only in the vicinity of the Gulf of Cambay. Mainly the aragonitic coprolites are responsible for those high Sr contents. The Mg contents of the carbonates are comparatively low; surprisingly enough the highest Mg concentrations were also measured in the coprolites. The maximum contents of organic matter (Core) were found along the upper part of the continental slope. They coincide with the highest porosity and water content of the sediments. Frequently the decomposition of organic matter by oxydation is responsible for the measured Corg contents. On the other side the quantity of originally deposited organic material is less important in most cases. The enrichment of the "bauxitophile" elements Fe, Ti, Cr and V in the carbonate- and quartz-free portions of the sediments is essentially due to the influence of coarse terrigenous detritus. For the elements Mn, Ni and Cu (in per cent of the carbonateand quartz-free sediment) a strong enrichment was observed in the deep-sea realm. The strong increase in Mn toward the deep-sea is explained by authigenesis of Mn-Fe-concretions. Mn-nodules form only under oxydizing conditions which obviously are possible only at very low rates of deposition. The Mg, B and, probably also Mn contents in the clay minerals increase with increasing distance from the continent. This can be explained by the higher adsorption of those elements from sea water because of increasing duration of the clay mineral transport. The comparison of median contents of some elements in our deep-sea samples with deep-sea sediments described by TUREKIAN & WEDEPOHL (1961) shows that clear differences in concentration exist only in the case of "bauxitophile" elements Cr and Be. The Cr and Be contents show a clear increase in the Indian Ocean deep-sea samples compared to those described by TUREKIAn & WEDEPOHL (1961) which can obviously be attributed to the enrichment in the lateritic and bauxitic parent rocks. The different behaviour of the elements Fe, Ti and Mn during decomposition of the source rocks, transport to the sea and during oxydizing and reducing conditions in the marine environment can be illustrated by Ti02/Fe and MnO/Fe ratios. The different compositions of the sediments off the Indus Delta and those of the remaining part of the area investigated are characterized by a different distribution of the elements Mn and Ti. 2. Chemical inhomogenities in the sediments Most longer cores show 3 intervals defined by chemical and sedimentological differences. The top-most interval is coarse-grained, the intermedial interval is fine grained and the lower one again somewhat coarser. At the same time it is possible to observe differences from interval to interval in the organogenic and detrital constituents. During the formation of the middle interval different conditions of sedimentation from those active during the previous and subsequent periods have obviously prevailed. Looking more closely at the organogenic constituents it is remarkable that during the formation of the finer interval conditions of a more intensive oxydation have prevailed that was the case before and after: Core decreases, whereas P shows a relative increase. This may be explained by slower sedimentation rate or by a vertical migration of the oxygen rich zone of the sea-water. The modifications of the elements from minerals in detrital portion of the sediments support an explanation ascribing this fact to modifications of the conditions of denudation and transportation which can come about through a climatic change or through tectonic causes. The paleontological investigations have shown (ZOBEL, in press) that in some of the cores the middle stratum of fine sedimentation represents optimal conditions for organic life. This fact suggests also oxydizing conditions during the sedimentation of this interval. In addition to the depositional stratification an oxydation zone characterized by Mn-enrichment can be recognized. The thickness of the oxidation zone decreases towards the coast and thins out along the middle part of the continental slope. At those places, where the oxydation zone is extremely thin, enrichment of Mn has its maximum. This phenomenon can probably be attributed to the migration of Mn taking place in its dissociated form within the sediment under reducing conditions. On the other side this Mn-migration in the sediment does not take place in the deep-sea, where oxydizing conditions prevail. 3. Interstitial waters in the sediments Already at very small core depths, the interstitial waters have undergone a distinct modification compared with the overlying sea water. This distinct modification applies both to total salinity and to the individual ions. As to the beginning of diagenesis the following conclusions can be drawn: a) A strong K-increase occurs already at an early stage. It may be attributable to a diffusion barrier or to an exchange of Mg-ions on the clays. Part of this increase may also originate from the decomposition of K-containing silicates (mica and feldspars). A K-decrease owing to the formation of illite (WEAVER 1967), however, occurs only at much greater sediment depth. b) Because of an organic protective coating, the dissolution of carbonate is delayed in recent organogenic carbonates. At the same time some Ca is probably being adsorbed on clay minerals. Consequently the Ca-content of the interstitial water drops below the Ca-content of the sea water. c) Already at an early stage the Mg adsorption on the clays is completed. The adsorbed Mg is later available for diagenetic mineral formations and transformations.

Chemical and strontium isotopic compositions of interstitial waters from sediments of ODP Site 135-841

The depth variations in the major chemical components dissolved in interstitial waters from the Tonga margin (ODP Site 841) are much more pronounced than those usually observed in deep-sea sediments. The extensive alteration of volcanic Miocene sediments to secondary minerals such as analcime, clays, and thaumasite forms a CaCl2-rich brine. The brine results from a high exchange of Ca to Na, K, and Mg and an increase in Cl concentrations due to removal of H2O from the fluid during the authigenesis of hydrous minerals. The formation of thaumasite could have partly controlled the concentration of dissolved SO4, HCO3, and Ca in the Miocene sediments. The strontium isotopic signature of the interstitial water suggests that alteration of the volcanic Miocene sediments occurred a long time after sedimentation. A transient diffusion model indicates that molecular diffusion was not prevented by lithologic barriers and that the formation of secondary minerals in the Miocene sediment occurred over a short period of time (e.g.,

Lithium and strontium isotopic compositions of sediments and pore waters at ODP Sites 152-918 and 152-919

The distribution of Li isotopes in pore waters to a depth of 1157 m below seafloor is presented for ODP Sites 918 and 919 in the Irminger Basin, offshore Greenland. Lithium isotope data are accompanied by strontium isotope ratios to decipher diagenetic reactions in the sediments which are characterized by the pervasive presence of volcanic material, as well as by very high accumulation rates in the upper section. The lowering of the 87Sr/86Sr ratio below contemporaneous seawater values indicates several zones of volcanic material alteration. The Li isotope profiles are complex suggesting a variety of exchange reactions with the solid phases. These include cation exchange with NH4+ and mobilization from sediments at depth, in addition to the alteration of volcanic matter. Lithium isotopes are, therefore, a sensitive indicator of sediment-water interaction. d6Li values of pore waters at these two sites vary between -42 and -25?. At shallow depths (<100 mbsf), rapid decreases in the Li concentration, accompanied by a shift to heavier isotopic compositions, indicate uptake of Li into alteration products. A positive anomaly of d6Li observed at both sites is coincident with the NH4+ maximum produced by organic matter decomposition and may be related to ion exchange of Li from the sediments by NH4+. In the lower sediment column at Site 918, dissolved Li increases with depth and is characterized by enrichment of 6Li. The Li isotopic compositions of both the waters and the solid phase suggest that the enrichment of Li in deep interstitial waters is a result of release from pelagic sediments. The significance of sediment diagenesis and adsorption as sinks of oceanic Li is evaluated. The maximum diffusive flux into the sediment due to volcanic matter alteration can be no more than 5% of the combined inputs from rivers and submarine hydrothermal solutions. Adsorption on to sediments can only account for 5-10% of the total inputs from rivers and submarine hot springs.

Geochemistry of lavas from the Lau-Tonga arc and back-arc system

Detailed comparison of mineralogy, and major and trace geochemistry are presented for the modern Lau Basin spreading centers, the Sites 834-839 lavas, the modern Tonga-Kermadec arc volcanics, the northern Tongan boninites, and the Lau Ridge volcanics. The data clearly confirm the variations from near normal mid-ocean-ridge basalt (N-MORB) chemistries (e.g., Site 834, Central Lau Spreading Center) to strongly arc-like (e.g., Site 839, Valu Fa), the latter closely comparable to the modern arc volcanoes. Sites 835 and 836 and the East Lau Spreading Center represent transitional chemistries. Bulk compositions range from andesitic to basaltic, but lavas from Sites 834 and 836 and the Central Lau Spreading Center extend toward more silica-undersaturated compositions. The Valu Fa and modern Tonga-Kermadec arc lavas, in contrast, are dominated by basaltic andesites. The phenocryst and groundmass mineralogies show the strong arc-like affinities of the Site 839 lavas, which are also characterized by the existence of very magnesian olivines (up to Fo90-92) and Cr-rich spinels in Units 3 and 6, and highly anorthitic plagioclases in Units 2 and 9. The regional patterns of mineralogical and geochemical variations are interpreted in terms of two competing processes affecting the inferred magma sources: (1) mantle depletion processes, caused by previous melt extractions linked to backarc magmatism, and (2) enrichment in large-ion-lithophile elements, caused by a subduction contribution. A general trend of increasing depletion is inferred both eastward across the Lau Basin toward the modern arc, and northward along the Tongan (and Kermadec) Arc. Numerical modeling suggests that multistage magma extraction can explain the low abundances (relative to N-MORB) of elements such as Nb, Ta, and Ti, known to be characteristic of island arc magmas. It is further suggested that a subduction jump following prolonged slab rollback could account for the initiation of the Lau Basin opening, plausibly allowing a later influx of new mantle, as required by the recognition of a two-stage opening of the Lau Basin.

Major oxides, trace elements and rare earth elements of selected basalt samples at DSDP Hole 83-504B

DSDP Hole 504B is the deepest section drilled into oceanic basement, penetrating through a 571.5-m lava pile and a 209-m transition zone of lavas and dikes into 295 m of a sheeted dike complex. To define the basement composition 194 samples of least altered basalts, representing all lithologic units, were analyzed for their major and 26 trace elements. As is evident from the alteration-sensitive indicators H2O+, CO2, S, K, Mn, Zn, Cu, and the iron oxidation ratio, all rocks recovered are chemically altered to some extent. Downhole variation in these parameters enables us to distinguish five depth-related alteration zones that closely correlate with changes in alteration mineralogy. Alteration in the uppermost basement portion is characterized by pronounced K-uptake, sulfur loss, and iron oxidation and clearly demonstrates low-temperature seawater interaction. A very spectacular type of alteration is confined to the depth range from 910 to 1059 m below seafloor (BSF). Rocks from this basement portion exhibit the lowest iron oxidation, the highest H2O+ contents, and a considerable enrichment in Mn, S, Zn, and Cu. At the top of this zone a stockwork-like sulfide mineralization occurs. The chemical data suggest that this basement portion was at one time within a hydrothermal upflow zone. The steep gradient in alteration chemistry above this zone and the ore precipitation are interpreted as the result of mixing of the upflowing hydrothermal fluids with lower-temperature solutions circulating in the lava pile. Despite the chemical alteration the primary composition and variation of the rocks can be reliably established. All data demonstrate that the pillow lavas and the dikes are remarkably uniform and display almost the same range of variation. A general characteristic of the rocks that classify as olivine tholeiites is their high MgO contents (up to 10.5 wt.%) and their low K abundances (-200 ppm). According to their mg-values, which range from 0.60 to 0.74, most basalts appear to have undergone some high-level crystal fractionation. Despite the overall similarity in composition, there are two major basalt groups that have significantly different abundances and ratios of incompatible elements at similar mg-values. The majority of the basalts from the pillow lava and dike sections are chemically closely related, and most probably represent differentiation products of a common parental magma. They are low in Na2O, TiO2, and P2O5, and very low in the more hygromagmaphile elements. Interdigitated with this basalt group is a very rarely occurring basalt that is higher in Na2O, TiO2, P2O5, much less depleted in hygromagmaphile elements, and similar to normal mid-ocean ridge basalt (MORB). The latter is restricted to Lithologic Units 5 and 36 of the pillow lava section and Lithologic Unit 83 of the dike section. The two basalt groups cannot be related by differentiation processes but have to be regarded as products of two different parental magmas. The compositional uniformity of the majority of the basalts suggests that the magma chamber beneath the Costa Rica Rift reached nearly steady-state conditions. However, the presence of lavas and dikes that crystallized from a different parental magma requires the existence of a separate conduit-magma chamber system for these melts. Occasionally mixing between the two magma types appears to have occurred. The chemical characteristics of the two magma types imply some heterogeneity in the mantle source underlying the Costa Rica Rift. The predominant magma type represents an extremely depleted source, whereas the rare magma type presumably originated from regions of less depleted mantle material (relict or affected by metasomatism).

Geochemistry of Loihi Seamount hydrothermal deposits

High-resolution bathymetric surveys, bottom photography and sample analyses show that Loihi Seamount at the southernmost extent of the Hawaiian ëhotspotí is an active, young submarine volcano that is probably the site of an emerging Hawaiian island. Hydrothermal deposits sampled from the active summit rift system were probably formed by precipitation from cooling vent fluids or during cooling and oxidation of high-temperature polymetallic sulphide assemblages. No exotic benthic fauna were found to be associated with the presently active hydrothermal vents mapped.

Geochemical and isotopic composition of pore fluids of ODP Hole 131-808C

At the Western Nankai Trough subduction zone at ODP Site 808, chemical concentration and isotopic ratio depth profiles of D, O, Sr, and He do not support fluid flow along the décollement nor at the frontal thrust. They do, however, support continuous or periodic lateral fluid flow: (1) at the base of the Shikoku Basin volcanic-rich sediment member, situated ~140 m above the décollement, and particularly (2) below the décollement. The latter must have been rather vigorous, as it was capable of transporting clay minerals over great distances. The fluid at ~140 m above the décollement is characterized by lower than seawater concentrations of Cl- (>=18% seawater dilution). It is 18O-rich and D-poor and has a non-radiogenic, oceanic, or volcanic arc Sr isotopic signature. It originates from "volcanic" clay diagenesis. The fluid below the décollement has also less Cl- than seawater (>20% dilution), is more enriched in 18O and depleted in D than fluid, but its Sr isotopic signature is radiogenic, continentalterrigenous. The source of this fluid is located arcward, is deep-seated, where illitization of the subducted clay minerals, a mixture of terrigenous and volcanic clays, occurs. The 3He/4He ratio below the décollement points to an ~25% mantle contribution. The nature of the physical and chemical discontinuities across the décollement suggests it is overpressured and is forming a leaky "dynamic seal" for fluid flow. In contrast with the situation at Barbados and Peru, where the major tectonic features are mineralized, here, although the complex is extremely fractured and faulted, mineralized macroscopic veins, fractures, and faults are absent. Instead, mineralized microstructures are widespread, indicating a diffuse mode of dewatering.

Element and Sr isotope composition of interstitial waters in sediments of ODP Leg 129

Interstitial water samples from Leg 129, Sites 800, 801, and 802 in the Pigafetta and Mariana basins (central western Pacific), have been analyzed for major elements, B, Li, Mn, Sr, and 87Sr/86Sr. At all sites waters show enrichment in Ca and Sr and are depleted in Mg, K, Na, SO4, B, alkalinity, and 87Sr compared to seawater. These changes are related to alteration of basaltic material into secondary smectite and zeolite and recrystallization of biogenic carbonate. Water concentration depth profiles are characterized by breaks due to the presence of barriers to diffusion such as chert layers at Sites 800 and 801 and highly cemented volcanic ash at Site 802. In Site 800, below a chert layer, concentration depth profiles are vertical and reflect slight alteration of volcanic matter, either in situ or in the upper basaltic crust. Release of interlayer water from clay minerals is likely to induce observed Cl depletions. At Site 801, two units act as diffusion barrier and isolate the volcaniclastic sediments from ocean and basement. Diagenetic alteration of volcanic matter generates a chemical signature similar to that at Site 800. Just above the basaltic crust, interstitial waters are less evolved and reflect low alteration of the crust, probably because of the presence in the sediments of layers with low diffusivities. At Site 802, in Miocene tuffs, the chemical evolution generated by diagenetic alteration is extreme (Ca = 130 mmol, 87Sr/86Sr = 0.7042 at 83 meters below seafloor) and is accompanied by an increase of the Cl content (630 mmol) due to water uptake in secondary hydrous phases. Factors that enhance this evolution are a high sediment accumulation rate, high cementation preventing diffusive exchange and the reactive composition of the sediment (basaltic glass). The chemical variation is estimated to result in the alteration of more than 20% of the volcanic matter in a nearly closed system.

Geochemistry and mineralogy of ODP Leg 107 basalts

During ODP Leg 107, the basement of the Tyrrhenian Sea was drilled at Site 650, located in the Marsili basin, and at Sites 651 and 655, both located in the Vavilov basin. In addition, a lava flow was drilled at Site 654 on the Sardinia rifted margin. Mineral and whole rock major and trace element chemistry, including rare earth element (REE) and Sr and Nd isotopic ratios, were determined in samples of these rocks. Site 654 lava was sampled within uppermost Pliocene postrift sediments. This lava is a basaltic andesite of intraplate affinity, and is analogous to some Plio-Pleistocene tholeiitic lavas from Sardinia. Site 650 basalts, drilled beneath 1.7-1.9-Ma-old basal sediment, are strongly altered and vesicular suggesting a rapid subsidence of the Marsili basin. Based on incompatible trace elements, these basalts show calc-alkaline affinity like some products of the Marsili Seamount and the Eolian arc. The basement of the two sites drilled within Vavilov basin shows contrasting petrologies. Site 655, located along the Gortani ridge in the western part of the basin, drilled a 116-m-thick sequence of basalt flows beneath 3.4-3.6-Ma-old basal sediments. These basalts are chemically relatively homogeneous and show affinity to transitional MORB. Four units consisting of slightly differentiated basaltic lavas, have been identified. Site 655 basalts are geochemically similar to the high Ti lavas from DSDP Leg 42, Site 373 (Vavilov Basin). The basement at Site 651, overlain by 40 m of metalliferous dolostone covered by fossiliferous sediments with an age of 2 Ma, consists of two basalt units separated by a dolerite-albitite intrusive body; serpentinized harzburgites were drilled for 30 m at the base of the hole. The two basalt units of Site 651 are distinct petrochemically, though both show incompatible elements affinity with high-K calc-alkaline/calc-alkaline magmas from Eolian arc. The cpx chemistry and high K/Na ratio of the lower unit lavas suggest a weak alkaline tendency of potassic lineage. Leg 107 basement rock data, together with data from DSDP Site 373 and from dredged samples, indicate that the deepest basins of the central Tyrrhenian Sea are underlain by a complex back-arc basin crust produced by magmas with incompatible element affinities to transitional MORB (Site 655 and DSDP Site 373), and to calc-alkaline and high-K calc-alkaline converging plate margin basalts (Sites 650 and 651). This petrogenetic complexity is in accordance with the back-arc setting of the Vavilov and Marsili basins. Other back-arc basin basalts, particularly those from ensialic basins such as the Bransfield Strait (Antarctica), show a comparable petrogenetic complexity (cf., Sounders and Tarney, 1984).

Contaminant concentrations in ivory gull (Pagophila eburnea) eggs from Svalbard and the Russian Arctic (2006-2007)

We found high levels of contaminants, in particular organochlorines, in eggs of the ivory gull Pagophila eburnea, a high Arctic seabird species threatened by climate change and contaminants. An 80% decline in the ivory gull breeding population in the Canadian Arctic the last two decades has been documented. Because of the dependence of the ivory gull on sea ice and its high trophic position, suggested environmental threats are climate change and contaminants. The present study investigated contaminant levels (organochlorines, brominated flame retardants, perfluorinated alkyl substances, and mercury) in ivory gull eggs from four colonies in the Norwegian Svalbard) and Russian Arctic (Franz Josef Land and Severnaya Zemlya). The contaminant levels presented here are among the highest reported in Arctic seabird species, and we identify this as an important stressor in a species already at risk due to environmental change.

Chemical composition of Fe-Mn nodules and host bottom sediments along the submeridional profile across the Brazil Basin

Sedimentation and ore formation were studied in sediments from nine stations located along the 24°W profile in the Brazil Basin of the Atlantic Ocean. Bottom sediments are represented by mio- and hemipelagic muds, which are variably enriched in hydrothermal iron and manganese oxyhydroxides. As compared to bottom sediments from other basins of the Atlantic Ocean, the sediments in study are marked by extremely high manganese contents (up to 1.33%) and maximal enrichment in Ce. It was shown that the positive Ce anomaly is related to REE accumulation on iron oxyhydroxides. Influence of hydrothermal source leads to decrease of Ce anomaly and LREE/HREE ratio. In reduced sediments preservation of positive Ce anomaly and/or its disappearance was observed after iron and manganese reduction. REE contents were determined for the first time in the Ethmodiscus oozes of the Brazil Basin. Ore deposits of the Brazil Basin are represented by ferromanganese crusts and ferromanganese nodules. Judging from contents of iron, manganese, REE, and other trace elements, these formations are ascribed to sedimentation (hydrogenic) deposits. They are characterized by a notable positive Ce anomaly in the REE pattern. Extremely high Ce content (up to 96% of total REE) was discovered for the first time in the buried nodules (Mn/Fe = 0.88).

Geochemistry of sands and sandstones in ODP Leg 126 sites

During Ocean Drilling Program Leg 126, six sites were cored in a young backarc rift basin and its flanks (rift onset 1.1-3.56 Ma) and in the forearc basin of the Izu-Bonin Arc. In the backarc area, strata are younger than about 4.5 Ma, whereas in the forearc, ages are about 0-31 Ma in sections punctuated by important Miocene unconformities. Bulk chemical analyses of volcaniclastic turbidite sands and sandstones, derived directly from the arc, were obtained from 271 atomic absorption analyses (major elements), 253 XRF analyses (trace elements) and 16 ICP-MS analyses (trace and rare-earth elements). Of the 271 samples, 78 come from the backarc area and the remainder from the forearc. The sands and sandstones reflect the igneous compositions of their sources. Most are formed of materials derived from subalkaline, low-K andesites, and dacites, although compositions range from basalt to rhyolite. Basic and acid andesites are predominant in Oligocene rocks; in contrast, Pliocene-Pleistocene sediments were derived from acid andesitic to rhyolitic sources. The oldest sandstones, estimated to have an age of about 31 Ma, were derived from an arc tholeiitic, not boninitic, source. The 26-31 Ma sandstones furthest to the north, at Sites 787 and 792, have higher relative concentrations of Ti, Zr, and Y than do those at southern Site 793. Data from younger samples indicate that, for more than 30 m.y., the average composition of volcaniclastic sediments and volcanism near Aoga Shima was more basic than to the south, near Sumisu Jima. Using the sandstones as igneous proxies, we conclude that magmas erupted along the arc have become more depleted in light-rare-earth elements (LREE) with time. There was a major change in rare-earth-element (REE) concentrations in the late Oligocene, from essentially flat patterns (normalized La/Yb about 1-1.5) to LREE-depleted patterns (normalized La/Yb about 0.5). At the same time, Zr/Y ratios decreased from about 2-4 to about 1.5-2.5. These changes may reflect a shift in provenance, or changes in the composition of the mantle wedge beneath the arc. In the backarc area, lithic clasts and glass shards of rift-facies basalt are present in sediments as old as 2.35-3.15 Ma. Two samples of mafic sand from the backarc basin have flat REE patterns (normalized La/Yb about 1.0), like some of the <1-Ma rift lavas and unlike pre-rift sand and sandstone samples. These possibly represent the local effects of sedimentary mixing of detritus from arc and backarc eruptions because no evidence from the arc itself exists to suggest a recent change in the REE content of magmas.