Large-scale Mantle Heterogeneity Research Articles

Automatic measurement and quality control of S3KS-SKKS differential traveltimes and the influence of mantle heterogeneity

SUMMARY The velocity structure of the uppermost part of the outer core can be investigated seismologically based on the differential traveltimes between the SKKS and S3KS waves (S3KS-SKKS). We develop a procedure for the automatic measurement and quality assessment of the S3KS-SKKS differential traveltime measurements. Four criteria with empirically determined thresholds are used to quantify the quality of the measurements in our automatic algorithm, which greatly simplifies the data quality control. Measurements obtained in the epicentral distance range of 120–140° display positive residuals relative to PREM with no obvious spatial dependency. Waveform simulations in both PREM and S40RTS models show that neither finite-frequency effects nor large-scale mantle heterogeneities can fully explain the magnitude of the observed residuals. Mantle heterogeneities have a significant contribution to the data uncertainty, but do not have sufficient effect on the mean finite-frequency correction. We thus conclude that the observed residuals in the S3KS-SKKS differential traveltimes after finite-frequency correction are mainly caused by the velocity structure in the uppermost outer core.

Dynamical links between small- and large-scale mantle heterogeneity: Seismological evidence

We identify PKP•PKP scattered waves (also known as P′•P′) from earthquakes recorded at small-aperture seismic arrays at distances less than 65°. P′•P′ energy travels as a PKP wave through the core, up into the mantle, then scatters back down through the core to the receiver as a second PKP. P′•P′ waves are unique in that they allow scattering heterogeneities throughout the mantle to be imaged. We use array-processing methods to amplify low amplitude, coherent scattered energy signals and resolve their incoming direction. We deterministically map scattering heterogeneity locations from the core–mantle boundary to the surface. We use an extensive dataset with sensitivity to a large volume of the mantle and a location method allowing us to resolve and map more heterogeneities than have previously been possible, representing a significant increase in our understanding of small-scale structure within the mantle. Our results demonstrate that the distribution of scattering heterogeneities varies both radially and laterally. Scattering is most abundant in the uppermost and lowermost mantle, and a minimum in the mid-mantle, resembling the radial distribution of tomographically derived whole-mantle velocity heterogeneity. We investigate the spatial correlation of scattering heterogeneities with large-scale tomographic velocities, lateral velocity gradients, the locations of deep-seated hotspots and subducted slabs. In the lowermost 1500 km of the mantle, small-scale heterogeneities correlate with regions of low seismic velocity, high lateral seismic gradient, and proximity to hotspots. In the upper 1000 km of the mantle there is no significant correlation between scattering heterogeneity location and subducted slabs. Between 600 and 900 km depth, scattering heterogeneities are more common in the regions most remote from slabs, and close to hotspots. Scattering heterogeneities show an affinity for regions close to slabs within the upper 200 km of the mantle. The similarity between the distribution of large-scale and small-scale mantle structures suggests a dynamic connection across scales, whereby mantle heterogeneities of all sizes may be directed in similar ways by large-scale convective currents.

Redox-induced lower mantle density contrast and effect on mantle structure and primitive oxygen

The origin of large-scale mantle heterogeneities remains enigmatic. Experiments show that different oxygen fugacities lead to density differences in lower-mantle materials, which lead to a heterogeneously oxidized mantle in simulations.

5125 Verifying CTV-PTV margins for isocentric breast cancer radiotherapy, using an off-line correction protocol and fixed couch height

Continental flood volcanism of the Paraná basin (Lower Cretaceous) is represented by two-pyroxene tholeiitic basalts (90 vol. %). The Northern Paraná Province (NPP) is dominated by basalts high in TiO2 and incompatible elements (HTiB), while the Southern Paraná Province (SPP) is dominated by basalts low in TiO2 and incompatible elements (LTiB).NPP basalts show relatively small variations of initial (120 Ma) 87Sr86Sr (R0) and 143Nd144Nd (Nd∗) ratios (R0=0.7051−0.7062 and Nd∗=0.5124−0.5125, respectively) relative to those occurring in SPP (R0=0.7046−0.7120 and Nd∗=0.5122−0.5128, respectively). The latter basalts show significant positive correlations between R0 vs. SiO2, K2O, Rb and Ba, and negative correlations between R0 vs. (Cr + Ni) and mg-value, believed to be due to crustal “granitic” contamination. In general, the effects of contamination are pronounced in SPP and tend to vanish towards NPP. δ18O-values range from +6.5 to +10.0‰ in the basalts from NPP and SPP and essentially reflect water-magma interactions.“Uncontaminated” LTiB and HTiB basalts from NPP are isotopically distinct from the “uncontaminated” analogues from SPP (LTiB) (R0=0.7055 vs. 0.7046; Nd∗=0.5124 vs. 0.5128, respectively). All chemical data consistently indicate distinct sources and a large-scale mantle heterogeneity. The NPP mantle source is expected to have relatively high content of “enriched” components, possibly related to small-volume melts and metasomatic fluids. Tentatively, the last stabilization age of the Paraná mantle heterogeneity is 0.5–1.0 Ga old.The existence of high- and low-TiO2 basalt suites in both Paraná and Karoo provinces in Brazil and southern Africa, respectively, indicates a large-scale heterogeneity in the subcontinental mantle, and suggests that basalt generation occurred in the lithospheric mantle.

Isotope systematics of Li, Sr, Nd, and volatiles in Indian Ocean MORBs of the Rodrigues Triple Junction: Constraints on the origin of the DUPAL anomaly

The DUPAL anomaly, a radiogenic isotope anomaly discovered in the Indian Ocean mantle, has been interpreted as due to a large-scale mantle heterogeneity. To provide new constraints on the DUPAL origin, we analyzed isotope ratios of Li, Sr, and Nd in fresh N-MORB glasses recovered from the Rodrigues Triple Junction in the Indian Ocean, and from the North Atlantic. The Li isotopic compositions of the Indian Ocean DUPAL N-MORBs were comparable to those of the North Atlantic non-DUPAL N-MORBs. The source of the DUPAL signature in Indian Ocean MORBs and the E-MORB-type enriched mantle source have quite different Li isotopic compositions. The 143Nd/ 144Nd values of both sources are significantly lower than those of the North Atlantic N-MORBs. The δ 7Li values of most oceanic island basalts with similar low 143Nd/ 144Nd signatures are also higher than those of the North Atlantic N-MORBs, except for several Koolau lavas. The Li isotope results support the recent proposal that significant amounts of recycled lower continental crust might produce the radiogenic isotope signatures of the Indian Ocean DUPAL source.

Constraining large-scale mantle heterogeneity using mantle and inner-core sensitive normal modes

Attempts to resolve density heterogeneity within the mantle using normal-mode data revealed an unexpected feature near the core-mantle boundary: regions of strong sheer velocity reduction, known as superplumes, are characterized by heavier than average material [Science 285 (1999) 1231]. Thus far, free-oscillation studies of mantle structure have relied upon modes which sample only the mantle (e.g., [ J. Geophys. Res. 96 (1991) 551; Science 285 (1999) 1231; J. Geophys. Res. 104 (1999a) 993; Geophys. J. Int. 143 (2000b) 478; Geophys. J. Int. 150 (2002) 162]). In order to better constrain mantle heterogeneity, we add inner-core sensitive modes to our data set, and invert for mantle structure and inner-core anisotropy simultaneously. The additional modes do not alter the pattern of the density anomalies relative to model SPRD6 [Geophys. J. Int. 145 (2001) 77]. However, they help constrain the amplitude of lateral variations in density. In a previous study, obtaining a density model with reasonable amplitudes required supplemental gravity data, which are not included in the current study. Nonetheless, the new density model is generally weaker, especially at the bottom of the mantle. The root-mean square (RMS) amplitude of the model exhibits two maxima around 600 and 2300 km depth. Experiments indicate that these are robust components of the model and neither artifacts of the choice of radial basis functions nor the damping scheme. Comparisons of the density model with shear- and compressional-velocity models show negative or nearly zero correlation in the transition zone and near the base of the mantle, where the root-mean square density amplitude is high. Finally, the anomalous features near the core-mantle boundary are confirmed: strong anti-correlation between shear and bulk-sound velocities, and increased density at the locations of superplumes.

The Early Proterozoic carbonatite complex of Angico dos Dias, Bahia State, Brazil: geochemical and Sr-Nd isotopic evidence for an enriched mantle origin

AbstractBorehole samples of carbonatites and phlogopite-pyroxenites from the Angico dos Dias (AdD) intrusive alkaline complex, State of Bahia, Brazil, have been investigated in terms of mineralogy, geochemistry and C-O-Sr-Nd isotopes. The AdD complex, of Early Proterozoic age (2 Ga), intrudes the northern side of the São Francisco Craton. Mineralogy and petrography indicate that the studied rocks only partially preserved their magmatic textural features owing to their metamorphic re-equilibration (greenschist facies). The REE contents and LREE/HREE ratios of the AdD carbonatites are very high (mean 3979±718 ppm and La/Yb = 215±23, respectively), as for most Precambrian magmatic carbonatites. The AdD carbonatites are also enriched in 18O (δ18O = 11.9 to 15.8‰), possibly due to secondary processes (e.g. metamorphism, alteration) whereas carbon isotopes are in the range of ‘primary carbonatites’ (δ13C = –5.7 to –7.1‰). Most of the initial 87Sr/86Sr and 143Nd/144Nd values of the studied carbonatites were not appreciably modified by secondary processes. Their εtSr and εtNd values (20.0 to 25 and 0.7 to –4.5, respectively) indicate enriched mantle sources very different from the ‘depleted’ ones related to many Precambrian carbonatites from North America (0.6 –2.6 Ga) and Africa (0.5 –2.0 Ga). The Early Proterozoic Sr-Nd isotopic signatures of the AdD carbonatites are similar to those of the Early Cretaceous carbonatites from the Paraná basin. The latter carbonatites show a great isotopic variability ranging from Bulk Earth to the related potassic magmatism from Asunción-Sapucai graben in the Eastern Paraguay (K-ASU magmatism: εtSr = 35 to 50 and εtNd = –12 to –20). The very similar isotopic compositions of Precambrian and post-Palaeozoic carbonatites worldwide indicate that the subcontinental mantle variability lasted for long periods of time and indicate a large-scale mantle heterogeneity.

Mantle Mixing: The Generation, Preservation, and Destruction of Chemical Heterogeneity

▪ Abstract Observations of the geochemical diversity of mid-oceanic ridge and ocean-island basalts have traditionally been attributed to the existence of large-scale mantle heterogeneity. In particular, the layered convection model has provided an important conceptual basis for discussing the chemical evolution of the Earth. In this model, a long-term boundary is assumed between a well-mixed and depleted upper mantle and a heterogeneous and more primitive lower mantle. The existence of high3He/4He in ocean-island sources has been used to argue for the preservation of a primitive component in the deep mantle. Nevertheless, a primitive deep layer is difficult to reconcile with the abundant lithophile isotopic evidence for recycling of oceanic crust and the lack of preservation of primitive mantle. In addition, the widespread acceptance of geophysical evidence for whole mantle flow has made straightforward application of the layered convection model problematic. Model calculations show that whole mantle convection with present day heat flow and surface velocities is sufficiently vigorous to mix large-scale heterogeneity to an extent that is incompatible with the geochemical observations. Several concepts have been proposed in recent years to resolve the apparent conflicts between the various observational constraints and theoretical interpretations. The suggestions include the presence of deeper layering, preservation of highly viscous blobs, core mantle interaction, and strong temporal variations in mantle dynamics. Although these models generally appear to solve parts of the puzzle, at present no single model is able to account for all of the major observations. The reconciliation of conflicting evidence awaits improvements in observational and experimental techniques integrated with better model testing of hypotheses for the generation and destruction of mantle heterogeneity.

Microanalysis of δ13C, δ15N, and N abundances in diamonds by secondary ion mass spectrometry

This paper describes in situ analysis methods for determination of nitrogen abundance, δ 13C, and δ 15N in diamond by secondary ion mass spectrometry (SIMS). The simple chemical composition of diamond results in the absence of matrix effects on instrumental mass fractionation (IMF) of carbon and nitrogen. Carbon isotopes are analyzed using extreme energy filtering (EEF), while nitrogen abundance and isotope ratios are measured at high mass resolution (HMR) without energy filtering. Because the variable surface properties of different sample mounts can result in variable measurements of IMF, especially using HMR techniques, precise application of IMF values derived from standard measurements requires the co-location of standards and samples on the same probe mount. With care given to instrument alignment and standardization, δ 13C in diamond can be determined to ±0.3‰ (1 σ) total uncertainty (precision+accuracy). Reproducibility of δ 15N measurements can be as good as ±1‰ (1 σ) depending on N abundance, but standard heterogeneity currently limits the accuracy of δ 15N data to ±3‰ (1 σ). Nitrogen abundance measurements, obtained during δ 15N measurements, have an accuracy of ±10% and a detection limit of 0.5 ppm. The 25-μm spatial resolution of this technique permits isotopic measurements at the scale of diamond growth zoning observed by cathodoluminescence (CL) imaging. Measurements on a central polished section of a Siberian diamond show variations in δ 13C (−8.8‰ to +1.7‰), δ 15N (−17.1‰ to +5.3‰) and N abundance (14–1390 ppm) associated with CL-imaged growth zones, which span a significant fraction of the total isotopic variability observed in bulk analyses of individual diamonds. In situ δ 13C, δ 15N, and N abundance measurements, in conjunction with CL imaging and FTIR N aggregation measurements, hold promise for determining the extent of isotopic variability which can be attributed to diamond formation processes, and the significance of isotopic variability in diamonds to large-scale mantle heterogeneity and the origin of terrestrial volatiles.

Whole Mantle Convective Model Interpretation of Oceanic Basalt Compositions: Are the Geodynamical Inferences of This Model Realistic?

The geochemical composition of oceanic basalts should represent a very interesting tool to discuss this problem (Hofmann, 1997). It can actually be thought that distinct geochemical characteristics should characterize oceanic basalts derived from mantle reservoirs evolving either through a layered or a whole mantle convective mode. In the large scale mantle convective mode for example, the time span from one convective turn to another would be rather large. According to present day plate velocities, and taking into account the generally admitted increase of mantle viscosity with depth, time span should be quite elevated and could reach half a billion year. Consecutively, in this convective mode, mixing of mantle materials within each convective cell and between convective cells will be minimized, allowing large scale mantle heterogeneities to persist. Mantle materials magmatically differentiated along the convective turns, such as the oceanic crust and complementary residual mantle could remain close one to another along several periods. If mid oceanic ridge basalts (MORB) and oceanic island basalts (OIB) form in the same convective cell from such magmatically differentiated sources, relationships between MORB and OIB compositions can be expected. The differentiation ages of the mantle sources indicated by the basalts compositions could be significantly old reaching possibly several billion years for differentiated sources which realized several convective turns without being involved in new differentiation processes. The geochemical approach, however, presents difficulties. Interpretations of the oceanic basalts data can be numerous depending on the assumed nature of their sources. Furthermore within a model frame, expected compositions can extend on a large range, depending, for example, on melt tapping processes and more generally mantle geodynamics. These factors remain nowadays a matter of debate. To check the plausibility of a model, an interesting work is ultimately to discuss the geodynamical constrains used in the modelling or obtained from it. Oceanic basalts database and whole convective mantle modelling

Peridotite xenoliths in alkali basalts from the Sikhote-Alin, southeastern Siberia, Russia: trace-element signatures of mantle beneath a convergent continental margin

Abundant spinel peridotite xenoliths occur in late Cenozoic alkali basaltic rocks in the Sikhote-Alin region at the Pacific margin of the Asian continent. Major- and trace-element compositions of representative peridotite xenolith are documented for four occurrences located in different structural units of the continental margin. In each locality, the majority of xenoliths have distinctive microstructures, modal and chemical compositions that are typical for a given xenolith suite. Significant textural and compositional differences between the four xenolith suites suggest that the upper mantle beneath the Sikhote-Alin consists of distinct domains with contrasting composition. The inferred large-scale mantle heterogeneities may be due to juxtaposition of lithospheric blocks of different provenance during accretion of the Sikhote-Alin to the Asian continent. Trace-element patterns of the xenoliths and their minerals obtained ICP-MS technique provide evidence of depletion and enrichment events and indicate contrasting behaviour of REE, HFSE and other incompatible trace elements. The HFSE behave non-concordantly, in particular, some xenoliths have highly fractionated Zr/Hf, Ti/Zr, Nb/Ta, La/Nb and U/Th ratios relative to their values in the primitive mantle. The fractionated compositions may be related to the interaction of evolved subduction-related fluids and melts with lithospheric mantle at the Mesozoic-early Cenozoic active continental margin or to metasomatism during later continental rifting.

Reading the Moon's volcanic record by ion microprobe analysis of Apollo 14 glass beads

The application of secondary ion mass spectrometry to lunar volcanic problems is demonstrated by individually analyzing representative glass from the seven Apollo 14 pyroclastic glass bead groups (black, orange, yellow, LAP, green A, VLT, and green B) for selected trace elements (e.g., rare earth elements [REE], Ba, Sr, Zr, V, and Co). The trace element characteristics of glass beads are useful for differentiating impact from volcanic glass beads and identifying other volcanic glass types (e.g., LAP). Trace element modeling indicates that the bead groups are unrelated by low-pressure fractional crystallization to each other, to Apollo 14 crystalline basalts, or to basalts from other landing sites. A possible exception is the relation between LAP and Apollo 14 aluminous basalts. The absence of evolved basalts derived from primary magmas with volcanic glass compositions suggests either that these evolved basalts have not been sampled or that fire fountaining tapped mantle sources subtly different from crystalline mare-basalt source regions. Hybridization of mantle source regions is preferred to assimilation-fractional crystallization processes to explain the incorporation of the evolved potassium-REE-phosphorus (KREEP) component identified in these primitive magmas. Apollo 14 volcanic glass and mare basalt trace element signatures indicate that the character of the Apollo 14 mantle source region is intrinsically different from that of other sites, which suggests that large-scale mantle heterogeneities exist.

Regional variations within the Paraná flood basalts (southern Brazil): Evidence for subcontinental mantle heterogeneity and crustal contamination

Continental flood volcanism of the Paraná basin (Lower Cretaceous) is represented by two-pyroxene tholeiitic basalts (90 vol. %). The Northern Paraná Province (NPP) is dominated by basalts high in TiO 2 and incompatible elements (HTiB), while the Southern Paraná Province (SPP) is dominated by basalts low in TiO 2 and incompatible elements (LTiB). NPP basalts show relatively small variations of initial (120 Ma) 87 Sr 86 Sr ( R 0) and 143 Nd 144 Nd (Nd∗) ratios ( R 0=0.7051−0.7062 and Nd∗=0.5124−0.5125 , respectively) relative to those occurring in SPP ( R 0=0.7046−0.7120 and Nd∗=0.5122−0.5128 , respectively). The latter basalts show significant positive correlations between R 0 vs. SiO 2, K 2O, Rb and Ba, and negative correlations between R 0 vs. (Cr + Ni) and mg-value, believed to be due to crustal “granitic” contamination. In general, the effects of contamination are pronounced in SPP and tend to vanish towards NPP. δ 18O-values range from +6.5 to +10.0‰ in the basalts from NPP and SPP and essentially reflect water-magma interactions. “Uncontaminated” LTiB and HTiB basalts from NPP are isotopically distinct from the “uncontaminated” analogues from SPP (LTiB) ( R 0=0.7055 vs. 0.7046; Nd∗=0.5124 vs. 0.5128, respectively). All chemical data consistently indicate distinct sources and a large-scale mantle heterogeneity. The NPP mantle source is expected to have relatively high content of “enriched” components, possibly related to small-volume melts and metasomatic fluids. Tentatively, the last stabilization age of the Paraná mantle heterogeneity is 0.5–1.0 Ga old. The existence of high- and low-TiO 2 basalt suites in both Paraná and Karoo provinces in Brazil and southern Africa, respectively, indicates a large-scale heterogeneity in the subcontinental mantle, and suggests that basalt generation occurred in the lithospheric mantle.

Lower mantle high-velocity zone beneath the Kurils as inferred from P-wave travel time and amplitude data.

We attempt to detect a high-velocity zone in the lower mantle beneath the Okhotsk Sea by analyzing travel time and amplitude data for teleseismic P-waves from intermediate and deep earthquakes in the Kurils. Travel time residuals with respect to the J-B tables are obtained by correcting ISC arrival time data for the Earth's ellipticity, station anomaly, and station elevation and are then inverted to obtain a P-velocity model for depths shallower than 1, 250 km in the Kuril subduction zone. The velocity model exhibits a dipping high-velocity zone down to a depth of 1, 200 km in the lower mantle immediately beneath the Kuril deepest seismicity. The velocity in the dipping zone is 2-3% higher than the ambient lower mantle. Resolution analysis supports the existence of a slablike high-velocity zone, but indicates that the depth extent of the zone cannot be well resolved by travel time data alone. We further constrain the velocity model by analyzing amplitude data, which are more sensitive to fine velocity structure than travel times. After correcting the amplitude data for the source radiation pattern, geometrical spreading, station anomaly, and instrument response, we obtain the amplitude anomaly due to laterally heterogeneous earth structure. The amplitude at a period of 1.5 s decreases by a factor of 2 to 4 at stations in the distance range between 50-65 and with azimuths perpendicular to the Kuril trench. To separate the effects of slab structure from those of large-scale mantle heterogeneity, we model the amplitude anomaly using a hybrid method combining a finite element (FEM) scheme with geometrical optics. The high-velocity zone obtained by the travel time inversion produces an amplitude reduction for the short-period P-waves which is comparable to the observed amplitude reduction. This suggests that the observed amplitude anomaly is primarily due to the lower mantle high-velocity zone. Comparison of the observed and theoretical amplitudes shows that a much larger velocity gradient is required near the upper boundary of the dipping high-velocity zone than near the lower boundary. A high-velocity zone with a maximum depth of 1, 000 km, thickness of 150 km, and velocity contrast of 3.5% gives the best fit to the amplitude data. Similarities in the velocity structure and the dip angle between the lower mantle high-velocity zone found by the present study and the upper mantle slab suggest that the former represents a lithospheric slab penetrating into the lower mantle.

Proterozoic Mantle Heterogeneity: Geochemical Evidence from Contrasting Basic Dykes

Summary Two swarms of early Proterozoic basic dykes occur in the Archaean craton of SW Greenland. The MD dykes are tholeiitic dolerites, while the BN dykes are norites. Virtually identical norites occur with tholeiitic dolerites in the Scourie swarm of NW Scotland and in the Laramie, Bighorn and Beartooth ranges of Wyoming, USA. Close geochemical similarities between typical coarse-grained norites and quenched varieties in SW Greenland and Wyoming suggest that the former type has not been strongly modified by crystal accumulation. The mineralogy and geochemistry of the quenched norites closely resemble those of boninites. Most of the norites have high MgO ( c. 13–18 wt%) and correspondingly high Cr and Ni, but they also have relatively high silica, alkali and large ion lithophile (LIL) element abundances. They are all light rare-earth element (REE) enriched (La = 40 times average CI chondrite, La N /Lu N c. 7). In contrast, most of the tholeiitic MD dolerites have fiat, unfractionated REE distributions. The tholeiites were probably derived from a fertile mantle source. However, the norites originated from a depleted harzburgitic mantle which had earlier borne Archaean komatiitic and tholeiitic magmas but which had been metasomatically replenished in LIL elements by fluids derived from Archaean oceanic and lower continental crustal material returned to the mantle during late-Archaean crustal thickening. Some major layered basic igneous intrusions, such as Bushveld and Stillwater, were derived by the mixing of two different basic magmas, which were equivalent to the tholeiitic and noritic (boninitic) dyke magmas of SW Greenland, NW Scotland and Wyoming. The apparently ubiquitous occurrence of tholeiite-norite associations suggests that similar large-scale mantle heterogeneities, due to the voluminous extraction of Archaean komatiitic and tholeiitic magmas, had become well established by the early Proterozoic.

20J. A quantitative model for olivine fractionation : Part I. Large scale mantle heterogeneity in Fe/Mg ratio.(Abstracts of Papers Presented at the Spring Meeting of the Society)

20J. A quantitative model for olivine fractionation : Part I. Large scale mantle heterogeneity in Fe/Mg ratio.(Abstracts of Papers Presented at the Spring Meeting of the Society)

Halogens in the mantle beneath the North Atlantic

F, Cl and Br contents of tholeiitic volcanic glasses dredged along the Mid-Atlantic Ridge from 53° to 28° N, including the transect over the Azores Plateau, are reported. The halogen variations parallel those of 87 Sr/ 86 Sr, La/Sm or other incompatible elements of varying volatility. The latitudinal halogen variation pattern is not obliterated if only Mg-rich lavas are considered. Variations in extent of low-pressure fractional crystallization or partial melting conditions do not appear to be the primary cause of the halogen variations. Instead, mantle-derived heterogeneities in halogens, with major enrichments in the mantle beneath the Azores, are suggested. The Azores platform is not only a ‘hotspot’ but also a ‘wetspot’, which may explain the unusually intense Azores volcanic activity. The magnitude of the halogen and incompatible element enrichments beneath the Azores appear strongly dependent on the size of these anions and cations, but independent of relative volatility at low pressure. The large anions Cl and Br behave similarly to large cations Rb, Cs and Ba, and the smaller anion F similarly to Sr and P. Processes involving crystal and liquid (fluid and/or melt), CO 2 rather than H 2 O dominated, seem to have produced these largescale mantle heterogeneities. Geochemical ‘anomalies’ beneath the Azores are no longer apparent for coherent element pair ratios of similar ionic size. Values of such ‘unfractionated’ coherent trace element ratios provide an indication of the mantle composition and its nature before fractionation event (s) which produced the inferred isotopic and trace element heterogeneities apparently present beneath the North Atlantic. The relative trace element composition of this precursor mantle does not resemble that of carbonaceous chondrites except for refractory trace element pairs of similar ionic size. It is strongly depleted in halogens, and to a lesser extent in large alkali ions Rb and Cs relative to refractory Ba. These relative depletions are comparable within a factor of 5 to Ganapathy & Anders’s estimates for the bulk Earth, with the exception of Cs. There is also evidence for removal of phosphorus into the iron core during its formation. With the exception of San Miguel, alkali basalts from the Azores Islands appear to have been derived from the same mantle source as tholeiitic basalts from the ridge transect over the Azores Platform but by half as much degree of partial melting. The Azores subaerial basalts seem to have been partly degassed in Cl, Br and F, in decreasing order of intensity. A working model involving metasomatism from release of fluids at phase transformation during convective mantle overturns is proposed to explain the formation of mantle plumes or diapirs enriched in larger relative to smaller halogen and other incompatible trace elements. The model is ad hoc and needs testing. However, any other dynamical model accounting for the 400 -1000 km long gradients in incompatible trace elements, halogens and radiogenic isotopes along the Mid-Atlantic Ridge should, at some stage, require either (1) some variable extent of mixing or (2) differential migration of liquid relative to crystals followed by re-equilibration (or both), as a diffusion controlled mechanism over such large distances is clearly ruled out, given the age of the Earth.

Archaean mantle heterogeneity: evidence from chemical and isotopic abundances in Archaean igneous rocks

The problem of compositional heterogeneity in the Archaean upper mantle is evaluated from the existing data of isotopic compositions (Sr, Pb and Nd), trace element abundances (rare earth elements (r.e.e.)) and some critical major element chemistry in Archaean igneous rocks. To assess a large-scale mantle heterogeneity, we have emphasized the chemical characteristics of undoubtedly mantle-derived komatiites. The initial isotopic compositions of Sr and Pb ( I Sr and I Pb ) in Archaean igneous rocks show a significant variation, suggesting that a long-term heterogeneity in Rb/Sr and U /Pb ratios might have existed in their mantle sources. In contrast, the available I Nd values show little variation from the expected evolution in a chondritic uniform reservoir. These contrasting results, i.e. the variable I Sr and I Pb values as against the uniform I Nd values, are in part attributed to the fact that mobile-refractory pairs in the Rb/Sr and the U /Pb systems but only the refractory-refractory pair in the Sm/Nd system have been involved in geochemical differentiation. The r.e.e. distributions in komatiitic rocks are not uniform. Many show strong depletion in light r.e.e. The degree of light r.e.e. depletion in these komatiitic rocks is comparable with that found in present mid-ocean ridge basalts. Regarding the major element chemistry, we have demonstrated that the CaO / Al 2 O 3 ratios are significantly different in the komatiitic suites of two adjacent cratons (Kaapvaal and Rhodesia) in southern Africa. Both trace and major elements data suggest that the Archaean upper mantle was heterogeneous, but whether the heterogeneity truly represents a long-term phenomenon remains to be resolved by more isotopic analyses.