Lower Mantle Source Research Articles

Major Element Records of Variable Plume Involvement in the North Atlantic Province Tertiary Flood Basalts

Major element variations in North Atlantic Tertiary Province primitive, early erupted, alkaline-transitional-tholeiite basalts, recalculated to a restricted value of MgO, give insights into the process of plume-related magmatism. Basalts primitive enough to be crystallizing only olivine were recalculated to a proposed primary magma composition of 15 wt % MgO. The recalculated dataset shows clear inter-element correlations including a strong, significant, negative correlation between Fe and Si indicating polybaric melt segregation. Overlap between basalt compositions and experimental melts from a fertile, Fe-rich, low mg-number (85.5) peridotite suggests that, relative to normal peridotite with mg-number > 89, the North Atlantic basalt source was Fe rich. Linear regression of the experimental data gives apparent pressures of magma segregation of 17·5–37 kbar, with intra-region variability in the depth derivation from the melt column for each sample, thus suggesting that lithospheric thickness ‘lid-effect’ control on magma generation may have been overemphasized in recent studies. Comparable source composition, magma segregation depth and calculated mantle potential temperature (1440–1460°C) throughout the Province supports the previously suggested plume impact model, arriving below East Greenland, derived from a variably enriched and depleted lower-mantle source. Given the good agreement between conclusions drawn from major element data and previously published results we suggest that restricted-MgO recalculated datasets may be usefully applied to study other large igneous provinces.

Existence of complex spatial zonation in the Galápagos plume

Basalts from intraplate or hotspot ocean islands (e.g., the Hawaiian, Galapagos, and Canary Islands) are believed to be formed by mantle plumes, which emanate from mantle boundary layers such as the coremantle boundary. The long-term chemical structure of mantle plumes, however, remains poorly constrained. Spatial variation in the chemical composition has long been recognized in lavas from the Galapagos Islands: Enriched plume material forms a horseshoe-shaped region with depleted mantle, similar in composition to mid-ocean ridge basalt, in its inner part. The enriched horseshoe-shaped region can be subdivided into three distinct geochemical domains. We show that these same domains occur in the same relative positions with respect to morphology in a geochemical profile across the Galapagos hotspot track off the coast of Costa Rica, indicating that the asymmetrical spatial zonation of the Galapagos hotspot has existed for at least 14 m.y. Combined with published He isotope data, the results of this study imply that plume material can ascend from the lower mantle, possibly from the core-mantle boundary, with little stirring occurring during ascent, and that zonation in hotspot lavas may in some cases reflect spatial heterogeneity within the lower mantle source.

A Lower Mantle Source for Central European Volcanism.

Cenozoic rifting and volcanism in Europe have been associated with either passive or active mantle upwellings. Tomographic images show a low velocity structure between 660- and 2000-kilometer depth, which we propose to represent a lower mantle upwelling under central Europe that may feed smaller upper-mantle plumes. The position of the rift zones in the foreland of the Alpine belts and the relatively weak volcanism compared to other regions with plume-associated volcanism are probably the result of the past and present subduction under southern Europe.

186Os– 187Os systematics of Hawaiian picrites

Primitive Hawaiian picrites have 187Os/ 188Os as high as ∼0.145 and are more radiogenic than the depleted upper mantle, reflecting a time-integrated suprachondritic Re/Os ratio. The high Re/Os may be explained either by an ancient recycled crustal component or an evolved outer core component in the Hawaiian plume. New high precision 186Os/ 188Os measurements for these picrites, combined with previous analyses, show that the Hawaiian plume source has 186Os/ 188Os that range from chondritic mantle values of ∼0.119834 to more radiogenic values as high as ∼0.119848. The higher 186Os/ 188Os reflects long-term suprachondritic Pt/Os and is coupled with higher 187Os/ 188Os in all but the Koolau picrites. The latter have near-chondritic 186Os/ 188Os but with radiogenic 187Os/ 188Os. The Pt/Re of crustal materials that may make up ancient recycled slabs ranges from ∼0.1 to 33. Recycled slab material with such Pt/Re ratios evolved for 1–3 Ga and added to the plume source may explain the Koolau Os isotopic compositions. Pt/Re ratios of 88–100, however, are required for the ancient recycled crust to generate the coupled enrichments of 186Os/ 188Os and 187Os/ 188Os in picrites from Loihi and Hualalai. These high Pt/Re ratios do not occur in any known crustal materials, but they are consistent with the observed partition coefficients for Os>Re>Pt, during metal crystallization from an initially chondritic molten core (Pt/Re ∼21–24). Such partitioning may have produced an evolved outer core with suprachondritic Pt/Os, Re/Os and Pt/Re, resulting in the production of suprachondritic 186Os/ 188Os and 187Os/ 188Os over time. Small amounts of outer core metal (≤1.2%) mixed into the Hawaiian plume source can explain the coupled 186Os/ 188Os and 187Os/ 188Os enrichment in some of the Hawaiian picrites. In addition, the most radiogenic 186Os/ 188Os in the Hawaiian picrites is correlated with higher 3He/ 4He, consistent with an undegassed, and likely, lower mantle source. These data provide compelling geochemical evidence that the Hawaiian plume was generated at the core-mantle boundary.

Central European Cenozoic plume volcanism with OIB characteristics and indications of a lower mantle source

Average concentrations of incompatible elements and isotopes of radiogenic Sr, Nd and Pb in a 350 km long belt of Central European volcanics (CECV) resemble those in OIB in general and in OIB from the N Atlantic in particular. This similarity allows to infer comparable sources for CECV and OIB which might have been located in the lower mantle according to seismic tomography and chemistry, with the latter unlike a MORB source. The incompatible element contribution of lower mantle origin can be modelled from primitive mantle minus continental crust and upper mantle inventories. Alkali basaltic magmas from the modelled source are close in composition to CECV and OIB. The continental crust contains almost half of the silicate Earth's content of Rb, K, Pb, Ba, Th and U, which were mobilized to a major extent through dehydration of subducted ocean crust. Related losses from the lower mantle had to be replaced by deep subductions of oceanic lithosphere recognized from their isotopic imprint. From a balance based on Nd isotopes it can be concluded that average CECV contains 60% matter from residual primitive mantle and 40% from deeply subducted lithosphere (including some young upper mantle materials). Plume products from separate CECV regions developed, within 45 Ma, from rather depleted to more primitive isotopic signatures. Four periods of volcanism from Eocene to late Quaternary time are explained as four pulses of an almost stationary ultrafast plume uprise as modelled by Larsen and Yuen (1997). Magma production has increased from the first to the third pulse with the peak during Miocene time in the Vogelsberg region. The final pulse produced the Quaternary Eifel volcanoes. Tectonism from the Alpine orogen has probably triggered the synchronous volcanism of CECV, Massif Central etc. The European lithospheric plate has moved under the control of the opening Atlantic almost in an eastern direction with a velocity of 1 cm per year and has shifted extinct volcanoes off their source channels.

A paleomagnetic analysis of Cambrian true polar wander

The latest Neoproterozoic through Cambrian is one of the most remarkable intervals in geologic time. Tectonically, the period from 580 to 490 Ma marks a time of rapid plate reorganization following the final stages of supercontinental breakup and Gondwana assembly. The apparent speed at which this reorganization occurred led some to propose a link between tectonic events, biologic changes and climatic changes. One of the more intriguing proposals is that the tectonic changes were triggered by an episode of inertial interchange true polar wander (IITPW) which resulted in a rapid (6°/m.y.) shift of the spin axis relative to the geographic reference frame. IITPW is a special case of true polar wander (TPW) that makes specific demands on the length of apparent polar wander paths (APWPs) recording the motion. Specifically, each path must allow for ∼90° of synchronous motion during the interval from 523 to 508 Ma. A review of paleomagnetic data for Laurentia, Baltica, Siberia and Gondwana indicates that none of the APWPs approaches the necessary length, each path is of a different length and the apparent motions are non-synchronous. Collectively, these observations negate the premise of a Cambrian IITPW event. Since the IITPW hypothesis was proposed as an alternative to rapid plate motion of Laurentia and Gondwana during the Neoproterozoic–Cambrian interval, any alternative model must account for this rapid motion. I suggest that a reasonable explanation for `anomalously' high rates of plate motion for some continents, possibly on the order of 20–40 cm yr−1, is enhanced plate motion driven by lower-mantle thermal anomalies and possibly true polar wander. In fact, the enhanced plate motions driven by these lower-mantle sources may provide a dynamic feedback triggering true polar wander.

Element partitioning between silicate perovskites and calcic ultrabasic melt

Element partitioning between Mg- and Ca-perovskites and anhydrous calcic ultrabasic melt was studied experimentally at 25 GPa and 2200°C. The observed partitioning relationships in this system are similar to the results of previous studies in magnesian ultrabasic and basaltic systems, and are consistent with predictions based on crystal chemical considerations. These results provide new insights into early global mantle differentiation, and kimberlite petrogenesis from the lower-mantle source regions.

Isotopic variations with distance and time in the volcanic islands of the Cameroon line: evidence for a mantle plume origin

The oceanic sector of the Cameroon line consists of three volcanic islands: Principe, São Tomé and Pagalu. New 40Ar 39Ar data for Pagalu basalts, combined with published KAr ages for Principe and São Tomé, indicate that all three islands have been active in the past 5 Ma. They have similar petrogenetic histories, with basements of basaltic flows capped by more evolved rocks. However, the age of the earliest exposed volcanic rocks decreases oceanward from Principe (31 Ma) to São Tomé (13 Ma) to Pagalu (4.8 Ma). This age progression is consistent with the suggested motion of the African plate over this period of time. The average incompatible trace element compositions of < 10 Ma lavas with ⩾ 4 wt% MgO on each island are very similar. However, ( 87 Sr 86 Sr ) t increases from 0.7029 to 0.7037 and ( 206 Pb 204 Pb ) t decreases from 20.2 to 18.9 from Principe through São Tomé to Pagalu for all samples younger than 10 Ma. In addition to the overall spatial isotopic variations, Principe and São Tomé display temporal isotopic variations, with Pb isotopic ratios becoming progressively more radiogenic. Pagalu shows no temporal geochemical or isotopic differences and the island has the least radiogenic Pb but most radiogenic Sr. These distinctive Pb, Sr and Nd isotopic compositions are also found in the early tholeiitic hyaloclastite breccia from Principe (31 Ma). Similarly, the Nd and Sr isotopic compositions are identical to those of the earliest São Tomé lavas (13 Ma) and the Pb isotopic compositions of early São Tomé samples are only slightly radiogenic relative to Pagalu. Therefore, it is probable that all these islands were initiated from a common source, similar to that of Pagalu, that migrated relative to the melt zone of each island with time. Since their initiation, the magma conduits at Principe and São Tomé have been gradually modified by the introduction of a HIMU component. The common source from which the islands were initially derived probably represents ambient upper mantle, entrained with the plume head during ascent. This entrained component is like ‘PREMA’, but the Nd and Sr isotopic data indicate that it represents variably mixed depleted and enriched components, such as DMM and EMI. The HIMU component is probably representative of a lower mantle source from which the plume head was derived. The long-lived episodic magmatism on Principe provides evidence that the initial melt migration paths from the upper mantle form a hot zone that can be re-activated after long periods (10 7 yr) of apparent quiescence. The progression to HIMU characteristics within each island probably reflects the gradual flattening of the contaminated plume head within this hot zone, near the base of the lithosphere, and the melting of a stem composed of relatively uncontaminated HIMU mantle.

Rapid drift of large continents during the late Precambrian and Paleozoic: Paleomagnetic constraints and dynamic models

Research Article| November 01, 1994 Rapid drift of large continents during the late Precambrian and Paleozoic: Paleomagnetic constraints and dynamic models Michael Gurnis; Michael Gurnis 1Seismological Laboratory, California Institute of Technology, Pasadena, California 91125 Search for other works by this author on: GSW Google Scholar Trond H. Torsvik Trond H. Torsvik 2Geological Survey of Norway, P. B. 3006 Lade, N-7002 Trondheim, Norway Search for other works by this author on: GSW Google Scholar Author and Article Information Michael Gurnis 1Seismological Laboratory, California Institute of Technology, Pasadena, California 91125 Trond H. Torsvik 2Geological Survey of Norway, P. B. 3006 Lade, N-7002 Trondheim, Norway Publisher: Geological Society of America First Online: 02 Jun 2017 Online ISSN: 1943-2682 Print ISSN: 0091-7613 Geological Society of America Geology (1994) 22 (11): 1023–1026. https://doi.org/10.1130/0091-7613(1994)022<1023:RDOLCD>2.3.CO;2 Article history First Online: 02 Jun 2017 Cite View This Citation Add to Citation Manager Share Icon Share Facebook Twitter LinkedIn MailTo Tools Icon Tools Get Permissions Search Site Citation Michael Gurnis, Trond H. Torsvik; Rapid drift of large continents during the late Precambrian and Paleozoic: Paleomagnetic constraints and dynamic models. Geology 1994;; 22 (11): 1023–1026. doi: https://doi.org/10.1130/0091-7613(1994)022<1023:RDOLCD>2.3.CO;2 Download citation file: Ris (Zotero) Refmanager EasyBib Bookends Mendeley Papers EndNote RefWorks BibTex toolbar search Search Dropdown Menu toolbar search search input Search input auto suggest filter your search All ContentBy SocietyGeology Search Advanced Search Abstract During the late Precambrian and early Paleozoic, Laurentia and Baltica moved at minimum drift rates of up to 23 cm/yr. These drift rates are computed from paleomagnetic apparent-polar-wander paths and represent minimum velocities, because there could have been significant undetected longitudinal motion. A pronounced burst in latitudinal velocity followed the breakup of the supercontinent Rodinia, which had been assembled for about 400-500 m.y. Finite-element models with tectonic plates show that the presence of deep continental roots can strongly influence the velocity of continents if the driving source of buoyancy is located in the lower mantle. When plates are driven by lithospheric cooling and subducted slabs, the presence of a root is less important. We argue that Laurentia and Baltica were either pushed off of a hot lower-mantle source or pulled toward cold lower-mantle anomalies and that the presence of continental roots enhanced this motion. This content is PDF only. Please click on the PDF icon to access. First Page Preview Close Modal You do not have access to this content, please speak to your institutional administrator if you feel you should have access.

The composition of peridotite tectonites from the Ivrea Complex, northern Italy: Residues from melt extraction

A total of forty-three samples from the Balmuccia, Baldissero, and Finero peridotite tectonites have been analyzed for major elements, forty minor elements, mineral composition, and a subset of samples have been investigated for δD, δ 18O, δ 34S, and 87Sr/ 86Sr ratios. The spinel Iherzolites from Balmuccia and Baldissero contain on average 56% olivine, 28% orthopyroxene, 14% clinopyroxene, and 1.5% spinel. They have a neodymium and strontium isotopic signature similar to MORB. These peridotites were moderately depleted in Al and Ca and highly depleted in incompatible elements by the separation of 4.5% P-MORB. The MORB in excess to the formation and subduction of the oceanic crust fractionated into a tonalitic continental crust and a pyroxenitic residue which was recycled into the mantle. The crustal mass fraction of the upper mantle-crust system is 2.8%. A primitive mantle composition can be calculated from 97.2% Balmuccia model peridotite plus 2.8% bulk crust. The new set of primitive mantle concentrations is in accordance, within about 10%, with the data from fertile peridotite xenoliths and primitive meteorites for the elements Li, Mg, Si, Ca, Sc, V, Mn, Fe, Co, Ni, Zn, Ga, Y, Zr, La, Ce, Nd, Eu, Tb, Dy, Yb, Lu, and Hf. Larger differences occur for the elements F, Na, P, Ti, Cr, and Cu. The bulk crust concentrations of the highly incompatible and mobile elements K, Rb, Ba, Th, and U exceed those being supplied from a primitive upper mantle reservoir of 660 km depth. An additional lower mantle source is required for about half of the crustal accumulation of these elements. Moderately depleted peridotite bodies with a cross section of 3 km can be sufficiently well homogenized in compatible and moderately incompatible elements to represent MORB producing upper mantle. Concentrations of Na, Sc, Ti, V, Ga, Y, Zr, and Yb are well correlated with the Al abundance so that primitive mantle values can also be estimated using the cosmically derived primordial Al concentration. The spinel harzburgites from Finero contain on average 74% olivine, 16% orthopyroxene, 3% clinopyroxene, 5% amphibole, 1% phlogopite, and 1% spinel. These peridotites were depleted by a separation of about 18% MORB. Subsequently they have undergone metasomatic alteration probably from dehydration of a subducted lithospheric slab. They received two pulses of water-rich fluids at temperatures of about 800°C to form amphibole low in incompatible elements or amphibole plus phlogopite high in incompatible elements without complete reequilibration of the preexisting rocks. The accumulation of elements is correlated to their incompatibility with the exception of a depletion of U, Th, Nb, and Ti. Isotopic ratios of H, O, S, Sr, Nd, and Pb, chlorine in phlogopite and a relative deficit of Nb and Ti are consistent with a crustal imprint in the metasomatic fluids.

Temporal Sr-, Nd- and Pb-isotopic variations in the Siberian flood basalts: Implications for the plume-source characteristics

The Sr-, Nd- and Pb-isotopic compositions of lavas from the 248-Ma-old Siberian Flood Basalt Province (SFBP) become increasingly uniform with height as the main phase of volcanic activity is approached. The late-stage lava sequences from the Putorana subprovince, which constitute approximately 90 vol% of the SFBP, require a homogeneous mantle as their source. Volcanostratigraphic and volume arguments imply that extreme isotopic compositions, displayed by the early lavas from Norilsk subprovince, are not the end-member compositions for the bulk of the SFBP. In 147/Sm 144Nd-initial ε Nd space, most of the Putorana basalts define a horizontal array with a majority clustering at (0.17, +1.8). In contrast, the Norilsk lavas show a positively correlated trend indicative of continental lithospheric contributions to the primary plume. Among the three isotopic systems considered, Nd shows the least effect of crustal contamination. We estimate an initial ε Nd value of +1.8 ± 0.7 (1 σ) for the uncontaminated Siberian plume, which points to a near-chondritic, slightly depleted lower mantle source. The Siberian basalts, which show an ε Nd of ⩾ 0, display a remarkably uniform Sr-isotopic composition with an average 87/Sr 86Sr of 0.7050 ± 3 (1 σ). However, in a ( 87/Sr 86Sr) i-ε Nd diagram these lavas are significantly displaced to the right-hand side of a mixing line between bulk earth and depleted mantle. We suggest that these basalts suffered selective contamination of Sr during ascent through the continental crust. This element-specific contamination, which may also be concluded from Nd—Pb and Sr—Pb isotopic variations in the basalts, is at a maximum for Pb-isotopes and at a minimum for Nd. The Pb-isotopic ratios of the contaminated basalts indicate significant contributions from the lower crust with little or no input from the upper crust. The early Norilsk lavas define the extreme compositions and help constrain the initial Pb-isotopic composition of the plume source as 206/Pb 204Pb= 18.3, 207/Pb 204Pb= 15.5 and 208/Pb 204Pb= 38.0, which falls slightly to the right of the geochron at 248 Ma B.P.. We propose a two-stage growth history of Pb for the plume source: An early core-mantle differentiation stage, with a μ value of 0.21 for the first 100–150 m.y. after the Earth formed, was followed in the second stage by an increase in the μ value to about 9.2 for the rest of Earth's evolution.

Binary mixing of enriched and undegassed (primitive?) mantle components (He, Sr, Nd, Pb) in Samoan lavas

We have measured He, Sr, Nd, and Pb isotope ratios and Rb, Sr, Sm, and Nd concentrations in stratigraphically controlled lavas from the Pago shield volcano on Tutuila, American Samoa. We interpret these lavas as products of mixing between two isotopically extreme mantle constituents. The first is a highly enriched component with very high Sr isotope ratios, low 3He/ 4He ratios, and high “Δ(74)” and “Δ(84)” Pb isotopic characteristics. This is probably recently recycled ( < 400 Ma) crustal material. The second component has high 3He/ 4He ratios ( > 24R A) and intermediate Sr, Nd, and Pb isotopic ratios. This material was derived from a largely undegassed mantle source, and, in conjunction with data from several other ocean islands, provides strong evidence for the existence of a high 3He/ 4He ratio mantle end member (primitive helium mantle, PHEM) with consistent Sr, Nd, and Pb isotopic characteristics: near bulk-earth 87Sr/ 86Sr (0.7042–0.7052) and 143Nd/ 144Nd (0.51265–0.51280,ε Nd = +0.2to+3.2), and radiogenic Pb ( 206Pb/ 204Pb∼ 18.5–19.0, 207Pb/ 204Pb∼ 15.5–15.57, 208Pb/ 204Pb∼ 38.4–39.2). Although this material cannot have been derived from a reservoir completely closed to elemental fractionation for the full 4.55 Ga duration of Earth's history, it may indicate the presence of a highly primitive mantle source. Erratic temporal variations in the isotopic composition of individual flows indicate sporadic and variable mixing of these two sources. We interpret these results using a model in which high 3He/ 4He ratio plume material, rising intermittently from a lower-mantle source, intercepts and melts recycled crustal matter in the upper mantle or lithosphere and erupts as a binary mixture of PHEM and the so-called “EM” components derived from this recycled material.

On the distribution of anomalous mass within the Earth: forward models of the gravitational potential spectrum using ensembles of discrete mass elements

SUMMARY We investigate forward models of the gravitational potential spectrum generated by ensembles of discrete sources of anomalous mass, having radial distributions with different statistical properties. Models with a random distribution of point source locations throughout the volume of the mantle produce spectra similar to that of the Earth only when the (absolute) source magnitudes increase strongly with depth, at least as d'.5. The effects of the geographic (latitude-longitude) distribution of source locations are generally unimportant in determining the general degree dependence of the potential spectrum. The dimensions of the sources are also of secondary importance, at least up to an angular diameter of about 40°, i.e., continent-sized. Sources of this size confined to the upper mantle do not appear capable of producing the degree dependence of the observed geopotential spectrum; the low harmonics (2-4) in particular appear to require lower mantle sources of considerable strength. Further, at least some of these deep sources must be largely monopolar in nature, (i.e., uncompensated) due to the stronger depth attenuation of dipole (compensated) sources. Because topography on the core-mantle boundary must be either isostatically or dynamically compensated, it may contribute little strength to the observed potential spectrum.

Global relationship between oceanic geoid and topography

In order to compare model predictions of upper mantle convection with surface observations, we have computed global residual depth and geoid anomalies on a 5° × 5° grid over all oceans, after removal of lithospheric cooling effects and long‐wavelength geoid components (&gt; 4000 km) related to lower mantle sources. We have derived the transfer function (admittance) of geoid over topography as a function of wavelength λ in the range 1000–4000 km. The observed admittance is positive at all λ and presents a maximum of 2.5–3.5 m/km at λ ≃ 2750 km. Comparing this result with the theoretical admittance computed for the lithospheric cooling plate model, we ascribe most of the admittance peak at λ = 2750 km to errors in the depth‐age and geoid‐age correction. This suggests that a standard age correction applied to all seafloor is not valid. At shorter wavelengths, the admittance remains roughly constant, on the order of 1.5 m/km, and can be explained in terms of sediment loading and crustal thickening effects. Comparing spatially the residual geoid and depth anomaly maps, we note a positive correlation over a few areas coinciding with hot spot swells, volcanic plateaus or seafloor deeps; the geoid versus depth ratio amounts then to ∼ 6–8 m/km in agreement with the values predicted above upwellings and downwellings by models of constant viscosity convection. However, the global admittance analysis leads to much lower values at all wavelengths, suggesting that these areas are too limited in number and in extent to produce any significant signal.

Hotspot magmas can form by fractionation and contamination of mid-ocean ridge basalts

Melts ascending from deep reservoirs and trapped beneath thick lithosphere exchange heat with and partially melt the shallow mantle as they crystallize. Basalts erupting through thick lithosphere are more likely to have experienced deep crystal fractionation and contamination than those at mature spreading centres. Geochemical data for ocean island and other hotspot magmas do not require a primitive or lower mantle source.