Keith Bell

Nd and Sr isotopic compositions of East African carbonatites: Implications for mantle heterogeneity

Sr and Nd isotopic data from some young carbonatites from East Africa exhibit an excellent negative correlation on an ϵSr - ϵNd diagram. A simple mixing model involving continental crust was tested and found not to fit the isotope patterns. The mixing of two mantle sources is preferred, one being depleted and the other slightly enriched in the large ion lithophile elements relative to bulk earth. The presence of a depleted source below the Tanzanian Shield is consistent with findings from some Canadian carbonatites.

Archean depleted mantle: Evidence from Nd and Sr initial isotopic ratios of carbonatites

Abstract Initial 87 Sr 86 Sr and 143 Nd 144 Nd ratios of some carbonatites and related alkalic rocks intruded into the Canadian Shield between 2700 Ma and 110 Ma ago define development lines on a plot of initial ratio versus age. The isotopic data show that these complexes have been derived from a mantle source with a Rb Sr ratio of 0.020 ± 0.002 and a Sm Nd ratio of 0.358 ± 0.008, which are lower and higher respectively than the values for the silicate portion of the Earth. An estimate of the mean age of formation of this LIL-depleted reservoir, based on the intersection of the development lines with those of bulk Earth, is about 2900 Ma, a figure similar to the age of most Superior Province rocks. A model is proposed that involves differentiation, ca. 2900 Ma ago, of mantle similar in chemical composition to bulk Earth, into granitoid rocks (the Superior Province), and a depleted, residual upper mantle. The source for the Nd and Sr contained in the carbonatites is thought to lie within this depleted sub-continental lithosphere. Measured Sm Nd ratios of about 0.15 for the carbonatites are much lower than the calculated Sm Nd ratio of their source, and support enrichment of the light REE either slightly before or at the same time as carbonatite magma generation.

U-Pb and Rb-Sr geochronology of the Western Ethiopian Shield

Five U-Pb zircon and four Rb-Sr whole-rock isochron dates constrain the ages of plutonic rocks and their subsequent history within the Western Ethiopian Shield. The U-Pb data show plutonic activity between 830 and 540 Ma, similar to the time span established for the Pan-African rocks of the Southern Red Sea Hills of the Sudan and the Jeddah terrane of Saudi Arabia. U-Pb zircon ages of 828 +9/-2 and 814 ± 2 Ma, obtained from pre-kinematic plutons of the Birbir Domain, date the time of their emplacement. One of these plutons (Goma) yielded a whole-rock Rb-Sr isochron date of 759 ± 18 Ma. Granite sheets within the high-grade Baro Domain, believed to be anatectic, were dated by U-Pb at 783 +19/-14 Ma. The 780-760 Ma event is interpreted as a period of Pan-African regional metamorphism that correlates with the age of low-grade metamorphism of arc-related rocks elsewhere in northeast Africa. Other reset Rb-Sr whole-rock isochrons from the Western Ethiopian Shield that cluster around 635 Ma provide evidence of significant isotopic homogenization, probably associated with major transcurrent movement parallel to a plate margin. A U-Pb lower-intercept age of 582 +29/-33 Ma from Baro granite sheets denotes a second high-grade metamorphic event and correlates with the age of Mozambiquean amphibolite-facies metamorphism in northwest Kenya. Late- to post-kinematic plutons at 571 +11/-3 and 541 +10/-16 Ma place a minimum age limit on deformation and metamorphism.

Non-depleted sub-continental mantle beneath the Superior Province of the Canadian Shield: Nd-Sr isotopic and trace element evidence from Midcontinent Rift basalts

Midcontinent Rift flood basalts represent a sample of the relatively shallow, sub-continental upper mantle beneath the Canadian Shield at 1.1 Ga. A thick sequence of olivine tholeiite lavas, including minor intermediate to rhyolitic lavas, from the Portage Lake Volcanics (PLV) in northern Michigan have initial Nd and Sr isotopic compositions which cluster near Bulk Earth values. The effects of assimilation of old LREE-enriched continental crust into mantle-derived fractionating liquids are isotopically discernible in evolved lavas as well as in olivine tholeiites from the lowest portion of the volcanic pile. However, the effects of crustal contamination decrease with stratigraphie height and are absent in more primitive lavas in the upper half of the section. Therefore, the youngest olivine tholeiites preserve the isotopic characteristics of their sub-continental mantle source area which, at 1095 Ma, had ϵNd(T) and ϵSr(T) values of about +0.8 and +2, respectively. Incompatible trace element compositions from the PLV olivine tholeiite suite support the interpretation of a mantle source slightly enriched in LIL elements relative to chondritic compositions as opposed to a suite of hybrid magmas resulting from mixtures between depleted mantle and enriched crustal end-members. The source for PLV tholeiites is substantially less depleted than previously reported mantle values from the Superior Province. An origin for the PLV source is compatible with either of several mantle evolution models. The PLV source may have been associated with upwelling of a LIL element-enriched, asthenospheric plume which emplaced non-depleted material from deeper sources into the shallow subcontinental mantle beneath the Midcontinent Rift during continental break-up. Alternatively, the PLV source may have originated by enrichment of refractory sub-continental lithospheric mantle which was previously depleted in incompatible trace elements during Archean-aged melt extraction and continental crust formation. Concurrent generation of carbonatite magmas in other areas beneath the Superior Province indicates the widespread presence of sub-continental mantle with substantially higher ϵNd(T) and lower ϵSr(T) than the PLV source. Combined tholeiite and carbonatite data indicates the presence of large, chemically distinct regions in the upper mantle beneath the Superior Province at 1.1 Ga.

Sr and Nd isotope data of apatite, calcite and dolomite as indicators of source, and the relationships of phoscorites and carbonatites from the Kovdor massif, Kola peninsula, Russia

A detailed Sr−Nd isotopic study of primary apatite, calcite and dolomite from phoscorites and carbonatites of the Kovdor massif (380 Ma), Kola peninsula, Russia, reveals a complicated evolutionary history. At least six types of phoscorites and five types of carbonatite have been identified from Kovdor by previous investigators based on relative ages and their major and accessory minerals. Isotopic data from apatite define at least two distinct groups of phoscorite and carbonatite. Apatite from the earlier phoscorites and carbonatites (group 1) are characterized by relatively low87Sr/86Sr (0.70330–0.70349) and143Nd/144Nd initial ratios (0.51230–0.51240) with F=2.01–2.23 wt%, Sr=2185–2975 ppm, Nd=275–660 ppm and Sm=31.7–96.2 ppm. Apatite from the second group has higher87Sr/86Sr (0.70350–0.70363) and143Nd/144Nd initial ratios (0.51240–0.51247) and higher F (2.63–3.16 wt%), Sr (4790–7500 ppm), Nd (457–1074 ppm) and Sm (68.7–147.6 ppm) contents. This group corresponds to the later phoscorites and carbonatites. One apatite sample from a carbonatite from the earlier group fits into neither of the two groups and is characterized by the highest initial87Sr/86Sr (0.70385) and lowest143Nd/144Nd (0.51229) of any of the apatites. Within both groups initial87Sr/86Sr and143Nd/144Nd ratios show negative correlations. Strontium isotope data from coexisting calcite and dolomite support the findings from the apatite study. The Sr and Nd isotopic similarities between carbonatites and phoscorites indicate a genetic relationship between the two rock types. Wide variations in Sr and Nd isotopic composition within some of the earlier carbonatites indicate several distinct intrusive phases. Oxygen isotopic data from calcite and dolomite (δ18O=+7.2 to +7.7‰ SMOW) indicate the absence of any low-temerature secondary processes in phoscorites and carbonatites, and are consistent with a mantle origin for their parental melts. Apatite data from both groups of phoscorite plot in the depleted quadrant of an εNd versus εSr diagram. Data for the earlier group lie along the Kola Carbonatite Line (KCL) as defined by Kramm (1993) and data from the later group plot above the KCL. The evolution of the phoscorites and carbonatites cannot be explained by simple magmatic differentiation assuming closed system conditions. The Sr−Nd data can best be explained by the mixing of three components. Two of these are similar to the end-members that define the Kola Carbonatite Line and these were involved in the genesis of the early phoscorites and carbonatites. An additional component is needed to explain the isotopic characteristics of the later group. Our study shows that apatite from rocks of different mineralogy and age is ideal for placing constraints on mantle sources and for monitoring the Sr−Nd evolution of carbonatites.

SR-ND-PB ISOTOPE RELATIONSHIPS IN LATE ARCHEAN CARBONATITES AND ALKALINE COMPLEXES : APPLICATIONS TO THE GEOCHEMICAL EVOLUTION OF ARCHEAN MANTLE

Isotopic studies of strontium, neodymium, and lead in young carbonatites from several continents have shown that the data generally plot within the fields of ocean island basalts in isotope correlation diagrams. These and other data suggest that the magmas originate in mantle sources and generally pass through continental crust with minimal contamination. Further studies have shown that initial isotopic ratios in carbonatites and alkaline complex rocks from the Canadian Shield spanning an age range from 0.1 to 2.7 Ga appear to trace the evolution of “depleted” subcontinental mantle over the past 2.7 Ga. However, the data were based upon carbonatites between 0 and 1.9 Ga, and substituted syenites at 2.7 Ga due to lack of known carbonatites of that age. This raised the question of possible crustal contamination in the syenites, particularly for lead. Recently two carbonatite bodies with ages of 2.68 Ga have been identified at the Lac Shortt mine and Dolodau dykes in south-central Quebec. Lead, strontium, and neodymium isotope data from these carbonatites fit the evolution patterns already established from the syenites, with slightly more radiogenic neodymium and slightly less radiogenic strontium and lead. The initial ϵ(Nd) from eight carbonatite samples is + 2.8 ± 0.3; ϵ(Sr) = −0.3 ± 0.3 (87Sr86Sr = 0.70127 ± 0.00002). Lead ratios from combined carbonatites and syenites define a regression line with a slope of 0.71 ± 0.04 in a 207Pb-206Pb correlation diagram, corresponding to radiogenic lead evolved in a closed system between ca. 3.8 and 2.7 Ga; however, we ascribe this trend to mixing of mantle components rather than to crustal contamination processes. Data from young (<0.2 Ga) carbonatites from five continents closely fit a model corresponding to mixing between EM1 and HIMU mantle components in isotope correlation diagrams. Similar diagrams for the Canadian Shield 2.7 Ga samples exhibit no clear mixing trends, and 87Sr86Sr-206Pb204Pb ratios tend to cluster closely around model “bulk silicate Earth” in an isotope correlation diagram. Mantle differentiation processes appear to have changed in fundamental ways around 3 Ga ago, with most sialic continental crust produced after that time.

Is there a mantle plume below Italy

Some of the most diverse igneous rocks found on Earth occur along the length of Italy and in many of the islands in the southeastern Tyrrhenian Sea, all the result of Cenozoic magmatism. Magmas extremely rich in alkalis, particularly potassium, and many undersaturated with respect to silica, were erupted, as well as others of calc-alkalic affinity (see legend in Figure 1). Their origin has been the subject of heated debate, and there is still no general consensus about how they formed. Most attribute them to subduction-related processes (see Beccaluva et al. [2004] for a review); others consider them to be the result of within-plate magmatism [e.g., Vollmer, 1976; Lauecchia and Stoppa, 1996]. Still others consider magmatism the result of a deep, mantle upwelling within a slab window coupled with mixing between isotopically different reservoirs [Gasperini et al., 2002].

Probing the mantle: The story from carbonatites

One of the most exciting advances in Earth science in the last several decades has been our increased understanding of the structure and composition of the mantle. Seismic tomography and isotope geochemistry have been major players in those advances. The isotopic studies used basalts from ocean basins to minimize the problems of possible crustal contamination. Data from mid-ocean ridge basalts (MORBs) and oceanic island basalts (OIBs) revealed a relatively detailed picture of the isotope geochemistry of the sub-oceanic mantle [e.g. Hofmann, 1997] that led to the recognition of four principal magma components that define end-member compositions. These are DMM, HIMU, EMI, and EM2 (see Table 1 for further details). All of these components except DMM have been attributed to subduction of different materials, such as oceanic and continental crust and lithosphere, down into the mantle. Several studies have indicated that the mantle is isotopically heterogeneous with heterogeneities that were probably established at least several billions of years ago.

Sm-Nd and Rb-Sr isotope systematics of scheelites: Possible implications for the age and genesis of vein-hosted gold deposits

A positive correlation between present-day {sup 143}Nd/{sup 144}Nd and {sup 147}Sm/{sup 144}Nd ratios from scheelite samples from the Hollinger-McIntyre-Coniaurum gold deposit of Ontario, Canada, corresponds to an age of 2403 {plus minus} 47 Ma and may date the age of gold mineralization. Initial isotope ratios of scheelites suggest that the fluid responsible for the scheelite and gold mineralization was probably of mantle origin and that it subsequently interacted with continental crust.

Lead and strontium isotope relationships in the Oka carbonatite complex, Quebec

Abstract Lead isotopic compositions are presented for 8 calcite, 1 pyrite and 1 melilite separates from the Oka, Quebec, carbonatite complex. Pyrite, melilite and two calcite samples with small in situ decay corrections yield similar initial isotope ratios, averaging 206 Pb 204 Pb = 19.65 , 207 Pb 204 Pb = 15.56 , 208 Pb 204 Pb = 39.00 . This composition shows a close affinity for Pb isotope ratios from ocean island basalts (OIB) and indicates an origin of the Pb in large-ion lithophile (LIL)-element depleted mantle sources. The Pb probably results from a mixture from a LIL-element depleted source and possibly, metasomatic fluids. Four calcites fail to yield accurate initial ratio information due to large in situ decay corrections. In addition three samples exhibit high initial 207 Pb 204 Pb ratios indicative of probable crustal contamination. The silicate fraction of one okaite and a sample of country rock gneiss near Oka contain Pb with isotopic compositions that illustrate the kinds of contamination that could be involved. Sr isotope data are reported for 2 whole-rock sovites, plus 1 nepheline, 4 calcite and 2 apatite separates. All samples yield nearly identical isotopic compositions, with a mean 87 Sr 86 Sr ratio of 0.70331 ± 0.00002 at the 2 sigma level. This result verifies an earlier value reported by Bell et al. (1982) and supports their suggestion of a LIL-element depleted mantle source for the Sr. The Sr data do not show the apparent crustal contamination detected by some of the Pb data, presumably because of the high Sr contents of the carbonate samples. The Oka data are compared with results of a similar study by Lancelot and Allegre (1974) on carbonatite complexes from eastern Uganda. We conclude that the African data, as presently known, are compatible with the model developed for the mantle-derived Oka suite.