| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Journal of Petrology Volume 42 Number 10 Pages 1927-1945 2001
© Oxford University Press 2001
Nd, Pb and Sr Isotopic Compositions of East African Carbonatites: Evidence for Mantle Mixing and Plume Inhomogeneity
1OTTAWA–CARLETON GEOSCIENCE CENTRE, DEPARTMENT OF EARTH SCIENCES, CARLETON UNIVERSITY, OTTAWA, ONTARIO, CANADA, K1S 2W8
2DEPARTMENT OF GEOLOGICAL SCIENCES, UNIVERSITY OF CALIFORNIA, SANTA BARBARA, SANTA BARBARA, CA 93106-9630, USA
Received February 16, 2000; Revised typescript accepted March 17, 2001
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONANALYTICAL PROCEDURESRESULTSBULK CHEMICAL COMPOSITIONDISCUSSIONNATURE OF THE EAST...MANTLE COMPONENTSPETROGENETIC MODELSCONCLUSIONSREFERENCES |
|---|
New Pb isotopic data are presented for 10 young Mesozoic to Cenozoic (0–116 Ma) carbonatites from a 1400 km long segment of the East African Rift. Patterns observed in Pb vs Pb, Sr vs Pb and Nd vs Pb isotope diagrams define unusual, nearly linear, trends that are interpreted as mixing between two components that are broadly similar to the two mantle end-member components, HIMU and EM1, which were first recognized from ocean-island basalts. The two plutons with isotope signatures closest to HIMU and EM1 crop out within 140 km of each other. From these data, EM1 and HIMU are now known to occur in both continental and oceanic settings that are associated with plumes or rifts. Moreover, these isotopic signatures tend to occur in regions where seismic tomography indicates prominent low-velocity zones in the lower mantle. For these reasons, we favour a model for the origin of the East African Rift carbonatites that involves melting and mixing of HIMU and EM1 components contained within an isotopically heterogeneous mantle plume. We consider the HIMU and EM1 sources to be stored within the deep (lower 1000 km) mantle, possibly the core–mantle boundary. The role that continental lithosphere plays in carbonatite generation is probably one of concentrating volatiles at the upper levels of an ascending mantle plume.
KEY WORDS: carbonatites; isotopes; rifts; plumes; FOZO
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONANALYTICAL PROCEDURESRESULTSBULK CHEMICAL COMPOSITIONDISCUSSIONNATURE OF THE EAST...MANTLE COMPONENTSPETROGENETIC MODELSCONCLUSIONSREFERENCES |
|---|
Although carbonatites are volumetrically insignificant, their widespread distribution on most continents and their variation in age can provide valuable insights into the nature of the sub-continental mantle over a period of
2·7 by. On the basis of stable isotope data (Deines, 1989
; Keller & Hoefs, 1995), Pb, Sr and Nd isotope compositions (Bell et al., 1982; Nelson et al., 1988; Bell & Blenkinsop, 1989; Kwon et al., 1989) and noble gas data (Sasada et al., 1997; Marty et al., 1998), there seems little doubt that the parental melts to carbonatites are generated within the mantle. One of the most intriguing features to emerge from the studies of young carbonatites is that they share isotope similarities with many oceanic island lavas (Basu & Tatsumoto, 1980; Bell et al., 1982; Bell & Blenkinsop, 1987a, 1987b; Nelson et al., 1988; Kwon et al., 1989), thus raising the question as to whether parts of the sub-continental mantle and the sub-oceanic mantle are essentially the same in terms of their isotopic signatures.
Evidence from melting studies has shown that carbonatites can be produced in three different ways: (1) direct partial melting of a metasomatized mantle source (e.g. Wyllie & Huang, 1975; Wallace & Green, 1988; Wyllie & Lee, 1998); (2) derivation by immiscible separation at low or high pressures from carbonated silicate melts such as a carbonated nephelinite (e.g. Koster van Groos & Wyllie, 1963; Kjarsgaard et al., 1995; Brooker, 1998); (3) crystal fractionation of a carbonated alkali silicate melt (e.g. King, 1949; Veksler et al., 1998). Dolomitic carbonatites are generally considered to be primary in origin (Bailey, 1993; Harmer & Gittins, 1998), whereas calcitic carbonatites can be generated from immiscible liquids or can form either as crystal cumulates or as late-stage products of crystal fractionation (see Wyllie & Lee, 1998). Primary carbonatites can also be modified by interaction with mantle peridotite (e.g. Wallace & Green, 1988; Sweeney, 1994). Natrocarbonatites (sodium-rich carbonatites) are considered to be the products of extreme differentiation involving immiscible separation from a carbonated peralkaline nephelinite at low pressures and low temperatures (Bell & Keller, 1995). The generation of carbonatitic melts at mantle depths has important consequences because they can act as metasomatic agents that could profoundly influence mantle chemistry (e.g. Green & Wallace, 1988). Because of their extremely high enrichments in Sr, Sm and Nd, and in some cases Pb, carbonatites have the potential for changing both the isotope ratios and daughter/parent ratios of mantle sources during metasomatism.
Approximately half of all known carbonatites occur in Africa, most are younger than 200 Ma, and many are associated with the East African Rift Valley System (Woolley, 1989). Carbonatites and associated nephelinites, phonolites, ijolites and syenites are known in both the eastern and western branches of the East African Rift. Sr and Nd isotope ratio measurements already obtained from several young carbonatites from East Africa (Bell & Blenkinsop, 1987a) define a linear array, termed the East African carbonatite line (EACL), that starts from well inside the depleted quadrant and extends into the enriched quadrant in the 
(Nd)–87Sr/86Sr diagram. On the basis of the similarity of the EACL to the LoNd array of Hart et al. (1986), Bell & Dawson (1995) suggested that the two end-members are very similar to HIMU and EM1, two mantle end-member components defined on the basis of Nd, Pb and Sr isotope ratios from ocean-island basalts (OIBs).
In this paper we present new Pb isotopic data for some carbonatites from East Africa that were previously analysed for Sr and Nd (Bell & Blenkinsop, 1987a; Morisset 1992), as well as seven new samples analysed for Sr, Nd and Pb. The sample localities are shown in Fig. 1. The enhanced resolution between DMM and HIMU end-members provided by the Pb isotope data helps to assess whether the origin of the East African carbonatites is indeed consistent with simple HIMU–EM1 mixing, and may possibly help to establish the roles that convecting and non-convecting mantle may have played in generating carbonatitic melts.
|
The East African carbonatites mark episodic periods of magmatic activity probably related to rift valley faulting because of their close spatial relationship. Their ages are given in Table 1. They range from the late Mesozoic to the Cenozoic although most are younger than 40 Ma. The oldest carbonatite in the present suite is Panda Hill, which has an age of 116 Ma (Cahen & Snelling, 1984). We assume that the nearby Sengeri Hill pluton is coeval with Panda Hill. The distribution of the complexes along with some of the features of the regional geology are shown in Fig. 1. New Pb isotope analyses are provided for carbonatites from Bukusu, Kalyango, Rusekere, Sukulu, Tororo and Toror (Uganda), Homa Bay (Kenya), Panda Hill, and Sengeri Hill (Tanzania). Also included are two new analyses of carbonatites from Oldoinyo Lengai, which is the only known currently active carbonatite volcano and which differs markedly in chemical composition from any of the other carbonatites found in East Africa in that it is at present erupting natrocarbonatites. In subsequent discussions, we have added data for Napak, Sukulu and Tororo from Nelson et al. (1988); Shombole from Bell & Peterson (1991); Napak from Simonetti & Bell (1994a); Chilwa from Simonetti & Bell (1994b); Kerimasi from Paslick et al. (1995); and various carbonatites from the extensive study by Kalt et al. (1997).
|
|
ANALYTICAL PROCEDURES |
|---|
|
TOPABSTRACTINTRODUCTIONANALYTICAL PROCEDURESRESULTSBULK CHEMICAL COMPOSITIONDISCUSSIONNATURE OF THE EAST...MANTLE COMPONENTSPETROGENETIC MODELSCONCLUSIONSREFERENCES |
|---|
The chemical procedures for the Nd, Pb and Sr isotope determinations have been given by Tilton & Bell (1994)
. Isotope measurements were made at both Carleton University and University of California (Santa Barbara) using Finnigan-MAT 261 multi-collector mass spectrometers. Pb isotope ratio measurements were corrected for fractionation using a value of 0·12 ± 0·01% per mass unit and reproducibility is considered to be 0·04% (2
) per mass unit for replicate analyses. Uncertainties for Th, U and Pb abundances, determined by isotope dilution, are considered to be ±0·5% of the quoted values. Further details are given in the footnotes for Table 2. In addition, 10 carbonatite samples were analysed for trace elements using inductively coupled plasma mass spectrometry (ICP-MS) by Activation Laboratories at Ancaster, Ontario. All samples were fused to ensure that all minerals were taken into solution. The estimated analytical uncertainties for samples with <100 ppm vary from element to element, but most are about ±10%.
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONANALYTICAL PROCEDURESRESULTSBULK CHEMICAL COMPOSITIONDISCUSSIONNATURE OF THE EAST...MANTLE COMPONENTSPETROGENETIC MODELSCONCLUSIONSREFERENCES |
|---|
Previously published and new data are given in Tables 1 and 2. Because sample ages range from 0 to 116 Ma, all ratios are corrected to present-day values using methods described in the footnotes to the tables.
Figure 2 summarizes all Nd and Sr isotope data available at present from the East African Rift. Also included for comparison are data for seven carbonatite locations from different parts of the Earth that we call ‘Reference Carbonatites’ as defined in the caption to Fig. 2, plus data for the Chilwa carbonatite from southern Tanzania, which does not belong to the East African magmatism, but is considered part of the Chilwa alkaline province. We will use those datasets in all following diagrams. For ages >13 Ma small, additional corrections were made to the initial 143Nd/144Nd ratios based upon the average evolution patterns for carbonatite sources beneath the Canadian Shield over the past 3 by, as given in Bell & Blenkinsop (1987b). We assume that these are the closest approximations we can make to the sources for the East African carbonatites. For 116 Ma plutons the adjustment added 0·00016 to the initial ratios. Using the Rb/Sr ratios in the sources from Bell & Blenkinsop (1987b), the initial 87Sr/86Sr ratios for the 116 Ma plutons increased by 0·00010.
|
The data now show considerably more scatter than was evident in the original data of Bell & Blenkinsop (1987a)
. Because the separation between the HIMU–DMM and HIMU–EM1 chords is not large compared with the spread of the data, it is difficult to distinguish between DMM and EM1 end-members. The lack of carbonatites with (Nd) values greater than, or 87Sr/86Sr values less than, those for HIMU suggests that DMM plays only a minor role, if any, in determining the isotope patterns of the data. This uncertainty is further addressed by considering the Pb data.
The new Pb isotope data (Table 1) are plotted in 206Pb/204Pb–207Pb204Pb and 206Pb/204Pb–208Pb204Pb diagrams in Fig. 3, together with data from the additional sources that are cited in Fig. 2. In all figures, the 206Pb/204Pb ratios were normalized to 0 Ma by assuming closed system evolution in reservoirs calibrated from 2·68 Ma (Tilton & Bell, 1994) and 0·11 Ga (Grünenfelder et al., 1986) samples, which result in an increase of 0·3 for the ratios from the 0·116 Ga plutons. Corrections to the 208Pb/204Pb ratios are the same. The data form near-linear arrays in both diagrams that clearly point to a HIMU end-member. In the diagram EM1 and DMM are too closely spaced to definitively identify one of them as an end-member. However, the separation is much better in the 206Pb/204Pb–208Pb204Pb diagram owing to the greater evolution of 208Pb compared with that of 207Pb. There the trend is away from DMM, seemingly disqualifying it as a major end-member.
|
Part of the Pb data from the East African carbonatites has already been presented in a preliminary form by Grünenfelder et al. (1986). They commented on the linear distribution of the data, and indicated that a regression line through the 207Pb/204Pb–206Pb/204Pb data, referred to but not given in their paper, yielded a slope of 0·098 ± 0·001, similar to that cited by Tatsumoto (1978) for oceanic basalts. The similarity between the Pb isotope data for Ugandan carbonatites (Bukusu, Napak and Sukulu) and the Group I African kimberlites was also noted and the linear array attributed to a mixing between two mantle components, one a large ion lithophile element (LILE)-depleted mantle source, and the other a metasomatic fluid with a high 206Pb/204Pb ratio.
The available data from the East African carbonatites in plots of 87Sr/86Sr vs 206Pb/204Pb and 143Nd/144Nd vs 206Pb/204Pb in Fig. 4a and b clearly show mixing patterns marked by HIMU and EM1 end-members rather than HIMU and DMM, although the spread about the mixing lines implies that additional sources are needed to account for all of the isotope ratios.
The fit to the two-component mixing model is especially good in Fig. 4b. In fact, we find that Pb–Nd diagrams generally provide the most definitive method for sorting out the possible mantle end-members among various sample suites by Pb–Nd–Sr isotopes. Interestingly, the reference and Chilwa samples do not fit such a two-component, HIMU–EM1 model, but plot out in the central field in both figures with HIMU as one end-member. Tilton et al. (1998) have already noted the trend for the Reference Carbonatites, but the Chilwa data add a new set of samples to the group. In addition, the Napak carbonatite, which forms part of the East African Rift group, appears to fit the reference sample group, especially in Fig. 4a and b. Those data raise the question of whether the trend signifies another widespread two-component source for carbonatite and alkaline silicate magmas. The identity of a second component is not apparent, but is similar to isotopic compositions suggested for FOZO (e.g. Hauri et al., 1994). We consider FOZO further below.
Figure 5 shows the patterns when 208Pb/204Pb is substituted for 206Pb/204Pb in Fig. 4. The results are similar to those found in Fig. 4, with Nd again fitting the two-component mixing model more closely than does Sr.
|
|
|
BULK CHEMICAL COMPOSITION |
|---|
|
TOPABSTRACTINTRODUCTIONANALYTICAL PROCEDURESRESULTSBULK CHEMICAL COMPOSITIONDISCUSSIONNATURE OF THE EAST...MANTLE COMPONENTSPETROGENETIC MODELSCONCLUSIONSREFERENCES |
|---|
In previous attempts to model the mantle source regions for basaltic melts, trace element data have been employed in addition to isotopic tracers. Hofmann (1997)
used data from Tristan and Inaccessible Island (Le Roex et al., 1990; Cliff et al., 1991) to characterize EM1, and data from Tubuai (Chauvel et al., 1992) to characterize HIMU. Although some differences have been found between HIMU and EM1 in terms of element ratios, the primitive mantle normalized concentration patterns are surprisingly similar and both show a decreasing smooth trend for the more compatible elements. Both differ from mid-ocean ridge basalt (MORB) and Hawaiian basalts by their marked enrichments in the incompatible elements.
We have obtained trace element analyses using ICP-MS for some carbonatite samples that are from the same complexes as used in our study. Normalized values have been plotted in Fig. 6, and included among these are data from Simonetti et al. (1997) for several natrocarbonatite samples associated with the June 1993 eruption of Oldoinyo Lengai. Concentrations are normalized to those of primitive mantle given in the Geochemical Earth Reference Model (McDonough & Sun, 1995; McDonough, 1999) and plotted in order of increasing compatibility. Also shown for comparison are estimated patterns for both EM1 and HIMU using the same databases as Hofmann (1997).
|
In general terms, there are large variations in the normalized values from one carbonatite complex to another for any given element, but the patterns shown by the samples from Oldoinyo Lengai and the other carbonatites from East Africa are all are broadly similar. All are characterized by a strong negative anomaly for Zr. In chondrite-normalized rare earth element (REE) diagrams, all of the East African carbonatites show similar, parallel patterns, other than those from the Oldoinyo Lengai lavas, which are more depleted in the heavy REE (HREE). This latter feature can be seen in Fig. 6. All involve light REE (LREE) enrichment, consistent with the findings for carbonatites from Europe, North America, Brazil, and South Africa (Hornig-Kjarsgaard, 1998). Although the degrees of enrichment in the East African samples are variable, they still fall within the range shown by other world-wide carbonatites. The total REE abundances vary from a maximum of 4520 ppm for one of the Homa Bay samples to a minimum of 388 ppm for one of the samples from Sukulu. These differences may also be reflected in the isotopic data, as the Homa Bay samples have isotopic compositions that lie closest to the EM1 end-member, whereas those from Sukulu lie close to HIMU.
We have not plotted all of the data in Fig. 6 because some of the abundances are close to detection levels. This is particularly true for Pb. However, if we use the isotope dilution analyses given in Table 2 the normalized Pb ratios range between 26·8 (Sukulu) and 1027 (Oldoinyo Lengai).
Comparison of the normalized data from the East African carbonatites and those estimated for HIMU and EM1 shows the enrichment of the REE (especially the LREE) and Sr for the carbonatites. The other marked feature is the large negative Zr anomaly, which appears to be lacking in those lavas bearing the HIMU (Tubuai) and EM1 (Tristan and Inaccessible Island) isotopic signatures.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONANALYTICAL PROCEDURESRESULTSBULK CHEMICAL COMPOSITIONDISCUSSIONNATURE OF THE EAST...MANTLE COMPONENTSPETROGENETIC MODELSCONCLUSIONSREFERENCES |
|---|
With the present data Figs 2–5 show that most of the African Rift data broadly fit a simple model involving two-component mixing between HIMU and EM1 consistent with an origin in the mantle, with possible addition of some material from the upper mantle or lithosphere in some cases. We have also noted that the Reference Carbonatites, together with several of the rift carbonatites, depart substantially from this model, raising the question of the nature of the additional source material(s). Figure 7 tests the fit of all of the data in Figs 2–5 to three mantle end-member sources described by DMM, HIMU and EM1 in Nd–Sr–Pb space, using the method of Zindler et al. (1982)
. The fit of most of the rift samples with the mantle sources is again good, as indicated by their patterns in Figs 2–5. An important observation is that the reference and Chilwa samples also plot along the mantle plane, showing that mixtures of DMM, HIMU and EM1 can explain their isotope patterns, and that both end-members for the Reference Carbonatites and Chilwa group could be in the DMM–HIMU–EM1 mantle plane instead of placing one end-member in the continental crust.
|
We emphasize that the observed mixing correlations between HIMU and EM1 shown in Figs 2–5 are unique when compared with oceanic basalt data in general. In fact, the great majority of OIB samples plot off the HIMU–EM1 mixing line within DMM–HIMU–EM1 space in Pb–Sr–Nd correlation diagrams such that they formally require mixtures of three instead of only two mantle components (Hart et al., 1992). The same is true for most other carbonatites as well, as illustrated by the Reference Carbonatite pattern, as summarized previously by Tilton et al. (1998), which contains samples from four continents (North America, Europe, Africa, Australia and New Zealand), and oceanic as well as continental settings. Furthermore, the Chilwa Island carbonatite data (Simonetti & Bell, 1994b) likewise plot off the two-component mixing lines in Figs 4 and 5, suggesting that the HIMU–EM1 relationship terminates at the southern end of the Rift. It thus appears that the isotopic data from the East African carbonatites, involving only two mantle end-members, are quite unlike those from many other carbonatites or any single OIB dataset.
If we compare our East African Rift data for carbonatites with data from other areas of igneous activity involving silicate lavas in East Africa, we again see that there are major differences. Isotopic data from Pliocene to Quaternary basic volcanic rocks in the Hurl Hills, in northwestern Kenya, were considered to be the result of binary mixing between a plume end-member with HIMU affinity and a second end-member that might be either another plume component or lithospheric mantle (Class et al., 1994). In southwestern Uganda in the Virunga province, Pb, Sr and Nd isotopic data from rocks considered the products of mantle melting are consistent with binary mixing (Vollmer & Norry, 1983a), but with end-members reflecting a mantle fluid characterized by positive Nd values that invaded a lithospheric reservoir with negative Nd values. Our two proposed end-members differ markedly from those of Vollmer & Norry (1983b). One of their end-members is similar to EM2 as defined by Hauri et al. (1994) in a Nd vs Sr diagram, whereas the other has a Nd ratio of 0·51275 and Sr ratio of 0·7047. 206Pb/204Pb and 207Pb/204Pb ratios are approximately 19·4 and 15·7. In the eastern Virunga province, basanites and nephelinites supposedly generated by partial melting of enriched mantle show large variations in Nd (0·51204–0·51267) and Sr (0·7045–0·7070) isotope ratios (Rogers et al., 1998). Such differences were attributed to isotopically heterogeneous sources reflecting different ages (1 and 0·5 Ga) for parts of the mantle lithosphere.
In their detailed isotopic studies of individual carbonatite centres from East Africa, Kalt et al. (1997) indicated a more complicated model than we propose here. They emphasized the absence of linear trends in Sr–Nd and Nd–Pb isotope diagrams, and the lack of consistent apparent end-member compositions. Those features were attributed to small-scale isotopic heterogeneities in presumed lithospheric mantle sources. We believe that the difference in views occurs from considering the data on an over-literal, too detailed, scale. Even if the isotopic data for the EM1 and DMM end-members and their variations were perfectly known, which they are not, it would seem unlikely to find a perfect fit for plutons that are found over hundreds of kilometres along the rift. Even if the end-members originate in the deep mantle, possibly the core–mantle boundary, as sometimes proposed, the isotopes would not necessarily indicate uniform parent materials. For example, several workers have argued for heterogeneity in the deep mantle on the basis of seismic tomography (Kellogg et al., 1999; Van der Hilst & Karason, 1999). This raises the possibility that magmas could develop variations in isotopic compositions before reaching the outer mantle, even if isotope end-member patterns were initially identical. We believe the broad trend of the African Rift carbonatites data results from mixing between two HIMU-like and EM1-like mantle components, with the smaller scatter superimposed on the magmas as a result either of heterogeneity in the deep mantle (Kellogg et al., 1999) or of small amounts of contamination at higher levels, possibly in the lithosphere. In general, we find the fit of the isotope data in the African Rift carbonatites to HIMU–EM1 mixing to be more impressive than the ‘second-order’ variations.
An interesting observation noted by Kalt et al. (1997), and also by Simonetti & Bell (1994a) for nephelinites from Mt. Elgon and Napak, was a probable decoupling of Sr from Nd and Pb in some cases. Tilton et al. (1998) noted that although Sr from the Tanzanian Balla Gully carbonatite described by Kalt et al. (1997) does not fit simple HIMU–EM1 mixing in the Pb vs Sr and Nd vs Sr diagrams, the fit is good in the Pb vs Nd diagram. A similar case exists in the Amba Dongar carbonatite, India, where Sr seems to have been decoupled from Pb and Nd (Simonetti et al., 1995). Some of the scattering of data in the rift samples might thus be attributed to a greater mobility of Sr compared with Pb and Nd that could occur after magma formation. Other cases of seeming mobility of Pb relative to Sr and Nd have been reported in Pakistan carbonatites (Tilton et al., 1998). Apparently, one cannot necessarily attribute all of the scatter in carbonatite isotope data to differences in the original sources.
|
NATURE OF THE EAST AFRICAN LITHOSPHERE |
|---|
|
TOPABSTRACTINTRODUCTIONANALYTICAL PROCEDURESRESULTSBULK CHEMICAL COMPOSITIONDISCUSSIONNATURE OF THE EAST...MANTLE COMPONENTSPETROGENETIC MODELSCONCLUSIONSREFERENCES |
|---|
Because most carbonatites are found in continental settings, the common assumption has been that metasomatized, continental lithosphere plays an important role in their generation. Bell et al. (1982)
, on the basis of only Sr isotopes, originally suggested that carbonatites from Canada were formed from magmas generated by melting of Archaean depleted sub-continental lithosphere. New data from the East African Rift carbonatites were recently used to locate the source of carbonatitic melts within an isotopically heterogeneous lithospheric mantle (Kalt et al., 1997). However, other models based on Pb, Sr and Nd isotope data were more consistent with mixing between asthenosphere and lithosphere (Kwon et al., 1989; Bell & Simonetti, 1996; Tilton et al., 1998). Nelson et al. (1988), on the other hand, favoured generation of carbonatitic melts from the asthenosphere.
The isotopic heterogeneity of the lithosphere below East Africa has been attributed to various fluid or melt incursions into the lithosphere, at least since the Proterozoic. Hydrous alkaline pyroxenite xenoliths from Uganda, found in Quaternary to Recent nephelinites, leucitites and melilites in the Katwe–Kikorongo field have bulk Sr and Nd isotope signatures comparable with those in their host magmas, but Pb isotope ratios that are quite different (Davies & Lloyd, 1989). The variable isotope ratios define a linear array with a slope of 0·10, and this was interpreted as a recent mixing event between two end-members, derived from convecting upper mantle as small degree melts, or fluids (relatively poor in H2O and rich in CO2, F and Cl) that became trapped in the sub-continental lithosphere. Variable U/Pb ratios were considered to be controlled by different assemblages of apatite, perovskite and titanite within the sub-continental lithosphere, produced during a regional enrichment event that extended for at least 200 km along the rift valley.
Paslick et al. (1995) have arrived at a very different model from a study of mafic magmas in northern Tanzania. They noted unusual isotope patterns in some members of a suite of alkali basalts and nephelinites that gave high 206Pb/204Pb ratios similar to HIMU and 87Sr/86Sr ratios similar to EM1, as shown in Fig. 8a. Because of the absence of any crustal components in the area known to have such radiogenic Pb, they proposed a model in which aged (2 Ga) OIB material accounts for the EM1-like Sr and mixtures of OIB and lithospheric material account for the radiogenic Pb. However, we have made an estimate of the average Pb isotope ratios of the upper-crustal rocks of the Tanzanian Shield, using a physical composite made up of 15 granitoid samples from Kenya, Tanzania and Uganda. This composite yielded values of 206Pb/204Pb = 22·78, 207Pb/204Pb = 16·46 and 208Pb/204Pb = 41·13 which are extremely radiogenic. The 87Sr/86Sr and 143Nd/144Nd ratios of the composite are 0·73723 and 0·51172 (Nd = -17·9), respectively. Thus theoretically it might be possible to produce the anomalous isotope pattern with small amounts of contamination from the Precambrian crustal rocks of the Tanzanian Shield.
We note that although approximately one-third of the samples fit the Sr–Pb anomaly in Fig. 8a, the same samples do not reproduce the anomaly in the Nd–Pb in Fig. 8b; instead, they plot along the HIMU–EM1 mixing line. That pattern is reminiscent of the same patterns in the Amba Dongar (Simonetti et al., 1995) and Bala Gully (Kalt et al., 1997) carbonatites cited above.
Interestingly, we do not find similar complications in the carbonatites from Kerimasi, Ndiru Mbili and Oldoinyo Lengai, which are from the same general area as the silicate rocks, and have data that plot close to the EM1–HIMU mixing line for both Pb–Sr and Pb–Nd in Figs 7 and 8. This observation seems to illustrate the advantages of using carbonatite over silicate rocks as probes of mantle sources because of their rapid rise to the surface, and buffering against crustal assimilation by their high concentrations of Sr, Nd and often Pb in the magmas.
|
Other data from mantle xenoliths in Tanzania also suggest a complex history for the sub-continental lithosphere, involving both depletion and enrichment events. Metasomatized mantle xenoliths from Pello Hill, a small cone about 10 km from Oldoinyo Lengai, are enriched in K, Fe, Ti, OH and REE (Dawson & Smith, 1988), and a large spread of Nd, Pb and Sr isotope ratios from lherzolite xenoliths from the Lashaine cone has been attributed to a fractionation event within the mantle below Tanzania that took place at 2 Ga (Cohen et al., 1984). Interestingly, that is approximately the age yielded by Pb/Pb isotope data for the MORB–OIB population. Metasomatism by carbonatites of a depleted, refractory peridotite from the Olmani cinder cone was also proposed by Rudnick et al. (1993). Isotopic differences and differences in modal mineralogy between some of the Olmani peridotites were attributed to at least two episodes of carbonatite metasomatism. Pb isotopic data for nephelinites from Nyiragongo in the western branch of the East African Rift also indicate enrichment of the sub-continental lithosphere at 500 Ma (Vollmer & Norry, 1983b), an age that coincides with the widespread thermotectonic activity associated with the Pan-African orogeny, found throughout much of Africa. However, it is significant that the Pb isotope trends defined in the Vollmer & Norry (1983b) study do not fit the Pb isotope trends in our Fig. 3 for the African Rift carbonatites. For a given 206Pb/204Pb ratio, the rift samples have much lower 207Pb/204Pb and 208Pb/204Pb ratios compared with the Nyiragongo samples. If those workers’ conclusion that the Nyiragongo Pb data are derived from partial melting of ancient lithospheric domains is correct, their data argue for a different source for the rift carbonatites.
The repeated incursion of fluids and melts into the East African lithosphere from at least 2 Ga to the present time should result in an extremely heterogeneous lithosphere, in terms of both trace element abundance data and isotope ratios. In addition, further changes have been brought about by significant depletion of the lithosphere below East Africa because of repeated melting events. However, when we view xenolith data from Uganda and Tanzania (Cohen et al., 1984; Davies & Lloyd, 1989), they lie close to the end-on projection of the mantle plane (Fig. 9), and are surprisingly well behaved. It would thus appear that the xenoliths are representatives of sources that have not been affected by any metasomatic processes or any depletion events. Our interpretation of the xenolith data is that they simply reflect the addition of plume material to the lithosphere.
|
|
MANTLE COMPONENTS |
|---|
|
TOPABSTRACTINTRODUCTIONANALYTICAL PROCEDURESRESULTSBULK CHEMICAL COMPOSITIONDISCUSSIONNATURE OF THE EAST...MANTLE COMPONENTSPETROGENETIC MODELSCONCLUSIONSREFERENCES |
|---|
There has been considerable speculation in the literature about the locations of the sources for the EM1 and HIMU mantle components. Both mantle components must have been isolated from convecting mantle, i.e. from mixing and stirring, and require isolation within the mantle for substantial periods of time, possibly up to billions of years. HIMU (high U/Pb) represents one extreme of the LoNd array of Hart et al. (1986)
(low Rb/Sr, high Sm/Nd, high U/Pb) and is marked by data from the islands of St. Helena and Tubuai. The other, EM1 (enriched mantle 1; high Rb/Sr, low Sm/Nd, low U/Pb), was defined by data from the Walvis Ridge (Zindler & Hart, 1986; Hart, 1988). Hart et al. (1986) first suggested that both of these mantle components might reside in the sub-continental lithosphere, but this idea was subsequently discarded, because of the absence of such signatures in mantle xenoliths. Hart (1988) subsequently confined HIMU and EM1 to the core–mantle boundary because their isotopic compositions seemed to encompass the estimated composition of bulk silicate Earth. The HIMU component has been considered to be: (1) metasomatized mantle that involved the removal of Rb and perhaps Pb (Zindler & Hart, 1986), or (2) the enrichment of the mantle in U, Rb, and to a lesser extent REE, by CO2-rich fluids (Nakamura & Tatsumoto, 1988). A third model attributes the isotope characteristics of HIMU to those of ancient altered oceanic crust, perhaps involving the addition of oceanic sediments (Chase, 1981; White & Hofmann, 1982; Chauvel et al., 1992). EM1 can reflect: (1) recycled lower crust or lithosphere (Hawkesworth et al., 1984; Gerlach et al., 1988; Nakamura & Tatsumoto, 1988), (2) mantle metasomatized by CO2-rich fluids with high Rb/Sr and low U/Pb ratios (Zindler & Hart, 1986), (3) mixtures of HIMU and ancient pelagic sediments (Weaver, 1991; Chauvel et al., 1992), or (4) plagioclase-rich cumulates associated with recycled plume heads (Gasperini et al., 2000).
From present data EM1 also seems to be approximately confined to a large-scale isotopic anomaly, the DUPAL anomaly (Hart, 1984), centred in the Indian Ocean and in the Southern Hemisphere. This anomaly has been attributed to the pollution of the Indian Ocean mantle reservoir by old continental crust, sediments or continental lithosphere of Precambrian age. Hart (1984) suggested that the age of the DUPAL reservoir may be >3·0 Ga, on the basis of enrichment in U/Pb, Th/Pb and Rb/Sr ratios relative to other parts of the mantle, although such a model seems not to fit the low 206Pb/204Pb and 208Pb/204Pb ratios found in EM1.
The LoNd array was defined on the basis of data from Walvis Ridge, San Felix, Comoro Islands, Northeast Seamounts, St. Helena and Tubuai (Hart et al., 1986). The HIMU component from Tubuai involves oceanites, mugearites, nephelinite and basanites, and few, on the basis of their mg-number, represent primary mantle melts. Lavas from St. Helena share many of these characteristics. The isotopic data from the Walvis Ridge, the EM1 representative, were obtained from aphyric quartz tholeiites on the ridge crests. In their preferred model, Richardson et al. (1982) derived the Walvis Ridge basalts by partial melting of an E-type MORB source that had become heterogeneous on a small scale as a result of the introduction of small-volume melts and metasomatic fluids.
Another EM1 source is found in the Pacific Ocean at Pitcairn Island. Together with its related seamounts, Pitcairn Island defines another significant two-component mixing pattern in both Pb–Sr and Pb–Nd, one EM1 and the other FOZO,with little or no evidence for the addition of a DMM component, as noted by Woodhead & McCulloch (1989) and Woodhead & Dewy (1993). Their model for Pitcairn proposes mixing between ancient subducted oceanic crust and ancient subducted sedimentary material, but this is inconsistent with the isotopic data. In a mantle end-on view (Fig. 9), all of the samples from Pitcairn and the seamounts plot closely along the projection of the mantle plane, and do not reflect the wide departures from the line that are shown by present-day ocean sediments and island-arc magmas. We are intrigued by the fact that the predominant isotopic compositions of Pb, Nd and Sr from Pitcairn fit well within the field of estimated values for another widespread mantle component, FOZO (Hart et al., 1992), towards which many OIB patterns tend to converge. We refer readers to the discussion by Hauri et al. (1994), which covers the possible sources for FOZO, which they place in the lower mantle. From the Pitcairn Island and our carbonatite data, we can conclude that both FOZO and EM1 are widespread mantle components, and that both components are capable of generating carbonatitic melts.
Although the HIMU and EM1 end-member signatures were defined on the basis of isotopic data from a wide range of silicate mantle rock types from oceanic islands, we now show that the same isotopic signatures can be documented from carbonatites that are scattered over much of East Africa. Furthermore, where carbonatite and silicate assemblages occur together within the same body, the carbonatite isotopic compositions commonly identify the mantle end-members, whereas the silicate rocks show much more variation in their isotopic compositions that may reflect contamination from other sources, possibly lower crust in some cases. Harmer & Gittins’ (1998) summary of data from several African carbonatites illustrates such trends.
The geological settings of the East African carbonatites may provide some constraints on the origin and evolution of HIMU and EM1. Attempts to estimate the formation depths of HIMU and EM1 lavas on the basis of major element data from oceanic islands indicate values of about 49 kbar for HIMU and 35 kbar for EM1 (Chauvel et al., 1992), suggesting that the HIMU melts are generated from a source that was hotter and started melting deeper in the mantle than the source that produced EM1 (Chauvel et al., 1992). However, Chauvel et al. noted that certain characteristics of the trace element patterns of the Tubuai (HIMU) lavas, such as the depletion in the most incompatible elements and the low values of Ba/Nb, U/Nb and Th/Nb relative to primitive mantle, cannot be attributed solely to melting. Of course, the depth of melting does not necessarily correspond to the depth of origin of the source material.
On the basis of the data from the East African carbonatites, HIMU and EM1 are spatially related on an intimate scale, yet both characterize large parts of the mantle underlying Kenya, Tanzania and Uganda. Even within the same volcano, mixing between HIMU and EM1 can take place (e.g. Shombole; Bell & Peterson, 1991). Apparently, what can happen at mantle depths over much of East Africa can also happen on a much more localized scale below individual centres (Bell, 1998). That these two spatially related end-members can interact with one another on a discrete scale during melting suggests that these components are closely related, with the result that the whole isotopic range of the HIMU and EM1 components can be produced within a relatively restricted part of the mantle. These two end-members are so intimately related that they might simply reflect streaks within a convecting mantle, an idea that has been proposed by others (e.g. Allègre & Turcotte, 1986).
Our findings, along with those documented in the literature, indicate that any model proposed for the evolution of the HIMU and EM1 end-members must be consistent with the following observations:
- HIMU and EM1 are not restricted solely to oceanic environments, a finding that has also been shown by European volcanism (Wilson & Downes, 1991). Melting in the sub-continental mantle can thus generate magmas with isotopic signatures similar to those found in OIBs.
- Both end-members are spatially associated with one another on an intimate scale. Lavas, even within the same eruptive centres, can show isotope signatures that reflect the mixing of both mantle end-members.
- Both end-members have similar Sr/Nd and Nd/Pb ratios according to the linear mixing patterns, although radioactive parent/stable daughter element ratios have varied substantially over long periods of time according to isotopic data. This observation suggests that the radioactive parents are more mobile than the daughter elements.
- Mantle with HIMU and EM1 isotope characteristics is capable of generating melts that can lead to a wide variety of silicate rocks, including melts enriched in CO2 (see Bell, 1998).
|
PETROGENETIC MODELS |
|---|
|
TOPABSTRACTINTRODUCTIONANALYTICAL PROCEDURESRESULTSBULK CHEMICAL COMPOSITIONDISCUSSIONNATURE OF THE EAST...MANTLE COMPONENTSPETROGENETIC MODELSCONCLUSIONSREFERENCES |
|---|
Most models proposed for the generation of the East African carbonatites involve metasomatized lithosphere. One such model involves two end-members, one represented by metasomatic incursions of carbonated-hyperalkaline silicate melts and their subsequent mixing with a depleted sub-continental upper mantle that acts as a host for the metasomatic veins (Meen et al., 1989
). The high U and Th contents of the metasomatic veins could generate highly radiogenic Pb in old metasomatic materials (Meen et al., 1989), thus the model uses lithospheric sources to generate magmas with HIMU affinities.
An alternative explanation for the isotopic characteristics of carbonatites and silicate rocks, based on data from Oldoinyo Lengai, involves a two-stage process in which a sub-continental lithosphere with an EM1 signature was metasomatized by fluids or melts similar to HIMU generated from an upwelling plume (Bell & Simonetti, 1996). The isotopic similarity of some mantle xenoliths from East Africa to HIMU led Bell & Simonetti (1996) to propose mixing between a sub-continental lithosphere containing the EM1 component and a HIMU-fluid generated from an upwelling plume. Such a model was used to explain the generation of the HIMU–EM1 signature for the carbonatitic flows from Oldoinyo Lengai. Hoernle & Tilton (1991) proposed a similar model to explain isotope patterns in carbonatites and silicate rocks from Fuerteventura Island of the Canary Islands.
Evidence based on pre-rift uplift, as well as the morphological and geochemical features of early basalts and flood basalts, led to a model proposed by Smith (1994) that involved migration of a mantle plume into the lithospheric mantle below Kenya during the early Miocene. Small in size, with a diameter of about 100–150 km, the plume was focused beneath a zone of weakness marking the boundary between the Tanzanian Craton and the adjacent mobile fold belt. The situation, however, may be slightly more complicated. Two plumes in East Africa have now been proposed by George et al. (1998), an older marked by the Kenyan plume (45–35 Ma), and a younger (19 Ma old) corresponding to the Afar plume. How far these plumes migrated laterally and their overall size still remain unknown. One possible model proposed by Ebinger & Sleep (1998) suggests that the Afar plume, the major plume in Africa, has affected a much larger region than was previously thought, and that plume material was channelled laterally, its migration controlled by steep gradients along the margins of the Tanzanian Craton and the mobile belt boundaries. A channelled plume, such as that envisaged by Ebinger & Sleep (1998), extends the regional coverage of plume activity from Afar in Ethiopia to as far as the Cameroon line at the Atlantic, and the Comoro Islands in the India Ocean. The recent finding of carbonatite metasomatism of oceanic upper mantle from lherzolite and wehrlite xenoliths from La Grille volcano, Grand Comore (Coltori et al., 1999) could be related to the younger Afar plume. If the lithosphere is impregnated by carbonatitic melts, and then remelted after any substantial period of time, it would seem to be extremely unlikely that the isotope ratios retain any of the linear relationships shown by the data from East Africa.
Included among the substantial amount of evidence for plume activity in East Africa is the domal uplift (‘superswell’) of the East Africa plateau, and the voluminous outpourings of plateau lavas. These are but two of the surface expressions of plume activity, but carbonatite activity might provide further indications, especially because carbonatitic melts or their parents are produced by low degrees of partial melting. If the East African plumes contain both the HIMU and the EM1 signatures, it would be one way of explaining the intimate relationships of both components over much of East Africa. The idea of an isotopically heterogeneous mantle plume (IHMP) is not new (e.g. Gerlach et al., 1988). We know that some plumes are characterized by strong HIMU-like signatures (e.g. St. Helena in the Atlantic Ocean and Tubuai in the Pacific Ocean) whereas others have some of the isotopic characteristics of EM1 (e.g. the Tristan and Gough plumes). If we allow that plumes can contain both components, then generation of magmas during decompression melting, coupled with mixing at the upper levels of the plume, for example within the plume head, could explain the intimate association of both end-members, and their consistent isotopic composition over relatively large areas. The wide range of isotopic signatures shown by many carbonatite provinces would be consistent with an IHMP model. A plume origin for some Brazilian carbonatites favoured by Toyoda et al. (1994) invoked a model in which carbonate-rich material from the Tristan plume was trapped in the lithosphere and reactivated by the Trindade hotspot. Inspection of their isotope ratio diagrams shows that, although most of the data cluster close to EM1, the spread in isotope ratios requires mixing between HIMU and EM1 components. Interestingly, their Sr vs Pb diagrams indicate no significant DMM component in the Trindade carbonatites, although it is formally required in other oceanic carbonatites at the Canary and Cape Verde Islands. This may be related to the fact that the Canary and Cape Verde Islands (part of our Reference Carbonatite group) hotspots are located outside of the DUPAL anomaly zone, which is home to most, if not all, EM1-rich magmas. It may be significant that five of the seven Reference Carbonatites are located in the northern hemisphere, outside of the DUPAL anomaly zone.
The presence of excess 129Xe in carbonatites (Sasada et al., 1997) provides some additional important constraints on the origin of carbonatites, and perhaps on HIMU and EM1. An interesting observation made by Sasada et al. is that carbonatites with ages varying between 0·2 and 1·8 Ga, and with Sr and Nd signatures that reflect both enriched and depleted sources, exhibit 129Xe/130Xe ratios between 7·2 and 8·6, all much higher than atmospheric ratios (6·5), and 40Ar/36Ar ratios of 304–42 420. It would be hard to imagine that the primordial 129Xe/130Xe signature in carbonatites, comparable with those in diamonds, would be preserved if the carbon in carbonatites were related to subducted materials, unless the subducted material exchanged gases with those of the deep mantle. If such a source were the deep mantle, e.g. the D'' layer at the core–mantle boundary, then migration of volatiles during plume migration and their interaction with either HIMU or EM1 components might produce the range of signatures observed in the East African carbonatites.
Sufficient evidence is now available to assume that many, perhaps all, carbonatites are related to plume activity. In that connection, it is interesting that on the basis of seismic tomography, East Africa has been shown to be one of two sites that indicate major upwelling from the core–mantle boundary, the other being under the SW Pacific Ocean (Grand, 1994; Van der Hilst et al., 1997). There are other smaller deep low-velocity volumes in the tomographic images, but the African Rift and the SW Pacific Ocean are the prominent ones. This evidence suggests that the sources for one or both of the two end-members of the African Rift carbonatites could originate in the lower mantle, possibly, but not necessarily, at the core–mantle boundary. We feel it unlikely that the HIMU and EM1 components with such a widespread geographical distribution and uniform isotopic composition in both continental and ocean environments can be attributed to far-flung, localized changes within the mantle involving subduction of crustally derived materials, a process that is intermittent in time and that must involve isotopically heterogeneous and different materials. In fact, Kellogg et al. (1999) and Van der Hilst & Karason (1999) suggested that compositionally distinct high- and low-velocity mantle domains that produce MORB and OIB are generally located within the bottom 1000 km of the mantle, but that plumes do not necessarily originate from the core–mantle boundary.
More recent detailed shear-wave velocity studies show the presence of a tilted low-velocity channel beneath Africa that extends diagonally from the core–mantle boundary below the southwestern tip of the continent into the upper mantle underlying the East African Rift (Ritsema et al., 1999). Covering an extensive region at the core–mantle boundary (4000 km x 2000 km), the anomaly then thins and forms patches of relatively low-velocity mantle beneath central and NE Africa and the southern Indian Ocean. This anomaly suggests that the Cenozoic volcanism in the Afar region and active rifting beneath the East African Rift is linked to an extensive anomaly at the core–mantle boundary. This finding now provides a connection between magmatism in East Africa and the core–mantle boundary, a continuity that offers direct access to mantle material stored within the lowermost parts of the mantle.
Recently, King & Ritsema (2000) have proposed another model for generating hotspot volcanic rocks within cratons by edge-driven convection in shallow (<660 km depth) mantle, which they offer as an alternative to deep plumes to explain intraplate African and South American hotspot volcanism. We do not believe that the process, if true, has affected the Ugandan and Tanzanian carbonatites we have studied. We would expect shallow mantle convection to give the maximum opportunity to produce DMM imprints in the resulting magmas (e.g. Hofmann, 1997; Albarède & Van der Hilst, 1999). In all of the isotope ratio diagrams we see no evidence for the involvement of DMM, and all of the carbonatites record dominant EM1–HIMU mantle mixing. Although some of the Tanzanian volcanic rocks do contain radiogenic Sr that could be either from oceanic sediments or perhaps recycled material from the craton itself, the marked absence of any DMM component argues against any involvement with upper-mantle material and hence shallow mantle convection. For these reasons we prefer the deep-mantle conduit model over the end-driven shallow convection model for the rift carbonatites in our study.
A slightly modified version of the Van der Hilst & Karason (1999) model for the mantle has been proposed by Albarède & Van der Hilst (1999) on the basis of an attempted best fit to geophysical and geochemical data. Briefly, an upper transition zone 400–1000 km deep divides the mantle into two dynamic regimes. The upper layer is well mixed, rather degassed and the source of MORB. The bottom layer (1000–2000 km) has a higher density with slower mixing, and contains undegassed material enriched in heat-producing elements formed early in Earth history. Albarède & Van der Hilst (1999) have placed the OIB source at the top of the lower layer, or in the boundary between the two layers. The boundary layer is also assumed to be the graveyard for lithospheric plates that were not subducted to the core–mantle boundary, and that are depleted in lithophile elements. It should be noted that according to this model, a ‘lithospheric imprint’ in a magma based upon isotopic data does not necessarily imply mixing of plume material with the shallow sub-continental lithospheric material. Rather, EM1-rich material may originate from part of the lower mantle, and then be conveyed towards the surface by rising HIMU-rich plumes.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONANALYTICAL PROCEDURESRESULTSBULK CHEMICAL COMPOSITIONDISCUSSIONNATURE OF THE EAST...MANTLE COMPONENTSPETROGENETIC MODELSCONCLUSIONSREFERENCES |
|---|
- Our study has shown that HIMU and EM1 mantle components, first recognized in OIBs, also exist beneath part of the African continent.
- Nd–Pb and Nd–Sr isotope data yield clear evidence that most East African Rift carbonatites originate predominantly from mixtures between the two mantle end-members, HIMU and EM1, an unusual pattern.
- If the isotopic patterns for the xenoliths from Kenya and Tanzania reflect those of the lithosphere, then the patterns appear to be controlled by plume interaction.
- We postulate that deep mantle seismic tomography, which indicates a connection between the general African Rift area and the core–mantle boundary with lower velocity material, is related to the unusual HIMU–EM1 mixing pattern of the Rift carbonatites. From that observation we place the source of the EM1 component in the deep (lower 1000 km) mantle, or possibly at or near the core–mantle boundary.
- The two carbonatites with isotope patterns closest to EM1 (Homa Bay) and HIMU (Sukulu) crop out within 140 km of each other, indicating heterogeneity within the mantle sources over very small distances.
- If we assume an isotopically heterogeneous mantle plume, then low-degree partial melts produced by decompression or release of volatiles can generate both carbonatites as well as a wide variety of alkaline silicate rocks that cover the complete isotopic spectrum between HIMU and EM1.
- We also suggest that a mantle with either HIMU or EM1 signatures, or a mixture of both, can not only generate OIBs but can also produce melts that can ultimately lead to carbonatite.
- The Chilwa carbonatites have isotope patterns that clearly fit the group which we call Reference Carbonatites, which are explained as mixtures between EM1 and FOZO. The Napak carbonatite may also belong to that group.
|
ACKNOWLEDGEMENTS |
|---|
Julia Bryce, Jianping Wen, Hassan Mirnejad and Ina de Jong are thanked for assistance with the analytical work and the preparation of the manuscript. A. Woolley kindly provided a copy of his most recent compilation and description of the carbonatites of East Africa, and additional samples from Bukusu and Sukulu were donated to us by the Natural History Museum (London). We thank Natalie Morisset for allowing us to use her Sr and Nd isotope analyses from Panda Hill. We are grateful to Steve Goldstein, Hassan Mirnejad, Jonathan Patchett, Alexey Rukhlov, Greg Yaxley, Marjorie Wilson, and an anonymous reviewer for their detailed comments. This work was partially supported by grant A78134 from the Natural Sciences and Engineering Research Council (Canada) awarded to K.B., and National Science Foundation grants to G.T.
|
FOOTNOTES |
|---|
*Corresponding author. Telephone: 613-520-2600 (ext. 4419). E-mail: kbell{at}ccs.carleton.ca
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONANALYTICAL PROCEDURESRESULTSBULK CHEMICAL COMPOSITIONDISCUSSIONNATURE OF THE EAST...MANTLE COMPONENTSPETROGENETIC MODELSCONCLUSIONSREFERENCES |
|---|
Albarède, F. & van der Hilst, R. D. (1999). New mantle convection model may reconcile conflicting evidence. EOS Transactions, American Geophysical Union 80, 535–539.
Allègre, C. J. & Turcotte, D. L. (1986). Implications of a two-component marble-cake mantle. Nature 323, 123–127.
Bailey, D. K. (1993). Carbonate magmas. Journal of the Geological Society, London 150, 637–651.
Barreiro, B. & Cooper, C. F. (1987). A Sr, Nd and Pb isotope study of alkaline lamprophyres and related rocks from Westland and Otago, South Island, New Zealand. In: Morris, E. & Pasteris, J. (eds) Mantle Metasomatism and Alkaline Magmatism. Geological Society of America, Special Paper 215, 115–125.
Basu, A. R. & Tatsumoto, M. (1980). Nd-isotopes in selected mantle-derived rocks and minerals and their implications for mantle evolution. Contributions to Mineralogy and Petrology 75, 43–54.
Bell, K. (1998). Radiogenic isotope constraints on relationships between carbonatites and associated silicate rocks—a brief review. Journal of Petrology 39, 1987–1996.
Bell, K. & Blenkinsop, J. (1987a). Nd and Sr isotopic compositions of East African carbonatites: implications for mantle heterogeneity. Geology 15, 99–102.
Bell, K. & Blenkinsop, J. (1987b). Archean depleted mantle—evidence from Nd and Sr initial isotopic ratios of carbonatites. Geochimica et Cosmochimica Acta 51, 291–298.
Bell, K. & Blenkinsop, J. (1989). Neodymium and strontium isotope geochemistry of carbonatites. In: Bell, K. (ed.) Carbonatites: Genesis and Evolution. London: Unwin Hyman, pp. 278–300.
Bell, K. & Dawson, J. B. (1995). Nd and Sr isotope systematics of the active carbonatite volcano, Oldoinyo Lengai. In: Bell, K. & Keller, J. (eds) Carbonatite Volcanism: Oldoinyo Lengai and the Petrogenesis of Natrocarbonatites. IAVCEI Proceedings in Volcanology 4. Berlin: Springer, pp. 100–112.
Bell, K. & Keller, J. (eds) (1995). Carbonatite Volcanism: Oldoinyo Lengai and the Petrogenesis of Natrocarbonatites. IAVCEI Proceedings in Volcanology 4. Berlin: Springer.
Bell, K. & Peterson, T. D. (1991). Nd and Sr isotope systematics of Shombole volcano, East Africa, and the links between nephelinites, phonolites, and carbonatites. Geology 19, 582–585.
Bell, K. & Simonetti, A. (1996). Carbonatite magmatism and plume activity: implications from the Nd, Pb and Sr isotope systematics of Oldoinyo Lengai. Journal of Petrology 37, 1321–1339.
Bell, K., Blenkinsop, J., Cole, T. J. S. & Menagh, D. P. (1982). Evidence from Sr isotopes for long-lived heterogeneities in the upper mantle. Nature 298, 251–253.
Brooker, R. A. (1998). The effect of CO2 saturation on immiscibility between silicate and carbonate liquids: an experimental study. Journal of Petrology 39, 1905–1915.
Cahen, L. & Snelling, N. J. (1984). The Geochronology and Evolution of Africa. Oxford: Clarendon Press.
Chase, C. G. (1981). Oceanic island Pb: two stage histories and mantle evolution. Earth and Planetary Science Letters 52, 277–284.
Chauvel, C., Hofmann, A. W. & Vidal, P. (1992). HIMU-EM: the French Polynesian connection. Earth and Planetary Science Letters 110, 99–119.[Web of Science]
Class, C., Altherr, R., Volker, F., Eberz, G. & McCulloch, M. T. (1994). Geochemistry of Pliocene to Quaternary alkali basalts from the Huri Hills, northern Kenya. Chemical Geology 113, 1–22.
Cliff, R. A., Baker, P. E. & Mateer, N. J. (1991). Geochemistry of Inaccessible Island volcanics. Earth and Planetary Science Letters 92, 251–260.
Cohen, R. S., O’Nions, R. K. & Dawson, J. B. (1984). Isotope geochemistry of xenoliths from East Africa: implications for development of mantle reservoirs and their interaction. Earth and Planetary Science Letters 68, 209–220.
Coltorti, M., Bonadiman, C., Hinton, R. W., Siena, F. & Upton, B. G. J. (1999). Carbonatite metasomatism of the oceanic upper mantle: evidence from clinopyroxenes and glasses in ultramafic xenoliths of Grande Comore, Indian Ocean. Journal of Petrology 40, 133–165.
Davies, G. R. & Lloyd, F. E. (1989). Pb–Sr–Nd isotope and trace element data bearing on the origin of the potassium subcontinental lithosphere beneath south-west Uganda. In: Ross, J., Jaques, A. L., Ferguson, J., Green, D. H., O’Reilly, S. Y., Danchin, R. V. & Janse, A. J. A. (eds) Kimberlites and Related Rocks. Sydney, N.S.W.: Geological Society of Australia, pp. 784–794.
Dawson, J. B. & Smith, J. V. (1988). Metasomatised and veined upper mantle xenoliths from Pello Hill, Tanzania: evidence for anomalously light mantle beneath the Tanzanian sector of the East African Rift Valley. Contributions to Mineralogy and Petrology 100, 510–527.
Deines, P. (1989). Stable isotope variations in carbonatites. In: Bell, K. (ed.) Carbonatites: Genesis and Evolution. London: Unwin Hyman, pp. 301–359.
Ebinger, C. J. & Sleep, N. H. (1998). Cenozoic magmatism throughout east Africa resulting from impact of a single plume. Nature 395, 788–791.
Gasperini, D., Blichert-Toft, J., Bosch, D., Del Moro, A., Macera, P., Télouk, P. & Albarède, F. (2000). Evidence from Sardinian basalt geochemistry for recycling of plume heads into the Earth’s mantle. Nature 408, 701–704.[Medline]
George, R., Rogers, N. & Kelley, S. (1998). Earliest magmatism in Ethiopia: evidence for two mantle plumes in one flood basalt province. Geology 26, 923–926.
Gerlach, D. C., Cliff, R. A., Davies, G. R., Norry, M. & Hodgson, N. (1988). Magma sources of the Cape Verdes archipelago: isotopic and trace element constraints. Geochimica et Cosmochimica Acta 52, 2979–2992.[Web of Science]
Grand, S. P. (1994). Mantle shear structure beneath the Americas and surrounding oceans. Journal of Geophysical Research 99, 11591–11621.[Web of Science]
Green, D. H. & Wallace, M. E. (1988). Mantle metasomatism by ephemeral carbonatite melts. Nature 336, 459–462.
Grünenfelder, M. H., Tilton, G. R., Bell, K. & Blenkinsop, J. (1986). Lead and strontium isotope relationships in the Oka carbonatite complex, Quebec. Geochimica et Cosmochimica Acta 50, 461–468.
Harmer, R. E. & Gittins, J. (1998). The case for primary, mantle-derived carbonatite magma. Journal of Petrology 39, 1895–1903.
Hart, S. R. (1984). A large-scale isotope anomaly in the southern hemisphere. Nature 309, 753–757.
Hart, S. R. (1988). Heterogeneous mantle domains: signatures, genesis and mixing chronologies. Earth and Planetary Science Letters 90, 273–296.
Hart, S. R., Gerlach, D. C. & White, W. M. (1986). A possible new Sr–Nd–Pb mantle array and consequences for mantle mixing. Geochimica et Cosmochimica Acta 50, 1551–1557.
Hart, S. R., Hauri, E. H., Oschmann, L. A. & Whitehead, J. A. (1992). Mantle plumes and entrainment: isotopic evidence. Science 256, 517–520.
Hauri, E. H., Whitehead, J. A. & Hart, S. R. (1994). Fluid dynamic and geochemical aspects of entrainment in mantle plumes. Journal of Geophysical Research 99, 24275–24300.[Web of Science]
Hawkesworth, C. J., Rogers, N. W., van Calsteren, P. W. C. & Menzies, M. A. (1984). Mantle enrichment processes. Nature 311, 331–335.
Hoernle, K. A. & Tilton, G. R. (1991). Sr–Nd–Pb isotope data for Fuerteventura (Canary Islands) basal complex and subaerial volcanics: applications to magma genesis and evolution. Schweizerische Mineralogische und Petrographische Mitteilungen 71, 3–18.
Hoernle, K., Tilton, G. R., Le Bas, M. J., Duggen, S. & Garb-Schoenberg, D. (2001). Geochemistry of oceanic carbonates compared with continental carbonatites: mantle recycling of ocean crust and marine carbonate? Contributions to Mineralogy and Petrology (in press).
Hofmann, A. W. (1997). Mantle geochemistry: the message from oceanic volcanism. Nature 385, 219–229.
Hornig-Kjarsgaard (1998). Rare earth elements in sövitic carbonatites and their mineral phases. Journal of Petrology 39, 2105–2121.
Kalt, A., Hegner, E. & Satir, M. (1997). Nd, Sr and Pb isotopic evidence for diverse lithospheric mantle sources of East African Rift carbonatites. Tectonophysics 278, 31–45.
Keller, J. & Hoefs, J. (1995). Stable isotope characteristics of recent natrocarbonatites from Oldoinyo Lengai. In: Bell, K. & Keller, J. (eds) Carbonatite Volcanism: Oldoinyo Lengai and the Petrogenesis of Natrocarbonatites. IAVCEI Proceedings in Volcanology 4. Berlin: Springer, pp. 113–123.
Kellogg, L. H., Hager, B. H. & van der Hilst, R. D. (1999). Compositional stratification in the deep mantle. Science 283, 1881–1884.
King, B. C. (1949). The Napak area of southern Karamoja, Uganda. Geological Survey of Uganda, Memoir V, 57.
King, S. D. & Ritsema, J. (2000). African hot spot volcanism: small-scale convection in the upper mantle beneath cratons. Science 290, 1137–1140.
Kjarsgaard, B. A., Hamilton, D. L. & Peterson, T. D. (1995). Peralkaline nephelinite/carbonatite liquid immiscibility: comparison of phase composition in experiments and natural lavas from Oldoinyo Lengai. In: Bell, K. & Keller, J. (eds) Carbonatite Volcanism: Oldoinyo Lengai and the Petrogenesis of Natrocarbonatites. IAVCEI Proceedings in Volcanology 4. Berlin: Springer, pp. 163–190.
Koster van Groos, A. F. & Wyllie, P. J. (1963). Experimental data bearing on the role of liquid immiscibility in the genesis of carbonatites. Nature 199, 801–802.
Kwon, S.-T., Tilton, G. R. & Grünenfelder, M. H. (1989). Lead isotope relationships in carbonatites and alkalic complexes: an overview. In: Bell, K. (ed.) Carbonatites: Genesis and Evolution. London: Unwin Hyman, pp. 360–387.
Le Roex, A. P., Cliff, R. A. & Adair, B. J. I. (1990). Tristan da Cunha, South Atlantic: geochemistry and petrogenesis of a basanite–phonolite lava series. Journal of Petrology 31, 779–812.
Marty, B., Tolstikhin, I., Kamensky, I. L., Nivin, V., Balaganskaya, E. & Zimmermann, J.-L. (1998). Plume-derived rare gases in 380 Ma carbonatites from the Kola region (Russia) and the argon isotopic signature in the deep mantle. Earth and Planetary Science Letters 164, 179–192.
McDonough, W. F. (1999). Earth’s core. In: Marshall, C. P. & Fairbridge, R. F. (eds) The Encyclopedia of Geochemistry. Amsterdam: Kluwer Academic, pp. 151–156.
McDonough, W. F. & Sun, S. S. (1995). The composition of the Earth. Chemical Geology 120, 223–253.[Web of Science]
Meen, J. K., Ayers, J. C. & Fregeau, E. J. (1989). A model of mantle metasomatism by carbonated alkaline melts: trace-element and isotopic compositions of mantle source regions of carbonatite and other continental igneous rocks. In: Bell, K. (ed.) Carbonatites: Genesis and Evolution. London: Unwin Hyman, pp. 388–404.
Morisset, N. (1992). Stable isotope and radioisotope geochemistry of the Panda Hill carbonatite, Tanzania. M.Sc. thesis, Carleton University, Ottawa, Ont., 91 pp.
Nakamura, Y. & Tatsumoto, M. (1988). Pb, Nd and Sr isotopic evidence for a multi-component source for rocks of the Cook–Austral Islands and heterogeneities of mantle plumes. Geochimica et Cosmochimica Acta 52, 2909–2924.
Nelson, D. R., Chivas, A. R., Chappell, B. W. & McCulloch, M. T. (1988). Geochemical and isotopic systematics in carbonatites and implications for the evolution of ocean-island sources. Geochimica et Cosmochimica Acta 52, 1–17.
Paslick, C., Halliday, A., Jame, D. & Dawson, J. B. (1995). Enrichment of continental lithosphere by OIB melts: isotopic evidence from the volcanic province of northern Tanzania. Earth and Planetary Science Letters 130, 109–126.
Richardson, S. H., Erlank, A. J., Duncan, A. R. & Reid, D. L. (1982). Correlated Nd, Sr and Pb isotope variations in Walvis Ridge basalts and implications for the evolution of their mantle sources. Earth and Planetary Science Letters 59, 327–342.
Ritsema, J., van Heijst, J. & Woodhouse, J. H. (1999). Complex shear wave velocity structure imaged beneath Africa and Iceland. Science 286, 1925–1928.
Rogers, N. W., James, D., Kelley, S. P. & De Mulder, M. (1998). The generation of potassic lavas in the eastern Virunga province, Rwanda. Journal of Petrology 39, 1223–1247.
Rudnick, R. L., McDonough, W. F. & Chappell, B. W. (1993). Carbonatite metasomatism in the northern Tanzanian mantle: petrographic and geochemical characteristics. Earth and Planetary Science Letters 114, 463–475.[Web of Science]
Sasada, T., Hiyagon, H., Bell, K. & Ebihara, M. (1997). Mantle-derived noble gases in carbonatites. Geochimica et Cosmochimica Acta 61, 4219–4228.
Simonetti, A. & Bell, K. (1994a). Nd, Pb and Sr isotopic data from the Napak carbonatite–nephelinite centre, eastern Uganda: an example of open-system crystal fractionation. Contributions to Mineralogy and Petrology 115, 356–366.
Simonetti, A. & Bell, K. (1994b). Isotopic and geochemical investigation of the Chilwa Island carbonatite complex, Malawi: evidence for a depleted mantle source region, liquid immiscibility, and open-system behavior. Journal of Petrology 35, 1597–1621.
Simonetti, A., Bell, K. & Viladkar, S. G. (1995). Isotopic data from the Amba Dongar carbonatite complex, west central India: evidence for an enriched mantle source. Chemical Geology (Isotope Geoscience Section) 122, 185–198.
Simonetti, A., Bell, K. & Shrady, C. (1997). Trace- and rare-earth element geochemistry of the June 1993 natrocarbonatite lavas, Oldoinyo Lengai (Tanzania): implications for the origin of carbonatite magmas. Journal of Volcanology and Geothermal Research 75, 89–106.[Web of Science]
Smith, M. (1994). Stratigraphic and structural constraints on mechanisms of active rifting in the Gregory Rift, Kenya. Tectonophysics 236, 3–22.
Sweeney, R. J. (1994). Carbonatite melt compositions in the Earth’s mantle. Earth and Planetary Science Letters 128, 259–270.
Tatsumoto, M. (1978). Isotopic composition of lead in oceanic basalt and its implication to mantle evolution. Earth and Planetary Science Letters 38, 63–87.
Tilton, G. R. & Bell, K. (1994). Sr–Nd–Pb isotope relationships in Late Archean carbonatites and alkaline complexes: applications to the geochemical evolution of the Archean mantle. Geochimica et Cosmochimica Acta 58, 3145–3154.
Tilton, G. R., Kwon, S.-T. & Frost, D. M. (1987). Isotopic relationships in Arkansas Cretaceous alkalic complexes. In: Morris, E. & Pasteris, J. (eds) Mantle Metasomatism and Alkaline Magmatism. Geological Society of America, Special Paper 215, 241–249.
Tilton, G. R., Bryce, J. G. & Mateen, A. (1998). Pb–Sr–Nd isotope data from 30 and 300 Ma collision zone carbonatites in northwest Pakistan. Journal of Petrology 39, 1865–1874.
Todt, W., Cliff, R. A., Houser, A. & Hofmann, A. W. (1995). Evaluation of a 202Pb–205Pb double spike for high precision analyses. In: Basu, A. & Hart, S. (eds) Earth Processes: Reading the Isotopic Code. Geophysical Monograph, American Geophysical Union 95, 429–437.
Toyoda, K., Horiyuki, H. & Tokonami, M. (1994). Dupal anomaly of Brazilian carbonatites: geochemical correlations with hotspots in the South Atlantic and implications for the mantle source. Earth and Planetary Science Letters 126, 315–331.
Van der Hilst, F. D. & Karason, H. (1999). Compositional heterogeneity in the bottom 1000 kilometers of Earth’s mantle: toward a hybrid convection model. Science 283, 1885–1888.
Van der Hilst, F. D., Widiyantoro, S. & Engdahl, E. R. (1997). Evidence for deep mantle circulation from global tomography. Nature 386, 578–584.
Veksler, I. V., Nielsen, T. F. D. & Sokolov, S. V. (1998). Mineralogy of crystallized melt inclusions from Gardiner and Kovdor ultramafic alkaline complexes: implications for carbonatite genesis. Journal of Petrology 39, 2015–2031.
Vollmer, R. & Norry, M. J. (1983a). Possible origin of K-rich volcanic rocks from Virunga, East Africa, by metasomatism of continental crustal material: Pb, Nd and Sr isotopic evidence. Earth and Planetary Science Letters 64, 374–386.
Vollmer, R. & Norry, M. J. (1983b). Unusual isotopic variations in Nyiragongo nephelinites. Nature 301, 141–143.
Wallace, M. E. & Green, D. H. (1988). An experimental determination of primary carbonatite magma composition. Nature 335, 343–346.
Weaver, B. L. (1991). Trace element evidence for the origin of ocean-island basalts. Geology 19, 123–126.
White, W. M. & Hofmann, A. W. (1982). Sr and Nd isotope geochemistry of oceanic basalts and mantle evolution. Nature 296, 821–825.
Wilson, M. & Downes, H. (1991). Tertiary–Quaternary extension-related alkaline magmatism in western and central Europe. Journal of Petrology 32, 811–849.
Woodhead, J. D. & Dewey, C. W. (1993). Geochemistry of the Pitcairn seamounts, I: source character and temporal trends. Earth and Planetary Science Letters 116, 81–99.
Woodhead, J. D. & McCulloch, M. T. (1989). Ancient seafloor signals in Pitcairn Island lavas and evidence for large amplitude, small length-scale mantle heterogeneities. Earth and Planetary Science Letters 94, 257–273.
Woolley, A. R. (1989). The spatial and temporal distribution of carbonatites. In: Bell, K. (ed.) Carbonatites: Genesis and Evolution. London: Unwin Hyman, pp. 15–37.
Wyllie, P. J. & Huang, W.-L. (1975). Peridotite, kimberlite and carbonatite explained in the system CaO–MgO–SiO2–CO2. Geology 3, 621–624.
Wyllie, P. J. & Lee, W.-J. (1998). Model system controls on conditions for formation of magnesiocarbonatite and calciocarbonatite magmas from the mantle. Journal of Petrology 39, 1885–1893.
Zindler, A. & Hart, S. R. (1986). Chemical dynamics. Annual Review of Earth and Planetary Sciences 14, 493–571.[Web of Science]
Zindler, A., Jagoutz, E. & Goldstein, S. (1982). Nd, Sr and Pb isotopic systematics in a three-component mantle: a new perspective. Nature 298, 519–523.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. B. Dawson Chapter 8 Regional comparisons, petrochemistry and petrogenesis Geological Society, London, Memoirs, January 1, 2008; 33(1): 79 - 89. [Full Text] [PDF]
|
|
|
|
 |
|
 T. FURMAN, K. M. KALETA, J. G. BRYCE, and B. B. HANAN Tertiary Mafic Lavas of Turkana, Kenya: Constraints on East African Plume Structure and the Occurrence of High-{micro} Volcanism in Africa J. Petrology, June 1, 2006; 47(6): 1221 - 1244. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. H. Mitchell CARBONATITES AND CARBONATITES AND CARBONATITES Can Mineral, December 1, 2005; 43(6): 2049 - 2068. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. R. Bell and R. O. Moore Deep chemical structure of the southern African mantle from kimberlite megacrysts South African Journal of Geology, June 1, 2004; 107(1-2): 59 - 80. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. EWART, J. S. MARSH, S. C. MILNER, A. R. DUNCAN, B. S. KAMBER, and R. A. ARMSTRONG Petrology and Geochemistry of Early Cretaceous Bimodal Continental Flood Volcanism of the NW Etendeka, Namibia. Part 1: Introduction, Mafic Lavas and Re-evaluation of Mantle Source Components J. Petrology, January 1, 2004; 45(1): 59 - 105. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. P. LE ROEX, D. R. BELL, and P. DAVIS Petrogenesis of Group I Kimberlites from Kimberley, South Africa: Evidence from Bulk-rock Geochemistry J. Petrology, December 1, 2003; 44(12): 2261 - 2286. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. PECCERILLO, M. R. BARBERIO, G. YIRGU, D. AYALEW, M. BARBIERI, and T. W. WU Relationships between Mafic and Peralkaline Silicic Magmatism in Continental Rift Settings: a Petrological, Geochemical and Isotopic Study of the Gedemsa Volcano, Central Ethiopian Rift J. Petrology, November 1, 2003; 44(11): 2003 - 2032. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Halama, R. Halama, T. Wenzel, B. G. J. Upton, W. Siebel, and G. Markl A geochemical and Sr-Nd-O isotopic study of the Proterozoic Eriksfjord Basalts, Gardar Province, South Greenland: Reconstruction of an OIB signature in crustally contaminated rift-related basalts Mineralogical Magazine, October 1, 2003; 67(5): 831 - 853. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. M. Coulson, I. M. Coulson, K. M. Goodenough, N. J. G. Pearce, and M. J. Leng Carbonatites and lamprophyres of the Gardar Province - a 'window' to the sub-Gardar mantle? Mineralogical Magazine, October 1, 2003; 67(5): 855 - 872. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Antonini, P. Antonini, P. Comin-Chiaramonti, C. B. Gomes, P. Censi, B. F. Riffel, and E. Yamamoto The Early Proterozoic carbonatite complex of Angico dos Dias, Bahia State, Brazil: geochemical and Sr-Nd isotopic evidence for an enriched mantle origin Mineralogical Magazine, October 1, 2003; 67(5): 1039 - 1057. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. WITT-EICKSCHEN, H. A. SECK, K. MEZGER, S. M. EGGINS, and R. ALTHERR Lithospheric Mantle Evolution beneath the Eifel (Germany): Constraints from Sr-Nd-Pb Isotopes and Trace Element Abundances in Spinel Peridotite and Pyroxenite Xenoliths J. Petrology, June 1, 2003; 44(6): 1077 - 1095. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. WAGNER, A. MOKHTARI, E. DELOULE, and F. CHABAUX Carbonatite and Alkaline Magmatism in Taourirt (Morocco): Petrological, Geochemical and Sr-Nd Isotope Characteristics J. Petrology, May 1, 2003; 44(5): 937 - 965. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. M. LARSEN, A. K. PEDERSEN, B. SUNDVOLL, and R. FREI Alkali Picrites Formed by Melting of Old Metasomatized Lithospheric Mantle: Manitdlat Member, Vaigat Formation, Palaeocene of West Greenland J. Petrology, January 1, 2003; 44(1): 3 - 38. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. B. TRUMBULL, B. BUHN, R. L. ROMER, and F. VOLKER The Petrology of Basanite-Tephrite Intrusions in the Erongo Complex and Implications for a Plume Origin of Cretaceous Alkaline Complexes in Namibia J. Petrology, January 1, 2003; 44(1): 93 - 112. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. E. JANNEY, A. P. LE ROEX, R. W. CARLSON, and K. S. VILJOEN A Chemical and Multi-Isotope Study of the Western Cape Olivine Melilitite Province, South Africa: Implications for the Sources of Kimberlites and the Origin of the HIMU Signature in Africa J. Petrology, December 1, 2002; 43(12): 2339 - 2370. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Bizzarro, A. Simonetti, R. K. Stevenson, and J. David Hf isotope evidence for a hidden mantle reservoir Geology, September 1, 2002; 30(9): 771 - 774. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||