Journal of Petrology Advance Access originally published online on November 24, 2004
Journal of Petrology 2005 46(3):505-522; doi:10.1093/petrology/egh085
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Systematic Variation of Sr-, Nd- and Pb-Isotopes with Time in Lavas of Mauritius, Réunion Hotspot
1 DEPARTMENT OF ENVIRONMENTAL SCIENCES, KUMAMOTO UNIVERSITY, KUMAMOTO 860-8555, JAPAN
2 EARTHQUAKE RESEARCH INSTITUTE, UNIVERSITY OF TOKYO, BUNKYO-KU, TOKYO 113-0032, JAPAN
3 INSTITUTE OF GEOSCIENCE, AIST, TSUKUBA 305-8567, JAPAN
RECEIVED FEBRUARY 28, 2003; ACCEPTED SEPTEMBER 28, 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONVOLCANIC GEOLOGYSAMPLES AND ANALYTICAL METHODSRESULTSDISCUSSIONAPPENDIX: SAMPLE LOCATIONSREFERENCES |
|---|
We report Sr-, Nd- and Pb-isotopic compositions for the lavas of Mauritius, the second youngest volcanic island in the Réunion hotspot. The lavas of the Older Series (7·8–5·5 Ma) have identical isotopic compositions (87Sr/86Sr = 0·70411 to 0·70422,143Nd/144Nd = 0·512865 to 0·512854, and 206Pb/204Pb = 19·016 to 19·041) to those of Réunion, where the center of volcanic activity is currently located. The lavas of the Intermediate Series (3·5–1·9 Ma) and Younger Series (0·70–0·17 Ma) are shifted to lower Sr-isotopic compositions (0·70364–0·70394, with 143Nd/144Nd = 0·512813 to 0·512948 and 206Pb/204Pb = 18·794 to 18·984). The Intermediate Series lavas have similar trace-element characteristics (e.g. Zr–Nb, Ba–Y) to those of Rodrigues, in both cases requiring the involvement of an enriched mantle-like component in the mantle source. During the volcanic history of Mauritius, the magmas lost the principal isotopic characteristics of the Réunion hotspot with time, and became gradually imprinted with the isotopic signature of a shallower mantle source that produced the Central Indian Ridge basalts.
KEY WORDS: hotspot; isotopes; Mauritius; Réunion; trace element
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONVOLCANIC GEOLOGYSAMPLES AND ANALYTICAL METHODSRESULTSDISCUSSIONAPPENDIX: SAMPLE LOCATIONSREFERENCES |
|---|
Isotopic analyses of ocean island basalts have led to the identification of geochemical domains on various scales (Zindler & Hart, 1986
; Hofmann, 1997). Many oceanic island basalt suites are isotopically heterogeneous, and their generation requires the involvement of several mantle components. In contrast, the lavas of Réunion have limited ranges of Sr-, Nd- and Pb-isotopes (Oversby, 1972; Albarède & Tamagnan, 1988; Fisk et al., 1988; Albarède et al., 1997). The active phase of the Réunion mantle plume began with the eruption of huge volumes of flood basalts at the end of the Cretaceous (Peng & Mahoney, 1995). Subsequent products from this hotspot can be traced through the Laccadives and Maldives Ridge, the Chagos Bank, and the Mascarene Plateau, and the mantle plume is at present active beneath Réunion (Duncan, 1990). The lavas erupted along the hotspot track are characterized by systematic variations of Sr-, Nd- and Pb-isotopic compositions with increasing age towards less mid-ocean ridge basalt (MORB)-like isotopic signatures (White et al., 1990) and to the unique values of the Réunion mantle plume.
In this study, we have focused on the basaltic rocks of Mauritius, which lie in the western Indian Ocean at 57°30'E, 20°20'S, about 1000 km east of Madagascar (Fig. 1a). The islands of Mauritius, Réunion (170 km WSW of Mauritius), and Rodrigues (650 km ESE) comprise the Mascarene Islands. Among them, Mauritius is the second-youngest island associated with the Réunion hotspot. The mapping work by Simpson (1950), which classified the lavas into an Older Series, Intermediate Series and Younger Series, has provided the basis for all subsequent work on Mauritius. McDougall & Chamalaun (1969) performed a systematic K–Ar geochronology and paleomagnetic study of the island, and confirmed the geological divisions of Simpson (1950). They also established the time span of the volcanic activity for each series as 7·8–5·5 Ma for the Older Series, 3·5–1·9 Ma for the Intermediate Series, and 0·70–0·17 Ma for the Younger Series (Fig. 1b). Mauritius lies on the Mascarene Plateau, which is considered to be the trace of the Réunion hotspot. According to a hotspot track calculation by Duncan & Pyle (1988), Mauritius passed through the hotspot at 7 Ma.
|
Sr- and Nd-isotopic data for each series were reported by Mahoney et al. (1989)
. Sheth et al. (2003) presented a geochemical model for the genesis of Mauritius magmas based on Sr–Nd–Pb isotopic and trace-element analyses. The Older Series lavas are isotopically identical to those of Réunion, whereas the Intermediate and Younger Series have slightly lower 87Sr/86Sr and higher 143Nd/144Nd isotopic signatures. There have been few Pb-isotopic data reported for the Intermediate and Younger Series lavas before the work of Sheth et al. (2003). In order to examine the temporal variation of isotopes within a single island of the Réunion hotspot, we have analysed Sr-, Nd- and Pb-isotopic compositions for the lavas of the three series on Mauritius. In particular, the newly reported Pb-isotope data for the Intermediate and Younger Series lavas should provide an important clue to understanding not only the genesis of Mauritius volcanism but also the isotopic characteristics of the Reunion hotspot in combination with Sr- and Nd-isotope data and trace-element concentrations. |
VOLCANIC GEOLOGY |
|---|
|
TOPABSTRACTINTRODUCTIONVOLCANIC GEOLOGYSAMPLES AND ANALYTICAL METHODSRESULTSDISCUSSIONAPPENDIX: SAMPLE LOCATIONSREFERENCES |
|---|
The three volcanic series on Mauritius were identified by mapping (Simpson, 1950
), and their time span has been established by K–Ar dating (McDougall & Chamalaun, 1969). Baxter (1975, 1976) reported geochemical data and petrogenetic interpretations. According to Baxter (1975), the Older Series represents the erosional remnants of a large shield volcano and is an extensively differentiated transitional basalt suite. Two distinct stages of activity have been recognized. The first, or shield-building, stage is composed principally of alternating picrite–basalt flows and agglomerate, whereas the second stage is composed of feldspar-phyric basalt, hawaiite, mugearite and high-level trachytic intrusive rocks. The Intermediate Series lavas, which are the products of reactivated volcanic activity following a hiatus of 2 myr after the eruption of the Older Series, occur in the southwestern part of the island. They fill valleys within Older Series lavas at relatively high altitudes, and consist of MgO-rich alkali basalts displaying a range of incompatible element concentrations. The Younger Series lavas, primarily alkali basalts with subordinate basanites, have relatively low incompatible-element abundances, reflecting higher degrees of melting (Baxter, 1976).
Based on the major- and trace-element characteristics, Baxter (1975) demonstrated the strong control of phenocryst mineralogy on the bulk-lava chemistry of the Older Series. Lava compositions with more than 5 or 6% MgO are distributed along a pronounced olivine + clinopyroxene control line, and more evolved lavas along a trend produced by the fractionation of olivine + clinopyroxene + plagioclase + titanomagnetite. As for the Intermediate Series lavas, the compositions of the incompatible elements K, Ti, P, Sr, Rb, Ba and Zr show well-defined enrichment trends that cannot be explained by olivine fractionation, or by any other likely low-pressure fractionation schemes (Baxter, 1976). The Younger Series lavas, on the other hand, show clear evidence of olivine control in most major-oxide variation diagrams, together with a correlation between increasing amounts of phenocryst olivine and bulk-rock MgO (Baxter, 1976).
|
SAMPLES AND ANALYTICAL METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONVOLCANIC GEOLOGYSAMPLES AND ANALYTICAL METHODSRESULTSDISCUSSIONAPPENDIX: SAMPLE LOCATIONSREFERENCES |
|---|
The samples used in the present study were collected during 1994 and 1995 for the purpose of Ar–Ar dating, noble-gas isotope geochemistry, and Sr-, Nd- and Pb-isotopic studies (Fig. 1b). Sample locations are listed in the Appendix and indicated in Fig. 1b. Results of a noble-gas study of the Réunion hotspot have been reported by Hanyu et al. (2001)
. The results of K–Ar age determinations for the lavas of Mauritius will be presented elsewhere, along with general descriptions of the samples, by Uto et al. (in preparation).
Major-element oxides and Rb, Sr, Ba, Nb, Zr, Y, Ni and Cr were analysed using a wavelength-dispersive X-ray fluorescence (XRF) spectrometer (Phillips PW 1404) at the Geological Survey of Japan (GSJ). We also analysed selected samples for rare earth elements (REE), Cs, Th, Ta and Hf at the GSJ by instrumental neutron activation analysis (INAA). Detailed analytical procedures for XRF and INAA were presented by Hoang & Uto (2003).
The Sr-, Nd- and Pb-isotopic compositions were analysed using a Finnigan MAT 261 mass spectrometer at Kumamoto University. The analytical procedures for Sr and Nd have been described by Nohda et al. (1988, 1992) and for Pb, by Nohda (1999) and Kani & Nohda (2002). Sr-isotopic ratios were normalized to 87Sr/86Sr = 0·1194, and Nd-isotopic ratios to 143Nd/144Nd = 0·7219. Recent average values of standard samples are 87Sr/86Sr = 0·710214 ± 0·000012 (n = 18, 2
means) for NBS-987, and 143Nd/144Nd = 0·511862 ± 0·000010 (n = 15) for the La Jolla Nd standard. Pb-isotopic data were corrected for fractionation of 0·133% per atomic mass unit (a.m.u.) for 206Pb/204Pb, 0·131% for 207Pb/204Pb, and 0·139% for 208Pb/204Pb, based on repeated analyses of NBS-981. Measured values for the NBS-981 Pb standard were 206Pb/204Pb = 16·8908 ± 0·0013, 207Pb/204Pb = 15·4285 ± 0·0013, and 208Pb/204Pb = 36·4982 ± 0·0034 (n = 15, 2 means).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONVOLCANIC GEOLOGYSAMPLES AND ANALYTICAL METHODSRESULTSDISCUSSIONAPPENDIX: SAMPLE LOCATIONSREFERENCES |
|---|
Major elements
Major-element oxide compositions, listed in Table 1, have been recalculated to 100 wt % volatile-free with total Fe as FeO. Total contents of raw XRF data for major oxides range from 96·30 to 100·11 wt %, with an average of 98·93 ± 0·86 wt %. We have not measured the loss in weight of samples on ignition (LOI), but minor weight loss is attributed to water evaporation during preparation of fused glass disks. No signs of any significant subaerial alteration were observed in the samples. MgO-contents and Mg-numbers are high for most of the samples that we analysed.
|
All samples of the Older Series analysed in the present study are alkali basalts; MgO-contents range from 6·77 to 18·63 wt %, and Mg-numbers from 54·62 to 74·77. Sample MR94-07 has a SiO2 content of 44·43 wt %, with a total alkali-content of 3·07 wt %, and plots at the boundary between basanite–tephrite and picrobasalt in the total alkalis vs SiO2 (TAS) diagram (Fig. 2). Sample MR95-04-02 plots on the alkalic–subalkalic boundary. Baxter (1975)
reported a marked change in gradient at about 5–6 wt % MgO in major-element variation diagrams; this coincides with the appearance of titanomagnetite and plagioclase as major phenocryst phases. The TiO2- and P2O5-contents of the Older Series lavas are similar to those of the Intermediate lavas, and those from Rodrigues and Réunion. K2O-contents are equivalent to those of the lavas of the Intermediate Series and the Oceanite Series of Réunion, but are distinctly lower than those of Rodrigues.
|
The MgO-content of sample MR95-21-01 (4·03 wt %) is the lowest among the Intermediate Series and this sample plots in the trachybasalt field of Fig. 2. All other Intermediate Series lavas analysed in the present study have high MgO-contents, ranging from 8·03 to 12·07 wt %, and are classified as alkalic basalts. As noted by Baxter (1976)
, the Intermediate Series has a relatively restricted compositional range, with the exception of K2O and TiO2, and lacks any consistent trends in major-element oxides.
MgO-contents of the Younger Series range from 6·32 to 12·35 wt %, with an average of 10·13 wt %, and Mg-numbers range from 55·85 to 66·92. In the TAS diagram (Fig. 2), they plot close to the boundary between subalkalic and alkalic fields, and exhibit no significant variations in major-element oxides, consistent with the data of Baxter (1976). The Younger Series is characterized by distinctly low contents of K2O, TiO2 and P2O5 compared with the Older and Intermediate Series and basalts from the adjacent islands of Rodrigues and Réunion.
Trace elements
Figure 3 shows primitive-mantle-normalized trace-element patterns for the lavas of Mauritius compared with those of the Réunion Oceanite Series, Rodrigues and representative mid-ocean ridge and oceanic-island basalts. The patterns of the Mauritius lavas (Fig. 3a–c) are characterized by small Nd peaks which are not obvious in those of the lavas of Réunion and Rodrigues (Fig. 3d and e). Similar peaks have been reported for other lavas associated with the Réunion hotspot, such as the Deccan Traps and Réunion lavas (Peng & Mahoney, 1995). Slight Nb-depletion relative to Ta may be a common characteristic of the Mauritius lavas. Lavas of the Older Series and the Intermediate Series show similar patterns to those of Réunion. The Rodrigues lavas show a wide range of variation, with the most enriched sample showing a similar pattern to that of EM I ocean-island basalt (OIB). The Younger Series is evidently the least enriched among the studied lavas, with an intermediate pattern between E-MORB and average OIB, distinct from those of the other series of Mauritius, Réunion and Rodrigues. Sample MR95-10 has the highest Nd-concentration among the Younger Series, but the concentrations of other elements are within the ranges of this series.
|
Sr-, Nd- and Pb-isotopic compositions
Strontium isotopes
87Sr/86Sr ranges from 0·70364 to 0·70422 (Table 2). The Older Series have values >0·7041, with an average of 0·70419, consistent with previously reported values of 0·70411 and 0·70433 (Mahoney et al., 1989
; Sheth et al., 2003). These values also fall within the reported range of Réunion in which 87Sr/86Sr varies from 0·70397 to 0·70436 throughout the entire 2 Myr volcanic history (Fisk et al., 1988). The lavas of Piton de la Fournaise (Réunion) have identical Sr-isotopic compositions that range from 0·70410 to 0·70426 (Albarède & Tamagnan, 1988). The Intermediate and Younger Series lavas clearly have lower 87Sr/86Sr (average value of 0·70373) than the Older Series, as shown in Fig. 4a.
|
|
Neodymium isotopes
143Nd/144Nd of Mauritius lavas ranges from 0·512813 to 0·512905 (Fig. 4b). The lavas of the Older Series show a limited range of Nd-isotopic compositions (0·51285 to 0·51287), which are identical to those of Réunion (Fisk et al., 1988
). In contrast, 143Nd/144Nd values of the Intermediate and the Younger Series have a slightly wider range of values, from 0·51281 to 0·51295. In particular, sample MR95-23 of the Intermediate Series has the lowest 143Nd/144Nd of all samples analysed in the present study, plotting in a unique position in the 87Sr/86Sr vs 143Nd/144Nd diagram (Fig. 5).
|
Lead isotopes
The measured Pb-isotope compositions also show significant temporal variations, similar to 87Sr/86Sr (Fig. 4a and c–e). However, we need to consider potential age effects on Pb-isotopic compositions as we are comparing data for different lava series. In order to evaluate the age effect on Pb-isotopic compositions, we made age corrections to the data of Peng & Mahoney (1995)
and Sheth et al. (2003) using the U-, Th- and Pb-concentrations reported by those workers. The maximum correction for the Pb-isotope ratios of the Older Series lavas is –0·028 for 206Pb/204Pb, –0·001 for 207Pb/204Pb and –0·035 for 208Pb/204Pb, assuming an age of 7 Ma (Peng & Mahoney, 1995), whereas for the Intermediate Series lavas it is –0·016, –0·002 and –0·018, respectively, assuming a possible maximum age of 3 Ma. These correction factors are clearly larger than the analytical errors, except for 207Pb/204Pb. However, the age-corrected average 206Pb/204Pb ratios still show distinctions between lavas from the three series: the Older Series (18·899), the Intermediate Series (18·861) and the Younger Series (18·837). Thus, Pb-isotope differences between the Older, Intermediate and Younger Series seem to reflect the geochemical characteristics of the magma sources, and suggest that the measured Pb-isotopic data are useful for genetic interpretations (Fig. 4). In addition, the Pb-isotopic data for Mauritius are subsequently discussed in comparison with those for ODP Leg 115 basalts (White et al., 1990) (Fig. 6). To be consistent between all the datasets used for our discussion, we have used the measured Pb-isotope data throughout.
|
The Pb-isotopic compositions of the Older Series are similar to the Oceanite Series of Réunion (K–Ar age 1–2 Ma), whereas the Younger Series, with slightly less radiogenic Pb-isotope compositions, are similar to the Differentiated Series of Réunion (0·1–0·3 Ma) (Oversby, 1972
). However, the Pb-isotope data for Réunion were obtained by a different method (using PbI2) from this study. Thus, for a more detailed comparison of Pb-isotopes between Réunion and the other related lavas, further Pb-isotopic determinations for Réunion lavas are required. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONVOLCANIC GEOLOGYSAMPLES AND ANALYTICAL METHODSRESULTSDISCUSSIONAPPENDIX: SAMPLE LOCATIONSREFERENCES |
|---|
Réunion hotspot
It is clear that the trace-element concentrations and isotopic compositions of Sr, Nd and Pb of the Mauritius lavas have changed substantially throughout the island's 8 Myr volcanic history (Table 2, Figs 3–8). In contrast, lavas from the adjacent volcanic islands of Réunion and Rodrigues, with much shorter periods of volcanic activity, lasting for 2 Myr, show limited ranges of Sr-, Nd- and Pb-isotopic composition, despite significant variations in their trace-element characteristics. Lavas of the shield-forming Older Series of Mauritius have Sr- and Nd-isotopic compositions that span the entire range of Réunion lavas (Fisk et al., 1988
).
|
|
Zr/Nb–Ba/Y
The strong geochemical similarity between the Older Series lavas and those of Réunion is evident on a Zr/Nb vs Ba/Y diagram (Fig. 7) based on the Réunion data of Fisk et al. (1988)
. Although Zr and Nb have very different bulk distribution coefficients in the upper-mantle melting regime, the correlation between Zr- and Nb-concentrations is very high for every OIB suite (Kamber & Collerson, 2000). It follows that a restricted range of Zr/Nb should be expected in lavas of an oceanic island such as Réunion or Mauritius. On the other hand, variations in Ba/Y might be the result of magmatic differentiation processes, or low-temperature alteration, which causes mobility of Ba. The Younger Series lavas show significant variation of Zr/Nb but only a narrow range of Ba/Y in comparison with the Older and Intermediate Series lavas and the adjacent islands of Réunion and Rodrigues. The low and limited Ba/Y of the Younger Series lavas may be attributed to a higher degree of partial melting producing the homogeneity (Baxter, 1976). However, considerable variations in Zr/Nb occur within the Younger Series lavas, which seems to require a process of binary mixing (Kamber & Collerson, 2000). The trend observed towards the Zr/Nb of MORB probably indicates the involvement of a depleted MORB-source mantle component in the genesis of the Younger Series of Mauritius.
Basalts drilled during ODP Leg 115 also plot in the same region as the lavas of the Younger Series. These samples are distributed along the Deccan–Réunion hotspot lineament, and have been interpreted as the products of the Réunion hotspot (Baxter, 1990). It is suggested that the linear trend defined by the samples of the Younger Series and ODP Leg 115 represents a mixing line between MORB-source mantle and the Réunion mantle plume. The vertical distribution of the lavas of Réunion and Rodrigues in Fig. 7 (variable Ba/Y with limited range of Zr/Nb) may also be attributed to a mixing process because involvement of a small amount of a MORB-source component (high Zr/Nb and low Ba/Y) will effectively lower Ba/Y.
The Older Series lavas overlap with those of Réunion in Fig. 7, suggesting a common petrogenesis. The Rodrigues lavas have the lowest Zr/Nb, along with some Intermediate Series lavas, which seem to require the involvement of another end-member mantle source component compared with the lavas of Réunion and the Older Series. According to the extensive trace-element study of the lavas of Piton de Fournaise, Réunion (Albarède et al., 1997), Zr/Nb converges within the range 6–9, with a limited range of Ba/Y from 3 to 8, similar to the lavas of the Older Series of Mauritius.
Older Series lavas
The Older Series lavas have the highest 87Sr/86Sr and lowest 143Nd/144Nd values of all the oceanic islands and submarine basalts associated with the Réunion hotspot (White et al., 1990), except for the Deccan Trap basalts, which show clear evidence for contamination by the continental crust (Peng & Mahoney, 1995). Systematic temporal Sr–Nd–Pb-isotopic variations are found in the basalts of the Réunion hotspot track during its 68 Myr history, from the modern Central Indian Ridge (CIR) normal-MORB to the Réunion OIBs (Figs 5 and 6). This range of variation has been interpreted as reflecting increasing entrainment of the asthenosphere (depleted MORB-source mantle) by the rising plume with decreasing age (Mahoney et al., 1989; White et al., 1990). In Fig. 5, the EM II mantle component (Hofmann, 1997) plots in a region that could form the extrapolated end-point of the trend line from the Younger to the Older Series. However, there have been no reports of basalts from the Réunion hotspot track having higher Sr- and lower Nd-isotopic compositions than the lavas of Réunion. Even if an EM II component is involved in the Réunion hotspot activity, it must have been completely assimilated into the source material of the plume. The Pb-isotopic data are also consistent with identification of an enriched end-member component in the petrogenesis of the magmas associated with the Réunion hotspot (Fig. 6). It can be inferred from Fig. 6 that the lavas of the Older Series were derived from a similar source component to that tapped presently by the Réunion activity; this is enriched in radiogenic Sr and Pb and has lower 143Nd/144Nd relative to the Younger Series lavas (Figs 5 and 6) and restricted values of Zr/Nb, around 6–8 (Fig. 7). This implies that the parental magmas of the Older Series lavas were derived from an upwelling mantle plume characterized by the unique isotopic compositions of the Réunion hotspot.
Other mantle components
As shown in the Sr–Nd-isotope diagram (Fig. 5), the Mauritious Intermediate and Younger Series basalts must have been sourced from a component with a long time-integrated history of lower Rb/Sr and higher Sm/Nd compared with Bulk Earth. The depleted mantle component currently tapped by the CIR MORB is one candidate for this component. Lavas of the Intermediate and Younger Series and those of Rodrigues have low 3He/4He values (6·49–8·39), equivalent to those of MORB (Hanyu et al., 2001). In particular, sample MR95-23 (Intermediate Series, K–Ar age 3·0 Ma) has a lower 3He/4He (6·49; normalized to atmospheric) than MORB (8–9). The lavas of the Intermediate Series plot close to those of Rodrigues and the Younger Series in the 87Sr/86Sr–143Nd/144Nd diagram (Fig. 5). However, some lavas from Rodrigues and the Intermediate Series (e.g. MR95-23) are displaced to low 143Nd/144Nd in this diagram, suggesting the involvement of an unknown mantle component with this character. The Zr/Nb vs Ba/Y diagram (Fig. 7) may provide an important clue in this respect. The lavas of Rodrigues, together with the majority of the Intermediate Series basalts, have much lower Zr/Nb ratios compared with the other series of Mauritius and Réunion (Fig. 7), and appear to require another end-member component characterized by lower Zr/Nb than that of Réunion. Such a low and limited Zr/Nb is characteristic of the mantle end-components HIMU and EM, derived from ancient recycled basaltic oceanic crust (Weaver, 1991). From the Zr/Nb vs 87Sr/86Sr diagram (Fig. 8), this component may be the result of a mixing process between HIMU and EM, or alternatively a small amount of a MORB–source component and EM. Thus, the lavas of Rodrigues and the Intermediate Series appear to be derived from a mantle source contaminated by an EM component. Sheth et al. (2003) concluded that all Mauritius and Réunion lavas can be explained as 2–8% partial melts of a Bulk-Earth-like mantle source that is 1·5–0·25% metazomatized by a 0·1% MORB-source melt, based on model calculations. However, it is difficult to envisage the involvement of a Bulk-Earth-like source component with high 3He/4He (Zindler & Hart, 1986) in the genesis of the Mauritius lavas. The Younger Series lavas are characterized by isotopic compositions closed to CIR MORB (Fig. 5) and plot in the same region as ODP Leg 115 basalts in the Zr/Nb–Ba/Y diagram (Fig. 7). In particular, the 3He/4He values of the Younger Series lavas are similar to those of MORB (Hanyu et al., 2001). In spite of the low MORB-like 3He/4He, the Intermediate Series and Rodrigues lavas have enriched geochemical characteristics (Fig. 3) and low Zr/Nb and high Ba/Y ratios (Fig. 7). These geochemical constraints require the involvement of an EM component with low 3He/4He in the petrogenesis of the magmas of the Intermediate Series and Rodrigues lavas.
Thus, it is suggested that the lavas of the Intermediate Series and Rodrigues may retain the fingerprint of an ancient, recycled, basaltic oceanic crust component which is the commonly inferred source for HIMU and EM mantle components. It is proposed that the EM I end-member is derived from this source, contaminated by ancient pelagic sediment, and, for EM II, by ancient terrigenous sediment (Weaver, 1991). If so, then the EM component for the Intermediate Series and Rodrigues lavas could possibly be provided by the source containing both ancient pelagic and terrigenous sediments, creating the fanlike pattern radiating from HIMU to EM I and EM II in the diagram of 87Sr/86Sr vs 143Nd/144Nd (Fig. 5). The HIMU-like component is similar to the FOZO (Hart et al., 1992) or C-component (Hanan & Graham, 1996). In considering the dominant source materials for the magmas of Rodrigues and the Intermediate Series of Mauritius, the upwelling Réunion hotspot must have entrained a CIR-component and EM-components to have such specific isotopic and chemical characteristics as described above.
Secondary µ values
Unlike the wide range of 3He/4He observed in Mauritius lavas (6·49–11·81), the Sr-, Nd- and Pb-isotope compositions are not strongly shifted from the Réunion field towards that of the CIR, and form relatively narrow ranges. The slight variation in Pb-isotopes (Fig. 6) may be associated with the limited range of the secondary µ values observed for Réunion, as pointed out by Chase (1981). Following the calculation method of Chase (1981), we obtained primary and secondary µ values for Mauritius lavas along with those for OIB with mantle end-member characteristics as shown in Table 3. At present, it is difficult to make any obvious geochemical interpretation of the results obtained for Tubuai (HIMU) (Chauvel et al., 1992), Tristan da Cunha (EM I) (Le Roex et al., 1990) and Tahaa (EM II) (White & Duncan, 1996). The primary µ value of the Mauritius samples is consistent with the average value (= 7·91) determined for selected oceanic islands (Chase, 1981). The calculated isochron age of 2·15 Ga is slightly younger than the age for Réunion (2·47 Ga) according to Chase (1981). In the case of Réunion, it has been pointed out that mass fractionation in the Pb-isotopic measurements by the PbI2 method may have contributed to the steep 207Pb/206Pb slope observed (Chase, 1981). Thus, we may be able to ignore age difference between Réunion and Mauritius. Both Réunion and Mauritius exhibit an extremely limited range of the estimated secondary µ value, from 11·2 to 11·9. This may be one of the factors contributing to the restricted Pb-isotopic range of the Mauritius lavas. In addition, the Th/Pb ratio may have been homogenized some considerable time before magma generation because the 208Pb/204Pb of the Intermediate (sample MR94-04, age 1·9 Ma) and Younger Series lavas also has a very narrow range. Moreover, as shown in Table 3, the isochron age and secondary µ values for the Réunion hotspot track (ODP Leg 115) are also consistent with such a scheme (White et al., 1990). In contrast to other OIBs, which generally display a wide variation in secondary µ values, the Réunion hotspot has produced lavas with extremely homogeneous Pb-isotopic compositions throughout its 65 Myr volcanic activity.
|
If mixing of the Réunion hotspot or Rodrigues component and the CIR-component is a recent process, then the uniform secondary µ and Th/Pb values do not provide any constraint for the limited range of the Pb-isotopes observed. As mentioned earlier, the mixing process is strongly supported by the compositions of trace elements and Sr- and Nd-isotopes presented in this study. The problem remains to account for the generally similar primary µ values reported for basaltic suites of oceanic islands.
Structure of the Réunion mantle plume
We have identified the end-member components for the volcanism of Mauritius and adjacent islands. We consider that the Older Series of Mauritius and the Réunion lavas are predominantly derived from the mantle plume of the Réunion hotspot. The source material for the Intermediate Series and Rodrigues lavas may be the product of mixing between a CIR-component and the Réunion mantle plume component, with a minor contribution from an EM-like component. The Younger Series lavas seem to have a larger contribution from the CIR-like component as the result of increased entrainment of depleted MORB-source asthenosphere with distance from the center of the Réunion hotspot. This leads us to a simpler scheme for the Réunion plume than, for example, for Hawaii, as shown in the cross-section in Fig. 9a, which presents a possible scheme for the shield-forming Older Series volcanism of Mauritius at approximately 7 Ma. A similar condition is applicable to the present (or 500 kyr ago) Réunion activity; the plume is located directly beneath Réunion Island, and the isotopic compositions of the Réunion lavas should reflect the plume's signature (Fig. 9c).
|
The lavas of the Intermediate Series of Mauritius and Rodrigues require involvement of a distinctive, relatively enriched component with low Zr/Nb and 143Nd/144Nd. This enriched component has not ubiquitously been involved with the Réunion hotspot. As suggested by Morgan (1978)
, the Rodrigues volcanism may have involved channelled asthenospheric flow from the Réunion hotspot towards the Mid-Indian Oceanic Ridge. It is plausible that such asthenospheric flow entrained the Réunion hotspot component in part, and also encountered an EM-like domain drifting within the asthenosphere during upwelling of the flow. The channelled asthenospheric flow might have extended to Mauritius and contaminated the magma source of the Intermediate Series of Mauritius (Fig. 9b). When the channelled asthenospheric flow left the Mauritius region and propagated to Rodrigues, the Intermediate Series volcanism terminated at Mauritius. The volcanic activity observed at Rodrigues is from 1·6 to 1·3 Ma.
During the volcanic hiatus, lasting for 1·5 Myr, between the Intermediate and Younger Series, the EM-like domain may have become removed from the asthenospheric flow region, thus producing Younger Series lavas that are characterized by Sr-, Nd- and Pb-isotopic signatures solely because of a mixing process between a CIR-component and a Réunion hotspot-type source (Fig. 9c). About 500 kyr ago, when the hotspot was located approximately 200 km to the southwest of Mauritius, the plume was still capable of supplying thermal energy to the surrounding CIR-source mantle by heat propagation because of the slow motion (3·7 cm/yr) of the overlying African Plate (Duncan & Hargraves, 1990). Magmas of the Younger Series were derived from the outer zone of the plume and were principally imprinted with CIR-isotopic signatures. Accordingly, the outer zone of the plume, representing a mixture of plume and MORB-source mantle, might have been sufficiently degassed so that the 3He/4He became equivalent to MORB. However, they also exhibit, to a considerable extent, the characteristics of the Réunion hotspot component, which was possibly dragged by the overlying lithosphere (Tatsumi & Nohda, 1990). The present results suggest that the Réunion hotspot may have a simpler composition and structure compared with other long-lived hotspots, such as Hawaii and Polynesia (Chen et al., 1996; White & Duncan, 1996; Dixon & Clague, 2001).
|
APPENDIX: SAMPLE LOCATIONS |
|---|
|
TOPABSTRACTINTRODUCTIONVOLCANIC GEOLOGYSAMPLES AND ANALYTICAL METHODSRESULTSDISCUSSIONAPPENDIX: SAMPLE LOCATIONSREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
ACKNOWLEDGEMENTS |
|---|
Sampling was carried out as an International Scientific Research Program in 1994 and 1995, sponsored by the Ministry of Education, Science, Culture and Sport, Japan (06041026). T. Fujii, S. Nakada, Y. Tatsumi and A. Yasuda assisted the fieldwork. An anonymous reviewer and R. J. Arculus kindly provided numerous constructive and very thoughtful editorial suggestions, which substantially improved the first version of the manuscript.
|
FOOTNOTES |
|---|
Present address: Institute for Frontier Research on Earth Evolution (IFREE), Japan Marine Science and Technology Center (JAMSTEC), 2-15 Natsushima-cho, Yokosuka 237-0061, Japan. 
Present address: Scottish Universities Environmental Research Centre, University of Glasgow, Scottish Enterprise Technology Park, East Kilbride, Glasgow G75 0QF, UK. E-mail: s.xu{at}suerc.gla.ac.uk.
* Corresponding author. Telephone and fax: +81-96-342-3467. E-mail: snohda{at}sci.kumamoto-u.ac.jp.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONVOLCANIC GEOLOGYSAMPLES AND ANALYTICAL METHODSRESULTSDISCUSSIONAPPENDIX: SAMPLE LOCATIONSREFERENCES |
|---|
Albarède, F. & Tamagnan, V. (1988). Modelling the recent geochemical evolution of the Piton de la Fournaise volcano, Réunion Island. Journal of Petrology 29, 997–1030.
Albarède, F., Luais, B., Fitton, J. G., Semet, M., Kaminski, E., Upton, B. G. J., et al. (1997). The geochemical regimes of Piton de la Fournaise volcano (Réunion) during the last 530 000 years. Journal of Petrology 38, 171–201.[CrossRef][Web of Science]
Baxter, A. N. (1975). Petrology of the Older Series lavas from Mauritius, Indian Ocean. Geological Society of America Bulletin 86, 1449–1458.
Baxter, A. N. (1976). Geochemistry and petrogenesis of primitive alkali basalt from Mauritius, Indian Ocean. Geological Society of America Bulletin 87, 1028–1034.
Baxter, A. N. (1990). Major and trace element variations in basalts from Leg 115. In: Duncan, R. A., Backman, J., Peterson, L. C., et al. (eds) Proceedings of the Ocean Drilling Program, Scientific Results, 115. College Station, TX: Ocean Drilling Program, pp. 11–21.
Baxter, A. N., Upton, B. G. J. & White, W. M. (1985). Petrology and geochemistry of Rodrigues Island, Indian Ocean. Contributions to Mineralogy and Petrology 89, 90–101.[CrossRef][Web of Science]
Chase, C. G. (1981). Oceanic island Pb: two-stage histories and mantle evolution. Earth and Planetary Science Letters 52, 277–284.[CrossRef][Web of Science]
Chauvel, C., Hofmann, A. W. & Vidal, P. (1992). HIMU-EM: the French Polynesian connection. Earth and Planetary Science Letters 110, 99–119.[CrossRef][Web of Science]
Chen, C.-Y., Frey, F. A., Rhodes, J. M. & Easton, M. (1996). Temporal geochemical evolution of Kilauea volcano: comparison of Hilina and Puna Basalt. In: Basu, A. & Hart, S. (eds) Earth Processes-Reading the Isotope Code. Geophysical Monograph, American Geophysical Union 95, 161–181.
Devey, C. W., Albarède, F., Cheminée, J.-L., Michard, A., Mühe, R. & Stoffers, P. (1990). Active submarine volcanism on the Society hotspot swell (west Pacific): a geochemical study. Journal of Geophysical Research 95 (B4), 5049–5066.
Dixon, J. E. & Clague, D. A. (2001). Volatiles in basaltic glasses from Loihi Seamount, Hawaii: evidence for a relatively dry plume component. Journal of Petrology 42, 627–654.
Duncan, R. A. (1990). The volcanic record of the Reunion hotspot. In: Duncan, R. A., Backman, J., Peterson, L. C., (eds) et al. Proceedings of the Ocean Drilling Program, Scientific Results, 115. College Station, TX: Ocean Drilling Program, pp. 3–10.
Duncan, R. A. & Hargraves, R. B. (1990). 40Ar/39Ar geochronology of basement rocks from the Mascarene Plateau, the Chagos Bank, and the Maldives Ridge. In: Duncan, R. A., Backman, J., Peterson, L. C., et al. (eds) Proceedings of the Ocean Drilling Program, Scientific Results, 115. College Station, TX: Ocean Drilling Program, pp. 43–51.
Duncan, R. A. & Pyle, D. G. (1988). Rapid eruption of the Deccan flood basalts at the Cretaceous/Tertiary boundary. Nature 333, 841–843.[CrossRef]
Fisk, M. R., Upton, B. G. J., Ford, C. E. & White, W. M. (1988). Geochemical and experimental study of the genesis of magmas of Reunion Island, Indian Ocean. Journal of Geophysical Research 93, 4933–4950.
Hanan, B. B. & Graham, D. W. (1996). Lead and helium isotope evidence from oceanic basalts for a common deep source of mantle plumes. Science 272, 991–995.[Web of Science][Medline]
Hanyu, T., Dunai, T. J., Davis, G. R., Kaneoka, I., Nohda, S. & Uto, K. (2001). Noble gas study of the Reunion hotspot: evidence for distinct less-degassed mantle sources. Earth and Planetary Science Letters 83, 83–98.
Hart, S. R., Hauri, E. H., Oschmann, L. A. & Whitehead, J. A. (1992). Mantle plumes and entrainment: isotopic evidence. Science 256, 517–520.
Hoang, N. & Uto, K. (2003). Geochemistry of Cenozoic basalts in the Fukuoka district (northern Kyushu, Japan): implications for asthenospheric and lithospheric mantle interaction. Chemical Geology 198, 249–268.[CrossRef][Web of Science]
Hofmann, A. W. (1997). Mantle geochemistry: the message from oceanic volcanism. Nature 385, 219–229.[CrossRef]
Kamber, B. S. & Collerson, K. D. (2000). Zr/Nb systematics of ocean island basalts reassessed: the case for binary mixing. Journal of Petrology 41, 1007–1021.
Kani, T. & Nohda, S. (2002). Pb isotope analyses using bead method: an application to BCR-1 and GSJ standard rock samples. Journal of the Mass Spectrometry Society of Japan 50, 199–203.
Le Bas, M. J., Le Maitre, R. W., Streckeisen, A. & Zanettin, B. (1986). A chemical classification of volcanic rocks based on the total alkali–Silica diagram. Journal of Petrology 27, 745–750.
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.
MacDonald, G. A. & Katsura, T. (1964). Chemical composition of Hawaiian lavas. Journal of Petrology 5, 83–133.
Mahoney, J. J., Natland, J. H., White, R. P., Poreda, P., Bloomer, S. H., Fisher, R. L., et al. (1989). Isotopic and geochemical provinces of the Western Indian Ocean spreading centers. Journal of Geophysical Research 94 (B4), 4033–4052.
McDougall, I. & Chamalaun, F. H. (1969). Isotopic dating and geomagnetic polarity studies on volcanic rocks from Mauritius, Indian Ocean. Geological Society of America Bulletin 80, 1419–1442.
Morgan, W. J. (1978). Rodriguez, Darwin, Amsterdam: a second type of hotspot island. Journal of Geophysical Research 83, 5355–5360.
Nohda, S. (1999). Precise isotopic measurements of lead by fused silica-gel. Geochemical Journal 33, 133–139.[Web of Science]
Nohda, S., Tatsumi, Y., Otofuji, Y., Matsuda, T. & Ishizaka, K. (1988). Asthenospheric injection and back-arc opening: isotopic evidence from northeast Japan. Chemical Geology 68, 317–327.[CrossRef][Web of Science]
Nohda, S., Tatsumi, Y., Yamashita, S. & Fujii, T. (1992). Nd and Sr isotopic study of Leg 127 basalts: implications for the evolution of the Japan Sea backarc basin. In: Tamaki, K., Suehiro, K., Allan,. J., McWilliams, M., et al. (eds) Proceedings of the Ocean Drilling Program Scientific Results, 127/128, Part 2. College Station, TX: Ocean Drilling Program, pp. 899–904.
Oversby, V. M. (1972). Genetic relations among the volcanic rocks of Reunion: chemical and lead isotopic evidence. Geochimica et Cosmochimica Acta 36, 1167–1179.[CrossRef][Web of Science]
Peng, Z. X. & Mahoney, J. J. (1995). Drillhole lavas from the northwestern Deccan Traps, and the evolution of Reunion hotspot mantle. Earth and Planetary Science Letters 134, 169–185.[CrossRef][Web of Science]
Sheth, H. C., Mahoney, J. J. & Baxter, A. N. (2003). Geochemistry of lavas from Mauritius, Indian Ocean: mantle sources and petrogenesis. International Geology Review 45, 780–797.[Web of Science]
Simpson, E. S. W. (1950). The geology and mineral resources of Mauritius. Colonial Geological and Mineral Resources 1, 217–238.
Sun, S.-S. (1980). Lead isotopic study of young volcanic rocks from mid-ocean ridges, ocean islands and island arcs. Philosophical Transactions of the Royal Society of London Part A 297, 409–445.[CrossRef]
Sun, S.-S. & McDonough, W. F. (1989). Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes. In: Saunders, A. D. & Norry, M. J. (eds) Magmatism in Ocean Basins. Geological Society, London, Special Publication 42, 313–345.
Tatsumi, Y. & Nohda, S. (1990). Geochemical stratification in the upper mantle: evidence from Leg 115 basalts in the Indian Ocean. In: Duncan, R. A., Backman, J., Peterson, L. C., et al. (eds) Proceedings of the Ocean Drilling Program, Scientific Results, 115. College Station, TX: Ocean Drilling Program, pp. 63–69.
Weaver, B. L. (1991). The origin of ocean island basalt end-member compositions: trace element and isotopic constraints. Earth and Planetary Science Letters 104, 381–397.[CrossRef][Web of Science]
White, W. M., Cheatham, M. M. & Duncan, R. A. (1990). Isotope geochemistry of Leg 115 basalts and inferences on the history of the Reunion mantle plume. In: Duncan, R. A., Backman, J., Peterson, L. C., et al. (eds) Proceedings of the Ocean Drilling Program, Scientific Results, 115. College Station, TX: Ocean Drilling Program, pp. 53–61.
White, W. M. & Duncan, R. A. (1996). Geochemistry and geochronology of the Society Islands: new evidence for deep mantle recycling. In: Basu, A. & Hart, S. (eds) Earth Processes: Reading the Isotope Code. Geophysical Monograph, American Geophysical Union 95, 183–206.
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.[CrossRef][Web of Science]
Zindler, A. & Hart, S. (1986). Chemical geodynamics. Annual Review of Earth and Planetary Sciences 14, 493–571.[CrossRef][Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. J. Pietruszka, E. H. Hauri, and J. Blichert-Toft Crustal Contamination of Mantle-derived Magmas within Piton de la Fournaise Volcano, Reunion Island J. Petrology, April 20, 2009; (2009) egp016v1. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. ZHANG, J. J. MAHONEY, J. MAO, and F. WANG Geochemistry of Picritic and Associated Basalt Flows of the Western Emeishan Flood Basalt Province, China J. Petrology, October 1, 2006; 47(10): 1997 - 2019. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||