Journal of Petrology Advance Access originally published online on March 4, 2005
Journal of Petrology 2005 46(7):1309-1344; doi:10.1093/petrology/egi016
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Volcanism in the Vitim Volcanic Field, Siberia: Geochemical Evidence for a Mantle Plume Beneath the Baikal Rift Zone
1 DEPARTMENT OF EARTH SCIENCES, UNIVERSITY OF CAMBRIDGE, DOWNING STREET, CAMBRIDGE CB2 3EQ, UK
2 DEPARTMENT OF GEOLOGICAL SCIENCES, UNIVERSITY OF DURHAM, SOUTH ROAD, DURHAM DH1 3LE, UK
RECEIVED JUNE 23, 2003; ACCEPTED JANUARY 18, 2005
 |
ABSTRACT |
|---|
TOP
ABSTRACT
INTRODUCTIONThe Baikal Rift is a zone of active lithospheric extension adjacent to the Siberian Craton. The 6–16 Myr old Vitim Volcanic Field (VVF) lies approximately 200 km east of the rift axis and consists of 5000 km3 of melanephelinites, basanites, alkali and tholeiitic basalts, and minor nephelinites. In the volcanic pile, 142 drill core samples were used to study temporal and spatial variations. Variations in major element abundances (e.g. MgO = 3·3–14·6 wt %) reflect polybaric fractional crystallization of olivine, clinopyroxene and plagioclase. 87Sr/86Sri (0·7039–0·7049), 143Nd/144Ndi (0·5127–0·5129) and 176Hf/177Hfi (0·2829–0·2830) ratios are similar to those for ocean island basalts and suggest that the magmas have not assimilated significant amounts of continental crust. Variable degrees of partial melting appear to be responsible for differences in Na2O, P2O5, K2O and incompatible trace element abundances in the most primitive (high-MgO) magmas. Fractionated heavy rare earth element (HREE) ratios (e.g. [Gd/Lu]n > 2·5) indicate that the parental magmas of the Vitim lavas were predominantly generated within the garnet stability field. Forward major element and REE inversion models suggest that the tholeiitic and alkali basalts were generated by decompression melting of a fertile peridotite source within the convecting mantle beneath Vitim. Ba/Sr ratios and negative K anomalies in normalized multi-element plots suggest that phlogopite was a residual mantle phase during the genesis of the nephelinites and basanites. Relatively high light REE (LREE) abundances in the silica-undersaturated melts require a metasomatically enriched lithospheric mantle source. Results of forward major element modelling suggest that melting of phlogopite-bearing pyroxenite veins could explain the major element composition of these melts. In support of this, pyroxenite xenoliths have been found in the VVF. High Cenozoic mantle potential temperatures (
1450°C) predicted from geochemical modelling suggest the presence of a mantle plume beneath the Baikal Rift Zone. KEY WORDS: Baikal Rift; mafic magmatism; mantle plume; metasomatism; partial melting
|
INTRODUCTION |
|---|
Magmatic activity frequently accompanies continental rifting, and is generally considered to be the consequence of adiabatic decompression melting of the convecting mantle as it upwells beneath the thinned lithosphere (McKenzie & Bickle, 1988
). Melting processes beneath continental rifts are not, however, as well understood as those beneath mid-ocean ridges where upwelling of the mantle is a passive response to lithospheric extension. Beneath continental rifts, it is sometimes unclear as to whether asthenospheric mantle upwelling is an active or passive process. Active rifting is initiated and driven by the impingement of hot asthenospheric mantle (i.e. a mantle plume) on the base of the overlying continental lithosphere. This can cause lithospheric thinning and uplift. Passive rifting is driven by tensional forces that extend the lithosphere and thin it, resulting in upwelling of the asthenospheric mantle. In reality, however, there are difficulties in differentiating between active and passive rifting processes because passive rifting may eventually produce the same geophysical and geological effects (e.g. alkaline magmatism, lithospheric extension and asthenospheric upwelling) as active rifting (Ruppel, 1995). It is probable that processes occurring in the evolution of most continental rifts fall between these two extremes, and there may also be a temporal change from passive to active rifting in some regions (e.g. Delvaux et al., 1997).
The asthenospheric and lithospheric mantles are both known to be important melt source regions during continental rifting (e.g. Leat et al., 1988; Thompson et al., 1990). The volume and composition of the magmas generated during rifting have been shown to be directly related to the degree and duration of extension, the temperature of the underlying asthenospheric mantle (McKenzie & Bickle, 1988) and the presence or absence of volatile/hydrous phases within the mantle source (Gibson et al., 1993). Extension and thinning of the lithosphere cause adiabatic decompression melting of the asthenospheric mantle. Additionally, melting within the lithospheric mantle may occur if it has been previously metasomatized. Heat conduction and/or heat advection by melts derived from the convecting mantle (McKenzie, 1989) may cause partial melting of volatile and K-rich veins within the lithosphere, resulting in magmas that are enriched in incompatible trace elements. Evidence for such a process is widely documented in continental rifts. For example, in the Rio Grande Rift of the western USA, strongly potassic magmatism (lamproites and minettes) on the rift flanks have been interpreted as the products of re-melting small melt fractions that had previously infiltrated the lithospheric mantle and solidified as veins (Gibson et al., 1993). Wholesale melting of the lithospheric mantle could, potentially, occur if small melt fractions were to react substantially with the surrounding anhydrous lithosphere (e.g. Hawkesworth et al., 1992).
The Baikal Rift is one of the least studied regions of currently active major lithospheric extension. In this study, we focus on the petrography, mineral and whole-rock chemistry of extension-related lavas from the Vitim Volcanic Field (VVF) of the Baikal Rift, and develop a petrogenetic model for magma generation processes operating during the formation of this major continental rift.
|
BAIKAL RIFT ZONE |
|---|
The Baikal Rift is located within the Palaeozoic Sayan–Baikal fold belt, near its boundary with the Siberian craton (Fig. 1). The fold belt is a collage of terranes (e.g. Precambrian microcontinents, fragments of oceanic crust and island arcs) that were accreted onto the craton in the Late Riphean and Early to Late Palaeozoic (Logatchev & Zorin, 1992
). The onset of rifting occurred in the Oligocene (35–30 Ma), with a major increase in the rate of lithospheric extension in the Late Pliocene (<3 Ma). This increased rate of extension was largely responsible for the present-day topography (Logatchev & Zorin, 1987). The S-shape of the rift zone appears to be controlled by the edge of the Siberian craton and the Sayan–Baikal foldbelt (e.g. Molnar & Tapponnier, 1975); Lake Baikal itself follows the southeastern edge of the craton (Fig. 1).
|
Cenozoic igneous activity in the Baikal Rift Zone (BRZ) is restricted to several small (<7000 km2) lava fields to the east and south of Lake Baikal, with a total volume of approximately 5000 km3 (Logatchev & Florensov, 1978
). To the south and west of the southern end of Lake Baikal, there is a diffuse zone of volcanism in East Sayan and Tuva, extending into the regions of Hangai, Gobi and the Mongolian Altai (Fig. 1). The igneous activity in these regions has been described by Rasskazov (1994) and Barry & Kent (1998). Within the BRZ, there are four regions of Cenozoic volcanism, located to the south and east of Lake Baikal (Fig. 1): Udokan Plateau, Vitim, Hamar-Daban and Bartoy. The largest of the four volcanic fields is Vitim, which is located approximately 200 km east of Lake Baikal. Hamar-Daban and Bartoy are the smallest in area, and are found close to the axial part of the rift that runs through Lake Baikal in a NE–SW direction. Cenozoic volcanism is absent from the central part of the rift system and most of its basins (Ionov, 2002).
The BRZ has been the focus of numerous geophysical studies. The results of a recent seismic investigation (Suvorov et al., 2002) indicate that the depth to the Moho along the rift axis varies between 35 and 50 km; the crust is thinnest beneath the southern part of the rift. Furthermore, a 200 km wide zone of crustal thinning appears to extend NE from the rift axis to the VVF, where the crust is 35 km thick (Suvorov et al., 2002). A teleseismic study by Gao et al. (1994) indicates that there is a broad zone of asthenospheric upwelling beneath Lake Baikal and the region to the east. This was not observed in the studies of Burov et al. (1994), Petit et al. (1998) or Zhang (1998); Artemieva & Mooney (2001), however, have shown a region of lithospheric thinning extending 200 km to the east of Lake Baikal and suggested that the base of the thermal lithosphere (defined by them as the depth to the 1300°C adiabat) is between 100 and 125 km depth beneath the VVF.
Two hypotheses have been put forward to explain the origin of the forces responsible for extension in the BRZ:
(1) Oligocene rifting in the Baikal region began 30–35 Myr ago and was contemporaneous with the early stages of the India–Asia collision. Molnar & Tapponnier (1975) suggested that the collision was responsible for most of the large-scale tectonics of Asia, and that this may have caused rifting in the Baikal area. Recently, Polyansky (2002) suggested that the rifting was caused by the northward movement of the Indian plate into Eurasia, the east–west convergence of the North American and Eurasian plates and the southeastward extrusion of the Amur plate into NE Asia.
(2) Several workers have discussed the possibility of a mantle plume or plumes lying beneath the Baikal region during the Cenozoic (Zorin, 1981; Logatchev & Zorin, 1987, 1992; Kiselev & Popov, 1992; Windley & Allen, 1993; Petit et al., 1998). Balijinnyam et al. (1993) proposed that the Baikal and Mongolia regions are underlain by a number of small plumes or ‘diapirs’ (with diameters of <100 km), which were partly responsible for the formation of the rifts. As well as reporting features that suggest interaction of a mantle plume with the lithosphere beneath Lake Baikal (e.g. high heat flow, uplifted topography, lithospheric thinning and alkaline magmatism), Windley & Allen (1993) discussed how other observations are not consistent with stresses linked to the India–Asia collision. For example, alkalic magmatism and high heat flow are not confined to localized rifts, but extend across the Mongolian Plateau, implying a laterally extensive heat source. A variety of rift orientations are observed in the plateau. These are inconsistent with east–west extension produced by the India–Asia collision, as only a single rift orientation would be expected.
|
VITIM VOLCANIC FIELD |
|---|
A simplified geological map of the VVF is shown in Fig. 2a, indicating the distribution of the main volcanic centres. The igneous activity of the VVF may be divided into two phases: the earlier, more voluminous, phase occurred during Miocene to Pleistocene times (Kiselev, 1987
), whereas the second phase of activity took place in the Pleistocene and Holocene. The majority of the volcanic centres (Fig. 2a) are located in the northwestern part of the volcanic field, in an area parallel to the axial part of the rift zone.
|
There are several places in the VVF where mantle xenoliths have been found within the eruptives; the most well known is the ‘tuff pit’ of Ionov et al. (1993)
, which has also been referred to as the ‘Bereya quarry’ by Glaser et al. (1999) and the ‘picrobasalt quarry’ by Litasov et al. (2000).
Cenozoic sediments occur mainly to the SW of the VVF, close to the Atalanga and Vitim rivers (Fig. 2a). Mesozoic trachybasalt and andesite lavas are also present in this area. Granitic bodies, of inferred Palaeozoic age (Litasov et al., 2000), form the basement to the VVF and crop out further west towards Lake Baikal (Logatchev & Zorin, 1992). These are widespread and form a considerable proportion of the underlying crust, extending to depths of 20 km (Suvorov et al., 2002).
There are very few available age data for lavas from the VVF. Esin et al. (1995) gave K–Ar ages ranging from 6·6 to 10·65 Ma for four Cenozoic lavas from drill-hole 3313 (Ekzar) and three from drill-hole 4431 (Antasey), as well as an estimate of 16·25 Ma for the basalt hosting mantle xenoliths at the ‘tuff pit’ locality (Ionov et al., 1993).
|
SAMPLING |
|---|
Approximately 200 lava samples were taken from 60 mm diameter drill-cores, as exposure is very limited because of dense forest covering the plateau. Many of the drill-holes reach the granitic basement, with one extending to a depth of
700 m. Those in the Atalanga region intersect Mesozoic lavas at a depth of 400 m. The drill-hole locations are given in Appendix A and are shown in Fig. 2b. |
ANALYTICAL TECHNIQUES |
|---|
Analyses of olivine, clinopyroxene, plagioclase and spinel were made using a CAMECA SX-50 electron microprobe in the Department of Earth Sciences, University of Cambridge. Energy-dispersive spectrometry (EDS) was used to analyse SiO2, TiO2, Al2O3, FeO, MnO, MgO, Na2O and Cr2O3 on carbon-coated, polished thin sections of the samples. Operating conditions for EDS were an accelerating voltage of 20 kV, a beam current of 25 nA, a beam diameter of 1 micron, and a live counting time of 60 s for each analysis. Calibration was made with reference to a cobalt standard. On-line peak stripping and corrections were performed using Link Analytical ZAF4/FLS software. Wavelength-dispersive spectrometry (WDS) was used to analyse NiO and CaO in olivine, using the same operating conditions and Link Analytical software as for EDS. Calibration was made with reference to pure nickel and wollastonite standards.
At the University of Durham, 142 samples were analysed for their whole-rock major and trace element chemistry, by X-ray fluorescence (XRF) for major elements and, for trace elements, by inductively coupled plasma mass spectrometry (ICP–MS). XRF analyses were made using a Philips PW1400 X-ray fluorescence spectrometer with a PW1500 72 automatic sample changer. For major elements, glass discs made from the powdered sample fused with lithium tetraborate were analysed; trace element analyses were carried out on freshly pressed powder pellets. The international standards AGV-1 and DST-1 were used for calibration (Potts et al., 1992), and analyses were repeated throughout the run to monitor analytical precision. Rare earth and trace element (Pb, Th, U, Nb, Ta, Zr, Hf, Y and Ba) concentrations were determined using a Perkin-Elmer SCIEX ELAN 6000 ICP–MS. Samples were prepared by digestion with HF/HNO3 at the University of Cambridge, following the method of Jarvis & Jarvis (1992). Blanks were prepared with each batch of samples, and analytical accuracy and reproducibility were estimated from measurements of international rock standards GSP-1, BCR-2 and AGV-1. One standard and one blank were analysed at several intervals throughout the whole analytical run, to monitor signal drift and contamination within the instrument.
Sr, Nd and Hf isotope ratios were determined using a ThermoFinnigan Neptune Plasma Ionisation Multi-collector Mass Spectrometer (PIMMS) at the University of Durham. Nowell et al. (2003) have described the procedure for analysis of Sr, Nd and Hf on this instrument. Separation of Sr, Nd and Hf for analysis was achieved using a two-column procedure (Dowall et al., 2003). Whole-rock geochemical and Sr–Nd–Hf isotope analyses of representative samples from the VVF are given in Tables 1 and 2, respectively (the complete dataset is available as Electronic Appendices A and B, which may be downloaded from the Journal of Petrology website at http://www.petrology.oupjournals.org); the internal errors on the isotope ratios are reported in Table 3. The Vitim samples were analysed in two analytical sessions, and reproducibility is therefore given for Sr, Nd and Hf in each session. Details of normalization values, mass bias and standards used are given in the legend to Table 3.
|
|
|
|
MAGMA TYPES |
|---|
We have classified the Vitim lavas into sub-alkalic basalts, alkali basalts, basanites, melanephelinites and nephelinites (Fig. 3), using the IUGS total alkali versus silica (TAS) system of Le Maitre (2002)
and the proposals of Le Bas (1989) for strongly silica-undersaturated compositions. Melanephelinites are distinguished from nephelinites on the basis of their normative nepheline contents: melanephelinites contain less than 20% normative nepheline, whereas nephelinites contain >20%.
|
Figure 4 shows variations in SiO2 within the lavas from the Ekzar 3315, Antasey 4404 and Muliha 4633 drill-cores. It is clear that the sub-alkaline (tholeiitic) and alkali basalts and more strongly alkalic (basanite–nephelinite) magmas were erupted without any temporal pattern; highly alkalic lavas both overlie and underlie the tholeiitic basalts. The only consistent feature is that the youngest lavas shown in Figs 2 and 4 are all strongly alkalic.
|
|
PETROGRAPHY |
|---|
Table 4 summarizes the average modal proportions of the various phenocryst phases in the Vitim lavas. Melanephelinites, nephelinites, basanites and alkali basalts all have porphyritic textures with olivine, clinopyroxene ± plagioclase phenocrysts set in a fine-grained groundmass consisting of olivine, clinopyroxene, opaque oxides and plagioclase laths. Olivine phenocrysts are equant, often euhedral, and are most abundant in the nephelinites and melanephelinites. They vary from 0·2 to 3·0 mm in size, are commonly altered around the rims to brown iddingsite, and often contain inclusions of Cr-spinel. Some basanites and alkali basalts contain kink-banded olivine crystals, which may be xenocrysts (see below). Clinopyroxene phenocrysts are often pinkish in colour because of Ti enrichment. Large plagioclase phenocrysts (up to 2 mm in length) are found in the tholeiitic basalts and some alkali basalts, but are rare or absent in the other magma types. The tholeiitic basalts are dominated by laths of plagioclase feldspar, with generally fewer olivine phenocrysts than the other rock types. Cr-spinel inclusions in olivine are rare in the tholeiitic basalts.
|
|
MINERAL CHEMISTRY |
|---|
A comprehensive description of the mineral chemistry of the Vitim lavas is beyond the aim of this paper. However, below, we give a brief summary of the chemical compositions of olivine, clinopyroxene and plagioclase feldspar phenocrysts present in the Cenozoic lavas, and describe how olivine compositions can be used to understand the processes operating in crustal magma chambers beneath the VVF.
We have studied the composition of approximately 600 olivine crystals from the Vitim lavas (Electronic Appendix C, available at http://www.petrology.oupjournals.org); Table 5), in order to understand whether they crystallized in equilibrium with the magmas, or were xenocrysts from the underlying lithospheric mantle. Olivine phenocrysts in the Vitim lavas have forsterite contents in the range Fo67–Fo87 (Fig. 5). A trend towards a higher frequency of olivine phenocrysts with increasing forsterite content is observed. There are no phenocrysts with Fo > 87%; olivines with Fo87–Fo91 are invariably xenocrysts (identified by their anhedral shape, embayed margins, kink-banding, CaO contents below 0·1 wt %, and high Fo content). There is no systematic variation in Fo content with rock type.
|
|
There is a general trend of increasing NiO content of olivine crystal cores with Fo content. The highest Fo contents (>Fo87) are observed in olivines from lherzolite xenoliths from the ‘tuff pit’ locality of Ionov et al. (1993)
. These also have the highest NiO contents (>0·4 wt %). All other olivines are believed to be phenocrysts (consistent with CaO>0·18 wt %; Brey et al., 1990); those with the highest NiO contents are likely to have crystallized from the most primitive magmas. The compositions of Vitim olivines fall in the range expected for oceanic picrites [compare with fig. 6 of Gibson (2002)], but the Fo contents extend to lower values because the lavas from the VVF are more evolved (the majority have MgO < 12 wt %, and are therefore not classified as picrites).
Analyses of representative clinopyroxene phenocrysts are given in Table 6 (the full dataset is given in Electronic Appendix D, available at http://www.petrology.oupjournals.org). Using the classification scheme of Morimoto (1988), all of the pyroxenes analysed (from both lavas and xenoliths from the VVF) are quadrilateral. Their compositions are restricted on a Wo–En–Fs triangular plot (not shown), with the clinopyroxene phenocrysts generally falling within or very close to the diopside field. Clinopyroxene phenocrysts have Mg-number [Mg-number = Mg/(Mg + Fe)] in the range 70·1–83·4, and are estimated to have crystallized from liquids with Mg-number = 35·0–53·6, using a partition coefficient for Fe–Mg partitioning between augite and melt of 0·23 (Grove & Bryan, 1983).
|
Analyses of representative plagioclase feldspar phenocrysts are given in Table 6 (the full dataset is given in Electronic Appendix E, available at http://www.petrology.oupjournals.org). Plagioclase feldspars typically have core compositions that fall in the labradorite field, in the range An60–An70 [An = atomic Ca/(Ca + Na) x 100].
|
WHOLE-ROCK CHEMISTRY |
|---|
Whole-rock geochemical analyses for representative samples from the VVF are given in Table 1; the full dataset can be found in the study by Garner (2002)
and Electronic Appendix A (available at http://www.petrology.oupjournals.org).
Weathering and alteration
The majority of samples from the VVF are petrographically fresh, and contain few alteration minerals, e.g. iddingsite. Loss on ignition (LOI) values range from 0·14 to 6·26, with the majority of samples having LOI values of <3. Some were found to have gained mass during ignition, as a result of oxidation of FeO to Fe2O3 (see Table 1). This occurs if there are large amounts of FeO and small amounts of OH present in a sample, such that the increase as a result of oxidation is greater than the OH loss. Rb and K are particularly mobile during alteration. However, there does not appear to be any direct correlation between LOI and Rb and K contents in the Vitim lavas, suggesting that alteration processes have not significantly affected their chemistry.
Major element variation
The Cenozoic lavas of the VVF have SiO2 contents in the range of 43–53 wt %, and form two series on a total alkali versus silica diagram: a strongly alkaline series (melanephelinites, nephelinites and basanites) and a mildly/sub-alkaline series (alkali and tholeiitic basalts; Fig. 3). SiO2 contents increase from nephelinites (43–49 wt %) through to tholeiitic basalts (48–53 wt %; Fig. 6). Conversely, MgO contents are lowest in the tholeiitic basalts (3·3–10·6 wt %) and highest in the basanites (7·5–12·0 wt %). The nephelinites have a narrower range of MgO contents than the other magma types (6·5–10·1 wt %). Sample 93VBS 4 (a melanephelinite) has anomalously high MgO (MgO = 14·6 wt %) and contains xenocrystic olivine.
|
CaO and
(total Fe) contents of the lavas range from 7·7 to 11·6 and 9·7 to 14·7 wt %, respectively (Fig. 6). Al2O3 abundances are highest in the tholeiitic basalts (13·4–16·3 wt %), and lowest in the basanites, nephelinites and melanephelinites (12·2–15·3 wt %). TiO2 contents range from 1·8 to 3·3 wt %, with the highest values in the basanites. Nephelinites have the highest contents of Na2O (3·7–5·4 wt %), K2O (1·7–3·0 wt %) and P2O5 (0·5–1·1 wt %) of all the rock types. The melanephelinites exhibit a small range in all the oxides, and contents are intermediate between those of the nephelinites and basanites. Despite the large number of samples analysed, the Vitim lavas do not always show clear correlations on Harker variation diagrams (Fig. 6), and there are no distinct inflections in the trends of the data suggesting the onset of crystallization of different phases. To account for the wide range of major element compositions in the lavas (Fig. 6), it is likely that there was a spectrum of parental magma compositions. The systematic decrease in Na2O, P2O5 and K2O (Fig. 6b, d and h) from nephelinites through to tholeiitic basalts for a range of MgO values cannot be explained by fractional crystallization and reflects variations in partial melting processes in the mantle source (see below).
Trace-element variation
Concentrations of compatible trace elements, such as Ni and Cr, are highest in the basanites. These range from 106 to 279 and 82 to 394 ppm, respectively. The lowest abundances are found in the alkali basalts, in which Ni varies from 39 to 238 ppm and Cr from 55 to 403 ppm. Although the overall ranges for Ni and Cr are large (Ni = 39–279 ppm, and Cr = 59–542 ppm), the majority of samples have concentrations of Ni of <100 ppm, and between 100 and 400 ppm Cr. They are therefore not representative of primitive magma compositions. Both Ni and Cr contents are positively correlated with MgO for all of the groups, reflecting olivine and spinel crystallization, respectively (Fig. 7).
|
Abundances of incompatible trace elements in the Vitim lavas generally increase from the tholeiitic basalts to the nephelinites, e.g. Rb (9–52 ppm), Th (1·41–6·59 ppm), Nb (18·4–83·7 ppm), Ta (1·01–5·07 ppm), Sr (385–1248 ppm), Zr (118–337 ppm), Hf (3·02–7·38 ppm). Pb contents are more variable, but there is a general increase observed from the tholeiitic basalts (1·18–4·06 ppm) to the nephelinites (3·76–5·00 ppm). Concentrations of Y are more constant, ranging from 20·2–31·9 ppm in the tholeiitic basalts to 22·2–27·4 ppm in the nephelinites.
MgO contents are plotted against selected incompatible trace elements in Fig. 8 to show the effects of fractional crystallization on the parental magmas. There is no distinct trend defined by the Vitim lavas, confirming that their evolution is more complex than simple fractional crystallization of a single parent magma. The wide range in abundance of incompatible trace elements for a given MgO content (i.e. between different groups) may, however, be explained by variation in the degree of partial melting or mantle source heterogeneity (Fig. 8; see below).
|
On normalized multi-element plots (Fig. 9), all the lavas from the VVF exhibit smoothly curved concave-upward profiles that peak at Nb and Ta. The most incompatible trace element enriched lavas (relative to chondrites) are the nephelinites and melanephelinites, which also have the lowest normalized abundances of the more compatible trace elements, e.g. Yb. Concentrations of strongly incompatible trace elements, such as Ba, are higher in the nephelinites, melanephelinites and basanites (273–594 ppm) than the alkali and tholeiitic basalts (201–571 ppm).
|
Slight K depletions are observed in Fig. 9 for some of the nephelinites and basanites ([Nb/K]n
1·2), whereas the alkali basalts and tholeiitic basalts usually have either no relative depletion or a slight relative enrichment in K (e.g. [Nb/K]n = 0·8 for 93VBS 314). In several samples, there are slight relative enrichments in P (e.g. 93VBS 356) and Ti (e.g. 93VBS 223). These are particularly marked for P in the nephelinites, melanephelinites, basanites and alkali basalts, and Ti in the alkali and tholeiitic basalts. Chondrite-normalized rare earth element (REE) patterns form smoothly curved trends from La to Lu, with no significant relative depletions or enrichments of individual elements (Fig. 10). Eu anomalies are absent, indicating that there has not been any significant plagioclase fractionation. Light REE abundances are relatively restricted in their overall range, increasing in the different rock groups in the following order: tholeiitic basalts, alkali basalts, basanites, melanephelinites, nephelinites.
|
The gradual flattening-out of the slope of the REE pattern from nephelinites through to tholeiitic basalts is clearly visible in Fig. 10. The highest light REE to heavy REE ratios (LREE/HREE) are observed in the nephelinites ([La/Lu]n = 20·3–38·0), and two samples have particularly enriched LREE abundances (93VBS 356 and 281). Notably high [La/Lu]n ratios (>36·0) are observed in these nephelinitic samples. These lava flows can be traced across the volcanic field, and usually lie at or near the top of the lava pile. In general, the gradient of the REE patterns is broadly correlated with SiO2 content (lavas with lower SiO2 contents tend to have steeper REE slopes, and hence higher [La/Lu]n and [La/Yb]n ratios).
Radiogenic isotopes
Sr and Nd isotope systematics
Lavas from the VVF exhibit a small range of initial Sr and Nd isotopic ratios (87Sr/86Sri = 0·7039–0·7049 and 144Nd/143Ndi = 0·51272–0·51287; Fig. 11). There does not appear to be any systematic correlation of isotopic composition with rock type or location within the volcanic field; for example, 87Sr/86Sri ranges from 0·7039 to 0·7041 for nephelinites, 0·7040 to 0·7041 for basanites, 0·7039 to 0·7042 for melanephelinites and 0·7040 to 0·7049 for alkali basalts, compared with 0·7040–0·7048 for tholeiitic basalts. 143Nd/144Ndi ranges are as follows: 0·51283–0·51285 for nephelinites, 0·51279–0·51287 for melanephelinites, 0·51283–0·51287 for basanites, 0·51272–0·51284 for alkali basalts and 0·51283–0·51285 for tholeiitic basalts. All of the lavas have more enriched Sr and Nd isotopic compositions than the peridotite mantle xenoliths from the VVF (Ionov et al., 1995).
|
Nd and Hf isotope systematics
On a diagram of
Nd versus Hf (Fig. 12), the Vitim lavas define a positive correlation and plot within the field for OIB and close to Bulk Earth. The highest Hf values are observed in the nephelinites (8·7) and the lowest in the alkali basalts (5·9). The ranges in Hf overlap between the different groups, and are as follows: 8·0–8·7 for the nephelinites; 6·9–7·8 for melanephelinites; 7·6–8·1 for basanites; 5·9–8·1 for alkali basalts; 7·6–8·2 for tholeiitic basalts. As in Fig. 11, sample 93VBS 92 (an alkali basalt) plots away from the other samples, at lower Nd and slightly lower Hf values. All of the lavas have lower Nd and Hf values than peridotite xenoliths from the VVF (D. A. Ionov, personal communication, 2003).
|
|
CRUSTAL PROCESSES |
|---|
Before we can model the mantle melting processes that gave rise to the parental magmas of the Vitim lavas, it is important to constrain their melt source characteristics. In order to do this, it is first necessary to establish the extent to which magma compositions have been affected by crustal processes such as fractional crystallization and crustal assimilation.
Parental magma composition
The forsterite contents of olivines from the Vitim lavas (see Electronic Appendix C for data) were used to estimate the Mg-number and the MgO content of the magma with which they were in equilibrium. In order to calculate the Mg-number of the melt, it is first necessary to estimate the Fe2O3/FeO ratio. The oxygen fugacity (fO2) was calculated using Mg2+ and Fe2+ exchange between coexisting olivine and spinel (Sack & Ghiorso, 1991). We estimated 
logfO2 values for some Vitim lavas from electron microprobe analyses of olivine phenocrysts and the spinels contained within them (see Electronic Appendix F, available at http://www.petrology.oupjournals.org, for data). logfO2 ranges from –3·8 to –8·4 units below the fayalite–magnetite–quartz (FMQ) buffer. Using the empirical expression of Kilinc et al. (1983) and the revised constants of Holloway et al. (1992), we estimated that the Fe2O3/FeO ratio of these magmas is close to 0·08. Assuming that the olivine–spinel pair with the highest calculated fO2 has undergone the least subsolidus re-equilibration, and therefore gives a minimum value for Fe2O3/FeO (Gibson, 2002), the ratio used in the following calculations is 0·1.
Equilibrium values for olivine phenocrysts for a range of whole-rock Mg-numbers are shown in Fig. 13. Many samples contain olivines whose Fo contents are too low to be in equilibrium with the whole rock (e.g. 93VBS 204 and 288). This suggests that they have been incorporated into a magma with lower Mg-number [Mg-number = Mg/(Mg + Fe)], thus increasing the whole-rock Mg-number, or that the most forsteritic olivines in these samples were not analysed. Figure 13 shows that the most mafic VVF lava containing equilibrium olivine is 93VBS 367 (an alkali basalt). Olivine phenocrysts in this lava have moderate forsterite contents (up to Fo86) and we have calculated that these were in equilibrium with magma with a Mg-number of 64·9 (using 
, from Roeder & Emslie, 1970). Nephelinite 93VBS 281 contains slightly less forsteritic olivine phenocrysts (Fo85) that are in equilibrium with the whole-rock Mg-number (63·0). Using the variation in whole-rock Mg-number and MgO contents for the VVF lavas (Fig. 14), parental magma compositions of 10·8 and 9·9 wt % MgO were estimated for the alkali basalt and nephelinite lavas, respectively. It is unlikely that a magma with <11 wt % MgO is representative of a primary magma, as worldwide primary magmas from intra-plate settings have comparatively higher MgO contents, e.g. MgO = 17·5 wt % for Mauna Loa, Hawaii (Garcia et al., 1995); MgO = 20–21 wt % for West Greenland (Larsen & Pedersen, 2000).
|
|
Ni and Cr partition coefficients can also be used to suggest whether parental magmas are primary. Primary magmas are generally expected to have Ni > 400–500 ppm and Cr > 1000 ppm, together with Mg-number >70 (Wilson, 1989
). Hart & Davis (1978) showed that the Ni partition coefficient is dependent on the MgO content of the melt (their fig. 5). These are related by the equation
 |
value of 10·6 gives an estimated Ni content of 260 ppm for the melt in equilibrium with an olivine of Fo86 (the most forsteritic olivine in 93VBS 367). The calculated Cr content of the melt is 800 ppm.
Polybaric fractional crystallization
The pressures and depths of fractional crystallization may be estimated using CIPW normative compositions. We have plotted the CIPW norms of the Vitim lavas on a Ne–Ol–Di–Hy–Qz projection and compared them with cotectics for basaltic liquids in equilibrium with olivine, plagioclase and clinopyroxene at different depths within the crust (e.g. Thompson, 1983; Thompson et al., 2001; Fig. 15). The large majority of lavas from the VVF are strongly silica-undersaturated (nepheline normative), but a few are hypersthene normative and quartz normative. All of the samples lie between the 1 atm and 9 kbar cotectics, indicating that the Vitim magmas fractionated over a wide range of pressures within the crust (Fig. 15). The nephelinites appear to have undergone fractional crystallization at the lowest pressures (1 atm), whereas some of the tholeiitic basalts plot between the 1 atm and 9 kbar cotectics and have undergone fractionation throughout the crust. A recent seismic study by Suvorov et al. (2002) suggests that the base of the Moho beneath the VVF is 35 km, and fractional crystallization of tholeiitic magmas at 9 kbar during the formation of the VVF may have caused underplating of the crust.
|
Crustal contamination
In the subset of samples analysed for their Sr–Nd–Hf isotopic ratios, three samples (93VBS 63, 92 and 370) have significantly higher 87Sr/86Sri ratios than the rest of the group (Fig. 11). This displacement to higher 87Sr/86Sri ratios may be caused by hydrothermal alteration and/or crustal contamination, as both the upper and lower crusts are known to have 87Sr/86Sri > 0·7045 (Taylor & McClennan, 1985
). These lavas have low U/Pb ratios (<0·2), which may also indicate contamination (Thompson et al., 2001). The majority of samples, however, have similar U/Pb ratios to oceanic basalts (0·2–0·4), Nd values of 3–5 and 87Sr/86Sri < 0·7045 (Fig. 11), which suggest that crustal contamination has not played a significant role in their petrogenesis. This is confirmed by combined variations in Hf and Nd isotopic ratios. Most continental upper crust has present-day Nd < 0 and Hf < 0, and plots in the lower left quadrant of the Nd vs Hf diagram in Fig. 12 (Vervoort et al., 1999). Figure 12 shows that, with one exception, the Vitim data are tightly clustered and show no sign of a trend towards crustal compositions. We therefore believe that the parental magmas of the Vitim lavas have not undergone significant interaction with the continental crust and that their compositions may be used to assess the nature of the underlying mantle. |
MANTLE SOURCE CHARACTERISTICS |
|---|
Evidence for mantle source heterogeneity
The variations in abundance of major, trace and REE in the Vitim lavas, and their Sr–Nd–Hf isotope systematics, indicate that the mantle source regions beneath the VVF may be heterogeneous in composition. Tangible evidence for lithospheric mantle heterogeneity comes from the wide variety of xenoliths that are found in the VVF, including garnet- and spinel-bearing peridotites, as well as pyroxenites (Ionov et al., 1993
; Ashchepkov et al., 1994; Litasov et al., 2000; Ionov, 2004). These have been collected from a limited number of places, such as the 16·5 Ma ‘tuff pit’ locality (Esin et al., 1995) (Fig. 2). The mineralogy of the mantle xenoliths includes olivine, clinopyroxene, orthopyroxene, garnet, spinel, amphibole, phlogopite and ilmenite (Ionov et al., 1993; Litasov et al., 2000).
Mineralogy of the mantle source
We can potentially identify mineral phases that might be residual in the mantle source region of the Vitim magmas by examining the behaviour of incompatible trace elements during melting, using published partition coefficient data for accessory minerals (e.g. McKenzie & O'Nions, 1991; LaTourrette et al., 1995; Green et al., 2000). Ratios between middle REE (MREE) and HREE may be used to assess the presence or absence of garnet during mantle melting. This is because the garnet–melt partition coefficients are higher for the HREE than the MREE (McKenzie & O'Nions, 1991). Chondrite-normalized Gd/Lu ratios are high in the Vitim samples (generally between 2·5 and 4·5) and therefore indicate that their parental melts were generated within the garnet stability field. Most nephelinitic samples have [Gd/Lu]n ratios in excess of 3, which suggests that their parental melts were formed at either: (1) greater pressures or (2) smaller degrees of partial melting than the other Vitim magmas. Sample 93VBS 4 has [Gd/Lu]n = 7, which is much higher than in all the other lavas. This was collected from the ‘tuff pit’ locality (Fig. 2), and contains numerous xenocrysts derived from disaggregated crustal and mantle xenoliths.
Amphibole, phlogopite and ilmenite are metasomatic phases and their presence in the Vitim mantle xenoliths provides important information about the nature of the subcontinental lithospheric mantle. Available partition coefficient data indicate that Ba is compatible in phlogopite, but only moderately incompatible in amphibole (e.g. Späth et al., 2001). Neither phlogopite nor amphibole is capable of fractionating Sr. Figure 16 shows Ba normalized to Al2O3, plotted against the incompatible element, Th. We chose to normalize Ba to Al2O3 because, during melting, the ratio of an incompatible element to Al2O3 will decrease systematically with increasing melting as a result of Al2O3 being buffered by residual garnet. If an element is behaving compatibly when two completely incompatible elements are plotted against each other, this will show as a deviation from the expected linear trend (Hoernle & Schmincke, 1993). The Vitim magmas plotted in Fig. 16 form a clear inflection in the trend, corresponding to the change from basanitic to alkali basaltic magmas. In contrast, Sr behaves as a strongly incompatible element in all magma types. A residual phase capable of fractionating Ba but not Sr (e.g. phlogopite) must therefore be present in the source at lower degrees of melting. Amphibole may also be residual, but its effect will be masked by phlogopite.
|
Figure 17 shows the varying compatibility of elements during melting to produce the Vitim lavas. If an element X is totally incompatible during melting, values for X/Al2O3 will lie on a straight line of positive slope, passing through the origin (Hoernle & Schmincke, 1993
). Deviation from this trend shows that element X is probably hosted by a residual accessory mineral during melting. On a plot of P2O5/Al2O3 vs Rb/Al2O3 (Fig. 17a), Rb behaves incompatibly in the alkali and tholeiitic basalts, but there is a deviation from this trend for the nephelinites and several basanites (at P2O5/Al2O3 = 0·04). Values of Rb/Al2O3 are lower than expected for a given degree of melting, which suggests that Rb is residing in a mineral phase within the mantle source of the strongly alkaline magmas. Partition coefficients for amphibole (pargasite) and phlogopite show that Rb is compatible in phlogopite (KD > 1) but incompatible in amphibole (KD < 1; LaTourrette et al., 1995); the compatible behaviour of Rb in the smaller-degree melts is therefore probably a result of residual phlogopite. K (Fig. 17b) shows a similar distribution to Rb, and it is also more compatible in phlogopite than in amphibole. TiO2 (Fig. 17c) also behaves similarly to Rb and K in the Vitim nephelinitic and basanitic magmas, suggesting that a Ti-bearing residual phase is present in the mantle source at small melt fractions. Ta appears to have been fractionated during the melting that produced the nephelinitic magmas (at P2O5/Al2O3 > 0·06; Fig. 17d).
|
In summary, there is strong evidence that phlogopite was a residual mantle phase during the melting process that produced the Vitim nephelinitic to basanitic magmas. The inferred presence of this phase is consistent with the occurrence of phlogopite in mantle xenoliths from the VVF (Ionov et al., 1993
; Litasov et al., 2000), but it also has implications for the location of the Vitim magma sources. Experimental studies have shown that the stability of phlogopite is dependent on volatile content (OH and F) as well as temperature and pressure (Foley et al., 1986; Sato et al., 1997). Ionov et al. (1993), Litasov et al. (2000) and Ionov (2002) used mineral equilibria studies to obtain temperature and pressure estimates for the equilibration of phlogopite-bearing garnet lherzolite xenoliths from the VVF. None was found to have equilibrated at T > 1200°C and P > 30 kbar (Fig. 18). This suggests that the minimum depth of the mechanical boundary layer (MBL) at the beginning of magmatism (16·5 Ma; Esin et al., 1995) was 100 km. For mantle of normal potential temperature (1300°C), this MBL thickness corresponds to a thermal boundary layer (TBL) thickness of 35 km (McKenzie & Bickle, 1988). Figure 18 shows that at these depths, phlogopite is stable in both the MBL and TBL but not the convecting mantle. The presence of residual phlogopite in the mantle source of the nephelinites and basanites indicates that they were generated at T < 1300°C. [The definitions that we have used for the lithosphere are from McKenzie & Bickle (1988) and White (1988). These state that: (1) the TBL is a transition zone that is neither totally rigid nor vigorously convecting, and separates the rigid plate (crust and subcontinental lithospheric mantle) from the underlying convecting mantle; (2) the MBL is the outer layer of the Earth, which responds elastically to a depth where the temperature reaches 550–600°C. Beneath this, the MBL behaves plastically to long-term loads. This definition of the MBL differs from those of Jordan et al. (1989) and Anderson (1994), who used the term ‘MBL’ as equivalent to only elastic thickness of the lithosphere, which approximately corresponds to the 600°C isotherm.]
|
|
MAGMA GENESIS |
|---|
Here, we use independent geochemical modelling techniques to assess the composition of the contributing melt source regions, and to suggest the depth and extent of partial melting, beneath the VVF.
Alkali and tholeiitic basalts
We have used a forward major element modelling method, similar to that developed by Langmuir et al. (1992) for MORB, but with modifications to make it more appropriate for lavas from the VVF. The resulting model calculates major element compositions (FeO and Na2O) for polybaric and isobaric melting paths for plausible mantle sources, and compares these with the major element abundances in the VVF rocks. Model parameters are given in Table B1, and melting equations in Table B2 (both in Appendix B). From this point onwards, the modified model will be referred to as the ‘hybrid’ model, as it also incorporates some of the ideas from the melting models of Kostopoulos & James (1992). The original Langmuir et al. (1992) model was based on Fe–Mg partitioning, assuming that Fe is partitioned only into residual olivine (it was used for modelling melt generation at shallow depths where garnet is not a residual phase). However, during melting beneath the Vitim region, MREE/HREE ratios indicate that garnet is a significant residual phase (see above), and the hybrid model is designed to incorporate this. It also uses melting proportions (from Kostopoulos & James, 1992) to recalculate the residue composition at each stage of melting.
The output from the hybrid melting model applies only to primary melts and the Vitim samples must be extrapolated to primary compositions in order to make a quantitative comparison. For this fractionation correction, we used the method of Turner & Hawkesworth (1995), in which data are corrected by fitting least squares linear regression lines. Figure 6 shows that VVF magma compositions converge towards a common value at 12 wt % MgO. This value is therefore likely to represent the MgO content of a primary magma. Only samples with MgO > 10 wt % are projected along the regression vectors, in order to minimize the effects of fractional crystallization.
Initially, we assumed an anhydrous fertile peridotite source (KLB-1) which has moderate abundances of FeO* (8·59 wt %) and Na2O (0·30 wt %) (Takahashi, 1986). Model curves for both batch melting and accumulated fractional melting are shown in Fig. 19. In accumulated fractional melting, several melt increments are collected together in a common reservoir, after isolation from the melt source. The results of our polybaric melting calculations are shown in Fig. 19a. The Na and Fe contents of the Vitim tholeiitic basalt melts can be modelled by 7% decompression melting of KLB-1 between 35 and 33 kbar, i.e. from 115 to 110 km, assuming an increase of 1 kbar pressure for every 3·3 km depth increase in the mantle. Similarly, the Na and Fe contents of the alkali basalts indicate 5% partial melting between 115 and 100 km. At 35 kbar, the KLB-1 solidus is at 1560°C; this corresponds to a mantle potential temperature of 1490°C, assuming a mantle adiabat with gradient of 0·6°C/km (McKenzie & Bickle, 1988).
|
An alternative approach to modelling the generation of the Vitim alkali and tholeiitic basalts is the method of REE inversion. We have followed the scheme described by McKenzie & O'Nions (1991)
, and subsequently modified by White et al. (1992), in which partial melt distributions were obtained using whole-rock REE concentrations. It assumes that the compositions of all melts and residues are governed by the depth and degree of partial melting. The alkali and tholeiitic basalts were modelled by single-stage melting of an asthenospheric mantle source, which has Nd = 4·5 (similar to the Nd values of the Vitim basalts, Table 2). This is equivalent to a mixture of 55% Primitive Mantle (PM) and 45% Depleted Mantle (DM; McKenzie & O'Nions, 1991). A good fit to the REE (Fig. 20a and c) and a reasonable fit to other incompatible elements (Fig. 20b and d) was obtained. After correction for fractionation, the inversion modelling predicts 5 and 3% melting to generate the tholeiitic and alkali basalts, respectively, with melting occurring initially within the garnet stability field (between 85 and 105 km; Fig. 20e). The predicted melt distribution closely follows the decompression-melting curve for a mantle potential temperature of 1450°C (Fig. 20e).
|
Nephelinites, melanephelinites and basanites
The results of experimental studies suggest that silica-undersaturated melts, such as nephelinites, melanephelinites and basanites, may be generated by: (1) high-pressure, small-degree melting of fertile peridotite (Takahashi & Kushiro, 1983
; Zhang & Herzberg, 1994), (2) melting of fertile peridotite in the presence of CO2 or H2O (Hirose, 1997; Kawamoto & Holloway, 1997), (3) melting of an amphibole or phlogopite-bearing peridotite (Wallace & Green, 1988; Mengel & Green, 1989; Thibault et al., 1992), or (4) partial melting of garnet pyroxenite (Yaxley & Green, 1998; Hirschmann et al., 2003; Kogiso et al., 2003). Figure 19a suggests that polybaric melting of a fertile peridotite source (such as KLB-1) cannot generate the high Na and Fe contents of the VVF nephelinites, melanephelinites and basanites, even at the smallest degrees of partial melting. In addition, both high-pressure melts of fertile peridotite and those generated in the presence of volatiles (from fertile peridotite) have considerably higher MgO contents [and Ca(Ca + Mg) ratios] and lower Al2O3 contents than the VVF silica-undersaturated magmas. These discrepancies in MgO and Al2O3 cannot be attributed to post-melt generation processes, such as fractional crystallization, and we therefore believe that melts derived from partial melting of fertile peridotite are unlikely to have made significant contributions to the silica-undersaturated magmas.
Pyroxenite xenoliths from the VVF were described by Litasov et al. (2000). We consider that pyroxenites within the lithosphere could be the dominant melt source region for the silica-undersaturated magmas. We have explored this possibility by plotting the fractionation-corrected compositions of the nephelinites, melanephelinites and basanites from the VVF together with those of experimental melts of garnet pyroxenite on a SiO2 versus FeO diagram (Fig. 21). It can be seen from this that the VVF silica-undersaturated melts and garnet pyroxenite melts have very similar compositions.
|
We have used the whole-rock composition of garnet pyroxenite, MIX1G (Hirschmann et al., 2003
) in our hybrid melting models. This has FeO* and Na2O contents of 7·80 and 1·40 wt %, respectively. The polybaric melting model that we discussed above, for the tholeiitic and alkali basalts, assumes that melting occurred by adiabatic decompression of the convecting mantle. However, this mechanism may not be strictly appropriate for generation of the nephelinitic, melanephelinitic and basanitic magmas if their contributing melts were generated from metasomatic veins in the lithospheric mantle (as we have inferred above on the basis of residual phlogopite). Because a vein consisting of different mineral phases may be melted in discrete increments at the same pressure, an isobaric melting model is probably more suitable for generation of these silica-undersaturated melts beneath the VVF.
We tested isobaric melting using the hybrid model, with both KLB-1 and MIX1G as source compositions. Isobaric melting of KLB-1 (Fig. 19b) cannot reproduce the high Fe (or Na) contents of some of the Vitim melts. A closer fit to these Vitim data is achieved, however, by isobaric melting of MIX1G (Fig. 19c). The melanephelinites and basanites predominantly fall between the MIX1G melting curves at 20 and 25 kbar, and at 18–30% melting. The fractionation-corrected nephelinites have higher Fe contents and fall along the 25 kbar melting curve, at similar or lower degrees of melting. From this, we conclude that the nephelinites were formed by partial melting at slightly higher pressures than the basanites and melanephelinites. This is consistent with the conclusions of Hirschmann et al. (2003), who showed that the most Fe-rich silica-undersaturated magmas were formed by melting garnet pyroxenite at higher pressures and/or lower temperatures than magmas with lower Fe contents.
Despite the close correlation between most of the major element abundances in the VVF silica-undersaturated melts and those of experimental partial melts of garnet pyroxenite (Hirschmann et al., 2003), we note that the former have much higher contents of K2O, TiO2 and P2O5. We have shown above that both phlogopite and garnet were present as residual phases in the melt source region of the silica-undersaturated VVF melts and propose that they were generated from a phlogopite-bearing garnet pyroxenite source. Hirschmann et al. (2003) showed that, at 25 kbar, the garnet pyroxenite solidus is between 1375 and 1400°C. However, this initial melting temperature would be reduced and the pressure increased if the amount of K2O was higher, as a result of of the presence of phlogopite (Tsuruta & Takahashi, 1998; Wang & Takahashi, 2000).
|
DISCUSSION |
|---|
Implications for the sub-Vitim lithospheric mantle
The results of our geochemical modelling of the tholeiitic and alkali basalts from the VVF indicate that adiabatic decompression melting (up to 7%) of the convecting mantle occurred between
85 and 105 km (from inversion modelling), or 100 and 115 km (from major element modelling). We have assumed that the top of the melting column was controlled by the thickness of the overlying rigid MBL.
Oligocene to Recent lithospheric extension associated with the Baikal Rift appears to have reactivated Mesozoic rift structures beneath the region and may have caused localized thinning of the MBL to 85 km. Our findings are consistent with geophysical estimates, which suggest that the base of the MBL is at 100 km depth (Burov et al., 1994) beneath the whole of the Vitim region. Furthermore, no mantle xenoliths from the VVF appear to have equilibrated at depths below 80–100 km (Ionov et al., 1993; Litasov et al., 2000; Fig. 18). Model age estimates from the available Nd and Os isotopic data (Ionov et al., 1993; Pearson et al., 1998, 2003; D. A. Ionov, personal communication, 2003) suggest that the peridotite xenoliths analysed to date only sample the MBL of the lithospheric mantle.
We have shown above that the nephelinitic and basanitic melts from the VVF were formed by partial melting of a phlogopite-bearing garnet pyroxenite source. Such material can reside in the convecting mantle as ‘streaks’ (Gibson, 2002) or in the TBL and/or MBL as veins. Our estimated depths of garnet pyroxenite melting (83–66 km; 25–20 kbar) are less than those for the top of the melting column in the convecting mantle, and this confirms our earlier suggestion (based on phlogopite stability) that the basanites and nephelinites were generated in the base of the MBL and/or the TBL.
Litasov et al. (2000) proposed a model in which melt, derived from pyroxenite veins at depth within the sub-Vitim MBL, trickles upwards and subsequently crystallizes at lower pressures. These veins are re-melted at a later stage of rifting, mobilized, and then erupted at the Earth's surface. In order to melt metasomatic veins in the MBL, heat must be transferred by conduction from the asthenospheric mantle or by advection from asthenospheric melts (McKenzie, 1989). At high mantle potential temperatures, metasomatized zones at the base of the MBL would melt in less than 10 Myr by heat conduction (Roberts, 2002), and heat advected by rising melts would cause immediate melting. Both processes would lead to the occurrence of small volumes of silica-undersaturated magmas interbedded with Miocene to Recent tholeiitic and alkali basalts in the VVF. Nevertheless, we note that the silica-undersaturated magmas in the VVF have different Sr, Nd and Hf isotopic ratios from those of entrained peridotite mantle xenoliths (Figs 11 and 12). This suggests that either garnet pyroxenite veins in the sub-Vitim MBL have different Sr, Nd and Hf isotopic ratios from the peridotite xenoliths or that the magmas were derived from garnet pyroxenites in the TBL.
During periods of tectonomagmatic quiescence, continental lithospheres will develop an underlying TBL by conductive cooling (McKenzie & Bickle, 1988). Thus, the sub-Vitim lithosphere would have undergone conductive thickening during the interval between the earlier phase of Mesozoic magmatism (Rasskazov, 1994) and the Miocene. Although the TBL is likely to undergo convective overturn on a time-scale of >10 Myr (McKenzie & O'Nions, 1995), its uppermost few kilometres will characteristically remain stable for several million years at temperatures (by definition) lower than those of the underlying convecting mantle. This thin zone at the top of the TBL is an ideal place for batches of very-small-fraction incipient melts to solidify as they leak from the asthenosphere (McKenzie, 1989; Wilson et al., 1995). Thompson et al. (2005) have suggested that substantial amounts of such melts can accumulate within less than 20 Myr.
Experiments conducted by Yaxley & Green (1998) have shown how veined mantle in the TBL might melt. Initial fusion of the veins produces Mg-poor liquids that react rapidly with the surrounding peridotite, enriching it in garnet and clinopyroxene. When this reaction zone in turn begins to melt with rising temperature, the liquids produced are picritic nephelinites and basanites with higher Fe contents than anhydrous peridotite melts. This process might explain the relative Fe enrichment of the strongly alkalic magmas observed in the VVF. Because all these processes would have taken place beneath Vitim within 10 Myr or less, the basanites and nephelinites would be expected to retain their OIB-like Sr–Nd–Hf isotopic ratios.
Implications for the cause of mantle melting
We have estimated the potential temperature (Tp) of the convecting mantle beneath the VVF in our forward major element models (1480°C; Fig. 19a) and REE inversion models (1450°C; Fig. 20e). A mantle potential temperature of 1450°C is considerably hotter than ambient mantle, which has Tp 1300°C [Thompson & Gibson (2000), based on the calculations for the entropy of melting by Kojitani & Akaogi, (1995)]. It is therefore necessary to suggest a source for the excess heat during the Cenozoic. Three possibilities were put forward by Barry et al. (2003) in relation to the petrogenesis of contemporaneous Mongolian basalts: (1) the Asian continent may have been acting as a thermal blanket, causing the upper mantle to warm up; (2) a large-scale deep mantle plume beneath Asia allowed hot asthenospheric material to reach shallow depths by feeding it into ‘thinspots’ on the base of the lithosphere (Thompson & Gibson, 1991); (3) a small-scale plume was active beneath the Baikal region during the early stages of magmatism but, after this, only the cooling head remained. The teleseismic tomographic data of Petit et al. (1998) strongly support hypothesis (3). They located a relatively narrow (100–200 km diameter) mantle plume (seismically slow) rising from at least 600 km depth beneath the Siberian Craton and Baikal Rift axis. Mantle plumes of similar dimensions have been located beneath parts of the Central European rift system (Granet et al., 1995; Ritter et al., 2001).
Artemieva & Mooney (2001) suggested that the regional base of the TBL beneath Baikal is between 110 and 125 km. The VVF is located above relatively thin crust (35 km thick), which extends in a zone (200 km wide) NE from Lake Baikal (Suvorov et al., 2002). This may correspond to localized thinning of the underlying lithospheric mantle. The VVF is also located above a Mesozoic rift system and reactivation of this may explain the preferential location of volcanism at Vitim, rather than in the axial zone of the Baikal Rift. In addition, there are many faults in the VVF (Fig. 2) that may have aided uprise of magmas in this part of the rift zone.
Figure 22 illustrates our main conclusions and their implications. The concept of mantle plume upwelling and outflow beneath the BRZ is taken from fig. 9 of Petit et al. (1998). Mantle xenolith studies show that the geothermal gradient changed during the later Cenozoic beneath the VVF (Ionov, 2002). The Miocene geotherm is 100°C colder at a given pressure than the Pleistocene geotherm, and this indicates that heating of the lower lithosphere occurred during the late Cenozoic. Our study suggests that this was caused by both lithospheric thinning and a mantle plume, rather than lithospheric extension alone.
|
|
CONCLUSIONS |
|---|
Extension-related magmatism in the VVF occurred during the Cenozoic, near the boundary between the eastern margin of the Siberian craton and the Sayan–Baikal fold belt. The VVF consists of
5000 km3 of melanephelinite, nephelinite, basanite, alkali basalt and tholeiitic basalt lavas. The nephelinites generally occur towards the top of the lava pile and represent a relatively small volume of the overall volcanic succession. A comparison between the CIPW normative compositions of the VVF magmas and experimental studies of basalts suggests that their parental magmas had undergone polybaric fractional crystallization in the sub-Vitim crust prior to eruption.
All of the magmas have similar 87Sr/86Sri (0·704–0·705), 143Nd/144Ndi (0·5127–0·5129) and 176Hf/177Hfi ratios (0·2829–0·2830). This suggests that their parental melts were derived from the convecting mantle and/or the recently enriched base of the lithospheric mantle. Major and trace element abundances and Sr, Nd and Hf-isotope systematics, combined with geochemical modelling, suggest that the source for the melanephelinitic, nephelinitic and basanitic magmas is predominantly the sub-Vitim lithospheric mantle. The estimated composition of the primary silica-undersaturated melts corresponds to those generated in partial melting experiments of garnet pyroxenite between 20 and 25 kbar. Ba/Sr ratios combined with relative depletions in K on normalized multi-element plots suggest that phlogopite was a residual phase in the melanephelinite, nephelinite and basanite mantle source. We envisage that melting occurred at the base of the MBL and/or top of the TBL. The precise nature of this mantle source awaits the isotopic study of pyroxenite xenoliths from the VVF. The results of our geochemical modelling agree well with geophysical estimates for the thickness of the MBL (100 km; Burov et al., 1994) and the TBL (110–125 km; Artemieva & Mooney, 2001) beneath the eastern flank of the Baikal Rift. In support of this, no mantle xenoliths from the VVF have been found to come from depths greater than 100 km (Ionov et al., 1993; Litasov et al., 2000).
Both forward major element and REE inversion models indicate that the VVF alkali and tholeiitic basalts are the product of larger degrees of adiabatic decompression melting (up to 7%) of a fertile peridotite source at between 115 and 85 km depth. The high mantle potential temperatures (1450°C) that we calculated suggest that melting occurred in the convecting mantle. We find it difficult to explain the presence of such anomalously hot mantle beneath the BRZ without invoking a mantle plume, and this concept is supported by the teleseismic tomographic data of Petit et al. (1998).
The location of melt generation beneath the VVF may have been influenced by the relatively thin underlying lithosphere, caused by reactivation of Mesozoic rift structures. A deep fault or faults may have aided the uprise of magmas beneath Vitim. The general lack of volcanism beneath the Baikal Rift could be due to the thicker crust than at Vitim; some magma may have underplated the axial zone of the rift at Moho depths [as Petit et al. (1998) suggested, based on seismic evidence]. The thick sediment infill in the rift axis (Logatchev & Zorin, 1987) might also have prevented magma eruption. The widespread contemporaneous magmatic activity, high heat flow, elevated geotherms and uplifted topography in northern Mongolia and the Baikal region are consistent with the presence of a mantle plume and active rifting inferred by our study.
|
SUPPLEMENTARY DATA |
|---|
Supplementary data for this paper are available on Journal of Petrology online.
|
APPENDIX A: COORDINATE SYSTEM USED IN THE VITIM VOLCANIC FIELD |
|---|
The table below shows the original coordinate data, the estimated Universal Transverse Mercators (UTMs) and the corresponding latitude/longitude values for the drill-holes. The location of the drilling region marked ‘???’ is unknown. A value of +111° was used for the Central Meridian, which places the Vitim region in zone 49 U of the UTM grid (Snyder, 1987
). Numbers in bold type are those that have been changed from the original data.
|
|
APPENDIX B: MODEL PARAMETERS AND EQUATIONS FOR THE ‘HYBRID’ MELTING MODEL |
|---|
|
|
Definitions of terms
- CL Weight concentration of a trace element in the liquid
Average weight concentration of a trace element in a mixed melt
- C0 Weight concentration of a trace element in the original unmelted solid
- CS Weight concentration of a trace element in the residual solid after melt extraction
- DRS Bulk distribution coefficient of the residual solids
- D0 Bulk distribution coefficient of the original solids
- F Weight fraction of melt produced during partial melting
|
ACKNOWLEDGEMENTS |
|---|
S. V. Rasskazov, I. V. Ashchepkov and A. V. Ivanov organized and assisted our fieldwork and sample collection in Siberia. We thank William and Mary Knowles for their hospitality in Moscow, and Ivan Mahotkin for his invaluable help with extracting and exporting our samples from Russia. We are also grateful to Ron Hardy, Chris Ottley and Stephen Reed for help with geochemical analyses, and to Dan McKenzie and Paula Smith for their assistance with inversion modelling. Dmitri Ionov and Graham Pearson gave helpful discussions and access to unpublished data. We thank Paula Smith for her perceptive comments on an earlier version of the manuscript. Reviews by Andy Saunders, Robert Trumbull and an anonymous reviewer, together with the editorial comments of Marjorie Wilson, have substantially improved the manuscript. The Royal Society generously funded SAG and RNT for fieldwork in Siberia. This work was supported by NERC studentship GT04/98/46/ES to JSJ, and the Department of Earth Sciences (University of Cambridge), Cambridge Philosophical Society and Clare College, Cambridge. This is Department of Earth Sciences, University of Cambridge contribution no. 8071.
* Corresponding author. Present address: British Antarctic Survey, High Cross, Madingley Road, Cambridge CB3 0ET, UK. Telephone: +44 (0)1223 221313. Fax: +44 (0)1223 361616. E-mail: jsj{at}bas.ac.uk
|
REFERENCES |
|---|
Anderson, D. (1994). Sublithospheric mantle as the source of continental flood basalts: The case against the continental lithosphere and plume head reservoirs. Earth and Planetary Science Letters 123, 269–280.[CrossRef][Web of Science]
Artemieva, I. M. & Mooney, W. D. (2001). Thermal thickness and evolution of Precambrian lithosphere: a global study. Journal of Geophysical Research 106, 16387–16414.[CrossRef][Web of Science]
Ashchepkov, I. V., Litasov, Y. D. & Dobretsov, N. L. (1994). Pyroxenites and composite garnet peridotite xenoliths from picrite–basalt. Vitim Plateau (Transbaikal): implication for thermobarometry and mantle reconstruction. Proceedings of the Vth International Kimberlite Conference, Rio de Janeiro, Brazil 1/A, 455–466.
Balijinnyam, I., Bayasgalan, A., Borisov, B. A., Cisternas, A., Dem'yanovich, M. G., Ganbaatar, L., Kochetkov, V. M., Kurushin, R. A., Molnar, P., Philip, H. & Vashchilov, Y. Y. (1993). Ruptures of Major Earthquakes and Active Deformation in Mongolia and its Surroundings. Geological Society of America, Memoir 181, 62 pp.
Barry, T. L. & Kent, R. W. (1998). Cenozoic magmatism in Mongolia and the origin of central and East Asian basalts. In: Flower, M. J., Chung, S. -L., Lo, C. -H. & Lee, T. -Y. (eds) Mantle Dynamics and Plate Interactions in East Asia. American Geophysical Union, Geodynamics Series 27, 347–364.
Barry, T. L., Saunders, A. D., Kempton, P. D., Windley, B. F., Pringle, M. S., Dorjnamjaa, D. & Saandar, S. (2003). Petrogenesis of Cenozoic basalts from Mongolia: evidence for the role of asthenospheric versus metasomatized lithospheric mantle sources. Journal of Petrology 44, 55–91.
Brey, G. P., Kohler, T. & Nickel, K. G. (1990). Geothermobarometry in 4-phase lherzolites. 1. Experimental results from 10 to 60 kb. Journal of Petrology 31, 1313–1352.
Burov, E. B., Houdry, F., Diament, M. & Déverchère, J. (1994). A broken plate beneath the North Baikal Rift Zone revealed by gravity modelling. Geophysical Research Letters 21, 129–132.[CrossRef][Web of Science]
Delvaux, D., Moeys, R., Stapel, G., Petit, C., Levi, K., Miroshnichenko, A., Ruzhich, V. & San'kov, V. (1997). Paleostress reconstructions and geodynamics of the Baikal region, Central Asia, Part 2. Cenozoic rifting. Tectonophysics 282, 1–38.[CrossRef][Web of Science]
Dowall, D. P., Pearson, D. G. & Nowell, G. M. (2003). Chemical pre-concentration procedures for high-precision analysis of Hf–Nd–Sr isotopes in geological materials by plasma ionisation multi-collector mass spectrometry (PIMMS) techniques. In: Holland, J. G. & Tanner, S. D. (eds) Plasma Source Mass Spectrometry: Applications and Emerging Technologies. Cambridge: Royal Society of Chemistry, pp. 321–337.
Esin, S. V., Ashchepkov, I. V., Ponomarchuk, V. A., Yamamoto, M., Travin, A. V. & Kiselyeva, V. Y. (1995). Petrogenesis of Alkaline Basaltoids from the Vitim Plateau (Baikal Rift Zone). Novosibirsk: UIGGM, Siberian Branch, Russian Academy of Sciences Press, pp. 1–58.
Foley, S. F., Taylor, W. R. & Green, D. H. (1986). The effect of fluorine on phase relationships in the system KAlSiO4–Mg2SiO4–SiO2 at 28 kbar and the solution mechanism of fluorine in silicate melts. Contributions to Mineralogy and Petrology 93, 46–55.[CrossRef][Web of Science]
Gao, S., Davis, P. M., Liu, H., Slack, P. D., Zorin, Y. A., Logatchev, N. A., Kogan, M., Burkholder, P. D. & Meyer, R. P. (1994). Asymmetric upwarp of the asthenosphere beneath the Baikal Rift Zone, Siberia. Journal of Geophysical Research 99, 15319–15330.[CrossRef]
Garcia, M. O., Hulsebosch, T. P. & Rhodes, J. M. (1995). Olivine-rich submarine basalts from the southwest rift zone of Mauna Loa Volcano: implications for magmatic processes and geochemical evolution. In: Rhodes, J. M. & Lockwood, J. P. (eds) Mauna Loa Revealed. Geophysical Monograph, American Geophysical Union 92, 219–239.
Garner, J. S. (2002). Magmatism of the Vitim Volcanic Field, Baikal Rift Zone, Siberia. Ph.D. thesis, University of Cambridge, 231 pp.
Gibson, S. A. (2002). Major element heterogeneity in Archean to Recent mantle plume starting-heads. Earth and Planetary Science Letters 195, 59–74.[CrossRef][Web of Science]
Gibson, S. A., Thompson, R. N., Leat, P. T., Morrison, M. A., Hendry, G. L., Dickin, A. P. & Mitchell, J. G. (1993). Ultrapotassic magmas along the flanks of the Oligo-Miocene Rio Grande Rift, USA: monitors of the zone of lithospheric mantle extension and thinning beneath a continental rift. Journal of Petrology 34, 187–228.
Gibson, S. A., Thompson, R. N. & Dickin, A. P. (2000). Ferropicrites: geochemical evidence for Fe-rich streaks in upwelling mantle plumes. Earth and Planetary Science Letters 174, 355–374.[CrossRef][Web of Science]
Glaser, S. M., Foley, S. F. & Günther, D. (1999). Trace element compositions of minerals in garnet and spinel peridotite xenoliths from the Vitim volcanic field, Transbaikalia, eastern Siberia. Lithos 685, 1–23.
Granet, M., Wilson, M. & Achauer, U. (1995). Imaging a mantle plume beneath the French Massif Central. Earth and Planetary Science Letters 136, 281–296.[CrossRef][Web of Science]
Green, T. H., Blundy, J. D., Adam, J. & Yaxley, G. M. (2000). SIMS determination of trace element partition coefficients between garnet, clinopyroxene and hydrous basaltic liquids at 2–7·5 GPa and 1080–1200°C. Lithos 53, 165–187.[CrossRef][Web of Science]
Grove, T. L. & Bryan, W. B. (1983). Fractionation of pyroxene-phyric MORB at low pressure: an experimental study. Contributions to Mineralogy and Petrology 84, 293–309.[CrossRef][Web of Science]
Hart, S. R. & Davis, K. E. (1978). Nickel partitioning between olivine and silicate melt. Earth and Planetary Science Letters 40, 203–219.[CrossRef][Web of Science]
Hart, S. R., Hauri, E. H., Oschmann, L. A. & Whitehead, J. A. (1992). Mantle plumes and entrainment: isotopic evidence. Science 256, 517–520.
Hawkesworth, C. J., Gallagher, K., Kelley, S., Mantovani, M., Peate, D. W., Regelous, M. & Rogers, N. W. (1992). Paraná magmatism and the opening of the South Atlantic. In: Storey, B. C., Alabaster, T. & Pankhurst, R. J. (eds). Magmatism and the Causes of Continental Break-up. Geological Society, London, Special Publications 68, 221–240.
Herzberg, C. & Zhang, J. (1996). Melting experiments on anhydrous peridotite KLB-1: compositions of magmas in the upper mantle and transition zone. Journal of Geophysical Research 101, 8271–8295.[CrossRef]
Hirose, K. (1997). Partial melt compositions of carbonated peridotite at 3 GPa and genesis of alkali basalt magmas. Geophysical Research Letters 24, 2837–2840.[CrossRef][Web of Science]
Hirose, K. & Kushiro, I. (1993). Partial melting of dry peridotites at high pressures: determination of compositions of melts segregated from peridotite using aggregates of diamond. Earth and Planetary Science Letters 114, 477–489.[CrossRef][Web of Science]
Hirschmann, M. M. (2000). Mantle solidus: experimental constraints and the effects of peridotite composition. Geochemistry, Geophysics, Geosystems 1.
Hirschmann, M. M., Kogiso, T., Baker, M. B. & Stolper, E. M. (2003). Alkalic magmas generated by partial melting of garnet pyroxenite. Geology 31, 481–484.
Hoernle, K. & Schmincke, H.-U. (1993). The role of partial melting in the 15-Ma geochemical evolution of Gran Canaria: a blob model for the Canary hotspot. Journal of Petrology 34, 599–626.
Holloway, J. R., Pan, V. & Gudmundsson, G. (1992). High-pressure fluid-absent melting experiments in the presence of graphite: oxygen fugacity, ferric/ferrous ratio and dissolved CO2. European Journal of Mineralogy 4, 105–114.
Ionov, D. (2002). Mantle structure and rifting processes in the Baikal–Mongolia region: geophysical data and evidence from xenoliths in volcanic rocks. Tectonophysics 351, 41–60.[CrossRef][Web of Science]
Ionov, D. (2004). Chemical variations in peridotite xenoliths from Vitim, Siberia: inferences for REE and Hf behaviour in the garnet-facies upper mantle. Journal of Petrology 45, 343–367.
Ionov, D. A., Ashchepkov, I. V., Stosch, H.-G., Witt-Eickschen, G. & Seck, H. A. (1993). Garnet peridotite xenoliths from the Vitim volcanic field, Baikal region: the nature of the garnet–spinel peridotite transition zone in the continental mantle. Journal of Petrology 34, 1141–1175.
Ionov, D. A., O'Reilly, S. Y. & Ashchepkov, I. V. (1995). Feldspar-bearing lherzolite xenoliths in alkali basalts from Hamar-Daban, southern Baikal region, Russia. Contributions to Mineralogy and Petrology 122, 174–190.[CrossRef][Web of Science]
Jarvis, I. & Jarvis, K. E. (1992). Plasma spectrometry in the earth sciences: techniques, applications and future trends. Chemical Geology 95, 1–33.[CrossRef][Web of Science]
Jordan, T. H., Lerner-Lam, A. L. & Creager, K. C. (1989). Seismic imaging of boundary layers and deep mantle convection. In: Peltier, W. R. (ed.) Mantle Convection: Plate Dynamics and Global Dynamics (The Fluid Mechanics of Astrophysics and Geophysics, Vol. 4). New York: Gordon and Breach, pp. 97–201.
Kawamoto, T. & Holloway, J. R. (1997). Melting temperature and partial melt chemistry of H2O saturated mantle peridotite to 11 gigapascals. Science 276, 240–243.
Kilinc, A., Carmichael, I. S. E., Rivers, M. L. & Sack, R. O. (1983). The ferric/ferrous ratio of natural silicate liquids equilibrated in air. Contributions to Mineralogy and Petrology 83, 136–140.[CrossRef][Web of Science]
Kiselev, A. I. (1987). Volcanism of the Baikal Rift Zone. Tectonophysics 143, 235–244.[CrossRef][Web of Science]
Kiselev, A. I. & Popov, A. M. (1992). Asthenospheric diapir beneath the Baikal Rift: petrological constraints. Tectonophysics 208, 287–295.[CrossRef][Web of Science]
Klemme, S. & O'Neill, H. StC. (2000). The near-solidus transition from garnet lherzolite to spinel lherzolite. Contributions to Mineralogy and Petrology 138, 237–248.[CrossRef][Web of Science]
Kogiso, T., Hirschmann, M. M. & Frost, D. J. (2003). High-pressure partial melting of garnet pyroxenite: possible mafic lithologies in the source of ocean island basalts. Earth and Planetary Science Letters 216, 603–617.[CrossRef][Web of Science]
Kojitani, H. & Akaogi, M. (1995). Measurement of heat of fusion of model basalt in the system Diopsite–Forsterite–Anorthite. Geophysical Research Letters 22, 2329–2332.[CrossRef][Web of Science]
Kostopoulos, D. K. & James, S. D. (1992). Parameterization of the melting regime of the shallow upper mantle and the effects of variable lithospheric stretching on mantle modal stratification and trace-element concentrations in magmas. Journal of Petrology 33, 665–691.
Langmuir, C. H., Klein, E. M. & Plank, T. (1992). Petrological systematics of mid-ocean ridge basalts: constraints on melt generation beneath ocean ridges. In: Morgan, J. P., Blackman, D. K. & Sinton, J. M. (eds). Mantle Flow and Melt Generation at Mid-Ocean Ridges. Geophysical Monograph, American Geophysical Union 71, 183–280.
Larsen, L. M. & Pedersen, A. K. (2000). Processes in high-Mg, high-T magmas: evidence from olivine, chromite and glass in Palaeogene picrites from West Greenland. Journal of Petrology 41, 1071–1098.
LaTourrette, T., Hervig, R. L. & Holloway, J. R. (1995). Trace element partitioning between amphibole, phlogopite, and basanite melt. Earth and Planetary Science Letters 135, 13–30.[CrossRef][Web of Science]
Leat, P. T., Thompson, R. N., Morrison, M. A., Hendry, G. L. & Dickin, A. P. (1988). Compositionally diverse Miocene–Recent rift-related magmatism in northwest Colorado: partial melting and mixing of magmas from three different asthenospheric and lithospheric mantle sources. Journal of Petrology, Special Lithosphere Issue, 351–377.
Le Bas, M. J. (1989). Nephelinitic and basanitic rocks. Journal of Petrology 30, 1299–1312.
Le Maitre, R. W. (2002). Igneous Rocks: a Classification and Glossary of Terms: Recommendations of the International Union of Geological Sciences Subcommission on the Systematics of Igneous Rocks. Cambridge: Cambridge University Press, 236 pp.
Litasov, K. D., Foley, S. F. & Litasov, Y. D. (2000). Magmatic modification and metasomatism of the subcontinental mantle beneath the Vitim volcanic field (East Siberia): evidence from trace element data on pyroxenite and peridotite xenoliths from Miocene picrobasalt. Lithos 54, 83–114.[CrossRef][Web of Science]
Logatchev, N. A. & Florensov, N. A. (1978). The Baikal system of rift valleys. Tectonophysics 45, 1–14.[CrossRef][Web of Science]
Logatchev, N. A. & Zorin, Y. A. (1987). Evidence and causes of the two-stage development of the Baikal Rift. Tectonophysics 143, 225–234.[CrossRef][Web of Science]
Logatchev, N. A. & Zorin, Y. A. (1992). Baikal Rift Zone: structure and geodynamics. Tectonophysics 208, 273–286.[CrossRef][Web of Science]
McDonough, W. F. & Sun, S.-s. (1995). The composition of the Earth. Chemical Geology 120, 223–253.[CrossRef][Web of Science]
McKenzie, D. (1989). Some remarks on the movement of small melt fractions in the mantle. Earth and Planetary Science Letters 95, 53–72.[CrossRef][Web of Science]
McKenzie, D. & Bickle, M. J. (1988). The volume and composition of melt generated by extension of the lithosphere. Journal of Petrology 29, 625–679.
McKenzie, D. & O'Nions, R. K. (1991). Partial melt distributions from inversion of rare earth element concentrations. Journal of Petrology 32, 1021–1091.
McKenzie, D. & O'Nions, R. K. (1995). The source regions of ocean island basalts. Journal of Petrology 36, 133–159.
Mengel, K. & Green, D. H. (1989) Stability of amphibole and phlogopite in metasomatized peridotite under water-saturated and water-undersaturated conditions. In: Ross, J. (ed.) Proceedings of the 4th International Kimberlite Conference. Perth: Geological Society of Australia, pp. 571–581.
Molnar, P. & Tapponnier, P. (1975). Tectonics of Asia: consequences and implications of a continental collision. Science 189, 419–426.
Morimoto, N. (1988). Nomenclature of pyroxenes. American Mineralogist 73, 1123–1133.[Abstract]
Nowell, G. M. & Parrish, R. R. (2001). Simultaneous acquisition of isotope compositions and parent/daughter ratios by non-isotope dilution solution-mode Plasma Ionisation Multi-collector Mass Spectrometry (PIMMS). In: Holland, G. & Tanner, S. D. (eds) Plasma Source Mass Spectrometry: The New Millennium. Royal Society of Chemistry, Special Publication 267, 298–310.
Nowell, G. M., Kempton, P. D., Noble, S. R., Fitton, J. G., Saunders, A. D., Mahoney, J. J. & Taylor, R. N. (1998). High precision Hf isotope measurements of MORB and OIB by thermal ionization mass spectrometry: insights into the depleted mantle. Chemical Geology 149, 211–233.[CrossRef][Web of Science]
Nowell, G. M., Pearson, D. G., Ottley, C. J., Schwieters, J. & Dowall, D. (2003). Long-term performance characteristics of a Plasma Ionisation Multi-Collector Mass Spectrometer (PIMMS): the ThermoFinnigan Neptune. In: Holland, J. G. & Tanner, S. D. (eds) Plasma Source Mass Spectrometry: Applications and Emerging Technologies. Cambridge: Royal Society of Chemistry, pp. 307–320.
Pearson, D. G., Ionov, D., Carlson, R. W. & Shirey, S. B. (1998). Lithospheric evolution in circum-cratonic settings: a Re–Os isotope study of peridotite xenoliths from the Vitim region, Siberia. Mineralogical Magazine 62A, 1147–1148.[CrossRef]
Pearson, D. G., Canil, D. & Shirey, S. B. (2003). Mantle samples included in volcanic rocks: xenoliths and diamonds. In: Turekian, K. K. & Holland, H. D. (eds) Treatise of Geochemistry, Volume 2: The Mantle and Core. Amsterdam: Elsevier.
Petit, C., Koulakov, I. & Déverchère, J. (1998). Velocity structure around the Baikal Rift Zone from teleseismic and local earthquake traveltimes and geodynamic implications. Tectonophysics 296, 125–144.[CrossRef][Web of Science]
Polyansky, O. P. (2002). Dynamic causes for the opening of the Baikal Rift Zone: a numerical modelling approach. Tectonophysics 351, 91–117.[CrossRef][Web of Science]
Potts, P. J., Tindle, A. G. & Webb, P. C. (1992). Geochemical Reference Material Compositions. Caithness: Whittles, 320 pp.
Rasskazov, S. V. (1994). Magmatism related to the eastern Siberia rift system and the geodynamics (Magmatisme associé au rift de Sibérie orientale: implications géodynamiques). Bulletin des Centres de Recherches Exploration–Production Elf-Aquitaine 18, 437–452.
Ritter, J. R. R., Jordan, M., Christensen, U. R. & Achauer, U. (2001). A mantle plume below the Eifel volcanic fields, Germany. Earth and Planetary Science Letters 186, 7–14.[CrossRef][Web of Science]
Roberts, M. A. (2002). The geochemical and volcanological evolution of the Mt. Meru Region, Northern Tanzania. Ph.D. thesis, University of Cambridge, 266 pp.
Roeder, P. L. & Emslie, R. F. (1970). Olivine–liquid equilibria. Contributions to Mineralogy and Petrology 29, 275–289.[CrossRef][Web of Science]
Rollinson, H. R. (1993). Using Geochemical Data: Evaluation, Presentation, Interpretation. Harlow: Longman.
Ruppel, C. (1995). Extensional processes in continental lithosphere. Journal of Geophysical Research 100, 24187–24215.[CrossRef][Web of Science]
Sack, R. O. & Ghiorso, M. S. (1991). Chromian spinels as petrogenetic indicators: thermodynamics and petrological applications. American Mineralogist 76, 827–847.[Abstract]
Sato, K., Katsura, T. & Ito, E. (1997). Phase relations of natural phlogopite with and without enstatite up to 8 GPa: implication for mantle metasomatism. Earth and Planetary Science Letters 146, 511–526.[CrossRef][Web of Science]
Smith, D., Griffin, W. L. & Ryan, C. G. (1993). Compositional evolution of high-temperature sheared lherzolite PHN1611. Geochimica et Cosmochimica Acta 57, 605–613.[CrossRef][Web of Science]
Snyder, J. P. (1987). Map Projections: a Working Manual. US Geological Survey Professional Paper 1395.
Späth, A., Le Roex, A. P. & Opiyo-Akech, N. (2001). Plume–lithosphere interaction and the origin of continental rift-related alkaline volcanism: the Chyulu Hills Volcanic Province, Southern Kenya. Journal of Petrology 42, 765–787.
Suvorov, V. D., Mishenkina, Z. M., Petrick, G. V., Sheludko, I. F., Seleznev, V. S. & Solovyov, V. M. (2002). Structure of the crust in the Baikal Rift Zone and adjacent areas from Deep Seismic Sounding data. Tectonophysics 351, 61–74.[CrossRef][Web of Science]
Takahashi, E. (1986). Melting of a dry peridotite KLB-1 up to 14 GPa: implications on the origin of peridotitic upper mantle. Journal of Geophysical Research 91, 9367–9382.
Takahashi, E. & Kushiro, I. (1983) Melting of a dry peridotite at high pressures and basalt magma genesis. American Mineralogist 68, 859–879.[Abstract]
Taura, H., Yurimoto, H., Kurita, K. & Sueno, S. (1998). Pressure dependence on partition coefficients for trace elements between olivine and the coexisting melts. Physics and Chemistry of Minerals 25, 469–484.[CrossRef][Web of Science]
Taylor, S. R. & McClennan, S. M. (1985). The Continental Crust: Its Composition and Evolution. Oxford: Blackwell, 312 pp.
Thibault, Y., Edgar, A. D. & Lloyd, F. E. (1992). Experimental investigation of melts from a carbonated phlogopite lherzolite: implications for metasomatism in the continental lithosphere. American Mineralogist 77, 784–794.[Abstract]
Thompson, R. N. (1982). Magmatism of the British Tertiary Volcanic Province. Scottish Journal of Geology 18, 49–107.[CrossRef][Web of Science]
Thompson, R. N. (1983). Thermal aspects of the origin of Hebridean Tertiary acid magmas, II. Experimental melting behaviour of the granites at 1 kbar PH2O. Mineralogical Magazine 47, 111–121.[Web of Science]
Thompson, R. N. & Gibson, S. A. (1991). Subcontinental mantle plumes, hotspots and pre-existing thinspots. Journal of the Geological Society, London 148, 973–977.
Thompson, R. N. & Gibson, S. A. (2000). Transient high temperatures in mantle plume heads inferred from magnesian olivines in Phanerozoic picrites. Nature 407, 502–506.[CrossRef][Medline]
Thompson, R. N., Leat, P. T., Morrison, M. A., Hendry, G. L., Dickin, A. P. & Gibson, S. A. (1990). Strongly potassic mafic magmas from lithospheric mantle sources during continental extension and heating: evidence from Miocene minettes of northwest Colorado, USA. Earth and Planetary Science Letters 38, 139–153.
Thompson, R. N., Gibson, S. A., Dickin, A. P. & Smith, P. M. (2001). Early Cretaceous basalt and picrite dykes of the Southern Etendeka Region, NW Namibia: windows into the role of the Tristan mantle plume in Paraná–Etendeka Magmatism. Journal of Petrology 42, 2049–2081.
Thompson, R. N., Ottley, C. J., Smith, P. M., Pearson, D. G., Dickin, A.- P., Leat, P. T., Morrison, M. A. & Gibson, S. A. (2005). Source of the Quaternary alkalic basalts, picrites and basanites of the Potrillo Volcanic Field, New Mexico, USA—Lithosphere or conrecting mantle? Journal of Petrology (in press).
Tsuruta, K. & Takahashi, E. (1998). Melting study of an alkali basalt JB-1 up to 12·5 GPa: behaviour of potassium in the deep mantle. Physics of the Earth and Planetary Interiors 107, 199–130.[CrossRef]
Turner, S. & Hawkesworth, C. (1995). The nature of the sub-continental mantle: constraints from the major-element composition of continental flood basalts. Chemical Geology 120, 295–314.[CrossRef][Web of Science]
Vervoort, J. D., Patchett, P. J., Blichert-Toft, J. & Albarède, F. (1999). Relationships between Lu–Hf and Sm–Nd isotopic systems in the global sedimentary system. Earth and Planetary Science Letters 168, 79–99.[CrossRef][Web of Science]
Wallace, M. E. & Green, D. H. (1988). An experimental determination of primary carbonatite magma composition. Nature 335, 343–346.[CrossRef]
Wang, W. & Takahashi, E. (2000). Subsolidus and melting experiments of K-doped peridotite KLB-1 to 27 GPa: its geophysical and geochemical implications. Journal of Geophysical Research 105, 2855–2868.[CrossRef]
White, R. S. (1988). The Earth's crust and lithosphere. Journal of Petrology, Special Lithosphere Issue, 1–10.
White, R. S. & McKenzie, D. (1995). Mantle plumes and flood basalts. Journal of Geophysical Research 100, 17543–17585.[CrossRef]
White, R. S., McKenzie, D. & O'Nions, R. K. (1992). Oceanic crustal thickness from seismic measurements and rare earth element inversions. Journal of Geophysical Research 97, 19683–19715.[CrossRef]
Wilson, M. (1989). Igneous Petrogenesis. London: Chapman & Hall, 466 pp.
Wilson, M., Rosenbaum, J. M. & Dunworth, A. D. (1995). Melilitites: partial melts of the thermal boundary layer? Contributions to Mineralogy and Petrology 119, 181–196.[Web of Science]
Windley, B. F. & Allen, M. B. (1993). Mongolian plateau: evidence for a late Cenozoic mantle plume under central Asia. Geology 21, 295–298.
Wood, D. A. (1979). Dynamic partial melting: its application to the petrogenesis of basalts erupted in Iceland, the Faeroe Islands, the Isle of Skye (Scotland) and the Troodos Massif (Cyprus). Geochimica et Cosmochimica Acta 43, 1031–1046.[CrossRef][Web of Science]
Yaxley, G. M. & Green, D. H. (1998). Reactions between eclogite and peridotite: mantle refertilisation by subduction of oceanic crust. Schweizerische Mineralogische und Petrographische Mitteilungen 78, 243–255.[Web of Science]
Zhang, Y.-S. (1998). Three dimensional upper mantle structure beneath east Asia and its tectonic implications. In: Flower, M. J., Chung, S.-L., Lo, C.-H. & Lee, T.-Y. (eds) Mantle Dynamics and Plate Interactions in East Asia. American Geophysical Union, Geodynamic Series 27, 11–24.
Zhang, J. & Herzberg, C. (1994). Melting experiments on anhydrous peridotite KLB-1 from 5·0 to 22·5 GPa. Journal of Geophysical Research 99, 17729–17742.[CrossRef]
Zorin, Y. A. (1981). The Baikal Rift: an example of the intrusion of asthenospheric material into the lithosphere as a cause of disruption of the lithospheric plates. Tectonophysics 73, 91–104.[CrossRef][Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J.H. Davies and H.-P. Bunge Are splash plumes the origin of minor hotspots? Geology, May 1, 2006; 34(5): 349 - 352. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||