Journal of Petrology Advance Access originally published online on October 16, 2007
Journal of Petrology 2007 48(12):2235-2260; doi:10.1093/petrology/egm058
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Geochemistry and Evolution of Subcontinental Lithospheric Mantle in Central Europe: Evidence from Peridotite Xenoliths of the Kozákov Volcano, Czech Republic
Ackerman1,2,*
1
1Institute of Geochemistry, Mineralogy and Mineral Resources, Faculty of Science, Charles University, Albertov 6, 128 43, PRAHA 2, Czech republic
2Institute of Geology v.v.i., Academy of Sciences of the Czech Republic, Rozvojová 269, 165 00, PRAHA 6, Czech Republic
3Department Of Geology and Geophysics, University of Wisconsin–Madison, WI 53706, USA
4Laboratories of the Geological Institutes, Faculty of Science, Charles University, Albertov 6, 128 43, Praha 2, Czech Republic
RECEIVED FEBRUARY 12, 2007; ACCEPTED SEPTEMBER 7, 2007
 |
ABSTRACT |
|---|
Neogene basanite lavas of Kozákov volcano, located along the Lusatian fault in the northeastern Czech Republic, contain abundant anhydrous spinel lherzolite xenoliths that provide an exceptionally continuous sampling of the upper two-thirds of central European lithospheric mantle. The xenoliths yield a range of two-pyroxene equilibration temperatures from 680°C to 1070°C, and are estimated to originate from depths of 32–70 km, based on a tectonothermal model for basaltic underplating associated with Neogene rifting. The sub-Kozákov mantle is layered, consisting of an equigranular upper layer (32–43 km), a protogranular intermediate layer that contains spinel–pyroxene symplectites after garnet (43–67 km), and an equigranular lower layer (67–70 km). Negative correlations of wt % TiO2, Al2O3, and CaO with MgO and clinopyroxene mode with Cr-number in the lherzolites record the effects of partial fusion and melt extraction; Y and Yb contents of clinopyroxene and the Cr-number in spinel indicate
5 to 15% partial melting. Subsequent metasomatism of a depleted lherzolite protolith, probably by a silicate melt, produced enrichments in the large ion lithophile elements, light rare earth elements and high field strength elements, and positive anomalies in primitive mantle normalized trace element patterns for P, Zr, and Hf. Although there are slight geochemical discontinuities at the boundaries between the three textural layers of mantle, there tends to be an overall decrease in the degree of depletion with depth, accompanied by a decrease in the magnitude of metasomatism. Clinopyroxene separates from the intermediate protogranular layer and the lower equigranular layer yield 143Nd/144Nd values of 0·51287–0·51307 (
Nd = +4·6 to +8·4) and 87Sr/86Sr values of 0·70328–0·70339. Such values are intermediate with respect to the Nd–Sr isotopic array defined by anhydrous spinel peridotite xenoliths from central Europe and are similar to those associated with the present-day low-velocity anomaly in the upper mantle beneath Europe. The geochemical characteristics of the central European lithospheric mantle reflect a complex evolution related to Devonian to Early Carboniferous plate convergence, accretion, and crustal thickening, Late Carboniferous to Permian extension and gravitational collapse, and Neogene rifting, lithospheric thinning, and magmatism. KEY WORDS: xenoliths; lithospheric mantle; REE–LILE–HFSE; Sr–Nd isotopes; Bohemian Massif
|
INTRODUCTION |
|---|
Studies of ultramafic xenoliths exhumed within the volcanic centers of the Cenozoic European Rift System (CERS) have provided substantial information regarding the physical and chemical characteristics of the subcontinental lithospheric mantle (SCLM) (Menzies & Bodinier, 1993
; Downes, 2001). Such studies have established a general pattern of incompatible element depletion in the peridotites through removal of basaltic melts, followed locally by cryptic or modal metasomatism via interaction with transient melts or fluids, the details of which vary between xenolith localities (Downes, 2001, and references therein).
The Pliocene Kozákov volcano in the Czech Republic is one of a number of eruptive centers located along the Ohre (Eger) Graben in the central European part of the CERS (Fig. 1). Spinel lherzolite xenoliths from Kozákov volcano yield a continuous range of equilibration temperatures from 680°C to 1065°C and have been estimated to originate from depths of 32–70 km, corresponding to the upper two-thirds of the SCLM in this region (Christensen et al., 2001). This suite of spinel lherzolite xenoliths from a single eruptive site provides a rare opportunity to evaluate the depth variation in physical and chemical characteristics of the SCLM in central Europe, similar to that provided for western Europe by the xenolith suite from the Ray Pic volcano in the French Massif Central (Zangana et al., 1997, 1999).
|
The Kozákov xenoliths were first described by Farsk
(1876), several bulk-rock and mineral analyses were reported by Fediuk (1971) and Vokurka & Povondra (1983), detailed geothermometry calculations were performed on several samples by Medaris et al. (1999), seismic properties were calculated from olivine petrofabrics by Christensen et al. (2001), and major element and rare earth element (REE) compositions for several bulk-rocks and clinopyroxene separates were determined by Kone
n et al. (2006). In this investigation we examine the depth variation in chemical composition of the Kozákov sample suite, including major and trace elements for whole-rocks, major elements for the constituent minerals, and trace elements and Nd and Sr isotopes for clinopyroxene separates. As for many spinel lherzolite xenolith suites elsewhere, the data indicate the decoupling of major and trace elements caused by melt extraction during partial fusion (e.g. a decrease in CaO and Al2O3) and subsequent metasomatism [e.g. an increase in light REE (LREE), large ion lithophile elements (LILE) and high field strength elements (HFSE)]. In the Kozákov xenolith suite, the degree of depletion tends to decrease with depth, which is a general phenomenon in the SCLM (Gaul et al., 2003), whereas in contrast the pattern of metasomatism is distinctive for the different textural types of xenoliths. |
LOCALITY AND GEOLOGICAL SETTING |
|---|
The central European lithosphere is a tectonic collage, resulting from the Devonian convergence and Carboniferous collision of Laurussia, Gondwana, and intervening continental and oceanic microplates, which juxtaposed disparate lithospheric fragments of various ages and provenance (Franke, 2000
; Matte, 2001). This collage assembly is reflected in Kossmat's (1927) prescient division of central Europe into the Moldanubian, Saxothuringian, and Rhenohercynian zones (Fig. 1), each of which itself is a composite of terranes. A collage assembly is also recorded by the lithospheric mantle, which exhibits divergently dipping anisotropic structures in the Moldanubian and Saxothuringian zones (Babuka & Plomerová, 2001; Plomerová et al., 2005).
Superimposed on this lithospheric collage is the CERS, which evolved in the Alpine foreland during late Eocene to Recent times (Fig. 1; Wilson & Downes, 1991; Ziegler, 1992). The distribution of grabens in the rift system is controlled by Cenozoic tensional reactivation of basement fracture systems, most of which originated during the late stages of the Middle to Late Paleozoic Variscan orogeny.
Late Cretaceous to Pleistocene intraplate volcanism in the Bohemian Massif is concentrated along the Ohre (Eger) Graben, which developed in late Eocene to sub-Recent time (Kopeck, 1986), and along the Labe Tectono-Volcanic Zone, which coincides approximately with the Lusatian (Luzice) fault (Fig. 1). Four episodes of alkaline magmatism are recognized in this region, one prior to development of the Ohre rift, and three related to rifting (Ulrych et al., 1999). The prerift volcanic rocks were erupted from late Cretaceous to middle Eocene times (79–49 Ma), whereas the synrift volcanic rocks were erupted during the late Eocene to early Miocene (43–16 Ma), middle Miocene to late Miocene (13–9 Ma), and Plio-Pleistocene (6–0·26 Ma).
Three major lithospheric blocks are juxtaposed along the Ohre Graben and Lusatian fault: the Saxothuringicum on the north, Bohemicum on the south and Lugicum on the east (Fig. 1). The Saxothuringicum in the vicinity of the rift consists of a polymetamorphic complex, locally containing eclogite and garnet peridotite, and late Variscan granitoids; the Bohemicum is composed of an anchimetamorphic Proterozoic basement and Lower Cambrian to Middle Devonian sedimentary cover; and the Lugicum is a complex mosaic, predominantly of metamorphic rocks, locally including eclogite and garnet peridotite, subordinate Cambrian to Carboniferous sedimentary rocks, and extensive Variscan granitoids. In addition to crustal distinctions between the three lithospheric blocks, the orientations of anisotropic structures in lithospheric mantle are different in each (Plomerová et al., 2005).
Kozákov volcano is situated along the Lusatian fault system (Fig. 1) about 45 km from the Litom
ice deep fault, which bounds the Ohre Graben on the south, and presumably taps mantle along the Lugicum–Bohemicum boundary. The crust diminishes in thickness from 40 km beneath the central Bohemian Massif to 31 km beneath the Ohre Graben and 32 km beneath Kozákov (ermák et al., 1991). Seismic velocity–depth profiles for crust in the Bohemicum and Lugicum blocks on either side of Kozákov reveal a relatively high-velocity layer (6·9 km/s) at the base of the crust (ermák, 1989) that may represent the crystallized products of underplated magma. The lithosphere also decreases in thickness from 140 km in the central Bohemian Massif to 90 km beneath the Ohre rift (ermák et al., 1991; Babuka & Plomerová, 1992).
|
KOZÁKOV XENOLITHS |
|---|
Mantle xenoliths were collected at Kozákov volcano from three quarries: Chuchelna, Slap, and Smrcí. Exposed in each of the quarries are two Early Pliocene nepheline basanite lava flows, whose ages are 3·95 Ma (upper flow) and 4·14 Ma (lower flow) (
ibrava & Havlíek, 1980). The sample suite was collected from the lower flow, which is notable for containing the highest concentration and largest size of mantle xenoliths among the many xenolith-bearing alkali basalts of the Bohemian Massif.
Lithology
The nepheline basanite lava flow at Kozákov contains abundant mantle xenoliths and rare, lower crustal xenoliths of olivine gabbronorite (Fediuk, 1971). Mantle xenoliths make up 2–3% of the lava flow, and olivine xenocrysts account for another 7–8%. Peridotite xenoliths are commonly 6–10 cm in diameter and rarely up to 70 cm; larger xenoliths are spheroidal to ellipsoidal in shape, and smaller ones tend to be subangular. The Kozákov mantle suite is anhydrous and consists of variable proportions of olivine, orthopyroxene, clinopyroxene and spinel. Spinel lherzolite is the predominant rock type in the mantle suite and is the focus of this investigation. Also found in the mantle suite are subordinate amounts of harzburgite and dunite, and, rarely, websterite, olivine clinopyroxenite, clinopyroxenite and orthopyroxenite (Fediuk, 1994).
Texture
Two principal textural varieties of lherzolite occur at Kozákov [following the classification scheme of Mercier & Nicolas (1975)]: medium-grained equigranular lherzolite, which contains discrete, intergranular spinel (Fig. 2a–c, g and h), and coarse- to very coarse-grained protogranular lherzolite, in which spinel occurs only in symplectic intergrowth with orthopyroxene and clinopyroxene (Fig. 2d–f). In both textural types small amounts of very fine-grained plagioclase, clinopyroxene and Al-rich spinel occur locally at spinel–pyroxene boundaries, as a result of incipient partial melting and subsequent quenching during eruption. No samples with porphyroclastic texture have been found so far at Kozákov.
|
In equigranular lherzolite, mineral grains are 1–4 mm in diameter, mosaic and triple-junction grain boundaries are common, and spinel occurs as discrete, dispersed grains (Fig. 2a, c, g and h). Most samples have a weak foliation and are devoid of phase layering, except for sample 95KZS4 (36 km), which displays prominent spinel layering and is a dunite, rather than a lherzolite (Fig. 2b).
Protogranular lherzolite is characterized by large grains (in some cases up to 2 cm in diameter), curvilinear grain boundaries, and prominent domains of spinel–pyroxene symplectite (Fig. 2d–f). Based on the reconstructed bulk chemical composition of the symplectites, Medaris et al. (1997) demonstrated that the symplectite represents the product of reaction between pre-existing garnet and matrix olivine.
A previous investigation by Christensen et al. (2001) indicated that the sub-Kozákov lithosphere is layered, consisting of a lower temperature equigranular layer at a depth of 32–43 km, an intermediate protogranular layer at 43–67 km, and a higher temperature equigranular layer below 67 km. This layered structure is thought to be inherited from Variscan orogenesis, during which garnet peridotite was tectonically injected into spinel peridotite. Subsequently, Neogene underplating and heating of the upper lithospheric mantle promoted recrystallization of the metastable garnet peridotite to spinel peridotite, in which the former presence of garnet is indicated by prominent spinel–pyroxene symplectites (Medaris et al., 1997).
|
ANALYTICAL METHODS |
|---|
Rock samples were crushed manually and then powdered using an agate mortar. Whole-rock major element analyses (wet chemistry technique) and trace-element analyses (by inductively coupled plasma mass spectrometry; ICP-MS) were performed at the Faculty of Science, Charles University, Prague. Replicate analyses of international reference whole-rock material (PCC-1; USGS) by the wet chemistry technique yield an average error (1
) for whole-rock analyses of ±5% (Table 1). Trace element ICP-MS analysis followed the methods of Strnad et al. (2005), and analysis of international peridotite reference material UB-N (CNRS) yields an average precision better than 11% for all corresponding elements (Table 1) with respect to recommended values (Govindaraju, 1989).
|
Clinopyroxene separates were obtained by hand-picking under a binocular microscope and subsequently acid-leached in hot HCl. The REE were extracted using ion exchange columns and then analyzed by ICP-MS.
Analyses of major elements in minerals from 21 xenoliths, which were utilized, but not tabulated, by Christensen et al. (2001), were obtained by Medaris by electron microprobe analysis (EMPA) using a CAMECA SX 50 system at the University of Wisconsin–Madison. Analytical conditions were an accelerating voltage of 15 kV, a beam current of 20 nA, and a beam diameter of 2 µm. Synthetic and natural minerals were used as internal standards for corresponding elements, and data reduction was performed using the Phi–rho–z program of Armstrong (1988). Minerals in two additional samples were analyzed by Ackerman by EMPA using a CAMECA SX 100 system equipped with a wavelength-dispersive spectrometry (WDS) analyzer at the Institute of Geology, Academy of Sciences of the Czech Republic, Prague. Analytical conditions were 15 kV accelerating voltage, 10 nA beam current and 2 µm beam diameter. Synthetic and natural minerals were used as standards, and data reduction was performed using the Merlet data reduction program (Merlet, 1994).
Clinopyroxene separates for Nd and Sr isotope analysis were prepared as described by Beard et al. (1992). Sample sizes ranged from 20 to 100 mg, and samples were spiked with Rb–Sr and Sm–Nd tracers for concentration and isotopic analyses prior to dissolution. Sample dissolution, chemical, and mass analysis procedures follow those of Johnson & Thompson (1991); all chemical separations and mass analyses were performed in the Radiogenic Isotope Laboratory at the University of Wisconsin–Madison. Strontium isotope compositions were measured by thermal ionization mass spectrometry (TIMS) with a Micromass Sector 54 instrument using a three-jump dynamic multi-collector analysis; 87Sr/86Sr isotope ratios were exponentially normalized to 86Sr/88Sr = 0·1194. Using this analysis method, the measured 87Sr/86Sr of NBS-987 was 0·710268 ± 14 (2 SD, n = 9) during the course of this study. Laboratory blanks were typically 350 pg for Sr and <150 pg for Rb, which are negligible. Neodymium was analyzed as NdO+ using single Re filaments and silica gel and phosphoric acid as the oxygen source, and 18O/16O and 17O/16O ratios of 0·002110 and 0·000387, respectively, were used to correct the data. Mass analysis was performed by TIMS using a Sector 54 instrument via a three-jump multi-collector dynamic analysis and a power-law normalization to 146Nd/144Nd = 0·7219. During the course of this study, the measured 143Nd/144Nd of the BCR-1 USGS rock standard was 0·512643 ± 9 (2 SD, n = 2), and this is taken to be equal to present-day CHUR (e.g. Jacobsen & Wasserburg, 1980). Laboratory blanks were 150 pg for Nd and <40 pg for Sm, which are negligible.
|
GEOTHERMOMETRY, THERMAL HISTORY, AND DEPTH ESTIMATES |
|---|
 TOPABSTRACTINTRODUCTIONLOCALITY AND GEOLOGICAL SETTINGKOZAKOV XENOLITHSANALYTICAL METHODS GEOTHERMOMETRY, THERMAL HISTORY,... ROCK COMPOSITIONSMINERAL COMPOSITIONSSR AND ND ISOTOPES...DISCUSSIONCONCLUSIONSREFERENCES |
|---|
The mineral compositions originally determined by Medaris were discussed, but not tabulated, by Christensen et al. (2001
); these compositions and our new mineral analyses are given in Table 2. Temperatures were calculated for several domains in each sample using three calibrations of the two-pyroxene geothermometer (Bertrand & Mercier, 1985; Brey & Köhler, 1990; Taylor, 1998), the Al-in-orthopyroxene geothermometer (Witt-Eickschen & Seck, 1991), and the olivine–spinel Mg–Fe2+ exchange geothermometer (Ballhaus et al., 1991). In general, there is good agreement between the different methods (Table 3), reflecting equilibrium distribution of most elements between coexisting olivine, orthopyroxene, clinopyroxene, and spinel at the time of exhumation in the host basalt [see Medaris et al. (1999) for a detailed comparison of results].
|
|
In particular, results from the three versions of the two-pyroxene geothermometer are in good agreement (Fig. 3a). Bertrand & Mercier (1985
) temperatures for the entire peridotite xenolith suite range continuously from 685°C to 1065°C (at P = 15 kbar), but the two different textural types of xenolith fall into three groups according to temperature (Fig. 3a): low-temperature equigranular (685–790°C), medium-temperature protogranular (835–1045°C), and high-temperature equigranular (1035–1065°C). It is unlikely that this temperature grouping is due to the effect of grain size on blocking temperatures, because some medium-grained equigranular samples yield higher temperatures than coarse-grained protogranular samples. The 10°C overlap in temperature estimates for the protogranular and high-temperature equigranular groups is within the estimated precision of the two-pyroxene geothermometers (20°C) and may not reflect a true overlap in temperatures for the two groups.
|
In the absence of a viable geobarometer for spinel peridotites, extraction depths for a suite of spinel peridotite xenoliths may be estimated from the intersection of temperature determined using an appropriate geothermometer with the estimated geotherm at the time of exhumation. Following this procedure, depths for the Kozákov mantle xenoliths have been estimated by combining temperatures from the Bertrand & Mercier (1985
) two-pyroxene geothermometer (taking into account the pressure dependence of the geothermometer) with a model geotherm at 5 Ma, based on a magmatic underplating scenario, whose thermal evolution is summarized in Fig. 4 and the details of which have been given by Christensen et al. (2001). Apart from the choice of boundary conditions for the underplating model, a precision of ±20°C in the two-pyroxene geothermometer translates to an uncertainty of ±4 km at a depth of 50 km. For the purposes of plotting, however, depth estimates for each sample are cited to the nearest kilometre (Table 3).
|
This approach suggests that the Kozákov xenoliths were derived from depths ranging from 32 km, near the crust–mantle boundary, to 70 km, representing
65% of the lithospheric mantle. In addition, it appears that the sub-Kozákov lithosphere is layered, consisting of an upper equigranular layer from 32 to 43 km, an intermediate protogranular layer from 43 to 67 km, and a lower equigranular layer below 67 km (Fig. 4), as previously noted by Christensen et al. (2001). The approximation of the boundary at 67 km arises from the overlap in estimated temperature for the highest temperature protogranular sample and the lowest temperature equigranular sample from the lower layer. The depth estimates for these samples are consistent with the presence of spinel as the principal aluminous phase in the xenoliths, rather than garnet, as predicted from the measured spinel compositions and experimental determination of the spinel–garnet phase transition (O’Neill, 1981).
Two petrological lines of evidence support the veracity of the underplating model for Kozákov and its use in estimating depths. The model predicts that shallow mantle will experience greater heating and faster cooling than the deeper mantle. These phenomena are reflected by the presence of thin exsolution lamellae in pyroxene in the low-temperature equigranular samples, but their absence in higher temperature protogranular and equigranular samples. Further support is provided by results from the Al-in-orthopyroxene (Witt-Eickschen & Seck, 1991) and two-pyroxene (Bertrand & Mercier, 1985) geothermometers, which are comparable above 950°C, but increasingly diverge at lower temperatures (Fig. 3b). Such divergence is probably due to the difference in blocking temperatures for Al (slower diffusion) and Ca–Mg (faster diffusion) in shallow, rapidly cooled pyroxene, compared with deeper, more slowly cooled pyroxene.
|
ROCK COMPOSITIONS |
|---|
Modes
Modal compositions (Fig. 5, Table 4) of the Kozákov mantle xenoliths were determined by mass-balance calculations from the major element compositions of the whole-rocks (Table 4) and the constituent minerals (Table 2), using the least-squares inversion method of Albarède (1995
). Of 14 analyzed rock samples, 12 are lherzolite, including eight protogranular samples from the intermediate layer and four equigranular samples from the lower layer (Fig. 5). The equigranular sample from the upper layer (33 km) plots on the boundary between harzburgite and lherzolite, and the layered equigranular sample (36 km) is dunite, which contains a larger proportion of clinopyroxene than orthopyroxene and has a high spinel content (7%). Although these calculated modes are generally representative, they may not be very precise, because of the difficulty in obtaining fully representative whole-rock analyses of relatively coarse-grained, but small-sized samples.
|
|
Major elements
With the exception of the equigranular dunite (at 36 km), the analyzed Kozákov xenoliths (Table 4) are typical of depleted lithospheric mantle, with high Mg-numbers (88·8–91· 0), relatively high Cr2O3 contents (0·38–1·37 wt %), and relatively low TiO2 contents (0·03–0·17 wt %). In most samples the concentrations of SiO2, TiO2, Al2O3, and CaO exhibit negative correlations with MgO (Fig. 6), which are similar to those described from other localities in the CERS (Downes, 2001
) and elsewhere (Griffin et al., 1998; Yaxley et al., 1991). Such negative correlations are commonly attributed to depletion of incompatible elements in the lithospheric mantle during partial fusion and melt extraction. The negative correlations of clinopyroxene mode with whole-rock Cr-number (Fig. 6) is also consistent with partial fusion and melt extraction.
|
The equigranular dunite (at 36 km) differs from the other samples in having a lower Mg-number (86·3), lower SiO2, and higher FeO (Fig. 6), and by a higher spinel content (7·2%) and higher proportion of clinopyroxene relative to orthopyroxene (5·6% vs 3·2%). These features and the prominent layering of spinel suggest that the composition of this sample may largely reflect cumulate, rather than partial melting, processes.
One protogranular sample (ORKZS6; 60·1 km) is anomalous for containing less MgO and more Al2O3 than primitive mantle (Fig. 6). It is likely that the analyzed split of this coarse-grained and small-sized sample contained a high, and non-representative, content of spinel, which resulted in its discrepant MgO and Al2O3 contents and high calculated spinel mode (7%).
Trace elements
All of the Kozákov mantle xenoliths are depleted in the heavy REE (HREE) relative to primitive mantle, but are enriched in LREE relative to HREE (Table 4; Fig. 7). Such patterns are characteristic for cryptic metasomatism subsequent to partial fusion and melt extraction. The protogranular xenoliths tend to have a concave-upward pattern, which is more pronounced in those samples with lower Yb and Lu contents. The equigranular xenolith from the upper layer (at 33 km), which is low in Lu and Yb, also has a pronounced concave-upward pattern. Equigranular xenoliths from the lower layer tend to have relatively flat patterns from Gd to Lu and marked LREE enrichment. The layered equigranular dunite from the upper layer (at 36 km) is distinct from the other samples in having a flat middle REE (MREE) and HREE pattern (Sm to Lu) and slight enrichment in LREE. Two protogranular xenoliths have marked positive Eu anomalies.
|
In addition to LREE enrichment, Kozákov peridotite xenoliths are enriched in the LILE and HFSE and have positive P, Zr, and Hf anomalies (Fig. 8), the magnitudes of which are negatively correlated with Yb contents. The upper equigranular sample also has a pronounced positive anomaly in Nb and Ti. Normalized U/Th ratios show a wide range in values (0·2–6·8), which are negatively correlated with normalized values of Th (16·5–0·9).
|
Compositional variation with depth
Disregarding the chemically distinct equigranular dunite (95KSZ4), plots of various chemical parameters with depth reveal subtle variations in the xenolith suite, despite appreciable overlap in the compositions of samples from the upper equigranular, protogranular, and lower equigranular layers. Regarding major elements (Fig. 9), there is a general tendency across the entire xenolith suite for whole-rock Mg-number and Cr-number to decrease with increasing depth and CaO (wt %) and modal clinopyroxene to increase. The single sample from the upper equigranular layer has the most depleted composition. Protogranular samples show the clearest trends of compositional variation with depth, and samples from the lower equigranular layer (with one exception) tend to be the least depleted. These patterns of compositional variation suggest that the degree of depletion in Kozákov lithospheric mantle decreases with depth, a feature that is also found in mantle xenoliths in southeastern Australia (Gaul et al., 2003
).
|
The patterns of trace element variations with depth (Fig. 10) are consistent with those for major elements. Yb contents are lowest in samples above 50 km and, with two exceptions, are highest in samples below 50 km. The degree of LREE enrichment, as measured by (Ce/Tb)N, tends to decrease with increasing depth. The magnitudes of Nb and Hf anomalies, represented by (Nb/La)N and (Hf/Sm)N, are appreciable in the shallow samples, but decrease with increasing depth in concert with the increase in Yb contents. The clearest trends in trace element variations with depth are seen within the protogranular layer, as is the case for major elements.
|
|
MINERAL COMPOSITIONS |
|---|
Major elements
Mineralogically, the Kozákov mantle xenoliths consist of variable proportions of magnesian olivine and orthopyroxene, Cr-diopside, and aluminous spinel, the compositions of which are summarized in Table 2. Each mineral species is homogeneous within a given xenolith, except for orthopyroxene and clinopyroxene in the upper equigranular layer, which contain thin exsolution lamellae of the complementary pyroxene and spinel. No significant differences in composition were found between cores and rims of grains, except locally in the vicinity of domains where incipient partial melting has occurred.
The Mg-number for olivine in the entire xenolith suite ranges from 90·2 to 91·7, except for that in the layered equigranular dunite, whose Mg-number is 88·2. The Mg-number is highest in the upper equigranular samples (91·1–91·7), shows the widest variation in the protogranular samples (90·2–91·4), and is relatively low in the lower equigranular samples (90·4–91·1). The NiO contents of olivine largely overlap between samples from the different layers, and the variation within the protogranular samples (0·37–0·46 wt %) encompasses the values from the equigranular samples.
The compositional range for spinel is similar to that in basalt-hosted mantle xenoliths elsewhere (Barnes & Roeder, 2001), exhibiting a variation in Mg-number from 69·3 to 81·2 and in Cr-number from 13·6 to 43·0 (Fig. 11). The most Cr-rich spinel occurs in the upper equigranular xenoliths, whereas spinel compositions from protogranular and lower equigranular xenoliths are more aluminous and largely overlap. Spinel in the layered equigranular dunite (95KZS4) is compositionally distinct from that in the other samples, with an Mg-number of 65·3. The compositions of olivine and spinel are well correlated (Fig. 11), with higher Mg-numbers in olivine being associated with lower Mg-numbers and higher Cr-numbers in spinel, a general relation demonstrated previously by Irvine (1965). TiO2 contents are generally low, ranging from 0·02 to 0·44 wt % with a mean value of 0·07 ± 0·09 wt %. The Fe3+-number [100 x Fe3+/(Fe3+ + Cr + Al)] of spinel is consistently low, having a calculated mean value of 3·9 ± 0·8, which shows no correlation with TiO2 or Cr2O3 contents. Values of oxygen fugacity, calculated from the compositions of spinel and coexisting silicates (Ballhaus et al., 1991), range from –0·04 to +0·65 log units relative to the fayalite–magnetite–quartz buffer (FMQ), which are similar to the values obtained by Konen et al. (2006) for other Kozakov samples and by Ballhaus et al. (1991) for slightly metasomatized mantle xenoliths elsewhere.
|
EMPA of pyroxene shows no compositional difference between grains in symplectites and those in the matrix of protogranular samples. For pyroxene from the upper equigranular xenoliths, which contain exsolution lamellae, only the host compositions are reported, because such compositions record conditions when the xenoliths were extracted from the mantle. Orthopyroxene and clinopyroxene are magnesian, having Mg-numbers of 90·4–92·0 and 89·4–95·3, respectively. The pyroxenes show a wide range in contents of Cr2O3 (0·43–0·80 wt %, orthopyroxene; 0·74–1·80 wt %, clinopyroxene) and Al2O3 (2·13–5·44 wt %, orthopyroxene; 2·05–6·54 wt %, clinopyroxene), which correlate well with the Cr-numbers in coexisting spinel (Fig. 12). Such a wide range in R2O3 contents and Cr/Al ratios in pyroxene reflect variations in equilibration temperatures and degrees of depletion. Al2O3 contents are lowest and Cr/Al ratios are highest in pyroxene from upper equigranular xenoliths, whereas these quantities largely overlap in pyroxene from protogranular and lower equigranular xenoliths (Fig. 12). Na2O contents in clinopyroxene range from 0·43 to 1·32 wt % (except for one anomalous sample) and correlate well with Cr-number. Clinopyroxene grains in upper equigranular samples have the lowest Na2O contents and highest Cr-numbers, whereas those from the protogranular and lower equigranular layers have overlapping values for these quantities.
|
Trace elements
In the absence of other phases, the bulk of the REE in four-phase spinel lherzolites are sequestered in clinopyroxene (Table 5), and the REE patterns in clinopyroxene and whole-rocks should be similar. However, this is not the case for the two uppermost samples at 33 and 36 km. In the upper equigranular xenolith (33 km) the REE pattern in clinopyroxene is relatively flat (Fig. 13), whereas that for the whole-rock is LREE enriched (see Fig. 7). Such a difference suggests that the LREE in the whole-rock are either located in a grain-boundary component or reside in an unidentified, LREE-enriched phase. A similar situation occurs in the layered equigranular dunite (36 km), in which the REE pattern is LREE depleted in clinopyroxene, but LREE enriched in the whole-rock (compare Figs 7 and 13). In the protogranular xenoliths the REE patterns in clinopyroxene and whole-rocks are similar (compare Figs 7 and 13), both showing a concave-upward configuration, although the minimum point in clinopyroxene is located between Sm and Gd, and the minimum point in whole-rocks lies between Gd and Ho. Another difference is seen in the HREE, which are relatively flat in clinopyroxene, but have a slight to moderate positive slope in the whole-rocks. In the lower equigranular xenoliths the REE patterns in clinopyroxene and whole-rocks are also similar, both showing a LREE enrichment, except that the HREE have a slight to moderate negative slope in clinopyroxene, but a slight positive slope in the whole-rocks.
|
|
Compositional variation with depth
The depth variations in the major element compositions of minerals are more clearly defined than those for whole-rocks, probably because of a greater number of analyses. Except for the compositionally distinct equigranular dunite (36 km) and one lower equigranular sample (70 km), the Mg-number in olivine, Cr-number in spinel, Cr-number in orthopyroxene, and Cr-number in clinopyroxene (not shown) decrease with increasing depth, whereas the Na2O content of clinopyroxene increases (Fig. 14). Samples from the upper equigranular layer are the most depleted, protogranular samples exhibit a decreasing depletion with increasing depth, and samples from the lower equigranular layer (with one exception) have the least depleted compositions. Such variations suggest that the degree of depletion in the Kozákov mantle decreases with depth, as was previously inferred from the variations in whole-rock compositions.
|
Despite the relatively small number of trace element analyses for clinopyroxene separates, the variation trends in trace element compositions with depth (Fig. 14) are consistent with those for major elements. The contents of Y and Yb in clinopyroxene are lowest in a single sample from the upper equigranular layer, increase with depth in the protogranular layer, and are highest in the lower equigranular layer. Apart from the two samples in the upper equigranular layer, the degree of LREE enrichment, as measured by (Ce/Sm)N, appears to decrease with increasing depth across the protogranular layer into the lower equigranular layer.
|
SR AND ND ISOTOPES IN CLINOPYROXENE |
|---|
Clinopyroxene separates from four protogranular and two lower equigranular xenoliths display a limited range of isotopic variation (Table 6; Fig. 15), with measured values of 87Sr/86Sr lying between 0·70328 and 0·70339, and 143Nd/144Nd ranging from 0·51287 to 0·51307 (
Nd = +4·6 to +8·4). These values plot within the Nd–Sr isotopic array defined by anhydrous spinel peridotite xenoliths from other central European localities and overlap the field for clinopyroxene in enriched peridotite xenoliths from Ray Pic, Massif Central (Zangana et al., 1997; Fig. 15).
|
|
An interesting feature of this isotopic array is the restricted range of 87Sr/86Sr between 0·7031 and 0·7037 for many xenoliths, including those from Kozákov and Ray Pic. These values are similar to those proposed for upwelling mantle beneath Europe (low-velocity component in Fig. 15), which has been identified by seismic tomography and is thought to be responsible for much of the alkaline magmatism associated with the CERS (Hoernle et al., 1995
).
Despite the limited amount of Kozákov isotopic data, it appears that values of 87Sr/86Sr and 143Nd/144Nd may be correlated with depth, as is the case for the major and trace elements. Two protogranular xenoliths at intermediate depths of 48 and 54 km have lower values of 87Sr/86Sr (0·703284–0·703285) and Nd (both at +4·6) compared with four other samples at depths below 64 km, which have higher values of 87Sr/86Sr (0·703338–0·703394) and Nd (+5·1 to +8·5). This isotopic variation may also be correlated with degree of LREE enrichment in clinopyroxene, the two samples at intermediate depths being more enriched, with (Ce/Sm)N = 7·6–7·9, compared with the deeper samples, in which (Ce/Sm)N = 2·2–2·8.
|
DISCUSSION |
|---|
Partial melting and depletion of the mantle sources of the xenoliths
Variations in the whole-rock and mineral compositions of the Kozákov mantle xenoliths are indicative of the progressive extraction of partial melts. Typical for the whole-rocks are a decrease in Al2O3, CaO, and TiO2 with increasing MgO, low HREE contents relative to primitive mantle, and a negative correlation between modal clinopyroxene and Cr-number. Complementary variations in mineral compositions include positive correlations between Mg-number and Cr-number between olivine, spinel, and pyroxenes, and decreases in TiO2, Na2O, and HREE with increase in Cr-number in clinopyroxene. Low values of whole-rock YbN, which are 0·10–0·54 relative to primitive mantle, indicate that partial melting probably occurred in the absence of garnet, otherwise values of YbN would remain close to 1·0 in residual garnet peridotite for any low to moderate degree of batch or fractional melting.
Because clinopyroxene was probably the principal host for HREE in spinel peridotite when partial melting occurred, the degree of partial melting can be estimated by assuming that clinopyroxene is the only phase effectively contributing to the bulk distribution coefficient for the Y and Yb in the residual assemblage (Norman, 1998). Ti and Na in Kozákov clinopyroxenes show a positive correlation with Y and Yb, and the Cr-number in clinopyroxene is negatively correlated with Y and Yb, suggesting that distribution of these elements in clinopyroxene was controlled by partial melting, rather than metasomatism, and that application of the Norman melting model is appropriate in this case. The modelling results (Table 5; Fig. 16) indicate that batch melting requires an unreasonably high degree of partial melting for the more refractory samples, as found for xenolith suites elsewhere (Norman, 1998; Beccaluva et al., 2001). Fractional melting thus seems the best choice for the melting mode in the Kozákov xenoliths, yielding degrees of melting, F, of 4·2–9·0% for lower equigranular samples, 6·7–13·2% for protogranular samples, and 16·9% for the single upper equigranular sample. Although the absolute values of F are sensitive to the choice of model parameters, especially the initial composition of the protolith and the initial clinopyroxene mode, the relative values are consistent with the results from the major element chemistry of whole-rocks and minerals, demonstrating that the lower equigranular samples are the least depleted, the upper equigranular samples are most depleted, and the protogranular samples exhibit intermediate in level of depletion. Similar conclusions were reached based on a partial melting model utilizing all the REE, following the method of Johnson et al. (1990), the results of which are shown in Fig. 17, where the calculated REE patterns of clinopyroxene in residual spinel peridotite at different degrees of partial melting are compared with those measured in Kozákov clinopyroxenes.
|
|
An alternative method for estimating the degree of partial melting in spinel peridotite is based on the spinel composition, the Cr-number of which is highly correlated with HREE contents of coexisting clinopyroxene in mid-ocean ridge peridotites (Hellebrand et al., 2001
). Application of this method to spinel compositions in the Kozákov xenoliths yields results that are roughly comparable (with one exception) with those based on the Y and Yb contents of clinopyroxene (Table 4). The differences in the results from the two methods arise from differences in model parameters; for example, the fraction of clinopyroxene in the source, which was taken to be 0·20 by Norman (1998) and 0·14 by Hellebrand et al. (2001), among other factors. Regardless, results from the spinel calibration indicate that shallow mantle beneath Kozákov volcano is more refractory than the deeper mantle, with estimates of the degree of partial melting being 14·3 ± 1·4% in five upper equigranular xenoliths, 8·0 ± 2·4% in 11 protogranular samples, and 6·7 ± 3·1% in three lower equigranular xenoliths (with one outlier at 14·8%).
Considering the results from the three methods described above, it appears that partial melting of the protolith of the four-phase spinel peridotite xenoliths varied between 5% and 15%. Such values and the paucity of harzburgite and dunite in the xenolith suite (Fediuk, 1971) suggest that the Kozákov lithospheric mantle experienced relatively low to moderate degrees of melt extraction.
Metasomatism of the xenoliths
The high concentrations of LREE, LILE, and HFSE, the occurrence of anhydrous mineral assemblages, and the absence of any recognized modal metasomatism in the Kozákov mantle xenoliths are consistent with cryptic metasomatism by a silicate melt, rather than a carbonatitic melt, H2O fluid, or CO2 fluid. Similar metasomatic effects occur in numerous other European lithospheric mantle xenoliths, which have been attributed to the influence of transient silicate melts (Downes, 2001). In addition, the Kozákov xenoliths exhibit positive Zr and Hf anomalies and, in some cases, positive Ti anomalies, in mantle-normalized trace element patterns, which have not been described so far from European peridotite xenoliths (e.g. Lenoir et al., 2000; Downes, 2001).
The concave-upward REEN patterns for the Kozákov peridotites (Fig. 7) and clinopyroxene separates (Fig. 13) probably reflect the effects of chromatographic fractionation as a result of percolation of LREE-rich melts through porous, previously depleted peridotite. Calculated REEN patterns for such a chromatographic process (see Navon & Stolper, 1987, fig. 4) resemble those of the Kozákov peridotites, and similar concave-upward REEN patterns were described for clinopyroxene from the Horoman peridotite, Japan, and for peridotite xenoliths from the Vogelsberg, Germany, where such patterns were also ascribed to chromatographic processes (Takazawa et al., 1992; Witt-Eickschen, 1993).
The REE patterns for xenoliths from the different mantle layers at Kozákov are compared with each other and with those in xenoliths from other localities by plotting (Ce/Tb)N vs (Tb/Yb)N for whole-rocks and (Ce/Sm)N vs (Sm/Yb)N for clinopyroxene (Figs 18 and 19). These plotting parameters are selected because the minimum points in the concave-upward REE patterns for Kozákov whole-rock peridotites and clinopyroxenes occur at Tb and Sm, respectively. In such plots, REE patterns with positive, concave-upward, negative, and convex-upward shapes plot in the lower left, upper left, upper right, and lower right quadrants of the figures, respectively. For whole-rocks, Fig. 18 illustrates that concave-upward REE patterns are more pronounced in Kozákov upper equigranular and protogranular peridotites than in lower equigranular peridotites, most samples of which have (Tb/Yb)N 1. Similar differences are found in Kozákov clinopyroxenes, with two of three protogranular samples having pronounced concave-upper patterns and three of four equigranular samples having negative slopes (Fig. 19).
|
|
Peridotite xenoliths in central Europe are mostly LREE enriched, and those from the Vogelsberg and Hessian Depression have REE patterns varying from negative slopes to concave-upward shapes that overlap with those for most of the Kozákov peridotites (Fig. 18). Clinopyroxenes from central European peridotite xenoliths show a wide variety of REE patterns, but those from the Vogelsberg (Witt-Eickschen, 1993
) are concave-upward and closely similar in shape and magnitude to those of the Kozákov protogranular samples (Fig. 19). Similarities include a predominance of anhydrous spinel lherzolite, protogranular textures with spinel–pyroxene symplectites after garnet, whole-rock major element depletion trends, comparable REE patterns in whole-rocks and clinopyroxenes (Figs 18and 19), and 143Nd/144Nd and 87Sr/86Sr values in clinopyroxene that closely bracket those for Kozákov (Fig. 15).
In addition to the LREE, the Kozákov peridotite xenoliths are enriched in the LILE, P, Nb, Zr, and Hf (Fig. 8). Although the LILE enrichment at Kozákov is similar to that in many other European mantle xenoliths, the HFSE enrichment coupled with positive anomalies in mantle-normalized trace element patterns appears to be unusual; where Zr and Hf anomalies occur in other European mantle xenoliths, they tend to be negative (Downes, 2001). The fractionation of HFSE from elements of similar compatibility cannot be accounted for by cryptic metasomatism, but more probably reflects modal metasomatism, the evidence for which remains unrecognized in the Kozákov xenoliths. Despite the lack of HFSE data for Kozákov clinopyroxenes, mass-balance calculations based on the REE demonstrate that clinopyroxene does not host all of the REE and that this REE ‘deficit’ is depth dependent, decreasing with depth. Lower equigranular and protogranular clinopyroxenes host generally more then 50% of the HREE and 30% of the LREE, but upper equigranular clinopyroxene accounts for only 40% of the HREE and only 10% of the LREE. This pattern of REE ‘deficit’ is accompanied by a decrease in (Hf/Sm)N and (Nb/La)N ratios with depth (Fig. 10). Such results may arise from the precipitation of LREE- and HFSE-enriched microphases during the fractionation of ascending metasomatic melts. Such enrichment in HFSE and LILE, in the absence of any visible metasomatic phase, is similar to that observed by Bodinier et al. (1996), who demonstrated the presence of microphases concentrated along spinel grain boundaries enriched in highly incompatible trace elements. The enrichment in the HFSE at Kozákov indicates that the metasomatic agent was probably a silicate melt, rather than a fluid, which would have low HFSE solubility.
Timing of depletion and metasomatism
The geochemical data presented here indicate that the Kozákov lithospheric mantle was depleted in incompatible elements through partial melt extraction and subsequently refertilized by cryptic metasomatism. Previously, Christensen et al. (2001) suggested that the tripartite layering of sub-Kozákov mantle originated during Variscan orogenesis, when deeper-level, protogranular garnet peridotite was tectonically emplaced into shallower-level, equigranular spinel peridotite, where it resided metastably until Neogene heating promoted recrystallization. If so, in this geological context it is likely that partial melting and depletion of the three peridotite lithospheric layers occurred prior to, or during, Variscan juxtaposition. In addition, the low Yb contents of the protogranular peridotites (Fig. 7) require that partial melting occurred in the spinel stability field, which could have occurred prior to stabilization of garnet in this unit (i.e. in a pre-Variscan spinel peridotite protolith).
The regular variation of trace elements with depth across the three mantle layers, as monitored by (Ce/Tb)N, (Nb/La)N, and (Hf/Sm)N in whole-rocks (Fig. 10) and (Ce/Sm)N in clinopyroxene (Fig. 14; except for the single upper equigranular sample), implies that cryptic metasomatism occurred after assembly of the layers (i.e. in post-Variscan times). Values of these elemental ratios generally decrease with increasing depth, which may reflect the proportionately smaller metasomatic signature in the less depleted, deeper samples, compared with the more depleted, shallower samples.
In a plot of 143Nd/144Nd vs 147Sm/144Nd, clinopyroxene from three protogranular and two equigranular samples yields an apparent age of 432 ± 26 Ma, if the single protogranular outlier is neglected. Inclusion of all six samples yields an apparent age of 329 ± 150 Ma. TDM model ages of range from 105 to 275 Ma. However, in view of the complex history of the Kozákov mantle, the intermediate position of Kozákov clinopyroxene within the Nd–Sr isotopic array for European peridotite xenoliths (Fig. 15), and the positive correlation of 143Nd/144Nd with 1/Nd (not shown), such results are unlikely to represent true ages, but rather are probably due to mixing. It has been suggested that the Nd and Sr isotopic compositions of clinopyroxene in enriched peridotite xenoliths from Ray Pic, Massif Central, are the result of mixing between the Cenozoic low-velocity component and lithospheric mantle with variable ratios of Sm/Nd, 1/Nd, and 143Nd/144Nd, as a result of previous melting events (Zangana et al., 1997). Such a mixing process could equally well be invoked for the isotopically similar Kozákov clinopyroxene.
|
CONCLUSIONS |
|---|
The Kozákov lithospheric mantle has a layered structure, consisting of an equigranular upper layer at depths from 32 to 43 km, a protogranular symplectite-bearing intermediate layer from 43 to 67 km, and an equigranular lower layer from 67 to 70 km. This layered configuration is thought to have originated during Variscan convergence, when deeper-level garnet peridotite was tectonically emplaced into shallower-level spinel peridotite, where it resided metastably, until Neogene heating promoted reaction of garnet and olivine to form spinel–pyroxene symplectite.
Variation patterns for major elements in spinel lherzolite whole-rocks and constituent minerals in the three layers are typical for peridotite that has undergone partial fusion and melt extraction, the degreee of which is estimated to have varied from 5% to 15%. Subsequent to this depletion event, spinel lherzolite in all three layers was refertilized by metasomatism, most probably by a transient silicate melt, which resulted in enrichment in the LILE, LREE, and HFSE and development of positive P, Zr, and Hf anomalies, as a result of a combination of chromatographic fractionation and precipitation of HFSE-enriched microphases at grain boundaries.
The timing of depletion and cryptic metasomatism in the Kozákov lithospheric mantle remains uncertain. If the layered structure is the product of Variscan tectonics, then depletion of the juxtaposed lithospheric layers probably occurred prior to their assembly. The regular variation of trace elements with depth across the three mantle layers suggests that metasomatism occurred after assembly of the layers, and the likely influence of the low-velocity component on the isotopic evolution of the Kozákov lithospheric mantle implies that metasomatism may have been associated with Neogene rifting and magmatism.
|
ACKNOWLEDGEMENTS |
|---|
We thank Orlando Vaselli and two anonymous reviewers for their constructive reviews on this manuscript. We are grateful to Anna Langrová for microprobe analysis of two samples in the Academy of Sciences of the Czech Republic, John Fournelle for direction in the electron microprobe analytical laboratory at the University of Wisconsin, and Shah Wali Faryad and Franti
ek V. Holub for constructive comments on the manuscript. This research was supported by the Grant Agency of the Academy of Sciences, project IAA3013403 and the Scientific Programme CEZ: Z3-013-912 of the Institute of Geology, Academy of Sciences of the Czech Republic, and MSM 0021620855 of the Charles University, Faculty of Science.
*Corresponding author. Telephone: + 420221951516. Fax: +420221951496. E-mail: ackerman{at}gli.cas.cz
|
REFERENCES |
|---|
Albarède F. Introduction to Geochemical Modelling (1995) Cambridge: Cambridge University Press.
Armstrong JT. Quantitative analysis of silicate and oxide materials: Comparison of Monte Carlo, ZAF, and 
(
z) procedures. In: Microbeam Analysis, Proceedings of the 23rd Annual Conference of the Microbeam Analysis Society—Newbury DE, ed. (1988) San Francisco, CA: San Francisco Press. 239–246.
Babuka V, Plomerová J. The lithosphere in central Europe—Seismological and petrological aspects. Tectonophysics (1992) 207:141–163.[CrossRef][Web of Science]
Babuka V, Plomerová J. Subcrustal lithosphere around the Saxothuringian–Moldanubian Suture Zone—a model derived from anisotropy of seismic wave velocities. Tectonophysics (2001) 332:185–199.[CrossRef][Web of Science]
Ballhaus C, Berry RF, Green DH. High pressure experimental calibration of olivine–orthopyroxene–spinel oxygen geobarometer: implications for the oxidation state of the upper mantle. Contributions to Mineralogy and Petrology (1991) 107:27–40.[CrossRef][Web of Science]
Barnes SJ, Roeder PL. The range of spinel composition in terrestrial mafic and ultramafic rocks. Journal of Petrology (2001) 42:2279–2302.
Beard BL, Medaris LG, Johnson CM, Brueckner HK, Mísa Z. Petrogenesis of Variscan high-temperature Group A eclogites from the Moldanubian Zone of the Bohemian Massif, Czechoslovakia. Contributions to Mineralogy and Petrology (1992) 111:468–483.[CrossRef][Web of Science]
Beccaluva L, Bonadiman C, Coltorti M, Salvini L, Siena F. Depletion events, nature of metasomatizing agent and timing of enrichment processes in lithospheric mantle xenoliths from the Veneto Volcanic Province. Journal of Petrology (2001) 42:173–187.
Bertrand P, Mercier J.-CC. The mutual solubility of coexisting ortho- and clinopyroxene: toward an absolute geothermometer for the natural system? Earth and Planetary Science Letters (1985) 76:109–122.[CrossRef][Web of Science]
Blusztajn J, Shimizu N. The trace-element variations in clinopyroxenes from spinel peridotite xenoliths from southwest Poland. Chemical Geology (1994) 111:227–243.[CrossRef][Web of Science]
Bodinier JL, Merlet C, Bedini RM, Simien F, Remaidi M, Garrido CJ. Distribution of niobium, tantalum, and other highly incompatible trace elements in the lithospheric mantle: The spinel paradox. Geochimica et Cosmochimica Acta (1996) 60:545–550.[CrossRef][Web of Science]
Brey GP, Köhler T. Geothermobarometry in four-phase lherzolites: II. New thermobarometers and practical assessment of existing thermobarometry. Journal of Petrology (1990) 31:1352–1378.
ermák V. Crustal heat production and mantle heat flow in central and eastern Europe. Tectonophysics (1989) 159:195–215.[CrossRef][Web of Science]
ermák V, Král M, Krel M, Kubík J, afanda J. Heat flow, regional geophysics, and lithosphere structure in Czechoslovakia and adjacent parts of central Europe. In: Terrestrial Heat Flow and Lithosphere Structure—ermák V, Rybach L, eds. (1991) New York: Springer. 33–165.
Christensen NI, Medaris LG, Jr, Wang HF, Jelínek E. Depth variation of seismic anisotropy and petrology in central European litosphere: A tectonothermal synthesis from spinel lherzolite. Journal of Geophysical Research (2001) 106:645–664.[CrossRef]
Downes H. Formation and modification of the shallow sub-continental lithospheric mantle: a review of geochemical evidence from ultramafic xenolith suites and tectonically emplaced ultramafic massifs of Western and Central Europe. Journal of Petrology (2001) 42:233–250.
Downes H, Embey-Isztin A, Thirlwall MF. Petrology and geochemistry of spinel peridotite xenoliths from the western Pannonian Basin (Hungary): evidence for an association between enrichment and texture in the upper mantle. Contributions to Mineralogy and Petrology (1992) 107:340–354.
Farsk F. Mineralogische Notizen I. Mineralien aus der Kosakover Basaltkugeln. Verhandlungen der Kaiserlichen und Königlichen Geologischen Reichsanstalt (Wien) (1876) 205–208.
Fediuk F. Ultramafics of Krkonoe–Jizerské hory region. Acta Universitatis Carolinae Geologica (1971) 4:310–343.
Fediuk F. Deep-origin xenoliths in volcanics of Czechoslovakia. In: Crustal Structure of the Bohemian Massif and the West Carpathians—Bucha V, Blí
kovsk M, eds. (1994) New York: Springer. 277–281.
Franke W. Variscan plate tectonics in Central Europe—current ideas and open questions. Tectonophysics (1989) 169:221–228.[CrossRef][Web of Science]
Franke W. The mid-European segment of the Variscides: tectonostratigraphic units, terrane boundaries and plate tectonic evolution. In: Franke, W. Haak, V. Oncken, O. & Tanner, D. (eds). Orogenic Processes: Quantification and Modelling in the Variscan Belt. Geological Society, London, Special Publications (2000) 179:35–61.
Gaul OF, O’Reilly SY, Griffin WL. Lithosphere structure and evolution in southeastern Australia. Evolution and Dynamics of the Australian Plate. Geological Society of America Special Paper—Müller RD, Hillis RR, eds. (2003) 372:185–202.
Govindaraju K. 1989 compilation of working values and sample description for 273 geostandards. Geostandards Newsletter (1989) 13((special issue):113.[Medline]
Griffin WL, Zhang A, O’Reilly SY, Ryan CG. Phanerozoic evolution of the lithosphere beneath the Sino-Korean Craton. In: Mantle Dynamics and Plate Interactions in East Asia. American Geophysical Union, Geodynamics Series—Flower M, Chung SL, Lo CH, Lee TY, eds. (1998) 27. 107–126.
Hartmann G, Wedepohl KH. Metasomatically altered peridotite xenoliths from Hessian depression (Northwest Germany). Geochimica et Cosmochimica Acta (1990) 54:71–86.[CrossRef][Web of Science]
Hellebrand E, Snow JE, Dick HJB, Hofmann AW. Coupled major and trace elements as indicators of the extent of melting in mid-ocean-ridge peridotites. Nature (2001) 410:677–681.[CrossRef]
Hoernle K, Zhang Y.-S, Graham D. Seismic and geochemical evidence for large-scale mantle upwelling beneath the eastern Atlantic and western and central Europe. Nature (1995) 374:34–39.[CrossRef]
Irvine TN. Chromian spinel as a petrogenetic indicator. Part I. Theory. Canadian Journal of Earth Sciences (1965) 2:648–672.
Jacobsen SB, Wasserburg GJ. Sm–Nd isotopic evolution of chondrites. Earth and Planetary Science Letters (1980) 50:139–155.[CrossRef][Web of Science]
Janouek V, Farrow CM, Erban V. GCDkit: New PC software for interpretation of whole-rock geochemical data from igneous rocks. Geochimica et Cosmochimica Acta (2003) 67:A186.
Johnson CM, Thompson RA. Isotopic composition of Oligocene mafic volcanic rocks in the northern Rio Grande rift—evidence for contributions of ancient intraplate and subduction magmatism to evolution of the lithosphere. Journal of Geophysical Research (1991) 96:13593–13608.
Johnson KTM, Dick HJB, Shimizu N. Melting in the oceanic upper mantle: an ion microprobe study of diopsides in abyssal peridotites. Journal of Geophysical Research (1990) 95:2661–2678.
Konen P, Ulrych J, Schovánek P, Huraiová M, 
anda Z. Upper mantle xenoliths from the Pliocene Kozákov volcano, NE Bohemia: P–T–fO2 and geochemical constraints. Geologica Carpathica (2006) 57:379–396.
Kopeck L. Geological development and block structure of the Cenozoic Ohre Rift. (Czechoslovakia). In: Proceedings of the 6th International Conference on Basement Tectonics—Aldrich MJ, Laughlin AW, eds. (1986) Salt Lake City: International Basement Tectonic Association Inc. 114–124.
Kossmat F. Gliederung des Variskischen Gebirgahauer. Abhandlungen Sächsischen Geologischen Landesamts (1927) 1:1–39.
Lenoir X, Garrido CJ, Bodinier JL, Dautria JM. Contrasting lithospheric mantle domains beneath the Massif Central (France) revealed by geochemistry of peridotite xenoliths. Earth and Planetary Science Letters (2000) 181:359–375.[CrossRef][Web of Science]
Matte P. The Variscan collage and orogeny (480–290 Ma) and the tectonic definition of the Armorica microplate: a review. Terra Nova (2001) 13:122–128.[CrossRef][Web of Science]
McDonough WF, Sun S. The composition of the Earth. Chemical Geology (1995) 120:223–253.[CrossRef][Web of Science]
Medaris LG, Jr, Fournelle JH, Wang HF, Jelínek E. Thermobarometry and reconstructed chemical compositions of spinel–pyroxene symplectites: evidence for pre-existing garnet in lherzolite xenoliths from Czech Neogene lavas. Russian Geology and Geophysics (1997) 38:277–286.
Medaris LG, Jr, Wang HF, Fournelle JH, Zimmer JH, Jelínek E. A cautionary tale of spinel peridotite thermobarometry: an example from xenoliths of Kozákov volcano, Czech Republic. Geolines (1999) 9:92–95.
Menzies MA, Bodinier JL. Growth of the European lithospheric mantle—dependence of upper-mantle peridotite facies and chemical heterogeneity on tectonics and age. Physics of the Earth and Planetary Interiors (1993) 79:219–240.[CrossRef][Web of Science]
Mercier JC, Nicolas A. Textures and fabrics of upper-mantle peridotites as illustrated by xenoliths from basalts. Journal of Petrology (1975) 16:454–487.
Merlet C. An accurate computer correction program for quantitative electron-probe microanalyses. Microchimica Acta (1994) 114:363–376.[CrossRef]
Navon O, Stolper E. Geochemical consequences of melt percolation: the upper mantle as chromatographic column. Journal of Geology (1987) 95:285–307.[Web of Science]
Norman MD. Melting and metasomatism in the continental lithosphere: laser ablation ICPMS analysis of minerals in spinel lherzolites from eastern Australia. Contributions to Mineralogy and Petrology (1998) 130:240–255.[CrossRef][Web of Science]
O’Neill HStC. The transition between spinel lherzolite and garnet lherzolite, and its use as a geobarometer. Contributions to Mineralogy and Petrology (1981) 77:185–194.[CrossRef][Web of Science]
Plomerová J, Vecsey L, Babuka V, Granet M, Achauer U. Passive seismic experiment MOSAIC—a pilot study of mantle lithosphere anisotropy of the Bohemian Massif. Studia Geophysica et Geodaetica (2005) 49:541–560.[CrossRef]
ibrava V, Havlíek P. Radiometric age of Plio-Pleistocene volcanic rocks of the Bohemian Massif. Vstník Ústedního Ústavu Geologického (1980) 55:129–139.
Stosch HG, Lugmair GW. Trace element and Sr and Nd isotope geochemistry of peridotite xenoliths from the Eifel (West Germany) and their bearing on the evolution of the subcontinental lithosphere. Earth and Planetary Science Letters (1986) 80:281–298.[CrossRef][Web of Science]
Stosch HG, Seck HA. Geochemistry and mineralogy of two spinel peridotite suites from Dreiser Weiher, West Germany. Geochimica et Cosmochimica Acta (1980) 44:457–470.[CrossRef][Web of Science]
Strnad L, Mihaljevi M, ebek O. Laser ablation and solution ICP-MS determination of rare earth elements in USGS BIR-1G, BHVO-2G and BCR-2G glass reference material. Geostandards and Geoanalytical Research (2005) 29:303–314.[CrossRef][Web of Science]
Szabó C, Taylor L. Mantle petrology and geochemistry beneath the Nógrád–Gömör Volcanic Field, Carpathian–Pannonian Region. International Geology Review (1994) 36:328–358.
Takazawa E, Frey FA, Shimizu N, Obata M, Bodinier JL. Geochemical evidence for melt migration and reaction in the upper mantle. Nature (1992) 359:55–58.[CrossRef]
Taylor WR. An experimental test of some geothermometer and geobarometer formulations for upper mantle peridotites with application to the thermobarometry of fertile lherzolite and garnet websterite. Neues Jahrbuch für Mineralogie, Abhandlungen (1998) 172:381–408.[Web of Science]
Ulrych J, Pivec E, Lang M, Balogh K, Kropáek V. Cenozoic intraplate volcanic rock series of the Bohemian Massif: A review. Geolines (1999) 9:123–129.
Vaselli O, Downes H, Thirlwall MF, Dobosi G, Coradossi N, Seghedi I, Szakacs A, Vannucci R. Ultramafic xenoliths in Plio-Pleistocene alkali basalts from the eastern Transylvanian Basin: depleted mantle enriched by vein metasomatism. Journal of Petrology (1995) 36:23–53.
Vaselli O, Downes H, Thirlwall MF, Vanucci R, Coradossi N. Spinel-peridotite xenoliths from Kapfenstein (Graz Basin, Eastern Austria): A geochemical and petrological study. Mineralogy and Petrology (1996) 57:23–50.[CrossRef][Web of Science]
Vokurka K, Povondra P. Geothermometry and geobarometry of lherzolite nodules from Kozákov, NE Bohemia, Czechoslovakia. Acta Universitatis Carolinae Geologica (1983) 4:261–272.
Wilson M, Downes H. Tertiary–Quaternary extension-related alkaline magmatism in western and central Europe. Journal of Petrology (1991) 32:811–850.
Witt-Eickschen G. Upper mantle xenoliths from alkali basalt of the Vogelsberg, Germany: implications for mantle upwelling and metasomatism. European Journal of Mineralogy (1993) 5:361–376.
Witt-Eickschen G, Kramm U. Mantle upwelling and metasomatism beneath central Europe: geochemical and isotopic constrains from mantle xenoliths from the Rhön (Germany). Journal of Petrology (1997) 38:479–493.[CrossRef][Web of Science]
Witt-Eickschen G, Seck HA. Solubility of Ca and Al in orthopyroxene from spinel peridotite: An improved version of an empirical geothermometer. Contributions to Mineralogy and Petrology (1991) 106:431–439.[CrossRef][Web of Science]
Yaxley GM, Crawford AJ, Green DH. Evidence for carbonatite metasomatism in spinel peridotites from western Victoria, Australia. Earth and Planetary Science Letters (1991) 197:305–317.
Zangana NA, Downes H, Thirlwall MF, Hegner E. Relationship between deformation, equilibration temperatures, REE and radiogenic isotopes in mantle xenoliths (Ray Pic, Massif Central, France): an example of plume–lithosphere interaction? Contributions to Mineralogy and Petrology (1997) 127:187–203.[CrossRef][Web of Science]
Zangana NA, Downes H, Thirlwall MF, Marriner GF, Bea F. Geochemical variation in peridotite xenoliths and their constituent clinopyroxenes from Ray Pic (French Massif Central): implications for the composition of the shallow lithospheric mantle. Chemical Geology (1999) 153:11–35.[CrossRef][Web of Science]
Ziegler PA. European Cenozoic rift system. Tectonophysics (1992) 208:91–111.[CrossRef][Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. Ladenberger, P. Lazor, and M. Michalik CO2 fluid inclusions in mantle xenoliths from Lower Silesia (SW Poland): formation conditions and decompression history European Journal of Mineralogy, August 1, 2009; 21(4): 751 - 761. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||