High-K calcalkaline metaluminous granitoids are abundant in many 'orogenic' to 'post-orogenic' tectonic settings. Models for their origins include: assimilation of crustal rocks by basaltic magmas, partial melting of enriched mantle, reactive assimilation of wall rocks by normal-K magmas, magma mixing, and melting of interlayered pelitic and amphibolitic source rocks. In many cases, multiple source and mixing origins can be ruled out on geological and geochemical grounds, and some writers prefer single-stage crustal melting as a model. Published experimental data show that metabasaltic rocks are not suitable sources; nor are any studied greywackes, metatonalites or metadacites. Roberts and Clemens (1993) hypothesized that the suitable protoliths would be high-K andesites - common rock types in magmatic arcs. Subsequently it has been shown that certain metaluminous Bt-Pl-Qtz gneisses can partially melt to produce peraluminous high-K melts.
The present preliminary experiments investigated the Roberts and Clemens hypothesis, using two unmetamorphosed high-K andesites from the Miocene (12 - 5 Ma) Cordillera Blanca batholith, northern Peru. Both contain around 55 wt% SiO2 and 2 wt% K2O. One has 10% Bt + 40% Hbl and the other 7% Bt + 20% Hbl. Experiments were at 1 GPa, fluid-absent, carried out in Au capsules, in a piston-cylinder apparatus with NaCl ± Pyrex cells. Oxygen fugacity was ~ FMQ. Runs at 890 and 910°C were for 70 to 125 h duration respectively.
Products include glass (quenched melt) + Pl + Bt + Hbl ± Qtz, and new Hbl + Cpx ± Opx. Electron probe (EDS) analyses of glasses show that melts are generally high-K, mostly monzogranitic, with one granodiorite. High-K granitoid melts can indeed be produced by partial melting of potassic arc andesites, but the experiments produced only relatively small melt proportions. If high-K magmas are formed in this way, crustal temperatures must be > 900°C. The single-source model is attractive because it relies only on the presence of rocks typical of arc settings, and does not demand that they be hydrated and metamorphosed as a prelude to partial melting. The main problem is to heat the crust sufficiently to induce partial melting. This can probably be accomplished through multiple injection of mafic magma. Given that high-K rocks are usually emplaced into the continental sides of arcs, in extensional settings, hot mantle itself may dome into the thinned crust.
The preliminary conclusion is that neither the protoliths nor the conditions involved in the genesis of the high-K series need be special or complicated. Single-stage melting of typical old arc crust may be all that is required to impart the observed chemical and isotopic signatures.
Roberts, MP & Clemens, JD, Geology, 21, 825-828, (1993).
The Cambrian Devil River Volcanics in the Takaka Terrane, New Zealand, comprise a complete arc/back-arc assemblage (Münker & Cooper, 1995). The rocks are basaltic to basaltic-andesitic in composition and show low to very low grade metamorphism. Here we present new data for the Baton Valley section of the Devil River Volcanics, a coherent succession of interbedded back-arc tholeiites and low- to high-K arc rocks. The rocks show Ti/Zr of 30-100, lower than typical values of MORB (>90). La/Yb and Th/Yb range from 5 and 0.2-1 (back-arc) up to 27 and 6 (high-K arc rocks), respectively. Ti/Zr, Zr/Nb and Nb/Th in the arc rocks decrease with increasing La/Yb and Th/Yb values. Nb/Ta values range from 11 to 22 and are positively correlated with La/Yb and Zr/Hf. Such enriched trace element contents in the arc rocks must reflect enrichment by slab melts of the subarc mantle, since particularly Th, REE and Zr are relatively insoluble in slab derived fluids. The observed correlations of Nb/Ta and Ti/Zr with REE and Th enrichment therefore indicate a significant addition of HFSE by slab melts to the mantle wedge. Superchondritic Nb/Ta of up to 22 in the high-K arc rocks suggest melting of the slab in the presence of rutile (DNb/DTa <1 in the rutile/melt system) as previously suggested by Münker (1998). In agreement with REE and HFSE systematics, U/Th ratios in the arc rocks (0.15-0.35) are consistently lower than in the back-arc rocks (0.25-0.6), although the Devil River Volcanics are altered. U/Th ratios decrease with Th contents. Such features are consistent with models for recent arc rocks (e.g. Hawkesworth et al., 1997) where low U/Th indicate slab melt and high U/Th fluid enrichment of the subarc mantle. Nb/Ta, Th/Yb and La/Yb mixing models can explain the trace element compositions of the arc rocks by addition of up to 15% slab melt. This slab melt must contain a significant (up 20%) sediment melt component in order to explain the Th enrichment of the subarc mantle. <epsilon> Nd values of the arc rocks ranging from +1 to +2, however, imply that up to 80% of the slab melt must be derived from the eclogitic portion of the subducted slab.
Münker, C, Chem. Geol., 144, 23, (1998).
Münker, C, Cooper, RA, Journ. Geol, 103, 687, (1995).
Hawkesworth, CJ, Turner, S, Chem. Geol, 139, 207, (1997).
The liquid line of descent of fractionally crystallizing hydrous, subalkaline magmas is determined experimentally at a pressure of 1 GPa, corresponding to a depth of approximately 30 km coincident with the minimum depth of the crust-mantle boundary (Moho) in suprasubduction 'island-arc' environments.
A natural calc-alkaline picrobasalt (containing 2.6 wt.% H2O) from the Adamello batholith (N-Italy) was used as the initial starting composition. Subsequent fractionation experiments were conducted with mixtures of synthetic oxides and hydroxides, synthetic fayalite and natural K-feldspar. The finely ground starting powders were loaded into graphite containers and sealed into Pt-capsules. The experiments were performed in end-loaded piston cylinder apparatuses using NaCl-pyrex assemblies. Run duration lasted from 24 to 48 hours. All charges were mounted in epoxy resin and ground to expose the center of the charge and analyzed by electron microprobe.
The liquidus temperature (at 1 GPa) of the picrobasaltic starting composition has been located at 1340°C with olivine as the liquidus phase. At 1230°C olivine and spinel are the liquid phases. The fractionation of forsteritic olivine leads to a rapid decrease in the Mg-content of the residual liquid. At 1200°C high-Ca clinopyroxene (cpx) joins the crystallizing assemblage. This is reflected in a rapid decrease of the Ca-content of the residual liquid. At 1170°C olivine ceases to crystallize leaving solely cpx and spinel on the liquidus; at 1140°C orthopyroxene (opx) joins the crystallizing assemblage and at 1110°C plagioclase occurs in addition. At 1080°C the differentiation trend is dictated by extensive plagioclase crystallization, which dominates over opx, cpx and spinel. This results in a tholeiitic differentiation trend characterized by Fe-enrichment at constant SiO2. In contrast, a calc-alkaline differentiation trend requires suppression of the plag crystallization to lower temperatures and concomitantly the stabilization of amphibole and/or magnetite to higher temperatures. The major factors controlling amph/magnetite versus plag crystallization are (i) the H2O-content of the magma and (ii) the oxygen fugacity at which the magma crystallizes. Both, high H2O-content and high fO2 should lead to the stabilization of amphibole/ magnetite and suppression of plagioclase crystallization. Increasing the H2O-content did not alter the observed crystallization sequence significantly. Currently, attempts are undertaken to control the oxygen fugacity with solid-solid buffers (Ni-NiO, Re-ReO) in H2O-undersaturated experiments in order to evaluate the role of fO2 in controlling calc-alkaline versus tholeiitic fractionation trends in subalkaline, hydrous, mantle-derived basaltic liquids in subduction zone environments.
The primary liquid composition of the Alpine orogenic magmatism is mostly inferred from the geochemical signature of Late Eocene - Oligocene dykes. Further insights can be provided by the mafic-ultramafic intrusive rocks associated with the Periadriatic calc-alkaline bodies. Amphibole and clinopyroxene from hornblendites and hornblende-gabbros of Adamello (Mt. Mattoni and Cornone di Blumone areas) and Bregaglia (Val Sissone area) intrusions were thus analysed for trace elements by ion microprobe. On the basis of a new set of experimentally determined AmPh/LD and Cpx/LD for an alkaline Mg-rich system (Tiepolo, PhD thesis, 1999), we have then calculated the composition of their equilibrium liquids.
The least evolved liquids calculated for the Adamello intrusion are slightly LREE-enriched (LaN/SmN ~ 2) with a nearly flat HREE pattern at about 20-30 times C1. The Bregaglia rocks yield significant variations in the liquid composition: hornblendites give nearly flat liquids at about 20 times C1, whereas the hornblende-gabbro liquids are slightly LREE-enriched, i.e. closely similar to those calculated from the Adamello rocks. In all cases, the liquids do not show significant depletion of HFSE relative to REE; if compared with N-MORB, however, they are markedly Ba- and Sr-enriched.
On the basis of textural relationships, e.g. different generations of both clinopyroxene and amphibole, it is possible to reconstruct, in the frame of a same rock-sample, a complex magmatic evolution. As a rule, the liquids became progressively LREE-enriched as crystallisation proceeded. Noticeably, an abrupt LREE enrichment over HREE characterise the late liquids calculated for the Mt. Mattoni rocks.
Many compositional variations observed for the parental liquids of clinopyroxene and amphibole cannot be ascribed to a fractional crystallisation process. For instance, the fractionating mineral assemblages are not able to produce a marked LREE-enrichment. The possible role of accessory zircon has to be excluded, because calculated liquids do not show significant variations in Zr concentrations.
The compositional liquid spectrum calculated from the mafic-ultramafic intrusives is in agreement with the composition of the dikes present in both Adamello and Bregaglia intrusions, ranging from picrite to basaltic andesite and shoshonite. Because isotopic data from literature indicate that no crustal assimilation affected the crystallisation of the mafic-ultramafic intrusive rocks, a mixing process among liquids derived from different primary melts has to be invoked. Such primary melts were most likely variably LILE- and LREE-enriched relative to N-MORB, thus suggesting that a heterogeneous mantle source was involved in the Alpine orogenic magmatism.
Volcanism in collisional belts is a useful tool in unravelling tectonic history as changes in composition, timing and, in a lesser extent, spatial distribution of the volcanism are strictly related to the evolving kinematics. Two examples of collision-related volcanism gained from the Greater Caucasus and the Alps are considered in order to draw some conclusions on the genesis and on the relationships of the volcanic products with geodinamical framework.
The Greater Caucasus is a recent-developed collisional belt resulting from the Oligocene collision of the northward-moving Arabian plate with Eurasia. The Late Cenozoic Caucasian volcanism is represented by well preserved calc-alkaline volcanic complexes (i.e. Elbrus) and ignimbrite sheets (i.e. Chegem) with associated minor intrusion (Mineralnije Vodi). This volcanism developed 25-30 Ma after the continental collision, ranging from Late Miocene to present. It is located along mayor discontinuities of an old thickened crust.
The main products are represented by trachybasalt, andesite and dacite in the Elbrus and in Kazbek area (Koronovski & Demina, 1996) whereas the volcanic rocks from the Chegem are essentially rhyolitic and dacitic ash flow (Lipman et al., 1993). All these volcanites display typical HK - calc-alkaline and shoshonitic characteristics.
The Alps is a more mature collisional belt than Greater Caucasus. Volcanic products are poorly preserved as remnants scattered throughout the chain (i.e. andesitic clasts and ash layers in sedimentary successions). Most of the primary records of the magmatism (intrusions and dikes) are located along a mayor discontinuity (Periadriatic lineament). The magmatism began approximatively in Late Eocene, i.e. 25-30 Ma after the collision took place and continues up to Late Miocene.
The volcanic products, mainly andesite with minor dacite and rhyolites, show typical HK calc-alkaline and shoshonitic characteristics. Some ultrapotassic dikes occur.
Comparing the volcanic products of the two chains allows to define some general features for developement of the volcanism in collisional belts:
1) Geochemical features suggest that the primary source of volcanites could be mantle material variably contaminated by a crustal component.
2) The time interval that exists between the beginning of the collision detected by geological observations and the emplacement of the volcanic products appear to be quite constant.
3) Most of collision-related volcanic and magmatic products occur along mayor discontinuities.
The ascent of magmas in geodynamic environments characterized by regional compressional regime is guided by deep transtensional structures acting during the whole collisional event.
Koronovski NV & Demina LI, Dokl. RAN, 350, n.4, 519-522, (1996).
Lipman PW, Bogatikov OA, Tsvetkov AA, Gazis C, Gurbanov AG, Hon K, Koronovsky NV, Kovalenko VI & Marchev P, Journ. Volc. Geotherm. Res, 57, 85-124, (1993).
Isotopic (Sr, Nd, Pb, O) and geochemical data are reported for a stratigraphically well-established sequence of calc-alkaline lavas and pyroclastics from the island of Salina (Aeolian arc, Italy). The studied samples are from the mid-Pleistocene volcanic cycle of the island (Keller, 1980). They cover the full compositional range of Salina volcanic rocks, which vary from 51.3 to 68.9 wt.% SiO2. Throughout the stratigraphic sequence early erupted basalts and basaltic andesites (Capo; Rivi; Corvo; Fossa delle Felci, group 1) are followed by the eruption of dacites, andesites and basaltic andesites (Fossa delle Felci, groups 2-8) in the final stages of activity (Gertisser and Keller, 1997). The observed compositional variations from primitive high-alumina basalt to dacite have been attributed to simple crystal-liquid fractionation under dominantly low pressure as the major process (Keller, 1980; Gertisser and Keller, 1997). To better constrain the role of crustal contribution in the genesis and evolution of this suite, 10 samples, that cover the entire compositional range, have been analysed for radiogenic (Sr, Nd, Pb) and oxygen isotopes. Throughout the stratigraphic sequence 87Sr/86Sr ratios display small but significant variations between 0.70410 and 0.70463 and show a well-defined negative correlation with 143Nd/144Nd (0.51275-0.51279). Pb isotope compositions of the same samples are distinctly radiogenic with relatively large variations in 206Pb/204Pb (19.30-19.66), fairly constant 207Pb/204Pb (15.68-15.76) and minor variations in 208Pb/204Pb ratios (39.15-39.51). Whole-rock 18O values, which range from +6.4‰ to +8.45‰, are accompanied by a slight increase in Sr isotope ratios. Sr isotope ratios generally exhibit a compositional dependence with the degree of differentiation of the rocks. SiO2, LILE (Rb, Ba, K), HFSE and ratios of elements with different incompatibility, such as Rb/Sr, Rb/Ba, Ba/La, Th/La show positive, Mg#, Sr and ferromagnesian elements negative correlations with Sr isotope ratios. Nd isotope ratios generally decrease with increasing degree of differentiation. 18O values are elevated in the evolved samples of the magmatic series. In conclusion, slight but significant Sr, Nd, Pb and O isotopic variations have been highlighted in the calc-alkaline basalt-dacite series of Salina. The observed variations are correlated with the degree of differentiation of the rocks, indicating small degrees of crustal assimilation, overprinting the dominant evolution by crystal-liquid fractionation. Thus, the isotopic data suggest the occurrence of AFC-type processes, giving rise to evolved magmas with elevated 18O and 87Sr/86Sr, and slightly decreased 143Nd/144Nd, 206Pb/204Pb and 208Pb/204Pb.
Gertisser R, Keller J, Terra Nova, 9, 474, (1997).
Keller J, Rend. Soc. Ital. Mineral. Petrol., 36, 489-524, (1980).
K-rich magmatism is known from active plate margins, and also from post-collisional and within-plate environments. The parental melts are thought to be derived from lithospheric mantle sources enriched by subduction-related fluids and melts. General characteristic of K-rich suites is a strong enrichment of LILE, as well as a relative depletion of HFSE. Discrimination between possible sources of K-rich magmas is difficult based on geochemistry alone, but some trace element ratios (e.g. Sr/Nd, Th/Nb) and isotopic signatures (e.g. Sr and Nd initial isotope ratios) are indicative.
There are several K-rich intrusions and dyke associations in the mid-European Hercynian chain, which are classified as shoshonitic or ultra-potassic based on their high K2O contents and K2O/Na2O ratios (e.g. Meissen Massif, Saxony; Ballons Massif, Vosges; Rastenberg, South Bohemia; Tabor-Milevsko, Central Bohemia; Punteglias and Giuv, Swiss Alps). The rocks display radiogenic initial Sr isotope ratios (mostly > 0.707) and low <epsilon> Nd values (mostly < -5). Low Sr/Nd combined with high Th/Nb ratios together with the isotopic signatures observed, point to similarities of the mid-European occurrences with post-collisional K-rich magmatism from Tibet, the Northwestern Alps, and Spain. High Th/Nb ratios are typically interpreted to reflect mantle source enrichment by subduction of crustal sediments (cf. Rogers et al., 1985). However, for some of the mid-European intrusions, such as the Meissen Massif monzonites, high Th and U contents are at least partly caused by crustal contamination as indicated by zircon relics with up to 2 wt% Th+U.
Direct evidence for K-enriched 335-340 Ma old (Rb-Sr whole rock ages) lithospheric mantle domains in the mid-European Hercynian chain is provided by glimmerite veins in exhumed garnet peridotites from southern Bohemia (Becker et al., 1999). The glimmerites show high LILE and LREE concentrations and negative HFSE anomalies, and crystallized from fluids released during the HT-HP breakdown of F-rich phlogopite of subducted felsic granulites. They represent analogues for potential source regions to the several K-rich intrusions.
The post-collisional character of the mid-European K-rich intrusions is in agreement with syn- to late-tectonic fabric patterns of the intrusions and the age relationships with high-pressure nappe units, which were subducted during the Upper Devonian or Lower Carboniferous (Wenzel et al., 1997).
Becker H, Wenzel T & Volker F, J Petrol, 40, in press, (1999).
Rogers NW, Hawkesworth CJ, Parker RJ & Marsh JS, Contrib Mineral Petrol, 90, 244-257, (1985).
Wenzel Th, Mertz DF, Oberhänsli R, Becker T & Renne PR, Geol Rundsch, 86, 556-570, (1997).
Mt. Erciyes stratovolcano which was built up after continental collision of Eurasian and African plates, constitutes the northern end of Cappadocian volcanic zone of central Anatolia. Alkaline basalts (SiO2 = 48.4 - 52.15) are represented by high Fe2O3, MgO, CaO, extremely low K2O content (K2O = 0.18% Wt.) and Nb, Ba and Rb depletion. Ni and Cr concentrations are close to the mantle values(106 and 201 ppm respectively). Alkaline basalts have low La/Yb (2.47-5.81), La/Nb (0.89-1.33) and high Zr/Nb (17.46-19.61), Zr/Ba (0.76-2.34) ratios whereas alkaline basaltic rocks are represented by high La/Yb (10.28 -14.52), La/Nb (1.35-1.55), Zr/Ba (0.46- 0.59) and low Zr/Nb ( 7.87-7.91) ratios, these differences indicate that alkaline basalts and alkaline basaltic rocks were produced by different partial melting processes from different mantle sources (Kurkcuoglu et al,1998). Calc-alkaline samples are represented by generally low MgO, Fe2O3, high K2O, Ni, Zr, Cr and high La/Yb (8.1-11.2), Zr/Nb ( 8.92-15.63) and Ba/Nb (12.61-16.30) ratios. Sr and Nd isotopic compositions range between 0.703344-0.705088 and 0.512920-0512630 for alkaline, 0.703434-0.70507 and 0.512942-0.512394 for calc-alkaline rock suites respectively. These low Sr and high isotopic signatures, trace element data and elemental ratios suggest an depleted asthenospheric and OIB-like mantle sources at the generations of the alkaline basalts and alkaline basaltic products. The low Ba, Nb, Rb and K2O content of alkaline basalts can be explained by the removal of these elements from mantle source region by subduction-related fluids whereas calc-alkalline basalts can possibly be generated by the mixing of asthenospheric partial melts with lithospheric mantle materials. In additon, high Ni, Cr, and Zr content of calc-alkaline basalts are interpreted as the inheritance of the mantle source region. All the geochemical data suggest that depleted asthenospheric and OIB-like mantle sources have undergone high and low degree partial melting episodes to generate alkaline basalts and alkaline basaltic rocks with different compositions respectively. Depleted asthenospheric mantle source region is partially modified by subduction-related fluids which leads to removal of Rb, Ba, Nb and K and the ascending magma that released from both of the mantle source regions, is interacted with lithospheric mantle materials to generate calc-alkaline basic products. While the high Ba/Nb (14.47-22.97) content of alkaline rocks may depend on the OIB-like mantle source, high Ba/Nb, 87Sr/86Sr and low 143Nd/144Nd ratios imply the lithospheric contribution and also AFC processes for the generations calc-alkaline basalts.
Kurkcuoglu,B., Sen, E., Aydar, E.,Gourgaud, A., Gundogdu, MN., Journal of Volcanology and Geothermal Research, 85 (1-4), 473-494, (1998).
Extensive volcanic activity followed an Eocene continent-arc collision in Western Anatolia. The early stage of collision-related volcanism, which was most evident during the Early Miocene (<21 Ma), produced a considerable volume of lavas and pyroclastic deposits of basaltic andesites to rhyolites composition. The volcanic activity continued into the Middle Miocene with a gradual change in eruptive style and magma composition. The Middle Miocene activity (>15 Ma) formed in relation to localised extensional basins and was dominated by lava flows and dykes of basalt to andesite composition. Both the Early and Middle Miocene rocks have calc-alkaline and shoshonitic character. The Late Miocene volcanism (<11 Ma) was marked by alkali basalts and basanites erupted along zones of localised extension.
The Early-Middle Miocene volcanic rocks exhibit enrichment in LILE and LREE relative to the HFSE and have high 87Sr/86Sr (0.707517-0.708681) and low 143Nd/144Nd (0.512318-0.512460) ratios. Modelling of these characteristics indicates a mantle lithospheric source region carrying a subduction component inherited from a pre-collision subduction event. Perturbation of this subduction-metasomatised lithosphere either by delamination of the thermal boundary layer or by slab detachment is the likely mechanism for the initiation of the post-collision magmatism.
Petrographic characteristics and trace elements systematics suggest that the Early-Middle Miocene magmas underwent hydrous crystallisation (dominated by plagioclase + pyroxene + pargasitic amphibole) in deep crustal magma chambers. Subsequent crystallisation in shallower magma chambers follows two different trends: (1) anhydrous (dominated by pyroxene + plagioclase); and (2) hydrous (dominated by edenitic amphibole + plagioclase + pyroxene).
AFC modelling shows that the Early-Middle Miocene magmas evolved through assimilation combined with fractional crystallisation, and that the effects of assimilation decrease gradually from the Early Miocene into the Middle Miocene. This may indicate a progressive crustal thinning related to the extensional tectonics.
In contrast, the Late Miocene alkaline rocks are characterised by low 87Sr/86Sr (0.703108-0.703253) and high 143Nd/144Nd (0.512929-0.512978) ratios and have OIB-type like trace element patterns characterised by enrichment in LILE, HFSE and L-MREE, and a slight depletion in HREE, relative to average N-MORB. REE modelling indicates that these rocks formed by small degree (~2% to ~10%) partial melting of a garnet-bearing lherzolite source. Trace element and isotope systematics are consistent with an origin by decompression melting of an enriched asthenospheric mantle source.
The Erzurum-Kars Plateau (EKP) in NE of Anatolia, contains the full record of a widespread collision-related volcanism (CRV) from Middle Miocene to Pliocene. Lavas and pyroclastics on the EKP are calc-alkaline in character and cover a broad compositional range from basalts to rhyolites. They exhibit a distinctive subduction signature inherited from a pre-collision subduction event. Collision started after the closure of the southern branch of Neotethys ocean between Eurasian and Arabian continents in Late Eocene (ca. 40 Ma). This created a regional block uplift of the Eastern Anatolia at around 13-15 Ma, postdating the collision some 25 My. Our radio-isotopic data indicate that the CRV started on the EKP immediately after the rapid uplift of the plateau at 11 Ma. Our fieldwork on the EKP revealed that most magma reached the surface along fissures opened up in areas of localised extension linked to pull-apart basins in strike-slip fault zones. The CRV on the EKP followed three diachronous stages and produced two distinct lava series, becoming younger towards the east. The Early (~11-6 Ma) and Late (5-2.7 Ma) Stages are represented by a bimodal volcanism dominated by an anhydrous (POAM) crystallisation assemblage (the high-Y series) whereas the Middle Stage (6-5 Ma) consists of unimodal (andesitic) volcanism dominated by hydrous (PAm) crystallisation (the low-Y series). We believe that the difference between low- and high-Y series is related to the depth of magma ponding in the crust. Our geochemical data indicate that the basic lavas of the Early Stage erupted from small, transient chambers while the acid lavas and pyroclastics erupted from large, shallow and zoned magma chambers with little or no assimilation. Low-Y lavas of the Middle stage, on the other hand, evolved in deeper chambers, underwent homogenisation with significant crustal assimilation. Lavas of the Late Stage are in general more basic than lavas of the Early and Late Stages. They display an insignificant degree of assimilation. This may be due to establishment of fractures in the crust created by strike-slip faults during the Late Stage. Isotopic and trace element data are consistent with derivation of the volcanic products of these three Stages from a similar parental magma generated from sub-continental lithosphere. Our field and geochemical data support the model suggested by Peace et al. (1990) in which both regional uplift and generation of magma was controlled by catastrophic delamination of a thickened, sub-continental, mantle lithosphere.
Pearce, JA, Bender, JF, De Long, SE, Kidd, WSF, Low PJ, Guner, Y, Saroglu, F, Yilmaz, Y, Moorbath, S & Mitchell, JG, Genesis of collision volcanism in Eastern Anatolia, Turkey, J. Vol. Geother. Res, 44, 189-229, (1990).
The western part of the island of Sulawesi (Indonesia) has been an active continental margin from approximately 60 to 10 Ma. Collision of this arc with the continental fragments of Sula and Buton caused cessation of subduction, but magmatism continued during and after these collisional events.
The pre-collisional deposits in both areas are typical of arc volcanics with high LILE/HFSE ratios and only slightly elevated Sr isotope ratios compared to MORB. Syn- collisional magmatism in South Sulawesi is alkaline (basalt to trachyte), and has higher Sr and Pb and lower Nd isotopic ratios compared to pre-collisional magmatism. Syn-collisional magmatism in Central Sulawesi ranges from lamprophyric to granitic. The magnitude of the isotopic shift is distinctly larger in Central than in South Sulawesi (Sr isotope ratios of up to 0.723 versus 0.711). Oxygen isotopic analyses suggest that upper crustal material was involved in the petrogenesis of the syn-collisional magmas from Central Sulawesi, although none of the samples analysed is peraluminous.
Basaltic to andesitic post-collisional magmatism in South Sulawesi (1-2 Ma) is characterised by low Ba/Nb and high Nb/Y ratios, with isotopic ratios only slightly more primitive than those of the syn-collisional magmas from this area. Post-collisional magmas in Central Sulawesi (1-6 Ma) are more silicic (Priadi et al., 1994) but the one available isotopic analysis suggests that the crustal component is less than in some of the syn-collisional magmas from this area.
We propose that the geochemistry of syn-collisional magmatism in South Sulawesi reflects addition of a larger sedimentary component to the mantle source, and that post-collisional magmatism in this area contains a contribution from the subcontinental lithospheric mantle. The interpretation of syn- and post-collisional magmatism in Central Sulawesi is more complex, but the isotopic data indicate the involvement of old crustal material. The first occurrence of this enriched isotopic signature is 11 Ma (our data), or as early as 17 Ma (Bergman et al., 1996). If the collision between Sula and Central Sulawesi indeed occurred no earlier than 10 Ma (Hall, 1996), it is unlikely that this collision is responsible for the observed geochemical characteristics.
The present study clearly reveals the capacity of magmatic systems to respond quickly to changing sources. It appears that characteristic arc geochemical signatures may have quite short survival times in the mantle wedge with some implications for models where the survival of ancient subduction-modified mantle is invoked
Bergman S, Coffield DQ, Talbot JP & Garrard RA, Tectonic evolution of Southeast Asia (Geological Society), 391-429, (1996).
Hall R, Tectonic evolution of Southeast Asia (Geological Society), 153-184, (1996).
Priadi B, Polvé M, Maury RC, Bellon H, Soeria-Atmadja R, Joron JL & Cotten JJ, J Southeast Asian Earth Sci, 9, 81-93, (1994).
Chlorine is known to strongly partition into hydrous fluids evolving from water-saturated magmas. Accordingly, chlorine is a useful tracer to monitor degassing processes in magmatic systems. However, the behavior of the heavy halogens bromine and iodine in these systems is virtually unknown. These elements could be particularly sensitive to the redox conditions prevailing during degassing, as they might be relatively easily oxidized from Br- and I- to Br2 and I2. However, the influence of magmatic degassing or hydrothermal fluid-magma interaction on the fractionation of Cl, Br, and I is still unknown. In order to constrain the behaviour of heavy halogens in magmatic-hydrothermal systems, we have experimentally measured the partition coefficients of Cl, Br and I between silicate melts and aqueous fluids. The starting materials were a glass of albitic composition (NaAlSi3O8) and solutions of NaCl, NaI, and NaBr with defined concentrations (from 0.1 to 20 g/l). Powders of glass and equal amounts of solution containing one of the halogens were sealed in Pt capsules. The experiments were run in rapid quench autoclaves at 900oC, 2 Kbars total pressure during a few days in order to reach equilibrium. The composition of the resulting glasses are measured with electron microprobe in those cases were the concentrations of the halogens were relatively large. For the low concentrations, with a magnitude of a few ppm, Br contents were measured using a nuclear probe (PIXE = Protons Induced X-Ray Emission). The concentrations in the fluid were calculated by mass balance and fluid/melt partition coefficients D were calculated according to the equation D = (C)fluid/(C)melt, where C is the concentration of Cl, Br and I.
Oligocene peralkaline volcaniclastic layers (VLs) are regionally extensive throughout the Tertiary Piemonte Basin (TPB, NW Italy). These layers represent one of the few records of peralkaline volcanism in the Alpine/Apennine belt; their study is thus crucial in the comprehension of Tertiary collisional volcanism. This contribution deals with their mineralogy, geochemistry, biostratigraphy, and zircons morphology in order to asses the nature and the origin of these volcaniclastic products.
The VLs occur within the Oligocene marly successions of the TPB. They are typically graded, lens shaped bodies, whose thickness ranges from 10 to 180 cm. Sedimentological studies suggest that these deposits are the reworked remnants of airfall volcanic ashes.
Biostratigraphical data based on calcareous nannofossil associations allow to refer the VLs to Zone CP 18 (Lower Oligocene) and CP 19a (Upper Oligocene) of Okada and Bukry (1980), in agreement with 39Ar/40Ar geochronological data (see d'Atri et al., 1999).
The VLs underwent an almost complete devetrification and diagenetic transformation with neoformation of smectites and zeolites, including uncommon Ba-zeolites.
Main volcanic minerals still recognizable are plagioclase rimmed by anorthoclase, sanidine, Ti-rich Fe-phlogopite, augite rimmed by aegirine-augite, kaersutitic hornblende and Fe-Ti oxides. The originary glass shards are almost completely altered in Ba-zeolites.
On magmatic discrimination diagrams these layers plot in the trachyte and the phonolite field, all samples having Nb/Y ratios greater than 2.0, that is typical of peralkaline magmas. A plot of Nb against Zr shows a distribution in the field of peralkaline composition. Chondrite normalized REE plots of these layers show a typical peralkaline patterns, with enrichment of light REE and a depletion of heavy REE.
Zircon crystals morphology, that reflects chemical composition and crystallization trend of magma, is in good agreement with petrographical and geochemical data. In fact, the zircon populations show a predominance of high temperature (>900°C) types, with well developed {211} pyramid, compatible with a trachytic and/or phonolitic magma.
The data exposed above suggest that these volcaniclastic layers represent the remnants of an extreme differentiation of calc-alkaline magmas related to collisional Alpine magmatism.
d'Atri et al., J. Conf. Abs., 4, (1999)
Okada H & Bukry D, Mar. Micropaleont, 5, 321-325, (1980).
The Plio-Quaternary Isparta volcanic zone, Turkey, is a component of the Eastern Mediterranean magmatic belt. At the apex of the so-called "Isparta Triangle", the volcanic activity is related to large scale extensional or collapse tectonic episodes that affected the whole Aegean domain after collision occurred between the African and European plates. Three major periods of activity have been recognised. The oldest one includes lava flows, lahars, domes and dykes yielding various compositions ranging from lamprophyre to trachyte. The two more recent periods of activity were confined to the area occupied by the Gölçuk volcano. After a period of effusive activity, major explosive events produced large pyroclastic deposits. The lower pyroclastic flows have trachytic compositions and are interlayered with paleosoils and lacustrine formations. The upper part of the Gölçuk volcano is made up by trachytic ash and pumice fall deposits alternating with base-surge deposits containing ballistic blocks, produced by phreato-Plinian to hydromagmatic eruptions of a maar-type tuff ring volcano. A particular attention was paid to a 10 meter-thick pumice deposit which contains lava debris and numerous xenoliths of country rocks (limestones and flysch) and also of plutonic fragments, that correspond to the result of sampling of the basement and of the top of the magma reservoir by downward penetration of the diatreme root of the volcano. The majority of the plutonic fragments are cumulates that yield ultramafic to mafic parageneses, whereas some samples are more evolved monzonite and syenite. Plutonic rocks are interpreted as comagmatic and affords a fine opportunity to test the nature of differentiation processes. Mixing calculations suggest that the series evolved through fractionation of pyroxene + plagioclase + apatite + biotite + K-feldspar, with minor amphibole, Fe-Ti oxides and titanite, and that evolution from the strongly silica-undersaturated absarokite and lamprophyre to the silica-saturated and oversaturated trachyte is controlled mainly by biotite precipitation. Whole rock compositions of the Gölçuk volcanic series yield higher contents of LILE, LREE relative to HREE, and HFSE, than the high-K arc series and are very similar to the continental rift-related K-alkaline series, such as the Virunga volcanoes, Rwanda. However, the Ta-Nb and Zr-Hf negative anomalies displayed by spidergrams suggest that the mantle source was metasomatised by fluids released from paleosubducted slab(s). We propose that the likely source for the high-K Isparta volcanic series lies in the subcontinental lithospheric mantle previously metasomatised by earlier subductions having affected the Aegean domain.
Numerous plutonic bodies constitute the Kaçkar batholith in an E-W rectangular area, 30-40 km in width and 400 km in length, between Ordu and Artvin towns, 20-30 km far away from the Black Sea coastal line in the eastern Black Sea region of Turkey. These discrete bodies are exposed to be emplaced within the Upper Cretaceous volcano-sedimentary unit. This paper deals with the Altiparmak Dagi and Soganli Dagi part of the Kaçkar batholith where the average altitude ranges from 2.000 m to 4.000 m. The Kaçkar batholith intrudes the basic volcanics and transformes them into albite - epidote - tremolite/actinolite hornfelses in addition to some skarn type ferriferous occurrences in this region. During the geological mapping to the scale of 1/25.000, the Kaçkar batholith in this area has been subdivided into four lithodem units such as Ayder K-feldspar megacrystalline monzogranite, Halkalitas quartz monzodiorite, Marselavat granodiorite and Sasmistal microgranite. Mineralogical-petrographical and geochemical studies reveal that there are three distinctive magmatic stages in this part of the batholith. The first stage with the characteristics of CAFEM/CALK and I-type has yielded the Ayder K-feldspar megacrystalline monzogranite. The Halkalitas and Marselavat units, representing the CAFEM/CALK and I-type features similar to that of first stage, have been derived from the fractional crystallisation of a single magma source. The leucocratic Sasmistal microgranite is formed from the third magmatic stage which shows an ALUM and S-type compositional feature. Among these magmatic stages, the second and third stages are considered to be part of Eastern Black sea arc magmatism and syn-collisional magmatism, respectively. As for the first stage, the continental crustal contribution is dominant in its genesis. It can be suggested to be part of arc magmatism because of having been intruded by the second stage. However, it should be pointed out that its certain setting could only be clarified by means of absolute age dating in future works.
This work was supported by the Research Fund of The University of Istanbul. Project number: T-212/050396
The rocks in the formation under consideration build elongated magmatic bodies of NNW- SSE strike along the fault ruptures particularly along the eastern margin of the Vardar zone (Lojane-Stip-Serta-Gradeska Planina-Plaus-Furka), seldom in the zone itself (Gradec-Gurnicet). The magmatic bodies intrude the Precambrian complex, the rocks of the Paleozoic Veles series and the Mesozoic volcanogenic sedimentary formation as well as the Dren gabbro-diabase massif (Ivanov et al., 1987).
Monzonitic granites grading into granodiorites or real granites prevail in the formation. The younger granitic phase characterized by intrusions into the older granites is present as leucocratic granites. Generally, there is tendency the most acidic varieties to occur in the area further south (poorer in biotite and richer in quartz), although the mineralogical composition and the chemistry are similar. Locally, granodiorites grade into dioritic or syenitic varieties (near Stip and Lojane) or most commonly into granodiorites with the occurrence of hornblende due to assimilation of surrounding rocks.
The mineralogical composition of the rocks is fairly uniform and consists of quartz, K-feldspar (orthoclase, seldom microcline), plagioclase (with about 25 to 35% anorthite or albite in leucocratic varieties), biotite and seldom muscovite. Accessory component parts are rare.
A characteristic feature of these magmatites is the warming of surrounding rocks and penetration of fluids rich in alkalis. This contact metamorphism possesses features of widespread intense migmatisation. The process ensued wide belts of biotite-cordierite (sillimanite)-andalusite gneisses, amphibolite-pyroxene schists, gneisses and similar rocks. It is worth mentioning that this type of metamorphism is related to the eastern belt of these granitoids, whereas the western belt, such as Dren (Bebien, 1982) and partially Lojane are characterized by lower metamorphism than the surrounding rocks.
The consolidation of these magmatites, based on isotopic geochronological analyses, took place either at the end of Jurassic or the beginning of Cretaceous (Marakis, 1970). This coincides well with data obtained in the terrain where the granites of Gradec and Grunicet intruded Jurassic basites, the granites occurring as alluvium (the Stip granites) in Alb-Cenomanian conglomerates near Mocarnik and the village of Goracino.
Bebien J, Bull. Soc. Geol. Fr, 35, 229-242, (1982).
Ivanov T, Misar Z, Boves D, Dudec A, Dumurganov N, Jaros J, Jelimak E, Pacesava M, Ofioliti, 12/3, 457-487, (1987).
Marakis G, Am. Geol. Pays Hell, 21, 121-152, (1970).
Karadag is a small strato-volcano covering an area of 220 km2 located at the inner border of Taurus chain, north of Karaman city. The main volcanic activities occurred during Plio-Quaternary period. The volcano which base is 1050 m above sea level has a horse-shoe caldera at its summit (2300 m). The main volcanic features are lava domes, lava flows, dikes, ignimbrites, air-fall deposits and phreatomagmatic deposits.
Karadag volcanic rocks exhibit generally potassic calc-alkaline features. They are basaltic andesite, basaltic trachyandesite, andesite, trachyandesite, dacite and trachydacite in composition and, consist primarly of plagioclase, amphibole, clinopyroxene, biotite, olivine and quartz phenocrysts. All samples are quartz normative. Reverse zoning in plagioclase, clinopyroxene, amphibole and olivine phenocrysts, and existence of olivine and quartz in the same samples as indicated by mineral chemistry show magma mixing. Variation diagrams of SiO2 with major- and trace- elements are consistent with fractional crystallization process. High 87Sr/86Sr (0.70551-0.70755) and low 143Nd/144Nd (0.51240-0.51264) ratios suggest crustal involvement in their petrogenesis. These data suggest various processes such that magma mixing, fractional crystallization and crustal contamination were effective in the formation of the Karadag volcanic rocks.
The Karadag strato-volcano is located near the southern end of a NNW-striking fault which turns to SE and divides the volcanic edifice into two parts. This fault is probably the result of an extensional tectonic regime which occurred within Anatolian plate due to the collision of Eurasian-Arabian plates during the evolution of Turkey in the Neotectonic period.
In northwestern Anatolia, the continent-continent collision occurred during the late Cretaceous-early Tertiary period following the total elimination of the Neo-Tethyan ocean floor. After the collision, the continental convergence and the compressional tectonism lasted to the Miocene. During this period, the crust was shortened and overthickened. After the collision, a severe magmatic activity developed during the Oligocene-early Miocene (25-15 ma). The magmatic rocks crop out extensively in the Biga peninsula.
The magmatism started with the emplacement of high-level, sub-volcanic granitic plutons (i.e. the Evciler, Ezine, Kozak, and Eybek plutons) and the associated hypabyssal rocks. They were followed by the partly co-eval volcanic associations. Thus, the magmatic rocks of this period are closely associated with each other in time and space. The volcanic association begins with the felsic products and pass up into intermediate volcanic rocks.
The geochemical and isotopic features of the magmatic rocks are similar. They are high-K, calc-alkaline and hybrid displaying the enrichment of LIL elements over the REE. They are depleted by Nb, and enriched by Ce, Pb, La. The geochemical features of the magmatic rock indicate that they were genetically associated, and derived from a similar source. Their geologic, petrographic and geochemical aspects are comparable with the collision-related hybrid magmas, as in the cases of the Eastern Anatolia and the Tibetan-type environments.
The studied volcanic rocks, ranging in composition from andesites through dacites to rhyolites, showing calc-alkaline affinity, were emplaced in several basins of the Rhodope Massif and Circum Rhodope Belt during the Upper Eocene to Lower Miocene.
Major and trace element geochemical data indicate that although all the studied volcanic rocks show typical volcanic arc characteristics, namely high LILE/HFSE ratios, Ti and Nb depletion, they also show remarkable compositional variation, analogous to those observed across- or along-arc in modern subduction zones, especially those involving ocean-continent collision. That compositional variation can not be explained by a simple fractional crystallisation process of a single magma chamber. Rather, a number of chemically distinct magma batches have evolved separately and source composition and mixing are the dominant petrogenetic controls.
Geochemical variations can mainly be attributed to a heterogeneously enriched, due to the subduction of an inhomogeneous oceanic slab, upper mantle source which underwent partial melting to various degrees, and to prolonged differentiation within the continental crust. Fractional crystallisation of mafic minerals and plagioclase was the main shallow level differentiation processes. Wall-rock contamination also contributed to the compositional variations.
The subducted oceanic slab may be represented by remnants of the Vardar Ocean which, at the time of the volcanic activity within the continental Rhodope promontory of the Eurasia, had already closed. Thus, the descending slab should be composed mainly of oceanic basic rocks, while subordinate amounts of continental crustal rocks possibly are existed. In an extensional regime, hot asthenospheric magma with mantle wedge and Vardar subduction components uplifted, differentiated during its ascent and finally extruded in rift bounded basins, which had already developed, forming the present volcanics in the area of North-Eastern Greece.
The Alpine magmatism in Bulgaria is generally related to different stages in evolution of Eurazian plate margin. The beginning is represented by Triasic, basic and intermediate volcanics. They are related to initial embrional rifting in Moesian platform. During the Upper Cretaceous intensive subduction - related magmatism formed Srednogorie volcano-intrusive zone. Collisional magmatic evidences with Late Paleogene age have been located in Rhodope Belt. Neogene intraplate volcanism marked the end of Alpine magmatism in Bulgaria (Dabovski et al, 1991). In East Rhodope - Egeian Trace region magmatic evidence migrated from North to South. The continent - continent collision beginning in Paleogene, and magmatic activity started long after that in Priabonian, and it ends by Upper Oligocene (39 - 24 Ma). This magmatism began after the main phase of Europa-Apulia continental collision in a lithospheric sector which had played the role of non-volcanic fore - arc during the closure of a Thetyan ocean, which northward subduction produced the Upper Cretaceous igneous activity in Srednogorie island arc. The collision related magmatism starts after an important phase of thickening/uplift and subsequent extensional collapse of Alpine chain (Yanev et al, 1998). The main phases are Priabonian - Lower Oligocene, the south ones are of Oligocene age. The rocks of North zone consist of Shoshonitic (rearely ultrapotasic) and high-K calcalkaline series, in the South they have high-K calcalkaline and calcalkaline characteristics. The NE-SW intrusive belt in East Rhodopes consists of gabbro-diorites, diorites, monzonites, quarz-monzonites, sienites with predominance of monzonites. This magmatism is related with first two phase of Volcanic activity in this area. It is presented by 6 intrusive bodies: Briastovo, Yabalkovo, Surnitsa, Karamantsi, Yolen dere, and Pesnopoi. They form Northeast - Southwest intrusive belt. This rocks were intruded in some volcanic structures (Briastovo, Yabalkovo, Surnitsa), volcanogenic-sedimentery and sedimentery (Karamantsi) or metamorphic (Yolen dere, Pesnopoi) rocks. This bodies outcrop along NE-SW trending narrow belt that extends for 45 km. The intrusive bodies of East Rhodopes, are belived to be of Priabon - Oligocen age. The northernmost (Yabalkovo 35.79 ± 1.41 Ma) marked the beginning of this intrusive magmatisum, that shows Shoshonitic and high-K calcalk-alkaline affinities.
Dabovski C, Harkovska A, Kamenov B, Mavrudchiev B, Stanisheva-Vasileva G & Yanev Y, Geologica Balcanica, 21, 3-15, (1991).
Yanev Y, Innocenti F, Manetti P & Serri G, Acta Vulcanologica, 10 (2), 279-291, (1998).
The Eastern Rhodopes Paleogene (Priabonian-Oligocene) magmatism took place after the complete closure of the Tethys ocean by microcontinents, derived from the African plate. This magmatism is bimodal: acid and intermediate in equivalent volumes with a few basalts. According to the contents of some trace elements these volcanics heve the typical features of the collision-related magmas f.i. after their Rb content (Pearce et al. diagram, 1984) and Nb/Zr ration (Thieblmont & Teguey diagram, 1994). One hypothesis is proposed for the origin of this magmas: they are generated from a metasomatized mantle on an earlier (Upper Cretaceous) ocean subduction (Yanev et al., 1995). The melting is probably provoked by raising of a thermal diapir as a result of break-off and sinking in the lower mantle of the subducted during the collision continental plate.
The presence of basalts, even thought in small quantity, as closest to the primitive magma, allows to characterize their eventual mantle source. The spidergrams, drawn up on the base of a FMM (Fertile MORB Mantle after Pearce & Parcinson, 1993) normalization, are characterized by high normalized values of Nb and Zr (but lower then these of OIB derived magmas), very low values of Cr, and especially of Ni. VHI (very high incompatible elements) >> HI (high incompatible) >MI (moderately incompatible). On the base of these data it could be supposed that the Åastern Rhodopes collision-related magmas are derived from no depleted FMM source with a low melting percentage (about 5%), similar to some subduction-related magmas (f.i. N. Fiji Basin and Scotia Sea). This confirms the hypothesis of their origin from metasomatized mantle.
Pearce JA, Harris NB & Tindle AG, J. Petrol, 956-983, (1984).
Pearce JA & Parkinson IJ, Geol. Soc. Spesial Pub, 76, 373-403, (1993).
Thieblmont D & Tegyey M, CR Acad. Paris, 319-II, 87-94, (1994).
Yanev Y, Innocenti F, Manetti P & Serri G, Proc XV Congr. CBGA, Athens, 4, 578-583, (1995).
Eocene nannofossil oozes and chalks recovered during Ocean Drilling Program Leg 171B at Blake Nose in the western Central Atlantic, off Florida, contain several ash layers that provide good markers for a site to site correlation. We characterize the ash layers sedimentologically, petrologically and geochemically to reconstruct the depositional environments of the volcaniclastic strata and to identify identical layers for correlation and temporal calibration of the sites along a passive margin depth transect. There is evidence that the volcaniclastic-rich ash layers comprise both, fall-out deposits with thicknesses around 1 cm, and turbidites with thicknesses of max. 2 cm. The geochemical discrimination of the differentiated dacitic-rhyodacitic ashes reflects a clear volcanic island arc origin, based on contents of trace elements. This points to a volcanic source in the Caribbean region, probably on, or close to, the Greater Antilles, where subduction and volcanism were active at least until the Middle-Late Eocene. Island arc-related Paleogene volcanic rocks from the the Pilon and El Cobre formations in the Sierra Maestra (eastern Cuba) comprise lava flows and pyroclastic rocks from an explosive volcanism with basaltic to rhyodacitic tholeiitic composition. Both types have been sampled and geochemically compared to the results obtained from the Blake Nose transect. Although the Cuban samples display a more basic composition, we assume that explosive volcanism and air-borne northward transport of pyroclastics provided the source for the volcaniclastic layers of Blake Nose. We discuss potential transport and sedimentation mechanisms and compare these with results from General Circulation Models.
The Joshua Flats - Beer Creek - Pluton (JBP) is situated at the southern end of the Inyo batholith in the White Inyo Mts (eastern Califonia), where it intruded a Neoproterozoic to Lower Cambrian metasedimentary sequence c. 180-160 Ma ago (McKee & Conrad, 1996). The intrusion consists of three distinct igneous phases: a monzodiorite (Marble Canyon Monzodiorite MCD), a quartz monzonite (Joshua Flats Quartzmonzonite JFM) and a granodiorite (Beer Creek Granodiorite BCG). Field relations suggest an intrusion sequence MCD -> JFM -> BCG. Structures in the pluton and its contact aureole document multiple emplacement mechanisms for the JBP, namely stoping, downward return flow, partial melting of host rock material and horizontal shortening of stratigraphic units. AMS, quartz-c-axes and strain measurements support the field observations and stress the significance of vertical material transfer (i.e. downward return flow and stoping) as the most important space-making process for the pluton. Contact metamorphism around the JBP reached hornblende-hornfels to lower amphibolite facies conditions (Ernst, 1996). Thermobarometric measurements were carried out at igneous rocks using the hornblende-plagioclase-thermometer (Blundy & Holland, 1990) and the Al-in-hornblende-barometer (Anderson & Smith, 1995) and at pelitic rocks from the aureole using the Na-in-cordierite-thermometer (Kalt et al., 1998). Hornblende crystallisation conditions were determined as T = 664 - 682°C and P = 2.30 - 2.52 kbar. The Na-in-cord-thermometer gave c. 730°C as the peak metamorphic temperature in the innermost aureole (0 - 500 m away from the contact) Such high temperatures in the aureole can be reached only by (fluid supported) convective heat flow and not solely by conduction. Further observations document the important rôle fluids played during the emplacement of the JBP: the increasing metamorphism towards the pluton is accompanied by dehydration reactions leading to the breakdown of chlorite, muscovite and biotite and to dehydration melting (documented by leucocratic veins) in the immediate contact area due to the mineral reaction sillimanite + biotite + plagioclase + quartz -> cordierite + alkali feldspar + melt; extensive copper mineralisations in marbles as well as skarns; myrmecites; numerous epidote veins; retrograde chloritisation of andalusite and biotite as well as extensive pinitisation of cordierite in pelitic rocks. Fluids derived from both the aureole and (Carrington & Harley, 1996) the pluton were probably important for the emplacement of the JBP, because the presence of a free fluid phase might have enhanced the effectiveness of the downward return flow. An unsolved problem in this scenario are relatively low contents of channel volatiles in cordierite (H2O = 0.76 - 0.85 weight% and CO2 = 0.06 - 0.13 weight%) , determined by IR spectroscopy measurements, which might be due to the equilibration of cordierite with a volatile undersaturated melt.
Anderson JL & Smith DR, Am. Mineral, 80, 549-559, (1995).
Blundy JD & Holland TJB, Contrib. Mineral Petrol, 104, 208-224, (1990).
Ernst WG, GSA Bulletin, 108, 1528-1548, (1996).
Carrington DP & Harley SL, Geology, 24, 647-650, (1996).
Kalt A, Altherr R & Ludwig T, J. Petrol, 39, 663-688, (1998).
McKee EH & Conrad EC, GSA Bulletin, 108, 1515-1527, (1996).
The Continental Flood Basalts from Putorana belong to the permian Siberian CFB platform and are located in the north-central of Siberia. The platform covered an area of about 1.500 000 km2 . The CFBs consist of columns and massive layers with a thickness of up to 2500 m which erupted in a very short period of 600 000 a (Campbell et al., 1992).
We studied in detail two stratigraphic profiles in N 69°24', E 93°29' and in N 69° 48', E 93°49'. They consist of 19 and 13 lava flows respectively, with variable thickness ranging from few meters to tens of meters. Lava flows are mainly typical tholeiitic basalts but also tuffs occur occasionally. In the second profile basalts are overlying a sequence of alkali picrites.
The tholeiitic basalts contain olivine, clinopyroxene, plagioclase, Ti-magnetite and ilmenite with ophitic to subophitic texture. Plagioclase accumulates appear in most of the samples. The flows are aphyric to porphyric. The upper parts of the massive flows have a high portion of amygdaloidal material. The alteration of the samples is moderate.
Major element chemistry shows that the Putorana CFBs have undergone significant amounts of fractionation of olivine, plagioclase and pyroxene. With a SiO2 content in a range of 50.2-47.5 wt% and MgO values of 8.0-5.8 wt% (# Mg in a range of 36.1-26.6) the tholeiitic basalts of Putorana are products of a highly evolved magma. They belong to the LPT-(Low-Phosphor-Titanium) basalts (P2O5 0.16 wt%, TiO2 1.51 wt%). Incompatible trace elements (normalized to the primitive mantle) show a pronounced negative anomaly of Nb and Ta and elevated Th relative to the primitive mantle as an evidences for a crustal contamination. Chondrite normalized REE show relatively flat patterns suggesting that these lavas arose from high degree melt fraction of the parental magma. In addition the negative Eu-anomaly confirms that lavas have undergone significant plagioclase fractionation. The very uniform geochemistry of major and trace elements indicates, that all flows from both profiles originated from a common parental magma which in turn were inherited directly from a uniform plume source as suggested by Sharma et al., 1991, 1992.
Campbell IH, Czamanske GK, Federenko VA, Hill RI, Stepanov V., Kunilov VE, Science, 258, 1760-1763, (1992).
Sharma M, Basu AR, Nesterenko GV, Earth Planet. Sci. Lett., 113, 365-381, (1992).
Sharma M, Basu AR, Nesterenko GV, Geochim. Cosmochim. Acta, 55, 1183-1192, (1991).
Index of EUG 10 Volume
Further EUG 10 Information
Index of the Journal of Conference Abstracts
Cambridge Publications Home Page
Last Updated on Wednesday, March 17, 1999.
© 1997 Cambridge Publications