Scholarly article on topic 'Magmatism at the Eurasian–North American modern plate boundary: Constraints from alkaline volcanism in the Chersky Belt (Yakutia)'

Magmatism at the Eurasian–North American modern plate boundary: Constraints from alkaline volcanism in the Chersky Belt (Yakutia) Academic research paper on "Earth and related environmental sciences"

CC BY-NC-ND
0
0
Share paper
Academic journal
Lithos
Keywords
{"Alkaline volcanism" / "Chersky seismic belt" / "Eurasian–North American plate boundary" / "Arctic Ocean spreading" / "Asthenospheric adiabatic decompression melting"}

Abstract of research paper on Earth and related environmental sciences, author of scientific article — Cornelius Tschegg, Michael Bizimis, David Schneider, Vyacheslav V. Akinin, Theodoros Ntaflos

Abstract The Chersky seismic belt (NE-Russia) forms the modern plate boundary of the Eurasian−North American continental plate. The geodynamic evolution of this continent−continent setting is highly complex and remains a matter of debate, as the extent and influence of the Mid-Arctic Ocean spreading center on the North Asian continent since the Eocene remains unclear. The progression from a tensional stress regime to a modern day transpressional one in the Chersky seismic belt, makes the understanding even more complicated. The alkaline volcanism that has erupted along the Chersky range from Eocene through to the Recent can provide constraints on the geodynamic evolution of this continental boundary, however, the source and petrogenetic evolution of these volcanic rocks and their initiating mechanisms are poorly understood. We studied basanites of the central Chersky belt, which are thought to represent the first alkaline volcanic activity in the area, after initial opening of the Arctic Ocean basin. We present mineral and bulk rock geochemical data as well as Sr–Nd–Pb–Hf isotopes of the alkaline suite of rocks combined with new precise K–Ar and 40Ar/39Ar dating, and discuss an integrated tectono-magmatic model for the Chersky belt. Our findings show that the basanites were generated from a homogeneous asthenospheric mantle reservoir with an EM-1 isotopic flavor, under relatively ‘dry’ conditions at segregation depths around 110km and temperatures of ~1500°C. Trace element and isotope systematics combined with mantle potential temperature estimates offer no confirmation of magmatism related to subduction or plume activity. Mineral geochemical and petrographical observations together with bulk geochemical evidence indicate a rapid ascent of melts and high cooling rates after emplacement in the continental crust. Our preferred model is that volcanism was triggered by extension and thinning of the lithosphere combined with adiabatic upwelling of the underlying mantle at 37Ma. This suggests that at that time, rift tectonics in the Mid-Arctic Ocean most likely had also affected the North-Asian continent, causing volcanic activity in the Chersky belt, before the regional geodynamic regime changed from a tensional to compressional. Our conclusions contribute not only to the understanding of volcanism in the Chersky seismic belt (NE-Russia) but also to general aspects of plate dynamics between the Eurasian and North American continent.

Academic research paper on topic "Magmatism at the Eurasian–North American modern plate boundary: Constraints from alkaline volcanism in the Chersky Belt (Yakutia)"

ELSEVIER

Contents lists available at ScienceDirect

Lithos

journal homepage: www.elsevier.com/locate/lithos

Magmatism at the Eurasian-North American modern plate boundary: Constraints from alkaline volcanism in the Chersky Belt (Yakutia)

Cornelius Tschegg a,*< Michael Bizimis b, David Schneiderc, Vyacheslav V. Akinin d, Theodoros Ntaflos a

a Department of Lithospheric Research, University of Vienna, Vienna, Austria b Department of Earth and Ocean Sciences, University of South Carolina, Columbia, SC, United States c Department of Earth Sciences, University of Ottawa, Ottawa, Canada

d North East Interdisciplinary Scientific Research Institute, Russian Academy of Science, Magadan, Russia

ARTICLE INFO

ABSTRACT

Article history: Received 3 March 2011 Accepted 30 April 2011 Available online 7 May 2011

Keywords: Alkaline volcanism Chersky seismic belt

Eurasian-North American plate boundary Arctic Ocean spreading Asthenospheric adiabatic decompression melting

The Chersky seismic belt (NE-Russia) forms the modern plate boundary of the Eurasian—North American continental plate. The geodynamic evolution of this continent—continent setting is highly complex and remains a matter of debate, as the extent and influence of the Mid-Arctic Ocean spreading center on the North Asian continent since the Eocene remains unclear. The progression from a tensional stress regime to a modern day transpressional one in the Chersky seismic belt, makes the understanding even more complicated. The alkaline volcanism that has erupted along the Chersky range from Eocene through to the Recent can provide constraints on the geodynamic evolution of this continental boundary, however, the source and petrogenetic evolution of these volcanic rocks and their initiating mechanisms are poorly understood.

We studied basanites of the central Chersky belt, which are thought to represent the first alkaline volcanic activity in the area, after initial opening of the Arctic Ocean basin. We present mineral and bulk rock geochemical data as well as Sr-Nd-Pb-Hf isotopes of the alkaline suite of rocks combined with new precise K-Ar and 40Ar/39Ar dating, and discuss an integrated tectono-magmatic model for the Chersky belt. Our findings show that the basanites were generated from a homogeneous asthenospheric mantle reservoir with an EM-1 isotopic flavor, under relatively 'dry' conditions at segregation depths around 110 km and temperatures of ~1500 °C. Trace element and isotope systematics combined with mantle potential temperature estimates offer no confirmation of magmatism related to subduction or plume activity. Mineral geochemical and petrographical observations together with bulk geochemical evidence indicate a rapid ascent of melts and high cooling rates after emplacement in the continental crust. Our preferred model is that volcanism was triggered by extension and thinning of the lithosphere combined with adiabatic upwelling of the underlying mantle at 37 Ma. This suggests that at that time, rift tectonics in the Mid-Arctic Ocean most likely had also affected the North-Asian continent, causing volcanic activity in the Chersky belt, before the regional geodynamic regime changed from a tensional to compressional. Our conclusions contribute not only to the understanding of volcanism in the Chersky seismic belt (NE-Russia) but also to general aspects of plate dynamics between the Eurasian and North American continent.

© 2011 Elsevier B.V. All rights reserved.

1. Introduction

The Chersky range in NE-Russia delineates the plate boundary between the North American and Eurasian continents (e.g. Fujita et al., 2009; Fig. 1). It is not only one of the seismically most active belts in the world, but also one of the least understood continental convergent plate margins (e.g. Parfenov et al., 1988). Numerous workers have attempted to characterize the geodynamic evolution of NE-Russia, with an emphasis on geophysical constraints focusing on this tectonically active zone (cf. Fujita et al., 2009; Grachev, 2003;

* Corresponding author. E-mail address: cornelius.tschegg@univie.acat (C. Tschegg).

0024-4937/$ - see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.lithos.2011.04.008

Nokleberg et al., 2000). One early model postulated that the spreading of the Arctic mid-ocean ridge (Gakkel Ridge) propagated into the Eurasian continent, forming the Eurasian-North American plate boundary (e.g. Grachev et al., 1970). This continental rift model has been a matter of debate, since a growing body of evidence supports a modern geodynamic setting characterized by compressional and transpressional stresses (e.g. Cook et al., 1986; Franke et al., 2000; Parfenov et al., 1988). A switch from an extensional Arctic ridge related geodynamic framework to a compressional one during the Eocene is now widely accepted for the Chersky area (Nokleberg et al., 2000 and references therein), the timing, however, remains equivocal. Widely distributed alkaline volcanism through the Cenozoic to Recent times, in and adjacent to the Chersky range, has been described as 'rift-related' and was linked to the spreading of the Gakkel Ridge and

Fig. 1. Schematic geological map of NE-Russia showing main tectonic and magmatic units, as well as localities mentioned in the text (Fujita et al., 2009; Nokleberg et al., 2000). The different gray shadings and hatchings comprise the main tectonic units; the investigated outcrop is indicated with an arrow. The inset illustrates the present-day stress regime in the area.

its propagation into the North Asian continent (Fujita et al., 1997 and references therein). Alternatively, the geodynamics at this plate boundary setting combined with its regional magmatic nature can be related to active, incipient plume-induced rifting (Grachev, 2003), instead of passive continental rifting. Additional, integrated and more profound geochemical and petrological data are required to further shed light on these competing models.

In this paper, we present new data for the timing, source and petrogenesis of the first known volcanic alkaline occurrence after the initiation of Early Eocene rifting along the Arctic mid-ocean ridge. Located in the Republic of Yakutia (NE-Russia), this occurrence was first discovered and described by Surnin et al. (1998) and called 'Volcano Ruditch'. The authors described the rocks from the region as alkali basanitic K-Na series, which contain abundant mantle xenoliths as well as eclogite xenoliths and augite/anorthoclase megacrysts. A whole-rock 40Ar/39Ar age of 36.65 ±0.16 Ma from these volcanic rocks has been reported in Layer et al. (1993), but without the corresponding methodological techniques, data tables or age spectrum. Apart from this date, no additional geochemical data were published on the volcanic rocks from Ruditch. In this contribution, we present mineral geochemical data on these alkaline rocks, including bulk geochemical major and trace elements, Sr-Nd-Pb-Hf isotopes, and K-Ar and 40Ar/39Ar geochronology; we then present a revised tectono-magmatic history of the region, integrating existing data. We address whether the volcanism in this highly debated area is related to active or passive rifting, whether it is influenced by plume or subduction activity and when it exactly occurred.

2. Geological and tectonic setting

Mantle xenolith-bearing alkali-basaltic remnants of a volcanic plug were sampled in Ruditch (N: 63° 44' 05", E: 143° 06' 33"), a locality in the vicinity of the Indigirka River, 90 km south of the village Ust-Nera (Yakutia Republic, Russian Far East; Fig. 1). The area is part of the Chersky Range, which is mainly built up by crustal thickening and arc-related Late Jurassic granites (Akinin et al., 2009; Layer et al., 2001). Those intruded into Triassic-Jurassic turbidite sequences during the Late Jurassic to Early Cretaceous collision of the Kolyma-Omolon superterrane (KO) against the Verkhoyansk (North Asian) Craton margin (Parfenov et al., 2009). The North Asian Craton (NAC) consists of an Archean to Proterozoic metamorphic basement, covered by undeformed Late Precambrian to Mesozoic sedimentary and volcanic successions. The Verkhoyansk fold and thrust belt (VFaTB) forms the eastern border of the North Asian Craton, comprising Devonian to Jurassic passive continental margin deposits. The Kolyma-Omolon superterrane originated primarily from the North Asian Craton, which rifted away in the Late Devonian to Early Carboniferous, before accreting again in the Late Jurassic. It consists of an Archean to Jurassic tectonic framework of continental margin, island arc and ophiolite terranes. Between the Kolyma-Omolon terrane and the Verkhoyansk fold and thrust belt lays the Verkhoyansk-Kolyma (VK) collage, a Late Paleozoic to Early Jurassic amalgamation zone, composed of distal formations of the passive continental North Asian margin (Parfenov et al., 2009). The Indigirka-Oloy (IO) assemblage crops out mainly on the margin of the Kolyma Omolon superterrane and consists of volcanic-plutonic belts

and numerous sedimentary basins of Late Jurassic to Early Cretaceous age (Nokleberg et al., 2000). In the Chersky Range, the Indigirka-Oloy assemblage is represented by the Uyandina-Yasachnaya arc, which was formed during migration of the KO superterrane towards the NAC and the subduction of an intermediate ocean basin. The Mid-Cretaceous to Paleogene Okhotsk-Chukotka Volcanic Belt forms a major Andean-type continental margin arc and extends over the modern southern continental border of Russian Far East to western Alaska. It is related to the subduction of the Paleo-Pacific ocean plate during Cretaceous beneath the modern North American continent. The Chersky Range forms the main collision belt, between the KO and the NAC superterrane and it is built up by a Late Jurassic to Early Cretaceous arc of plutonic and volcanic rocks (Layer et al. 2001).

The present-day North America-Eurasia plate boundary trends from the northern termination of the Mid-Atlantic Ridge to the southern margin of Russia's Far East, defined by a diffuse but highly active seismic zone (Fujita et al., 2009). Rifting along the Arctic mid-ocean ridge or Gakkel Ridge (north extension of the Mid-Atlantic ridge) initiated at about 50 Ma and extended through the Laptev Sea to the modern Russian Far East continent (Vogt et al., 1979). At that time, sea floor spreading extended from the Arctic Ocean, resulting in the North American-Eurasian Plate boundary and continued until the Eocene (Nokleberg et al., 2000 and references therein). The original idea of Grachev et al. (1970) and many other Soviet investigators (see Fujita et al., 1990) that the Arctic mid-ocean ridge continues into the continent, with the Moma "Rift" as logical continuation of the Gakkel Ridge, was mainly based on geophysical and structural observations. It was later disproven by many authors (e.g. Cook et al., 1986; Parfenov et al., 1988), when it was demonstrated that the focal mechanisms determined for large parts of the Chersky Belt show thrust faulting with compression directions perpendicular to Chersky Mountains (SW-NE). The compressional stress regime in the area allows thrusting with minor components of strike-slip, but without exten-sional components (Fujita et al., 1990). Recent seismo-tectonic observations from Franke et al. (2000) indicate that the rift structure terminates at the continental shelf and that the area of the Moma "Rift" is presently under compressional or transpressional stress. Already during the Cenozoic, the Euler pole position has migrated (Nokleberg et al., 2000 and references therein), triggering these changes in the stress regime of the region. However, alkaline volcanism within or adjacent to the Chersky Range has been reported from Eocene through recent, interpreted as compositionally consistent with continental rift basalts and probably being related to rifting (Fujita et al., 1997 and references therein; Grachev, 2003). Some of the Quaternary volcanic activity occurs up to historical time, and the Balagan-Tas is among one of the well-known volcanoes. Grachev (2003) describes the Quaternary volcanic activity in the area as nearly coeval fissure eruptions, which are related to incipient active mantle upwelling, more than to historic or recent geodynamic processes in the area. On the continent, the present-day extensional regime of the Gakkel Ridge has changed to transpression in the Chersky seismic belt and bifurcates into two branches, a southeastwards trending left-lateral transpressional structure towards Kamchatka, and southwards trending right-lateral structure towards Sakhalin Island (Nokleberg et al., 2000; Fujita et al., 1997, 2009).

Our sampling locality lies adjacent to the main central Chersky seismic zone, the most seismically active part of the entire belt (Fujita et al., 2009). Active faulting in the area of the Chersky Range is very complicated, preserving mainly NW-SE trending dextral and sinistral strike-slip faults as well as thrusts and axial normal faults, postulated as being probably related to the Mid-Arctic ridge spreading (Kozhurin, 2004). The crust is thinnest along the central Chersky Range (33 km on average) compared to the ~40 km crustal thickness in the adjacent areas (Fujita et al., 1997 and references therein). Mackey et al. (1998) report a crustal thickness of 35 km in the area specific to our study. The aseismicity of the southern Okhotsk sea and the large strike-slip

faults along the southern Eurasian and North-American Plate boundaries suggest the existence of an independent south-eastward extruding Okhotsk plate (Fujita et al., 2009, see inset of Fig. 1).

3. Analytical methods

3.1. Mineral composition

Mineral phases and glasses were analyzed on a Cameca SX-100 EPMA (electron-probe microanalyzer; Department of Lithospheric Research, University of Vienna) equipped with one EDS and four WDS detectors on carbon coated polished thin sections. All measurements were performed against natural standards using an acceleration voltage of 15 kV as well as a beam current of 20 nA. Feldspar, feldspathoids and glass analyses were performed using a 6 |jm defocused beam, minimizing the loss of K and Na. Olivine trace elements (Al, Ca, Ti, Cr, Ni, Mn and P) were analyzed in special runs using a 200 nA beam current at 20 kV acceleration voltage and 3 |jm beam diameter. Matrix corrections were done based on Si, Mg and Fe measured previously on the same grains. The San Carlos olivine was repeatedly analyzed to control the quality of the data. Calculated limits of detection yielded 12 ppm for Al, 28 ppm for Ca, 30 ppm for Ti, 40 ppm for Cr, 65 ppm for Mn, 85 ppm for Ni and 32 ppm for P. For element distribution maps of P, WDS scans with a resolution of 256x256,1 |am steps and 130 ms dwell-time were performed (beam conditions: 20 kV acceleration voltage and 200 nA beam current); Mg element maps were acquired with the EDS detector.

3.2. Major and trace elements

Before analysis, xenoliths and sparsely distributed xenocrysts were carefully sorted out of the crushed host rock material. Bulk major element abundances of the alkaline rocks were analyzed with an X-ray fluorescence spectrometer Philips PW2400 (Department of Lithospheric Research, University of Vienna) equipped with a Rh excitation source. Fused glass beads were produced mixing 1 part sample with 5 parts flux (Li2B4O7). Geo-reference sample GSR-3 was repeatedly measured to control accuracy and precision of the analytical procedure (< 2% relative error for all elements). Bulk trace elements were acquired on a Perkin Elmer Optima 5300DV ICP-OES (Department of Environmental Geo sciences, University ofVienna) and a Perkin Elmer Elan 6100DRCICP-MS (Department of Lithospheric Research, University ofVienna in cooperation with the Division of Analytical Chemistry, University of Natural Resources and Applied Life Sciences). Before analysis, the samples were digested in closed Teflon beakers using a mixture of HNO3-HF. Procedural errors < 2% for all elements except for Th and Pb (< 6%) were yielded, evaluated by replicate analysis of the geo-reference sample BHVO-2 and BCR-2.

3.3. Sr, Nd, Pb and Hf isotopes

Strontium and Neodymium isotopes of the alkaline rocks from Ruditch were analyzed on a ThermoFinnigan Triton TI TIMS (Department of Lithospheric Research, University of Vienna) following the method detailed described in Thoni et al. (2008). Measured standard values were 87Sr/86Sr = 0.710273±0.000003 (n=6) for NBS987 and Nd = 0.511846±0.000001 (n = 6) forLaJolla.

Lead and hafnium isotopes were analyzed at the Department of Earth and Ocean Sciences (University of South Carolina). The samples were dissolved in a distilled HF:HNO3 (3:1) mixture in Teflon vessels. Lead was extracted first using HNO3-HBr acids in anion resin (e.g. Abouchami et al., 1999). The eluent containing Hf and the bulk rock matrix was dried down repeatedly with 6 N HCL and picked up in 3 N HCl + 0.05 M ascorbic acid for the Hf separation using a modified version of the Munker et al. (2001) technique.

The Pb isotope compositions were determined on a Thermo Neptune MC-ICPMS with the Tl addition technique. The Pb sample solution was introduced with a 50 Teflon nebulizer (ESI, USA) running in a self aspirating mode, coupled to an APEX-Q(ESI, USA) system. The Pb isotope ratios were corrected for fractionation using 203Tl/205Tl = 0.418911 and the exponential law. The measured average ratios for the NBS 981 standard were: 206Pb/204Pb = 16.9367 ±0.0011, 207Pb/204Pb = 15.4915 ±0.0011, and 208Pb/204Pb = 36.6989 ± 0.0028 (2 standard deviations, n = 7, for a 40 ppb Pb solution, consuming ~30 ng Pb per run). The Pb/Tl ratio in the samples and Pb signal were kept near identical to the standard by first checking the Pb signal intensities and then adding Tl at the appropriate levels. Pb isotope ratios were then corrected for instrumental bias, based on the measured average ratios of the NBS 981 and the values reported by Todt et al. (1996).

Hf isotopes were determined with the above instrument configuration. During the course of the measurements, the JMC-475 standard was determined at 176Hf/177Hf=0.282159±0.000013 (2 standard deviations, n = 9, 36 ng runs). No additional external correction was applied to the data.

3.4. 40Ar/39Ar dating and K-Ar geochronology

Carefully separated groundmass aliquots of the same samples that have been used to analyze mineral compositions and bulk geochemistry were prepared for 40Ar/39Ar analysis. Separation techniques included hand-picking of unaltered groundmass material in the size range 0.21.0 mm. Individual separates were loaded into aluminum foil packets along with a single grain of Fish Canyon Tuff Sanidine (FCT-SAN) to act as flux monitor (apparent age = 28.03 ± 0.28 Ma; Renne et al., 1998). The sample packets were arranged radially inside an aluminum can and then irradiated for 12 h at the research reactor of McMaster University, Canada, in a fast neutron flux of approximately 3 x 1016 neutrons/cm2. CO2 laser 40Ar/39Ar step-heating analysis was carried out at the Geological Survey of Canada laboratories in Ottawa, Canada. Upon return from the reactor, samples were removed from individual packets, loaded into individual 1.5 mm diameter holes in a copper planchet. The planchet was then placed in the extraction line, and the system evacuated. Heating of sample aliquots in steps of increasing temperature was achieved using a Merchantek MIR1010 WCO2 laser equipped with a 2 mmx2 mm flat-field lens. The released Ar gas was cleaned over getters for 10 min and then analyzed isotopically using the secondary electron multiplier system of a Nu Noblesse gas source mass spectrometer; details of data collection protocols can be found in Villeneuve and MacIntyre (1997) and Villeneuve et al. (2000). Error analysis on individual steps follows numerical error analysis routines outlined in Scaillet (2000); error analysis on grouped data follows algebraic methods of Roddick (1988). Blank measurements were made between samples or aliquots and a running average used to correct data. Neutron flux gradients throughout the sample canister were evaluated by analyzing the sanidine flux monitors included with each sample packet and interpolating a linear fit to arrive at a calculated J factor for a given sample position. The error on individual J factor values is conservatively estimated at ±0.6% (2ct). Because the error associated with the J factor is systematic and not related to individual analyses, correction for this uncertainty is not applied until the calculation of dates from isotopic correlation diagrams (Roddick, 1988). Uncertainties on decay constants and apparent age of monitor are not included in the quoted error. Our criteria for the determination of plateau ages are as follows: plateaus must include at least 60% of 39Ar released, respectively and they should be distributed over a minimum of three consecutive steps agreeing at 95% confidence level. Plateau ages are given at the 2ct level and are calculated using the mean of all the plateau steps, each weighted by the inverse variance of their individual analytical error. Integrated ages (2ct) are calculated using the total gas released for each Ar isotope.

Potassium-argon isotopic measurements were performed at NEISRI (Russian Academy of Science, Magadan). Before analysis, phenocrysts and xenocrysts of the seven analyzed samples were picked out, and the mineral fraction 0.1-0.25 mm was sieved. The fraction was processed using standard mineral separation techniques, including Frantz magnetic separation and heavy liquids. Nonmagnetic fractions containing olivine phenocrysts were removed to avoid potential excess argon. A magnetic fraction (~5 g) was cleaned under a binocular microscope and separated into two equal 100 mg parts. Potassium analysis was done by atomic absorption spectroscopy analysis on an AAS-1 spectrometer. Argon isotopes were measured on an improved MI-1201 IG mass spectrometer using a reconstructed metal melting reactor, and glass vacuum tubes were used for extracting radiogenic 40Ar, 36Ar, and 38Ar spike with lower detection limit around 0.5-1 ng. Geo-reference samples MSA-11 and MSA-12 (Russian Academy of Science inter-laboratory standards, 290 Ma) were repeatedly measured to control accuracy and precision of the analytical procedure, giving uncertainties < 1% for potassium and <2-5% for argon (age errors estimated as <2% for 75% of measurements). The decay constants of Steiger and Jäger (1997) were used to regress the ages.

4. Results

4.1. Petrography and mineral compositions

The alkaline xenolith-bearing host rocks from Ruditch have a very fine-grained groundmass with olivine as the only phenocryst phase. All present minerals, phenocrysts and groundmass phases, have little variation in composition in individual rock samples and within the whole rock suite. Representative EPMA mineral analyses demonstrating the narrow range of compositional variation of the minerals are summarized in the Online supplement Table 1. Olivines are present as compositionally zoned phenocrysts (with Fo 83 in the core and around Fo 72 in the rim) and as fine grained homogeneous crystals (Fo 70) in the groundmass. Typically, the majority of the phenocrysts are euhedral and their size varies between 200 and 250 |am in diameter whereas groundmass olivines are anhedral and their size does not exceed 100 |am in diameter. The zonation from a more forsteritic core to a fayalitic rim composition can clearly be seen in the BSE image of Fig. 2A and B and the Mg element map of Fig. 2C. The high MgO abundances in the core (~45 wt.%) are illustrated by the bright contrast due to high excitation, and darker gray at the rim (36 wt.% MgO). In general, the rim composition of the phenocrystic olivine is very close to the composition of the groundmass olivine generation. MnO concentrations in the phenocrystic olivine population range from around 0.2 wt.% in the core to 0.6 wt.% in the rim and Ni from 0.2 to 0.1 wt.%, respectively. An oscillatory, crystal lattice preferred zoning of phosphorous can be observed in the phenocrysts with P2O5 concentrations around 0.05 wt.%, which is absent in groundmass olivine (Fig. 2D). The core of the phenocrysts is generally enriched the most, with P2O5 between 0.2 and 0.3 wt.%.

Minerals of the groundmass are represented by clinopyroxene, nepheline, feldspar, Ti-magnetite, accessory apatite, and glass. The clinopyroxene laths are normally smaller than 50 |am and of En4i-40 Wo49-48Fs12-10 composition. TiO2 concentrations range from 1.6 to 2.4 wt.%, whereas Cr2O3 values average around 0.23 wt.%. Nephelines (Ne26-24Ks5-4Qy1-70) and feldspars (Abss^Ans^O^^) grow in-terstitially and are generally smaller than 30 |am. The feldspars are close to anorthoclase composition, with 0.4-0.5 wt.% BaO and 0.80.9 wt.% SrO. The glass phases reflect a narrow compositional field with Na2O: 9.5 wt.%, Al2O3: 25 wt.% and SiO2: 54 wt.%, and low concentrations of Ba, Sr. Chlorine was not detected in any of the analyzed minerals and glasses. From petrographic observations, a crystallization sequence from olivine phenocrysts to groundmass Ti-rich clinopyroxene, Ti-magnetite, feldspar over nepheline to glass can be assessed.

Fig. 2. EPMA backscattered electron images and element mappings of representative olivine phenocrysts of the basanites from Ruditch. BSE image of an (A) olivine phenocryst and (B) groundmass olivine minerals. (C) shows a Mg mapping and (D) a P mapping of the olivine crystal of A.

4.2. Major and trace element composition

Major and trace element bulk rock concentrations of the host lavas from Ruditch are summarized in Table 1. The samples show narrow compositional variability with high total alkalis (Na2O: 5.3-5.6 wt.%and K2O: 2.9-3.4 wt.%) and low SiO2 (SiO2: 44.6-46.0 wt.%), all plotting in the basanite field of the total alkalis vs. silica (TAS) diagram (Fig. 3A, after Le Maitre et al., 1989). They have modestly high MgO (9.4-10.3 wt. %) and 13.7-15.8% normative olivine, consistent with a rather primitive nature and limited fractionation, which is confirmed by the fact that only olivine phenocrysts are present. The rocks have normative nepheline contents ranging from 19% to 23%, on the transition of basanites to nephelinites, based on the classification of Le Maitre (2002). However, due to the fact that SiO2 concentrations are not extremely low in these rocks (~45.5 wt.%) and feldspar minerals seem to be the predominant phase in the mineral paragenesis, the alkaline rocks from Ruditch are more appropriate to be classified as basanites.

The homogenous nature of the rocks is also confirmed by the small variability in trace element concentrations. The homogeneity and high enrichment of trace-elements in the basanites from Ruditch is illustrated by the multi-element diagram of Fig. 3B and the REE diagram of Fig. 5A. The compatible trace elements Cr and Ni vary from 307 to 358 ppm and from 213 to 253 ppm, respectively. Barium concentration varies between 288 and 343 ppm, Rb between 42 and 50 ppm. The high field strength elements Zr (299-319 ppm), Nb (9096 ppm), Th (8.1-9.2 ppm) and U (2.6-2.9 ppm) again indicate low compositional variation at relatively high element abundances. Lanthanum is 213-222 xC1 (Chondrite C1) and Lu ranges from 4.7 to 5.1 xC1, resulting in steep REE patterns, enriched in light rare earth elements (LREE) when normalized against chondritic composition ((La/Yb)N: 38-41.3).

Tectonic discrimination diagrams (e.g. Zr vs. TiO2 or Ba/Nb vs. Nb/Y — not shown) assign the Ruditch basanites to typical intraplate lavas (e.g. Pearce and Cann, 1973). According to Weaver (1991), the overall trace element geochemical patterns as well as specific trace element ratios indicate OIB affinity (e.g. Zr/Nb: 3.32-3.39, Ba/Th: 35.24-40.17, Th/La: 0.16-0.18). Using the more recent discrimination trace element ratio systematics ofWillbold and Stracke (2006), the Ruditch basanites show somewhat mixed characteristics with both HIMU-type ratios (Ba/Nb: 3.2-3.7 and Rb/Nb: 0.46-0.53), and EM-type ratios (i.e. high Rb/La: 0.812-0.953, Rb/Sr: 0.041-0.048,1/Sr: 0.0009-0.001).

Trace element compositions and element ratios in the studied rocks indicate that no crustal contamination processes occurred (e.g. Nb: 89.5-95.6 ppm vs. Nb/U: 32.3-35.6 and Ce: 95-98.7 ppm vs. Ce/ Pb: 24-26; Hofmann et al., 1986). Further, Ce/Pb and Nb/U ratios fall entirely within the range of OIB and show no influence from upper or lower crustal components (Taylor and McLennan, 1985).

4.3. Sr, Nd, Pb and Hf isotopes

Measured Sr, Nd, Pb and Hf isotope ratios of the alkaline basalts from Ruditch are reported in Table 2. The isotope data shows very little variability on all four isotopic systems. For a 87Sr/86Sr vs. sNd plot (Fig. 4A), the Ruditch basanites fall entirely within the OIB field with more radiogenic Nd (0.512896 to 0.512905) and unradiogenic Sr (0.703503 to 0.703571) isotope compositions compared to Bulk Earth, suggesting the involvement of a time-integrated depleted reservoir in their source. Compared to the geographically most adjacent alkaline rocks from the Viliga Volcanic field and the Balagan-Tas Volcano (Fig. 1), the Ruditch lavas are isotopically enriched with lower 143Nd/144Nd and higher 78Sr/86Sr.On a 176Hf/177Hf vs. 143Nd/144Nd plot (Fig. 4B), the samples plot near the center of the terrestrial basalt array. On

Table 1

Whole-rock composition of the basanites from Ruditch.

RU1 RU2 RU4 RU4-1 RU6 RU7

XRF (wt.%)

SiO2 45.78 45.23 45.91 45.63 45.55 45.96

TiO2 1.85 1.87 1.84 1.82 1.83 1.86

AI2O3 14.55 14.25 14.46 14.20 14.48 14.29

FeO1 10.25 10.43 10.31 10.20 10.32 10.47

MnO 0.17 0.18 0.17 0.17 0.17 0.17

MgO 9.45 9.95 9.55 9.35 9.65 10.31

CaO 7.88 7.95 7.85 7.88 7.90 7.96

Na2O 5.55 5.42 5.29 5.33 5.61 5.29

K2O 3.31 3.28 3.34 3.43 3.16 2.86

P2O5 0.95 0.96 0.96 0.96 0.97 0.97

&2O3 0.05 0.05 0.04 0.04 0.05 0.05

Total 99.8 99.6 99.7 99.0 99.7 100.2

LOI 0.77 0.73 0.90 0.84 0.95 0.68

Mg# 62.2 63.0 62.3 62.1 62.5 63.7

ICP-OES (ppm)

Sr 1005 1089 1061 1107 1043 995.7

Zn 123.1 121.1 119.1 126.5 118.2 120.6

Cu 30.6 29.98 29.52 33.67 28.13 29.49

Ni 214.3 233.1 217.4 213.2 219.3 252.5

V 149.5 148.7 142.7 150.4 144.1 149.3

Ba 301.6 326.9 342.3 343.3 314.1 287.9

ICP-MS (ppm)

Sc 12.74 12.94 11.52 12.30 12.87 11.93

Ga 23.73 24.47 24.13 24.20 24.87 21.82

Rb 48.56 45.75 49.90 45.80 43.49 41.49

Y 20.52 19.83 20.57 20.70 20.65 20.09

Zr 314.9 301.0 311.5 318.8 319.2 299.4

Nb 94.40 90.72 93.81 95.59 94.26 89.50

Mo 24.74 25.09 25.74 25.39 22.29 22.25

Cs 0.80 0.67 0.91 0.84 0.68 0.80

La 50.98 50.49 52.62 52.65 52.65 51.07

Ce 95.50 95.03 98.66 98.55 98.60 95.84

Pr 10.72 10.57 11.01 11.03 11.00 10.64

Nd 44.09 43.77 45.45 45.43 45.56 43.74

Sm 7.65 7.63 7.82 7.93 7.92 7.65

Eu 2.41 2.34 2.43 2.46 2.43 2.35

Gd 6.81 6.75 6.93 7.03 6.99 6.85

Tb 0.88 0.86 0.87 0.89 0.90 0.87

Dy 4.08 3.94 4.08 4.14 4.11 3.99

Ho 0.65 0.63 0.64 0.65 0.65 0.63

Er 1.47 1.45 1.48 1.49 1.49 1.45

Tm 0.18 0.17 0.17 0.18 0.18 0.17

Yb 0.96 0.88 0.94 0.94 0.92 0.93

Lu 0.13 0.12 0.12 0.12 0.13 0.12

Hf 6.03 5.69 5.88 6.02 6.06 5.69

Ta 5.78 5.22 5.26 5.37 5.20 4.82

Pb 4.00 3.79 4.09 4.01 3.82 3.97

Th 8.52 8.14 9.22 8.63 8.88 8.17

U 2.66 2.55 2.91 2.70 2.89 2.55

FeOt = total iron; LOI = Loss on ignition; Mg# = 100 x molar MgO/(MgO + FeOt).

a 206Pb/204Pb vs. 87Sr/86Sr plot (Fig. 4C), the Ruditch samples fall on the unradiogenic end of the OIB and MORB array and have lower 206Pb/204Pb ratios than any of the FOZO compositions (see Stracke et al., 2005 for discussion). On diagram of Fig. 4C, however, the Ruditch samples show a clear displacement towards the so-called EM-1 mantle end-member component (e.g. Zindler and Hart, 1986) with relatively low 206Pb/204Pb for a given 87Sr/86Sr. However the lack of spread in the isotope data prevents us from better constraining the possible end-member components of the Ruditch magma source.

4.4. 40Ar/39Ar and K-Ar geochronology

We obtained six 40Ar/39Ar plateau ages (Online supplement Fig. 1, Online supplement Table 2 and Table 3) ranging from 36.0 ±0.8 to 38.3 ± 0.6 Ma (2ct) with MSWD ranging from 4 to 16. The early heating increments for four of the six spectra yield older apparent ages until a plateau is reached, while the remaining two spectra are characterized by a single younger increment (~10% of 39Ar) before the plateau is reached. The oldest sample (RU7) has the most poorly defined plateau almost possessing a hump-shape; nevertheless, its age is indistinguishable from the other samples. The Ca/K spectra exhibit a similar shape in all six samples. We note that the Ca/K spectra show slight error increase toward the higher temperature steps but this is likely to reflect compositional zonation of the material dated. The relatively high Ca/K ratios of the Ruditch groundmass samples, which are consistent with the low whole-rock K2O wt.% of our samples, tend to lower the precision on the age mainly because of the more important corrections for Ca interferences and the smaller 39Ar ion beams. Five of the six samples yield reliable Late Eocene (Priabonian) ages. Potassium-argon measurements yielded ages between 38 and 41 Ma (Online supplement Table 3), which is in tolerable agreement with the higher precision 40Ar/39Ar geochronology.

5. Discussion

5.1. Rapid emplacement and cooling

Olivine occurs as euhedral phenocrysts (Fig. 2A) and as fine-grained anhedral groundmass crystals (Fig. 2B). The phenocrysts have Fo around 83 in the core decreasing to 72-71 in the rims, compositionally similar to the unzoned groundmass olivine, with Fo around 70. Calculating the theoretical olivine composition that is in equilibrium with the present bulk rock composition (olivine partition coefficients after Beattie et al., 1991), a Fo index of between 84.5 and 85.6 is calculated. As this composition is very similar to the core composition of the phenocrystic

Fig. 3. Major and trace element compositions of the Ruditch basanites (plotted with star-symbols). (A) Total alkalis vs. silica (TAS) diagram (after Le Maitre et al., 1989). (B) Primitive mantle normalized multi-element diagram (PM composition after McDonough and Sun, 1995). Data from the most adjacent locations with alkaline magmatism, the Viliga Volcanic Field (VVF; Tschegg et al., 2011) and the Balagan-Tas volcano (BG; Grachev, 2003), are plotted for comparison.

Table 2

Sr-Nd-Pb-Hf isotope ratios of the basanites from Ruditch.

TIMS measured

87Sr/86Sr 143Nd/144Nd e Nd

Initial

87Sr/86Sr e Nd

MC-ICPMS measured

206pb/204pb

207Pb/204Pb

208pb/204pb

208Pb/206Pb

207Pb/206Pb 176и(-,177щ

Initial

6Hf/1 e Hf

0.703540 0.512899

0.703487 5.5

18.028 15.476 38.013 2.108 0.858 0.283010

0.703503 0.512904 5.2

0.703457 5.6

0.703571 0.512897 5.0

0.703519 5.5

18.030 15.477 38.033 2.109 0.858 0.283015

0.703511 0.512905 5.2

0.703449 5.6

0.703507 0.512901 5.1

0.703464 5.6

18.037 15.474 38.018 2.108 0.858 0.283016

0.703549 0.512896 5.0

0.703502 5.5

olivines, we conclude that the cores of the phenocrysts, during initial growth, have been in equilibrium with a melt corresponding to the present whole-rock composition. The homogeneous groundmass olivine grew rapidly and contemporaneously with the phenocrystic

• Indian MORB . Atlantic MORB

• Pacific MORB

- HIMU

■ EM-I + EM-II

° NON-HIMU TYPE OIB M Ruditch

O Viliga Volcanic Field

• Balagan-Tas

27 : Щ ВШ .

7 EM-II

■ • • •

EM-I -8 .......

-7-5 -3 -1 1 3 5 7 9 11 13 ENd

0 702 -,-,-,-,-,-,-,-,-,-

16,5 17,5 18.5 19.5 20,5 21,5 ™РЬГРЬ

Fig. 4. Sr-Nd-Pb-Hf isotope plots for basanites from Ruditch. (A) 87Sr/86Sr vs. epsilon Nd, (B) epsilon Nd vs. epsilon Hf and (C) 206pb/204pb vs. 87Sr/86Sr. The plotted data compilation is from Stracke et al. (2003), plus some recent data from the GEOROC database. The Ruditch data are from this study, Viliga Volcanic Field data from Tschegg et al. (2011) and Balagan-Tas data from Grachev (2003). Please see the online version of the paper for a full color version.

rims in a very late state of cooling that can be inferred from their similar composition. The presence of phosphorous in the olivine phenocrysts indicates that olivine crystallized from a melt, enriched in phosphorous and depleted in SiO2. These olivines furthermore crystallized from this melt very fast and/or have been crystallized in a rapidly cooling unequilibrated system (Milman-Barris et al., 2008). The fact that P exhibits an oscillatory zoning, in contrast to Mg and Fe, is related to its ionic radius and the way of substitution within the crystal lattice (Boesenberg et al., 2004). While Fe and Mg are distributed without preferences in the very similar M1 and M2 octahedral sites in a continuous reaction, P substitutes for Si in the tetrahedral position in association with an increase of the vacancy in the octahedral site. Therefore the oscillatory zoning of P is a function of substitution mechanisms in combination with crystal growth processes and the fact that P diffusion rates are very low allowing the preservation of the P oscillation. Fig. 2 shows that the zoning of P appears to be asymmetrical in olivine compared to the one of Fe and Mg, which, in contrast to the P concentration, has the highest Mg and the lowest Fe contents coinciding with the morphological center of the crystal. We interpret this feature as an additional evidence for differential diffusion rates of Fe-Mg and P that furthermore confirms a system rapidly crystallizing with high cooling rates (Milman-Barris et al., 2008).

Both olivine phenocrysts and olivine groundmass minerals are relatively small grains due to the rapid crystal growth and fast cooling, which is confirmed also by the extremely fine-grained matrix (<50 |jm). That all samples are ne-normative precludes significant fractionation in a crustal level and rules out contamination/assimilation by significant amounts of crustal material (MacDonald et al., 2001 and references therein). From petrographic and mineral chemical observations as well as from bulk geochemical evidence, a rapid ascent of the magma through the crust and high cooling rates during emplacement of the magma can be inferred. The fact that the basanites are xenolith-bearing further indicates fast ascent from their segregation depth to the surface, and absence of magma chamber processes. The flat 40Ar/39Ar plateaus and consistent Late Eocene age for all the samples further support rapid emplacement and cooling of the Ruditch basanite suite.

5.2. Source characteristics and melt evolution

The basanitic xenolith-bearing rocks from Ruditch are extremely homogeneous in terms of mineral, bulk major- and trace-element and even isotopic composition. The lack of any significant compositional variation, does not allow us to resolve differentiation (e.g. fractionation) or mixing trends. The compositional variation of MgO and SiO2, often used as basic differentiation indicators, is less than 1 wt.%. The same is true for the trace element abundances, which do not show any

-7 -■-■-■-1-1-■-■-1-1-

0.702 0.703 0.704 0.705 0.706 0.'

"srrsr

Table 3

40Ar/39Ar ages of the basanites from Ruditch.

RU1 RU2 RU4 RU4-1 RU6 RU7

Total gas age (Ma) 36.83±1.05 36.94±0.61 37.71 ±1.12 38.21 ±0.90 36.78±1.14 38.51 ±0.88

Plateau age (Ma) 37.10 ±0.50 36.59 ±0.39 35.95 ±0.77 37.13 ±0.84 37.08 ±0.45 38.15 ±0.54

Plateau MSWD 15.26 11.57 6.97 16.72 12.99 4.06

Plateau 39Ar Gas % 88 84 59 69 87 62

significant differences in the studied rocks. However, certain important aspects about the geodynamic setting, source and petrogenetic historyof the volcanic suite can be developed.

The primitive mantle normalized multi-element diagram of Fig. 3 illustrates decreasing element abundances with increasing element compatibility, patterns very typical for the different types of OIB (e.g. Fitton and James, 1991; Willbold and Stracke, 2006). The lack of negative Eu anomalies argues against significant plagioclase fraction-ation. The high concentrations and slight positive anomalies (with respect to neighboring elements on a compatibility diagram) of Nb, Zr and Y imply that these melts were produced by a mantle astheno-spheric source without significant involvement of crustal material. The steep normalized REE patterns (Fig. 5A) as well as the very high (La/Yb)N: 38-41 and (Dy/Yb)N: 2.8-3.0 ratios confirm this and a priori suggest extremely small melt fractions from a garnet-bearing mantle source (e.g. Blundy et al., 1998).

Semi-quantitative melting models were produced to estimate the source composition of the studied rocks and to assess the degree of partial melting. Fig. 5A illustrates the results of a melting model and the composition of 0.5-2% melt fractions of a primitive mantle source in a chondrite-normalized REE multi-element diagram. Mantle composition, melting proportions and distribution coefficients are given in the caption of Fig. 5A. Based on their REE contents, the basanites from Ruditch can be modeled as ~1% partial melts, from a dry garnet peridotite. Taking the behavior of Zr, Nb and Y during partial melting into account and using the same parameters as for the latter model, an ~0.5% melting of a garnet lherzolite source is suggested (Fig. 5B). Neither petrographic (presence of phlogopite or amphibole) nor geochemical criteria (K- or Rb-negative anomalies in normalized multi-element diagrams) indicate the presence of hydrous and/or K-bearing phases during partial melting or fractionation during magma evolution, thereby suggesting that this garnet peridotite source must be also relatively dry. On grounds of the SiO2-CaO criteria proposed by Herzberg and Asimow (2008), the presence of significant CO2 during melting can also be

excluded. Our modeling suggests that the alkaline rocks of Ruditch represent quite primitive small volume melts from an anhydrous garnet peridotite source, without significant melt mixing or mingling, nor other types of compositional alteration and contamination.

Based on bulk major and trace element behavior and systematics, en-route crystal fractionation processes of olivine and eventually clinopyroxene during melt evolution of the parental magma are very difficult to assess due to the lack of compositional variation within the studied volcanic suite.

5.3. Pressure and temperature of melt generation

Estimates of mantle potential temperatures are very useful to evaluate whether excess heat existed in the source during melt generation or rather ambient mantle temperatures were present. This is crucial to assess whether melting and magmatism occur due to lithospheric extension and passive asthenospheric upwelling, or alternatively thermally-induced active upwelling and melting. Mantle potential temperature for the primary magmas of the alkaline rocks from Ruditch were calculated using PRIMELT2 (Herzberg and Asimow, 2008) yielding temperatures of ~ 1500 °C. Since a pyroxenite source or pyroxene fractionation of the primary magma is indicated during calculation (based on MgO vs. CaO criteria; for further reading see Herzberg and Asimow, 2008), this temperature is likely to be biased. Our source modeling indicates a dominantly asthenospheric garnet perido-tite source, but small degrees of clinopyroxene fractionation during melt evolution cannot be excluded. If the melts have lost some clinopyroxene during melt evolution due to high-pressure fractionation, even if only in small amounts, the temperatures are overestimated (Herzberg and Asimow, 2008) and the mantle potential temperature for the source of the Ruditch melts is ~ 1500 °C, but more likely less. Hence, the calculated temperatures are in accordance with the average mantle geotherm (1376-1532 °C), indicative for upper mantle sources, likely to be thermally undisturbed by subduction or active mantle upwelling

"I—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i 6 —i—i—i—i—i—i—i—i—i—i—i—i—r

Fig. 5. Melt modeling of the source of the Ruditch basanites. (A) Chondrite normalized REE diagram (C1 composition from Sun and McDonough, 1989). Dashed lines show modeled melts (0.5%, 1%, 2%) from a garnet lherzolite using an incremental batch melting model at 0.1% melt increments. Average partition coefficients for olivine, garnet and orthopyroxene are taken from the high-pressure experiments (2.4-3.4 GPa) of Salters and Longhi (1999), and Salters et al. (2002), summarized in Keshav et al. (2005). Clinopyroxene partition coefficients are from experiment 1097-5 of Salters et al. (2002), where clinopyroxene coexists with orthopyroxene and garnet. Source mineralogy: olivine = 0.62, orthopyroxene = 0.15, clinopyroxene = 0.15, garnet = 0.08, and melting mode olivine = 0.028, orthopyroxene = —1.18, clinopyroxene = 1.66, garnet = 0.492, following the arguments of Salters and Longhi (1999). (B) Zr/Nb vs. Nb/Y illustrating melting at 0.5,1 and 2%, using the same model parameters as described for A.

(Putirka et al., 2007 and references therein). Even if lower mantle potential temperatures are considered for the ambient mantle (13001400 °C, e.g. favored by Hastie and Kerr, 2010), a mantle plume source would show excess heat, fundamentally exceeding 1500 °C, and generating large degree partial melting at deep levels without the need of lithospheric extension (Hastie and Kerr, 2010; Putirka et al., 2007).

During adiabatic upwelling, an ~1500 °C or cooler upper mantle would have to ascend to at least ~ 110 km to cross its dry solidus and initiate partial melting (e.g. Hastie and Kerr, 2010; McKenzie and Bickle, 1988). A hotter mantle (with excess potential heat) would intersect the dry mantle solidus at much deeper levels, enhancing melting (Hastie and Kerr, 2010). To crosscheck and better constrain the pressure and temperature of melting, we calculated the melt segregation temperature and pressure using the methods of Albarede (1992), Putirka (2005) and Lee et al. (2009). Thermobarometry after Lee et al. (2009) yield segregation conditions of 1484-1499 °C (1490 °C in average) at 3.8-4.0 GPa (3.9 GPa in average) for the Ruditch basanites. Temperature estimates using the thermometer of Putirka (2005, model A) vary between 1451 and 1464 °C (1458 °C in average), very similar to the results from Lee et al. (2009). Pressure calculations based on the barometer proposed by Albarede (1992) yield 2.6-2.9 GPa (2.7 GPa in average), approximately 1 GPa lower than the results from Lee et al. (2009). An underestimation of P estimates based on Albarede (1992) was already noted by Lee et al. (2009) for pressures exceeding 3 GPa. Both pressure estimates were converted to depths using the formula provided by Scarrow and Cox (1995), resulting in an average melt segregation depth of 133 km for the Lee et al. (2009) and 88 km for the Albarede (1992) barometer. Averaging all our calculated melt segregation depths, we obtain a depth of 110 km ±25, which is in agreement with the hypothetical melting depths of a garnet peridotite source with < 1500 °C potential temperature (based on the decompression pathways). These depths are consistent with the trace element systematics that require melting in the presence of garnet.

5.4. Mantle components

Due to the isotopic homogeneity of our samples, it is difficult to delineate potential mantle end-members or mantle components in the source of the Ruditch basanites. However, the lack of any isotopic trend is consistent with a homogenous source for these basalts. Furthermore, the Sr, Nd, and Hf isotope ratios suggest a source that is depleted relative to bulk earth but more enriched relative to Depleted MORB Mantle (DMM), the source of MORB basalts. Lead isotope compositions combined with Sr or Nd-Hf isotopes suggest some contribution from an EM-I type source. In contrast, the proximal but much younger Neogene volcanics of the Viliga Volcanic Field (Tschegg et al., 2011 and references therein) and the Late Quaternary Balagan-Tas volcano (Grachev, 2003) indicate derivation from a more depleted source. The Miocene rocks from the Bering Sea Volcanic Province (Anrdonikov and Mukasa, 2010) also indicate a closer affinity to MORB, without being affected by an isotopically enriched component.

5.5. Tectono-magmatic implications

The tectonic evolution and timing (initiation, duration and cessation) of specific geodynamic events in the Chersky belt is highly debated and sometimes controversial. Some authors interpret the geodynamic setting in the region as well as the formation of the North American-Eurasian Plate boundary as a result of rifting. This rifting, related to the Arctic Ocean spreading, is thought to spatially continue into the Asian continent, resulting in the complex stitch work of sutures and scars (Kozhurin, 2004), with the Moma "Rift" as the continental continuation of the Arctic Rift (Grachev et al., 1970). The sporadic alkaline volcanism in the area of the Chersky belt is thought

to have evolved due to an intra-continent rift-like zone (Fujita et al., 1997) that continued until Eocene, forming the North American-Eurasian plate boundary (Nokleberg and references therein). Franke et al. (2000) however postulated that Mid-Arctic rifting terminates at the transition from the oceanic to the continental setting (in the Laptev Sea), reorganizing into the present-day transform and even compressional tectonic regime.

Based on the present data, which elucidates the source and petrogenesis of the alkaline rocks in the Ruditch region, it is challenging to decide whether the Eocene magmatism in the area could be related to a typical continental rift. Continental rift zone magmatism is in general triggered by either plume activity or lithospheric stretching (Turcotte and Emerman, 1983). As a large number of (sub-)lithospheric sources may be involved in the generation of these magmas, which additionally may be metasomatized, depleted, recycled, contaminated and/or mixed, an extremely large variety of rock types (from nephelinites, carbonatites and lamprophyres plus all their differentiates) can be generated in such tectonic settings (e.g. Kumar and Rathna, 2008 and references therein). Furthermore, large amounts of melts from chemically anomalous mantle sources, able to form massive magmatic provinces, are usually generated by typical rift-related alkaline magmatism, particularly when mantle plume activity is involved (Jung and Masberg, 1998 and references therein).

In contrast to typical intra-continental rift related rocks that tend towards extensive differentiation and variability (e.g. MacDonald et al., 1994), the Ruditch basalts indicate small degrees of partial melting of a relatively homogeneous, EM-I type dry garnet lherzolite source, without significant contribution of any other mantle or crust component, or signatures of advanced differentiation during magma ascent and crustal piercing. The ambient mantle potential temperature estimates do not suggest abnormally hot mantle (i.e. plume), able to promote deep seated and extensive partial melting (Hastie and Kerr, 2010). The lack of any subduction or crustal related geochemical signatures (e.g. HFSE depletions with accompanying LILE enrichments, such as high La/Nb or low Ce/Pb) argues against such tectonic processes and/or crustal interaction. Lava sources from the Basin and Range province of western North America have been distinguished based on La/Nb and Ce/Pb criteria, resulting in younger episodes of sub-lithospheric OIB like magmatism, and older volcanics that preserve a lithospheric mantle source signature (Fitton, 1995). The basanites from Ruditch correlate with melts generated from a sub-lithospheric mantle source based on the La/Nb vs. Ce/Pb diagram (Fig. 6), once more confirming an asthenospheric source. The model of Grachev (2003) that incipient mantle plume activity is responsible for the plate boundary in NorthEast Russia cannot be supported. Instead, the melt segregation temperatures/pressures as well as the calculated mantle potential temperatures are more consistent with passive asthenospheric mantle

Fig. 6. La/Nb vs. Ce/Pb in the basanites from Ruditch. Sub-lithospheric and lithospheric mantle fields comprise rocks of the Basin and Range province (Fitton, 1995).

upwelling, and magmatism in Ruditch that is due to decompression melting.

Comparing the Ruditch basanites to other alkaline rocks of the region is challenging, because of the paucity of alkaline magmatic rocks in the area, and the variable ages of eruption. The Neogene basanites and nephelinites from the Viliga Volcanic Field, recently studied by Tschegg et al. (2011), are almost 30 Ma younger, and also suggest asthenospheric sources. Fig. 3B shows that the trace element patterns of the samples from the Viliga Volcanic field are similar to the Ruditch samples, although those rocks were generated by slightly higher degrees of garnet peridotite mantle melting with melt contribution of shallower spinel peridotite partial melts. Based on the similar melt segregation conditions of 1495-1510 °C at 3.33.8 GPa, the basanites from the Viliga Volcanic Field (Tschegg et al., 2011) have been formed under comparable P-T conditions as the rocks from Ruditch. The lavas from the Pleistocene Balagan-Tas Volcano (Fig. 4), which are possibly related to destructive plate processes between the North American-Eurasian boundaries (Grachev, 2003), are isotopically and geochemically more similar to lavas from the Viliga volcanic field than those documented here.

Our findings suggest that the alkaline magmatism in the Chersky Mountains is more likely the result of lithospheric stretching, passive mantle upwelling and decompression melting, probably related to the Arctic mid-ocean spreading, than the result of the subsequent lithospheric transpression and thrusting processes. The timing of volcanism suggests that the opening of the Mid-Arctic Ocean and its continuation in the Laptev Sea controlled rifting in the Chersky belt area at least until ca. 37 Ma. Isotopes, bulk major and trace element compositions as well as our melting scenarios and ambient mantle potential temperature calculations fit a tectonic model that indicates no involvement of subduction-related geochemical fingerprints or plume-related excess temperatures. To find an alternative explanation for magmatism is difficult based on our evidence and model calculations as melting would have to be triggered by deformation-related weakening of the lithosphere, in a geodynamic frame of a convergent crustal thickening. High seismicity and active tectonics (e.g. thrusting and strike-slip) are documented in the area, however, therefore melt generation under those conditions is unlikely.

6. Summary

I) Bulk geochemical, petrographic and mineral geochemical features suggest a rapid ascent of the magmas from Ruditch and high cooling rates after emplacement in the crust.

II) No evidence for crustal contamination processes, large amounts of crystal fractionation or mixing with other sub- or lithospheric components are exhibited in the Ruditch basanite suite.

III) Petrogenetic modeling suggests that melts were likely produced by very low degrees of partial melting of an anhydrous garnet bearing asthenospheric source at around 110 km depth.

IV) Sr-Nd-Pb-Hf and trace elements support the derivation of melts from a deep-seated homogeneous asthenospheric source that has been slightly enriched by an EM-I mantle component.

V) Mantle potential temperature estimates < 1500 °C essentially preclude subduction and/or mantle plume disturbed geotherms, and hence indicate passive upwelling of astheno-spheric material.

VI) The passive upwelling promoted decompression melting and is in all probability related to tensional stress and thinning of the lithosphere. Reliable emplacement ages of the alkaline rocks from Ruditch are ca. 37 Ma, thus opening of the Arctic Ocean could have had affected the region of the Chersky seismic belt at that time, before the stress regime changed to transpression that continues today.

Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.lithos.2011.04.008.

Acknowledgments

We thank the Austrian Science Fund FWF (Project: I210-N10, P.I. Th. Ntaflos) for the financial support of this study. Parts of this work were supported by grant NSF-OCE: 0852488 to M. Bizimis. The ICPMS instrumentation at the University of South Carolina was funded by grant NSF-OCE: 0820723 to R. Thunnel. Aleksandr V. Alshevski is thanked for his assistance in the field. For ICP-OES trace element analysis, Sr-Nd isotope measurements and EPMA analyses, we kindly appreciate the help of our colleagues Wilfried Körner, Monika Horschinegg, Martin Thöni and Franz Kiraly. Beth Bair, Carl Frisby and Tarun Khanna (University of South Carolina) are thanked for their help with the Pb and Hf isotopes. Constructive reviews of Michael Roden and Ioan Seghedi are gratefully acknowledged.

References

Abouchami, W., Galer, S.J., Koschinsky, A., 1999. Pb and Nd isotopes in NE Atlantic Fe-Mn crusts: proxies for trace metal paleosources and paleocean circulation. Geochimica et Cosmochimica Acta 63,1489-1505. Akinin, V.V., Prokopiev, A.V., Toro, J., Miller, E.L., Wooden, J., Goryachev, N.A., Alshevsky, A.V., Bakharev, A.V., Trunilina, V.A., 2009. U-Pb SHRIMP ages of granitoides from the main batholith belt (North East Asia). Doklady Earth Sciences 426 (4), 605-610. Albarede, F., 1992. How deep do common basaltic magmas form and differentiate?

Journal of Geophysical Research 97,10997-11009. Anrdonikov, V., Mukasa, S.B., 20 1 0. 40Ar/39Ar eruption ages and geochemical characteristics of Late Tertiary to Quaternary intraplate and arc-related lavas in interior Alaska. Lithos 115,1-14. Beattie, P., Ford, C., Russell, D., 1991. Partition coefficients for olivine-melt and orthopyroxene-melt systems. Contributions to Mineralogy and Petrology 109, 212-224.

Boesenberg, J.S., Ebel, D.S., Hewins, R.H., 2004. An experimental study of phosphoran olivine and its significance in main group pallasites. Lunar and Planetary Science Conference XXXV. Lunar and Planetary Institute, Houston. Abstract 1368. Blundy, J.D., Robinson, J.AC., Wood, B.J., 1998. Heavy REE are compatible in clinopyroxene on the spinel lherzolite solidus. Earth and Planetary Science Letters 160,493-504.

Cook, D.B., Fujita, K., McMullen, C.A., 1986. Present-day plate interactions in northeast Asia: North American, Eurasian, and Okhotsk plates. In: Johnson, G.L., Kaminuma, K. (Eds.), Polar Geophysics, Journal of Geodynamics, 6, pp. 33-51. Fitton, J.G., James, D., 1991. Basic magmatism associated with Late Cenozoic extension in the Western United States: compositional variations inspaceand time. Journal of Geophysical Research 96 (B8), 13693-13711. Fitton, J.G., 1995. Coupled molybdenum and niobium depletion in continental basalts.

Earth and Planetary Science Letters 136, 715-721. Franke, D., Krüger, F., Klinge, K., 2000. Tectonics of the Laptev Sea-Moma 'Rift' region: investigation with seismologic broadband data. Journal of Seismology 4, 99-116. Fujita, K., Cook, D.B., Hasegawa, H., Forsyth, D., Wetmiller, R., 1990. Seismicity and focal mechanisms of the Arctic region and the North American plate boundary in Asia. In: Grantz, A., Johnson, L., Sweeny, J.F. (Eds.), The Geology of North America: The Arctic Ocean Region: Geological Society of America, L. 644 pp. Fujita, K., Stone, D.B., Layer, P.W., Parfenov, L.M., Koz'min, B.M., 1997. Cooperative program helps decipher tectonics of northeastern Russia: Eos. American Geophysical Union Transactions 78, 252-253. Fujita, K., Koz'min, B.M., Mackey, K.G., Riegel, S.A., McLean, M.S., Imaev, V.S., 2009. Seismotectonics of the Chersky Seismic Belt, eastern Sakha Republic (Yakutia) and Magadan District, Russia. Stephan Müller Special Publication Series 4,117-145. Grachev, A.F., Demenitskaya, R.M., Karasik, AM., 1970. The Mid-Arctic Ridge and its

continental continuation. Geomorphology 1, 30-32. Grachev, A.F., 2003. The Arctic rift system and the boundary between the Eurasian and North American plate tectonic theory. Russian Journal of Earth Sciences 5,307-345. Hastie, A.R., Kerr, A.C., 2010. Mantle plume or slab window?: physical and geochemical constraints on the origin of the Caribbean oceanic plateau. Earth-Science Reviews 98, 283-293.

Herzberg, C., Asimow, P.D., 2008. Petrology of some oceanic island basalts: PRIMELT2. XLS software for primary magma calculation. Geochemistry Geophysics Geosys-tems 9 (9), 1525-2027. Hofmann, A.W., Jochum, K.P., Seufert, M., White, W.M., 1986. Nb and Pb in oceanic basalts: new constraints on mantle evolution. Earth and Planetary Science Letters 79, 33-45.

Jung, S., Masberg, P., 1998. Major- and trace-element systematics and isotope geochemistry of Cenozoic mafic volcanic rocks from the Vogelsberg (central Germany) — constraints on the origin of continental alkaline and tholeiitic basalts and their mantle sources. Journal of Volcanology and Geothermal Research 86, 151-177.

Keshav, S., Corgne, A., Gudfinnsson, G.H., Bizimis, M., McDonough, W.F., Fei, Y., 2005. Kimberlite petrogenesis: insights from clinopyroxene-melt partitioning

experiments at 6 GPa in the CaO-MgO-Al2O3-SiO2-CO2 system. Geochimica Cosmochimica Acta 69, 2829-2845.

Kozhurin, A.I., 2004. Active faulting at the Eurasian, North American and Pacific plates junction. Tectonophysics 380, 273-285.

Kumar, K.V., Rathna, K., 2008. Geochemistry of the mafic dykes in the Prakasam Alkaline Province of Eastern Ghats Belt, India: implications for the genesis of continental rift-zone magmatism. Lithos 104, 306-326.

Layer, P., Parfenov, L.M., Surnin, A.A., Timofeev, V.F., 1993. First 40Ar/39Ar estimation of age of the magmatic and metamorphic rocks of Verkhoyano-Kolyma mesozoides. Doklady Earth Science 329 (N 5), 621-624 (in Russian).

Layer, P.W., Newberry, R., Fujita, K., Parfenov, L., Trunilina, V., Bakharev, A., 2001. Tectonic setting of the plutonic belts of Yakutia, northeast Russia, based on 40Ar/ 39Ar geochronology and trace element geochemistry. Geology 29, 167-170.

Le Maitre, R.W., Bateman, P., Dudek, A., Keller, J., Lameyre, J., Le Bas, M., Sabine, P., Schmid, R., Sorensen, H., Streckeisen, A., Woolley, A., Zanettin, B., 1989. In: Le Maitre, R.W. (Ed.), A Classification of Igneous Rocks and Glossary of Terms: Recommendations of the International Union of Geological Sciences Subcommission on the Systematics of Igneous Rocks. Blackwell, Oxford. 193 p.

Le Maitre, R.W., (Ed.), 2002. Igneous Rocks: A Classification and Glossary of Terms, 2nd Edition. Recommendations of the International Union of Geological Sciences Subcommission on the Systematics of Igneous Rocks. Cambridge University Press, 219 p.

Lee, C.-T.A., Luffi, P., Plank, T., Dalton, H., Leeman, W.P., 2009. Constraints on the depth and temperatures of basaltic magma generation on Earth and other planets using new thermobarometers for mafic magmas. Earth and Planetary Science Letters 279, 20-33.

Mackey, K.G., Fujita, K., Ruff, L.J., 1998. The crustal thickness of northeast Russia. Tectonophysics 284, 283-297.

McDonough, W.F., Sun, S., 1995. The composition of the Earth. Chemical Geology 120, 223-253.

MacDonald, R., Williams, L.A.J., Gass, I.G., 1994. Tectono-magmatic evolution of the Kenya rift valley: some geological perspectives. Journal of the Geological Society, London 151, 879-888.

MacDonald, R., Rogers, N.W., Fitton, J.G., Blade, S., Smith, M., 2001. Plume-Lithosphere interactions in the generation of the basalts of the Kenya Rift, East Africa. Journal of Petrology 42 (5), 877-900.

McKenzie, D., Bickle, M.J., 1988. The volume and composition of melt generated by extension of the lithosphere. Journal of Petrology 29, 625-679.

Milman-Barris, M., Beckett, J., Baker, M., Hofmann, A., Morgan, Z., Crowley, M., Vielzeuf, D., Stolper, E., 2008. Zoning of phosphorus in igneous olivine. Contributions to Mineralogy and Petrology 155, 739-765.

Münker, C., Weyer, S., Scherer, E., Mezger, K., 2001. Separation of high field strength elements (Nb, Ta, Zr, Hf) and Lu from rock samples for MC-ICPMS measurement. Geochemistry Geophysics Geosystems 2. doi:10.1029/2001GC000183.

Nokleberg, W.J., Parfenov, L.M., Monger, J.W.H., Norton, I.O., Khanchuk, A.I., Stone, D.B., Scotese, C.R., Scholl, D.W., Fujita, K., 2000. Phanerozoic tectonic evolution of the Circum-North Pacific. Professional Paper 1626, U.S. Geological Survey. 122 pp.

Parfenov, L.M., Koz'min, B.M., Grinenko, O.V., Imaev, V.S., Imeava, L.P., 1988. Geodynamics of the Chersky seismic belt. Journal of Geodynamics 9, 15-37.

Parfenov, L.M., Badarch, G., Berzin, N.A., Khanchuk, A.I., Kuzmin, M.I., Nokleberg, W.J., Prokopiev, A.V., Ogasawara, M., Yan, H., 2009. Summary of Northeast Asia geodynamics and tectonics. Stephan Müller Special Publication Series 4, 11-33.

Pearce, J.A., Cann, J.R., 1973. Tectonic setting of basic volcanic rocks determined using trace element analyses. Earth and Planetary Science Letters 19, 290-300.

Putirka, K.D., 2005. Mantle potential temperatures at Hawaii, Iceland, and the midocean ridge system, as inferred from olivine phenocrysts: evidence for thermally driven mantle plumes. Geochemistry, Geophysics, Geosystems, 6.

Putirka, K.D., Perfit, M., Ryerson, F.J., Jackson, M.G., 2007. Ambient and excess mantle temperatures, olivine thermometry, and active vs. passive upwelling. Chemical Geology 241, 177-206.

Renne, P.R., Swisher, C.C., Deino, A.L., Karner, D.B., Owens, T.L., DePaolo, D.J., 1998. Intercalibration of standards, absolute ages and uncertainties in 40Ar/39Ar dating. Chemical Geology 145,117-152.

Roddick, J.C., 1988. The assessment of errors in 40Ar/39Ar dating. Radiogenic Age and Isotopic Studies: Report 2: Geological Survey of Canada, Paper, 88-2, pp. 7-16.

Salters, V.J.M., Longhi, J., 1999. Trace element partitioning during the initial stages of melting beneath mid-ocean ridges. Earth and Planetary Science Letters 166, 15-30.

Salters, V.J.M., Longhi, J.E., Bizimis, M., 2002. Near mantle solidus trace element partitioning at pressures up to 3.4 GPa. Geochemistry Geophysics Geosystems 3, 7.

Scaillet, S., 2000. Numerical error analysis in 40Ar/39Ar dating. Chemical Geology 162, 269-298.

Scarrow, J.H., Cox, G., 1995. Basalts generated by decompressive adiabatic melting of a mantle plume: a case study from the Isle of Skye, NW Scotland. Journal of Petrology 36 (1), 3-22.

Steiger, R.H., Jäger, E., 1997. Subcommission on geochronology: convention on the use of decay constants in geo- and cosmochronology. Earth and Planetary Science Letters 36, 359-361.

Stracke, A., Bizimis, M., Salters, V.J.M., 2003. Recycling oceanic crust: quantitative constraints. Geochemistry Geophysics Geosystems 4, 3.

Stracke, A., Hofmann, A.W., Hart, S.R., 2005. FOZO, HIMU, and the rest of the mantle zoo. Geochemistry Geophysics Geosystems 6.

Sun, S., McDonough, W.F., 1989. In: Saunders, A.D., Norry, M.J. (Eds.), Chemical and Isotopical Systematic of Oceanic Basalts: Implications for Mantle Composition and Processes: Magmatism in the Ocean Basins: Geological Society Special Publication, 42, pp. 313-345.

Surnin, A.A., Okrugin, A.V., Zaitzev, A.I., 1998. Deep-seated xenoliths in basalts of Eastern Yakutia. Otechestvennaya geologia 6,44-48 (in Russian).

Taylor, S.R., McLennan, S.M., 1985. The Continental Crust: its Composition and Evolution. Blackwell Scientific Publications. 312 pp.

Thöni, M., Miller, C., Blichert-Toft, J., Whitehouse, M.J., Konzett, J., Zanetto, A., 2008. Timing of high-pressure metamorphism and exhumation of the eclogite type-locality (Kupplerbrunn-Prickler Halt, Saualpe, south-eastern Austria): constraints from correlations of the Sm-Nd, Lu-Hf, U-Pb and Rb-Sr isotopic systems. Journal of Metamorphic Geology 26, 561 -581.

Todt, W., Cliff, R.A., Hanser, A., Hofmann, A.W., 1996. Evaluation of a 202Pb-205Pb double spike for high-precision lead isotope analysis. In: Hart, S.R., Basu, A. (Eds.), Earth Processes: Reading the Isotope Code: Geophysical Monographs, 95, pp. 429-437. AGU.

Tschegg, C., Ntaflos, Th., Akinin, V.V., 2011. Polybaric petrogenesis of Neogene alkaline magmas in an extensional tectonic environment: Viliga Volcanic Field, northeast Russia. Lithos 122,13-24.

Turcotte, D.L., Emerman, S.H., 1983. Mechanisms of active and passive rifting. Tectonophysics 94, 39-50.

Villeneuve, M.E., MacIntyre, D.G., 1997. Laser 40Ar/39Ar ages of the Babine porphyries and Newman Volcanics, Fulton Lake map area, west-central British Columbia. Radiogenic age and isotopic studies, Report 10: Geological Survey of Canada, Current Research 1997-F, pp. 131-139.

Villeneuve, M.E., Sandeman, H.A., Davis, W.J., 2000. A method for the intercalibration of U-Th-Pb and 40Ar/39Ar ages in the Phanerozoic. Geochimica et Cosmochimica Acta 64,4017-4030.

Vogt, P.R., Taylor, P.T., Kovacs, L.C., Johnson, G.L., 1979. Detailed aeromagnetic investigation of the Arctic Basin: Journal of Geophysical Research 84, 1071 -1090.

Weaver, B.L., 1991. The origin of ocean island basalt end-member composition: trace-element and isotopic constraints. Earth and Planetary Science Letters 104,381 -397.

Willbold, M., Stracke, A., 2006. Trace element composition of mantle end-members: Implications for recycling of oceanic and upper and lower continental crust. Geochemistry Geophysics Geosysystems 7,4.

Zindler, A., Hart, S., 1986. Chemical geodynamics. Annual Reviews of Earth and Planetary Science 14,493-571.