Petrogenises of Triassic Gabbroic and Basaltic Rocks from Chukotka, NE Russia: Eastern End of the ‘Arc–Type’ Siberian LIP?

Journal of Mineralogical and Petrological Sciences, Volume 110, page 249–275, 2015

Petrogenises of Triassic gabbroic and basaltic rocks from Chukotka, NE Russia: Eastern end of the ‘arc–type’ Siberian LIP?

Minyahl Teferi DESTA*,§, Akira ISHIWATARI**,§§, Sumiaki MACHI**,***, Shoji ARAI***, Akihiro TAMURA***, † † † ‡ Galina V. LEDNEVA , Sergey D. SOKOLOV , Artem V. MOISEEV and Boris A. BAZYLEV

*Department of Earth Science, Graduate School of Science, Tohoku University, Sendai 980–8578, Japan **Center for Northeast Asian Studies, Tohoku University, Sendai 980–8576, Japan ***Department of Earth Sciences, Faculty of Science, Kanazawa University, Kanazawa 920–1192, Japan †Geological Institute, Russian Academy of Sciences, Moscow 119017, Russia ‡Vernadsky Institute of Geochemistry and Analytical Chemistry, Russian Academy of Sciences, Moscow 119991, Russia §Present address: Department of Earth Science, College of Science, Bahir Dar University, P.O.Box 79, Bahir Dar, Ethiopia §§Present address: Nuclear Regulation Authority, Tokyo 106–8450, Japan

The Triassic gabbroic intrusions and associated basaltic lavas from Chukotka are mainly tholeiitic with both ocean island basalt (OIB) and island arc basalt (IAB)–type geochemical signatures. Mg–number [Mg# = 100 × Mg/(Mg + Fe2+)] is around 40 for OIB–type gabbros, ranges from 48 to 66 for IAB–type hornblende–rich gabbros, and 43 to 65 for IAB–type basaltic rocks (ankaramites, lamprophyres, pyroxene–phyric basalts and hornblende–phyric basaltic andesite). TiO2 contents of the IAB–type gabbros and basaltic rocks are low (<2 wt%), but are high in OIB–type gabbros (4.3–5.3 wt%). OIB–type gabbros are typically enriched in FeO* (16–18 wt%) as compared to IAB–type gabbros (10–14 wt%) and IAB–type lavas (ankaramites, ~ 10 wt%; lamprophyres, ~ 14; pyroxene–phyric basalt, 11 wt% and basaltic andesite, 9–10 wt. %). In the primitive mantle normalized trace element patterns, IAB–type basalts and gabbros are characterized by depletion in HFSE (Nb, Ta, Zr and Hf) and enrichment in LILE. OIB–type gabbros can be distinguished from the rests by the absence of HFSE depletion, with strong negative Sr anomaly. The positive Ti anomaly in the OIB–type gabbros can be attributed to high content of ilmenite in these rocks. Trace element characteristics of IAB–type gabbroic rocks and basalts are compatible with their magmas derived from subduction inﬂuenced melts, whereas OIB–type gabbros show within–plate geochemical characteristics. IAB–type gabbros and basaltic rocks display similar geochemical features to the low–Ti Nadezhdinsky suit (Noril’sk region) and Bel’kov dolerite (New Siberian Islands) of the Siberian large igneous province (LIP) in view of HFSE depletion and high H2O content of the magma to crystallize abundant hornblende not only in gabbros but also as phenocrysts in basalts. The Triassic gabbroic and basaltic rocks of both OIB and IAB types may as a whole represents the eastern end of the Siberian LIP.

Keywords: Island–arc geochemistry, Intra–plate geochemistry, Ankaramite, Hornblende basalt, HFSE

INTRODUCTION However, on the basis of high field strength element (HFSE) contents of LIP lavas Puffer (2001) described Large igneous provinces (LIPs) are huge volume, the existence of both island arc basalt–type and intra–plate short duration intraplate–type magmatic events consisting basalt or ocean island basalt–type LIPs (hereafter we de- mainly of flood basalts and their associated plumbing sys- noted as IAB–type and OIB–type LIPs, respectively). Si- tems (Coffin and Eldholm 2005; Bryan and Ernst 2008). berian LIP is one of the most voluminous (~ 4 × 106 km3; Ivanov, 2007) and the representative examples of IAB– doi:10.2465/jmps.150504 type LIPs, because of the majority of the erupted low–Ti M.T. Desta, [email protected] Correspnding author basalts are characterized by IAB–type trace element pat- 250 M.T. Desta, A. Ishiwatari, S. Machi, S. Arai, A. Tamura, G.V. Ledneva, S.D. Sokolov, A.V. Moiseev and B.A. Bazylev terns (Puffer, 2001; Ivanov, 2007). The high–Ti series croplate (AACM). rocks (e.g., meimechites; Arndt et al., 1998; Sobolev et al., 2009) are volumetrically insignificant and commonly GEOLOGICAL SETTING display OIB–type geochemical features. The IAB–type geochemical feature of the low–Ti basalts of Siberian The study area (Fig. 1B) is situated within the Arctic LIP are traditionally attributed to a fingerprint of litho- Alaska–Chukotka microplate (AACM) or Chukotka mi- spheric contamination (Lightfoot et al., 1993; Reichow crocontinent (Parfenov et al., 1993). The AACM is com- et al., 2005). However, the presence of strong arc–type posed of a Neoprotrozoic crystalline basement (Natal’in geochemical signatures and primary magmatic water– et al., 1999, Amato, et al., 2009), overlain by Carbonif- bearing minerals in the low–Ti series rocks of Siberian erous and Permian platform strata (P’yankov, 1981; LIP (Reichow et al., 2002; Ivanov 2007) indicate that con- Natal’in et al., 1999). These units are, in turn, overlain tamination of lithospheric mantle cannot be the control- by a Permian to upper Triassic terrigenous continental ling factor. Therefore, the dominant IAB–type, low–Ti ba- shelf sedimentary sequence (Tuchkova et al., 2009) and salts within the Siberian LIP have been attributed to wet a Jurassic volcanic–terrigenous foreland sedimentary mantle upwelling, driven by dehydration of the Mongo- package (Vatrushkina and Tuchkova, 2014). lia–Okhotsk oceanic slab at the mantle transition zone To the south, the AACM is bounded by the South (Ivanov and Litasov, 2014). Anyui Suture (SAS) zone, which was formed by the clo- Eastern Chukotka is interpreted as a part of Arctic sure of a late Jurassic–early Cretaceous oceanic basin Alaska–Chukotka microplate, which formed during the and the collision of the North Asian craton with AACM Cretaceous opening of the Amerasian basin (Figs. 1A (Seslavinsky, 1979; Parfenov, 1984). This suture zone is and 1B). It has been proposed that it rifted either from buried under the Cretaceous Okhotsk–Chukotka volcanic the Arctic Canadian margin (Lawver and Scotese, 1990; belt (Tikhomirov et al., 2012). In western Chukotka, the Lawver et al., 2002) or from the Siberian craton (Miller et suture is marked by Mesozoic terrigenous turbidite and al., 2006). The latter is supported by U–Pb dates of de- fragments of ophiolite sequences (Sokolov et al., 2002), trital zircons from the Triassic terrigenous sediments of but in eastern Chukotka the suture corresponds to the eastern Chukotka (Miller et al., 2006). The hypabyssal Velmai terrane (Parfenov et al., 1993; Nokleberg et al., mafic rocks (dolerites) of the New Siberian Archipelago 1998; Sokolov et al., 2009). (Bel’kov Island) belongs to north–eastern margin of Si- In the study area the gabbroic intrusions occur as berian LIP and display IAB–type geochemical features flat–lying tabular bodies ranging in thickness from a few (Kuz’michev and Pease, 2007). The mafic magmatism meters to a few hundred meters, and are mainly confined in eastern Chukotka was also thought to be an eastern to the upper Permian–lower Triassic and lower–mid- extension of Siberian LIP (Gel’man 1963; Kuz’michev dle Triassic strata of the AACM cover (Fig. 1C). Howev- and Pease 2007). Gabbros and associated basaltic rocks er, the upper Triassic sedimentary units overlying these have been reported from eastern Chukotka (Ledneva et older sequences are devoid of tabular bodies of gabbroic al., 2011, 2014) and are generally display intra–plate ba- intrusions (Til’man and Sosunov, 1960; Gel’man, 1963; salt type geochemical features. We found that the IAB– Degtyaryov, 1975). In the western Chukotka these intru- type gabbros and basalts are dominant in eastern Chukot- sions extend for ~ 350 km (the Keperveem and Raucha ka. This may strengthen geological connection between Uplifts); in central and eastern Chukotka they are traced to the Siberian LIP and the eastern Chukotka mafic rocks, a distance of ~ 900 km (the vicinity of Cape Schmidt and and provide further insights into the LIPs magma genesis the interfluves of the Amguema and Vel’may rivers and and tectonics. the Kolyuchinskaya Bay area) (Fig. 1B; Ledneva et al., In this study, we present XRF whole rock geochem- 2014). These gabbroic rocks were recognized as parts of ical analyses, electron microprobe analyses of hornblende, a large regional complex (Til’man and Sosunov, 1960; clinopyroxene, plagioclase and Fe–Ti oxide, LA–ICP–MS Gel’man, 1963) named as the Amguema–Anyui Igneous analyses of bulk–rock, hornblende and clinopyroxene for Province (Degtyaryov, 1975). the Triassic gabbros and associated basaltic rocks of Chu- The Permian to Early–Middle Triassic age of these kotka in order to place constraints on the origin of their intrusions is proven by (1) their stratigraphic position parental magmas. We compared our results with the pre- within the Permian–lower Triassic and lower–middle viously published data from Siberian LIP in search for Triassic terrigenous sediments, contemporaneous joint genetic relationships between these igneous provinces. deformation of gabbroic bodies and their country rocks Our findings also provide additional constraints for the (Til’man and Sosunov, 1960; Gel’man, 1963), (2) bulk– much debated origin of the Arctic Alaska–Chukotka mi- rock K–Ar determinations of 250, 231, and 223 Ma Siberian LIP 251

Figure 1. (A) A position of the AACM among major geological structures of Arctic region (modiﬁed after Miller et al., 2006). AACM is shown in solid line and its inferred continuation is in dotted line; solid lines with ticks mark other signiﬁcant suture zones. Abbreviations: AACM, Arctic Alaska–Chukotka microplate; AN, Angayucham belt; BR, Brooks Range; CR, Chersky Range; KB, Kolyuchinskaya Bay; LI, Lisburn Hills; NSA, New Siberian Archipelago; SAS, South Anyui suture; WI, Wrangel Island; WHA, Wrangel, Herald Arch; WSB, West Siberian Basin. (B) General geology of Chukotka microplate after Ledneva et al. (2014) and position of study area. Abbreviations: AACM, Arctic Alaska–Chukotka microplate; OChVB, Okhotsk–Chukotka volcanic belt; VT, Vel’may terrane. (C) Sketch geological map of the Chukotka peninsula showing the sampling location of the Triassic basaltic and gabbroic rocks. 252 M.T. Desta, A. Ishiwatari, S. Machi, S. Arai, A. Tamura, G.V. Ledneva, S.D. Sokolov, A.V. Moiseev and B.A. Bazylev

Figure 2. (A) Field photographs showing the occurrence of the studied gabbros as plat formal ediﬁce tabular hills behind the ﬁeld vehi- cle. (B) Fine grained gabbroic dike occurring within coarse grained melanocratic gabbro. (C) Details showing coarse–grained IAB–type hornblende gabbro. (D) OIB–type gabbro cut by felsic veins. (E) Pil- low lava of basalt that is closely associated with dolerite and gabbro in the S4 area. (F) Hand specimen of hornblende basaltic andesite showing large brown amphibole phenocrysts with a size as large as 1 cm (the coin is 22 mm in diameter).

(Ivanov and Milov, 1975; Degtyaryov, 1975), and (3) and siltstone, and mudstone and bedded tuff. The most mag- the U–Pb TIMS igneous zircon age of 252 ± 4 Ma ob- nesian samples (i.e., ankaramites, Sample#69, #70, and tained for the Kolyuchinskaya Bay gabbro (Sokolov et #71) of this study come from the sampling locality 4 (Fig. al., 2009). 1C; Tables 1 and 2). The gabbroic intrusions described by Ledneva et al. (2014) are characterized by intra–plate (i.e. OIB–type) ba- ANALYTICAL TECHNIQUES saltic geochemical features. However, our mineralogical and geochemical data confirm the presence of both island Electron microprobe analyses were conducted on horn- arc basalt (IAB)–type and OIB–type mafic intrusions. blende, clinopyroxene, plagioclase and Fe ± Ti oxide min- These two gabbro types were likely derived from two erals in polished thin section using an energy dispersive different mantle sources. Our OIB–type gabbros are anal- X–ray spectrometer Oxford Link ISIS equipped on the ogous to those described by Ledneva et al. (2014) in JEOL JSM–5410 Scanning electron microscope (SEM) terms of petrography and geochemistry. In the study area, at the Earth Science department, Tohoku University. these gabbroic rocks and the associated basaltic units The operating conditions were at 15 kV acceleration volt- (i.e., ankaramites, pyroxene–phyric basalts, lamprophyres age and a beam current of 1 nA on Co standard. Major and and basaltic andesites) commonly occur together and ex- some trace elements (V, Cr, Ni, Rb, Sr, Ba, Y, Zr, Nb) were posed as platformal edifices (Fig. 2A). The basaltic lavas obtained using X–ray fluorescence spectrometer (XRF– rarely exhibit pillow structure (at S4 of Fig. 1C, see Fig. RIX 2100) on fused glass discs at the same institute. 2E). They are commonly interbedded with sandstone, The loss on ignition (LOI) was determined following Siberian LIP 253

Table 1. Major and trace element compositions for the gabbroic and basaltic rocks of Chukotka, NE, Russia

the procedure described by Desta et al. (2014). The trace roxene gabbro. They display a notable grain size varia- element analysis of whole rock and mafic minerals (i.e., tion; from medium grained (subordinately fine grained) clinopyroxene and amphibole) have been carried out by a to coarse crystals with size reaching about 20 mm. They laser ablation (193 nm ArF exciemer: MicroLas GeoLas have a granular texture with hornblende, pyroxene and Q–plus)–inductively coupled plasma spectrometer Agilent plagioclase (Fig. 3A). Magnetite and Ti–magnetite min- 7500S (LA–ICPMS) at Kanazawa University (Ishida et erals occur dominantly with variable size and shape. Rare al., 2004). apatite grains are also observed as accessory minerals in pyroxene–hornblende gabbro. PETROGRAPHY The IAB–type lavas comprised of basaltic rocks, such as ankaramites (Sample #69, #70, and #71), pyrox- On the basis of petrography, together with the results of ene–phyric basalt (Sample #67) and basaltic andesites mineral and bulk–rock major and trace element analyses (Sample #55 and #56). We have also identified three the Triassic mafic rocks of Chukotka are subdivided into lamprophyre samples based on their high modal abun- two types: (1) IAB–type gabbros and basalts (with ankar- dance of amphibole phenocrysts (Sample #59, #74, and amite, pyroxene–phyric basalt, lamprophyre and basaltic #77) (hereafter we refer these samples as basaltic rocks andesite sub types) and (2) OIB–type gabbros. for the sake of simplicity). Ankaramites are highly porphyritic with clinopyroxene (~ 40 vol%) as the only phe- IAB–type gabbroic and basaltic rocks nocryst. The groundmass mainly composed of plagioclase, amphibole, clinopyroxene, and Fe–Ti oxide. Pyrox- IAB–type gabbroic rocks comprised of amphibole (30–60 ene–phyric basalt is porphyritic with phenocrysts consist- vol%)–bearing gabbroic rocks such as pyroxene–horn- ing predominantly of clinopyroxene (~ 25 vol%). Clino- blende gabbro, hornblende gabbro, and hornblende–py- pyroxene phenocrysts commonly occur as broken pieces 254 M.T. Desta, A. Ishiwatari, S. Machi, S. Arai, A. Tamura, G.V. Ledneva, S.D. Sokolov, A.V. Moiseev and B.A. Bazylev

Table 1. (Continued)

* Total Fe as FeO. n.d., not determined; LOI, loss on ignition. suggesting an agglutinate and exhibit well developed thetic twinning. Fe–Ti oxides are represented by elongat- zoning (Fig. 3D). Lamprophyres are dark grey to green ed grains of ilmenite (5–10 vol%). in color with highly porphyritic texture. They are characterized by the occurrence of abundant amphibole phenoc- WHOLE–ROCK MAJOR AND TRACE rysts to megacrysts, which sometimes reaches as large as ELEMENT RESULTS 2 cm long, set in a groundmass composed of mainly amphibole, plagioclase, K–feldspar and clinopyroxene. The Representative, major element chemical data for the Tri- amphibole phenocrysts sometimes clustered in a glomer- assic gabbroic and basaltic rocks of Chukotka are pre- oporphyritic texture (Fig. 3E). Basaltic andesites are char- sented in Table 1. All the analyzed samples are tholeiitic acterized by porphyritic texture with amphibole, clinopy- in composition, although, one basaltic andesite sample roxene and plagioclase pheocrysts. Amphibole pheno- plot on the dividing line between the calc–alkaline and crysts sometimes reach as large as 1 cm (Fig. 2F). The tholeiitic fields on the SiO2–FeO*/MgO diagram of groundmass is composed of plagioclase, amphibole, Fe– Miyashiro (1974) (Figure not shown). In the AFM Ti oxide, rare biotite, quartz and pyroxene. (Na2O+K2O–FeO*–MgO) diagram, the studied samples showing a tholeiitic affinity, following iron enrichment OIB–type gabbros trend (Fig. 4A). These rocks have Nb/Y ratios of 0.09– 0.59, plotting in the sub–alkaline basalt field on the Nb/Y This rock type consists mainly of clinopyroxene and pla- versus Zr/Ti diagram of Winchester and Floyd (1977) and gioclase (locally transformed to sericite) (Fig. 3B). They Pearce (1996) (Fig. 4B), except one sample of OIB–type contain rare to no amphiboles. Plagioclase forms euhe- gabbro falling close to or in the alkaline field. dral and subhedral crystals and usually display polysyn- Variation diagrams for major element oxides against Siberian LIP 255

Table 2. Bulk–rock trace element analyses of maﬁc rocks of Chukotka, NE Russia

pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Note: Eu/Eu* = EuN/ SmN GdN, subscript N denotes normalized to Primitive mantle (McDonough and Sun, 1995).

MgO wt% are presented in Figures 5A–5C. Mg–numbers are low (<2 wt%) except OIB–type gabbros (4.3–5.3 [Mg# = 100 × Mg/(Mg + Fe2+)] range from 48 to 66 for wt%). IAB–type gabbros have higher CaO contents (13– IAB–type gabbros, around 40 for OIB–type gabbros, 43– 19 wt%) (except Sample #78, ~ 7 wt%) compared to 65 for IAB–type basaltic rocks (Table 1). The low to in- OIB–type gabbros (5–9 wt%). OIB–type gabbros typical- termediate Mg–number of the studied samples suggests ly enriched in FeO* (16–18 wt%) as compared to IAB– that these rocks do not represent primary, mantle–derived type gabbros (10–14 wt%), IAB–type basalts (~ 10 wt%, melts. This is consistent with the low concentrations of ankaramites; 11 wt%, pyroxene–phyric basalt; ~ 14 wt%, compatible trace element (such as Cr and Ni) in the stud- lamprophyres) and IAB–type basaltic andesite samples ied samples. The TiO2 contents of all the studied samples (9–10 wt%). Ankaramites have higher CaO/Al2O3 (0.99– 256 M.T. Desta, A. Ishiwatari, S. Machi, S. Arai, A. Tamura, G.V. Ledneva, S.D. Sokolov, A.V. Moiseev and B.A. Bazylev

Figure 3. Thin section photomicro- graphs of the Triassic gabbroic and basaltic rocks of Chukotka, NE Rus- sia. (A) Coarse–grained hornblende– gabbro (IAB–type) showing euhedral crystals of amphibole in contact with clinopyroxene and altered plagioclase (Open Nicols). (B) OIB– type gabbros showing ophitic texture and ilmenite grains (Crossed nicols). (C) Crossed–nicols view of ankaramite (Sample #69) showing porphyritic texture. (D) Fragments of clinopyroxene phenocrysts with broken sides in pyroxene–phyric basalt (Sample #67), showing normal zoning. (E) Lamprophyre (Sample #77) displaying a cluster of euhedral to sub–hedral phenocrysts of amphibole (Crossed nicols). (F) Back scat- tered electron (BSE) image showing hornblende phenocryst in basaltic andesite (Sample #55; the two circu- lar injuries were caused by LA–ICP– MS beams). Amp, amphibole; Cpx, clinopyroxene; Pl, plagioclase; Mt, Ti–Magnetite; Ilm, ilmenite.

1.04) ratios compared to pyroxene–phyric basalt, lamp- amite have low Ni (~ 81 ppm) at high Cr (>500 ppm). rophyres, and basaltic andesites (0.83, 0.82–0.86, and Incompatible elements (Such as Zr and Nb) are very 0.43–0.44, respectively). Moreover, ankaramite samples low in abundance for IAB–type gabbros and basalts, but have high Cr/Ni (6.5–7.6), and Sc/Ni (~ 0.9) ratios, char- are high in OIB–type gabbros and IAB–type basaltic an- acteristic of arc ankaramites (Barsdell and Berry, 1990). desite, and increase with decreasing MgO. Total rare earth IAB–type gabbroic rocks generally have lower Na2O than element (ΣREE) is low in IAB–type gabbros (32–72 ppm) the volcanic rocks. The Na2O concentrations in the stud- and basalts (ankaramite, ~ 38 ppm; pyroxene–pyric basalt, ied samples increase with deceasing MgO content, where- ~ 63 ppm; and lamprophyres, 52–56 ppm) but is high in as the K2O content of gabbroic rocks (except basaltic basaltic andesites (~ 174 ppm) and OIB–type gabbros rocks) shows more scatter (Figure no shown). In general, (138–175 ppm). the concentration of major oxides in the OIB–type gab- In the primitive mantle normalized multi–trace ele- bros is similar to the gabbroic rocks of eastern Chukotka ment plots (Fig. 6); IAB–type gabbroic and basaltic rocks (Ledneva et al., 2014), but the former is relatively higher are characterized by depletion of high ﬁeld strength ele- in Na2O and lower in K2O contents. ments (HFSE), such as Nb, Ta, Zr, and Hf, relative to the Trace element compositions of the analyzed samples neighbouring elements, similar to those of arc–type igne- are provided in Tables 1 and 2. Variation diagrams (Figs. ous rocks (Rudnick, 1995; Taylor and McLennan, 1995). 5D–5F) show that the compatible elements (Ni and Cr) They are depleted in HREE compared with N–MORB. In decrease with decreasing MgO. The Ni content of the addition, these rocks exhibit depletion in Rb and Th rela- studied samples are low, and even the most maﬁc ankar- tive to Ba and U, respectively, and show positive anoma- Siberian LIP 257

OIB–type gabbroic rocks can be distinguished from the rest of the studied samples based on the absence of HFSE depletion and presence of strong negative Sr anomalies (Fig. 6A). Moreover, their HFSE are slightly enriched relative to LREE [(La/Nb)N = 0.69–1.01] (Fig. 10A) and Nb/U = 33–58. In the primitive mantle–normalized REE diagram (Fig. 6B), these gabbro samples have similar LREEs contents to that of the ocean island basalt (OIB), but higher HREEs contents. The positive Ti anomaly in the OIB–type gabbros can be attributed to their high content of ilmenite.

MINERAL CHEMISTRY

Representative, major element compositions of minerals from the studied gabbroic and basaltic rocks are available in the supplementary Tables. IAB–type gabbros. The composition of clinopyroxene crystals ranges from Wo39En38Fs9 to Wo51En48Fs13. Generally, on a pyroxene quadrilateral diagram of Mori- moto et al. (1988) (figure not shown) the clinopyroxene Figure 4. Classification diagrams for the mafic rocks of Chukotka: compositions exhibit an alkaline trend (Wo increase with – (A) AFM (Na2O+K2O–FeO*–MgO) triangle diagram; (B) im- Fs). They contain relatively low TiO2 (0.17 1.25 wt%, mobile element ratios such as Nb/Y versus Zr/Ti distinguishing average 0.70 wt%) and high CaO (22–25 wt%, average subalkaline and alkaline basalts (Winchester and Floyd, 1977, 22.9 wt%) and Mg# (76.9–90.2, average 78.3). Amphi- modified by Pearce, 1996). boles are pargasite following the classification of Leake et al. (1997). Their Mg# ranges from 63.6 to 70.4. Mg– numbers in amphibole are mostly lower than those of lies of Ba and K (and Pb in one IAB–type gabbro sam- clinopyroxene, and interpreted as a late magmatic miner- ple). However, the concentrations of Th and U in IAB– al. Plagioclase compositions are An100–46. Biotites have type gabbros are variable. All the analyzed IAB–type Mg# values ranging from 64 to 66. The TiO2 content gabbroic and basaltic rocks have ThN/LaN < 1, and two of magnetite varies from 0.17 to 0.43 wt%. Titano–mag- samples (Sample #75 of IAB–type gabbro and Sample netite contains TiO2 ranging from 3.4 to 5.1 wt%. #69 of IAB–type basaltic rocks) have ThN/LaN ~ 1.2 (sub- IAB–type lavas. Clinopyroxene in basaltic rocks is script N denotes primitive mantle normalized values; nor- diopside and salite (Wo45–50En36–50Fs4–14), with the aver- malization values are from McDonough and Sun (1995). aged TiO2 contents ranging from 0.23 to 0.85 wt% and The most basic intrusive rock (Sample #75 of IAB–type Al2O3 from 1.83 to 4.84 wt%, whereas clinopyroxene in gabbros) is depleted in LREE compared to MREE similar basaltic andesite is made largely of augite (Wo29–42En31–40- to NMORB, but the former is more depleted in HREE Fs17–37) with 0.59–0.77 wt% TiO2 and 2.22–2.71 wt% (Fig. 6B). We see strong positive Sr anomalies in IAB– Al2O3. Furthermore, majority of the clinopyroxene from type gabbroic and basaltic rocks. basaltic rocks define a clear arc cumulate trend in the AlZ IV Basaltic andesites share similar patterns of depletion (Al × 100/2) versus TiO2 (wt.diagram (Loucks, 1990; in HFSE (Nb, Ta) to those displayed by basaltic rocks. see later section). Amphibole phenocrysts in lampro- However, they do not show negative Zr and Hf anomalies phyres and basaltic andesites are pargasite, whereas (Fig. 6C). They show enrichment in Th [(Th/La)N > 1.5] groundmass amphiboles in ankaramite are edenitic in and U relative to HFSE (Nb/U = 7.8–9.3). In the primitive composition following the classification of Leake et al. mantle normalized REE plot (Fig. 6D), they display ele- (1997). Plagioclase has composition An65–74 in lampro- vated trace element and total REE concentration compared phyres and An56–85 in hornblende basaltic andesite. Most to the basaltic rocks. Moreover, they exhibit strong LREE of the plagioclases from ankaramite and phyroxene–phyric enrichment [(La/Yb)N = 7.4–7.7], with a slightly concave– basalt are albite and probably they are alteration product. upward heavy REE (HREE) distribution and have slight Biotites in the basaltic andesite (Sample #56) have a com- negative Eu anomalies (Eu/Eu* ~ 0.85) (Fig. 6D). positional range of Mg# (51–53) and the K2O ranges from 258 M.T. Desta, A. Ishiwatari, S. Machi, S. Arai, A. Tamura, G.V. Ledneva, S.D. Sokolov, A.V. Moiseev and B.A. Bazylev

Figure 5. Plots of MgO (wt%) versus TiO2 (wt%), Al2O3 (wt%), FeO* (wt%), Cr (ppm), Nb (ppm) and Y (ppm) for Triassic maﬁc rocks of Chukotka, NE Russia. The shaded area shows the bulk–rock composition of eastern Chukotka gabbroic rocks reported by Ledneva et al. (2014). Symbols are in Figure 5B.

6.9 to 9.4 wt%. Opaque minerals from the pyroxene– wt% of TiO2) is a typical oxide phase in basaltic andesite phyric basalt are predominantly iron–sulphide minerals. samples. Lamprophyres includes both Ti–bearing (0.3 < TiO2 < OIB–type gabbros. The composition of analyzed 19.5 wt%) and Ti–free magnetite. Ilmenite (46.5–50.0 clinopyroxenes is: Wo39–46En34–39Fs17–23 and plots across Siberian LIP 259

Figure 6. Primitive mantle–normalized trace element and REE patterns for the gabbroic (A), (B) and basaltic rocks (C), (D) of Chukotka, NE Russia. Mantle–normalization values are from McDonough and Sun (1995). Multi–element and REE patterens of Siberian LIP (average value of Nadezhdinsky suit; Wooden et al., 1993), Kamchatka arc (average value of eastern volcanic front; Churikova et al., 2001), OIB and N–MORB (Sun and McDonough, 1989) are shown for comparison. Logarithmic scale. the diopside–augite field boundary in the pyroxene quad- pared to those in IAB–type gabbros. The rare amphibole rilateral diagram of Morimoto et al. (1988) (Figure not is edenite or hornblende following the classification of shown). Mg# is low, ranging between 53 and 71 (average, Leake et al. (1997). Plagioclase compositions vary from 64). They contain higher TiO2 (0.7–2.7 wt%, average 1.0 labradorite to alkali–feldspars (i.e., An1–57Ab99–42Or0–25). wt%) and lower CaO (18.3–21.1, average 20.1 wt%) com- Biotite has Mg# about 30. Opaque minerals in OIB–type 260 M.T. Desta, A. Ishiwatari, S. Machi, S. Arai, A. Tamura, G.V. Ledneva, S.D. Sokolov, A.V. Moiseev and B.A. Bazylev gabbros are mainly ilmenite. phase in the IAB–type gabbros and basalts, but amphibole in basaltic andesites. DISCUSSION OIB–type gabbros. As shown in Table 1, the OIB– type gabbros have a low MgO content and a low Mg num- Fractional crystallization ber (Mg#) ranging from 29 to 40, suggesting an evolved compositional feature. They display a positive correlation Primary basaltic magmas resulting from partial melting between Cr and CaO versus MgO, indicating fractional of upper mantle peridotites should have Mg# in the range crystallization of clinopyroxene. TiO2 and FeO* correlate of 68–75 (Green, 1976; Hanson and Langmuir, 1978). positively with MgO (Fig. 5) indicates fractionation of The studied gabbroic and basaltic rocks display different Fe–Ti oxide. They have strong negative Sr anomalies in evolutionary trends, suggesting that they followed differ- multi–element spider diagram (Fig. 6A), indicating frac- ent paths of fractional crystallization or accumulation. tionation of plagioclase. However, absence of Eu anoma- IAB–type gabbros and lavas. The lack of negative lies in the trace element plot argues against significant Eu anomalies and the progressive increase in Al2O3 with crystallization of plagioclase. decreasing MgO for IAB–type gabbros and basalts sug- In summary, the IAB–type gabbros and basalts and gests an absence of significant plagioclase fractionation OIB–type gabbros experienced different fractionation (Figs. 5B and 6). A negative correlation between MgO processes. This, together with their distinct geochemical and FeO* for the IAB–type gabbros and basalts suggests characteristics, suggests that these two groups of mafic no fractionation of olivine (Fig. 5C). This interpretation rocks cannot be related by crystallization along a single is supported by the absence of this mineral in these rocks. liquid line of descent from a common parental magma. Most of the samples show trends of decreasing Dy/Yb P–T–f with differentiation index (SiO2) (Fig. 7A), this is inter- O2 conditions of magma chamber preted as significant amphibole fractionation (Davidson et al., 2007; Smith, 2014), because amphibole incorpo- The pressure, temperature, and oxygen fugacity condi- rates preferentially MREE compared to HREE (Tiepolo tions of amphibole crystallization can be obtained from et al., 2007). However, this interpretation is inconsistent amphibole compositions using the recent thermobaromet- with the absence of concave–upward REE patterns in the ric formulation of Ridolfi and Renzulli (2012). The ob- studied basaltic rocks (Fig. 6) that would be expected as a tained pressure for amphibole crystallization range from result of amphibole fractionation observed in some arc 4.1 to 8.5 kbar (average 6.8 kbar) in IAB–type gabbros lavas (e.g., Jolly et al., 2002; Figueroa et al., 2009), there- and range from 0.1 to 0.3 kbar (average 0.2 kbar) in fore, amphibole fractionation is insignificant. OIB–type gabbros, indicating an intrusion depth of about The slight negative Eu anomalies indicate fractiona- 22 km and 0.7 km respectively. Application of Ridolfi and tion of plagioclase in the IAB–type basaltic andesite. This Renzulli (2012) thermobarmeter for amphiboles from interpretation is supported by the presence of marked neg- IAB–type basaltic rocks yield different pressure estimates: ative Eu anomalies exhibited by amphibole and clinopy- 2.1–5.8 kbar for basaltic andesite, 3.6–6.9 kbar for lamp- roxene phenocrysts in basaltic andesites (Figures not rophyre, and generally lower (0.2–2.5 kbar) for ground- shown). The REE patterns of the basaltic andesite display mass amphiboles from ankaramite. a slightly concave–upward shape (Fig. 6D), that has been Equilibration temperatures for amphiboles were cal- a feature classically attributed to amphibole fractionation culated using the amphibole geothermobarometer of (Green and Pearson, 1985). This inference is confirmed by Ridolfi et al. (2010). Results show temperature values that the presence of amphibole phenocrysts in these rocks. It is range from 940–1010 °C for basaltic andesites, 900–1080 important to note that the effects of clinopyroxene fractio- °C for basaltic rocks, 930–1000 °C for IAB–type gab- nation are broadly similar to those of amphibole (David- bros and 720–800 °C for OIB–type gabbros. Ridolfi et son et al., 2007). In addition, variation of Zr/Sm ratio is al. (2010) reported uncertainties of ±22 °C for the amphi- also commonly attributed primarily to amphibole fractio- bole geothermometer. These temperature estimates are nation (Fig. 7B); however, augite fractionation (Thirlwall generally lower than the crystallization temperature of a et al., 1994) and magnetite fractionation (Tribuzio et al., normal basaltic magma (>1200 °C) (Lee et al., 2009). fl f 1999) may also exert an in uence on the Zr/Sm ratios. Calculated oxygen fugacity ( O2 ) values using the The relatively constant and negative correlation between Ridolfi and Renzulli (2012) yields results varying from TiO2, FeO* and MgO (Figs. 5A and C, respectively) im- NNO + 2.2 to NNO + 4.5 for IAB–type gabbros, whereas ply no significant fractionation of Fe–Ti oxides in the OIB–type gabbros ranges from NNO–3.9 to NNO + 0.5. f magma. Overall, clinopyroxene is the main fractionated In addition, the calculated O2 values show a large range Siberian LIP 261

Figure 7. Plots of SiO2 (wt%) versus Dy/Yb and Zr/Sm. Variation of Dy/Yb and Zr/Sm primarily attributed to amphibole fractionation. However, clinopyroxene fractionation may also exert an inﬂuence on these ratios. Fractionation trends for amphibole are after Davidson et al. (2007). The geochemical data of Bel’kov Island dolerite (Siberian LIP; after Kuz’michev and Pease, 2007) and eastern Chukotka gabbros (Ledneva et al., 2014) also shown for comparison.

from NNO + 8.4 to NNO + 11.3 for lamprophyres, from trace elements of the calculated liquid relative to the stud- NNO + 6.0 to NNO + 7.2 for ankaramites and from ied basaltic rocks due to the increase in the value of clino- NNO–5.4 to NNO + 7.9 for basaltic andesites. pyroxene/liquid partition coefficient. This interpretation is consistent with the absence of olivine in the IAB–type gab- Parental magmas bros, indicating the evolved nature of the parental melt. Ankaramites of Chukotka (Sample #69 and #71) IAB–type gabbros and lavas. The trace element compo- have Mg# of 63–65, Cr–contents of 466–530 ppm, and sition of parental melts in equilibrium with clinopyroxens Ni–contents of 67–81 ppm. They are also characterized from the studied samples have been calculated using clino- by very low REE abundance with a strong depletion of pyroxene/melt partition coefficients after Hart and Dunn, HFSE. They include high–Mg clinopyroxene (Mg# ≤93), (1993), Adam and Green, (2006), and Tiepolo (1999). On although olivine is absent. Moreover, they are character- the primitive mantle normalized trace element diagram ized by high CaO/Al2O3 ~ 1.04, which are common fea- (Fig. 8), the calculated melt from clinopyroxene in IAB– tures of primitive or parental arc ankaramite (Della–Pas- type gabbros exhibits a marked enrichment of Th, U and qua and Varne, 1997). Hence, these features make them a LREE over HREE [(La/Yb)N average is 10.24] and deple- likely candidate for a near–primary mantle melt for IAB– tion of HFSE (Nb, Ta, and Zr) relative to LREE (e.g., NbN/ type gabbros and basaltic rocks. LaN = 0.14–0.21). These geochemical features are a char- OIB–type gabbros. The calculated melt in equilib- acteristic of subduction zone magmatism (Pearce and rium with clinopyroxene from OIB–type gabbros display Peate, 1995). The general trace element patterns of calcu- no negative Zr and strong Nb–Ta depletion relative to lated melt are similar to basalts in terms of strong depletion LREE (NbN/LaN = 1.51) (Fig. 8); they are unlikely to have of HFSE, but higher contents of overall trace elements for a subduction modified mantle source. LREE are slightly the former. These similarities of trace element patterns be- enriched relative to HREE (LaN/YbN = 7.6). In addition, tween calculated melts for IAB–type gabbros and the bulk the less fractionated pattern of HREE (Fig. 8), may sug- rock composition of IAB–type basalts may suggest that gest that the limited role of garnet in their mantle source. they are genetically related. The relatively higher concen- The calculated melt also display strong negative anomaly tration of the calculated melt suggests that magmas of the of Sr, which may indicates plagioclase fractionation. IAB–type gabbros are slightly more evolved. However, However, considering very low–Mg# and Cr content of experimental data show that the clinopyroxene/liquid par- clinopyroxene, the calculated liquid may not represent tition coefficients for REE, Y and Zr, markedly increase the parental magma. So far, we did not find the volcanic with silica content in the liquid (Sisson, 1991). Thus, it equivalent of our OIB–type gabbro. seems plausible that the relatively higher concentration In Summary, the calculated composition of the liq- 262 M.T. Desta, A. Ishiwatari, S. Machi, S. Arai, A. Tamura, G.V. Ledneva, S.D. Sokolov, A.V. Moiseev and B.A. Bazylev

Figure 8. Primitive mantle normalized multi–element patterns for the calculated liquid in equilibrium with clinopyroxene compared with volcanic rock trace element composition. Normalization factors for the calculated liquid are after Hart and Dunn (1993), Tiepolo (1999), and Adam and Green (2006). The bulk– rock composition of basalts also plotted for comparison. Mantle–normalization values are from McDo- nough and Sun (1995).

uid in equilibrium with clinopyroxene from the OIB–type significant crustal contamination. In addition, previous re- gabbros differs from that of the IAB–type gabbros, indi- searchers thought that minor crustal contamination may cating that they are not genetically related. produce positive Zr–Hf anomalies due enrichment of these elements in crustal material (Sun and McDonough, Constraints on mantle sources 1989). Thus, the depletion of HFSEs and enrichment of LILEs of the IAB–type gabbros and basalts may reflect the Several possible mechanisms have been suggested for the metasomatism of mantle sources by the influx of slab–re- generation of gabbroic magmas, such as: (1) partial melt- leased fluids and /or melts derived from partial melting of ing of asthenospheric mantle, (2) metasomatized litho- subducted slab (Stern, 2002; Zhao and Zhou, 2007). Com- spheric mantle, and (3) some combination of both (McDo- pared with LILE, both REE and HFSE (e.g., Nb, Ta, Zr, nough, 1990 and McKenzie and O’Nions, 1995). Most of Hf and Th) are relatively immobile in aqueous fluids the IAB–type gabbroic and basaltic rocks of Chukotka (Tatsumi, 1989; Turner et al., 1997). Thus, enrichment have lower Nb/Ta values (11.3–17.9) than those of prim- of REE and HFSE in a mantle wedge suggests the intro- itive mantle (mean value, 17.5), and higher La/Nb (2.6– duction of slab melts rather than aqueous fluids (Elliott et 3.8; i.e., >1.5, Fig. 10) and La/Ta (39.3–57.1; i.e., >30) al., 1997; Plank and Langmuir, 1992). Previous investiga- values than those of basalts derived from asthenospheric tions have shown that magmas derived from the sources mantle (Thompson and Morrison, 1988). However, the modified by subducted slab–derived melts display elevat- OIB–type gabbros have significantly low content of La/ ed Th and LREE contents relative to N–MORB as well as Nb (0.68–0.99) and La/Ta (10.3–14.3) ratios similar to high Th/La (>0.2) and Th/Yb (>2) ratios (Woodhead et those of basalts derived from unaltered ansthenospheric al., 2001 and Richards and Kerrich, 2007). Most of the mantle. The above mentioned points indicate that the IAB– IAB–type gabbros and basalts of Chukotka display deple- type gabbros and basalts have similar values observed in tion of Th relative to Ba as well as variable Th/La (0.02– basalts derived from lithospheric mantle or altered asthe- 0.15) and Th/Yb (0.1–0.6), indicating that the addition of nospheric mantle, whereas the values of OIB–type gabbros subducted sediments into the mantle source is insignifi- are similar to asthenospheric derived magma. The low cant. However, the high Th/La values (0.21–0.26) and HREE abundances in IAB–type gabbros and basalts reflect Th concentrations (7.4–9.4 ppm) in the basaltic andesites the presence of residual garnet in their mantle source, may reflect a subducted sediment component in the man- whereas the relatively flat HREE patterns of OIB–type tle source (e.g., Plank, 2005) or small amount (<10 wt%) gabbros indicates that mantle melting probably occurred of contamination of crustal materials (Kimura and Yoshi- within the spinel stability field, or that, if melting occurred da, 2006). OIB–like rocks can be generated in subduc- deeper, garnet have been melted out in the source. tion settings by melting of amphibole– and rutile–bearing The IAB–type gabbros and basalts are depleted in Nb HFSE–enriched sources (Allen et al., 2013); however, and Ta (Figs. 6 and 8), although such negative anomalies there is no evidence for subducted sediment related fluid can be caused by crustal contamination, these rocks have activity in our OIB–type gabbros, as these samples have obvious negative Zr–Hf anomalies in the primitive man- low Th/La ratios (0.10–0.12; Plank, 2005) and no Ce tle–normalized trace element diagram, arguing against the anomalies on the normalized plots (Fig. 8). Siberian LIP 263

source have been metasomatized by ﬂuid derived from the subducted oceanic slab, whereas the subduction related components is insigniﬁcant for the generation of OIB– type gabbros.

Evaluation of tectonic setting

The AlZ/TiO2 ratios (where AlZ indicates the percentage of tetrahedral sites occupied by Al) in clinopyroxene are an effective indicators for discriminating the tectonomag- matic affiliations of gabbros (Loucks, 1990). Clinopyr- oxne from subduction–related magmas yield an AlZ/ TiO2 slope that is twice as steep as that in clinopyroxnes from rift–related magmas (Loucks, 1990). Clinopyroxene of IAB–type gabbros and basalts clearly follow the trend defined by volcanic rocks and cumulates from island arc magmas (Loucks, 1990; Himmelberg and Loney, 1995; Krause et al., 2007), whereas those of OIB–type gabbros follow rift cumulate trend (Figs. 11A and 11B). In the (La/ Nb)N versus (Th/Ta)N and Nb/Yb versus Th/Yb diagrams (Figs. 10A and 10B), the studied IAB–type gabbroic and basaltic rocks have trace element compositions similar to island–arc basalts, particularly similar to the eastern volcanic front lava of Kamchatka arc (Churikova et al., 2001), also suggesting subduction–related environment. Therefore, above features indicate that the Triassic system of Chukotka includes both within plate basalt–type and Figure 9. (A) (Hf/Sm)N versus (Ta/La)N diagram (after LaFlèche et al., 1998) and (B) Th/Yb versus Sr/Nb (after Woodhead et al., island arc basalt–type mafic rocks coexist together. 1998), illustrating that the source region of the IAB–type mafic rocks of Chukotka has been influenced by subduction–related Comparison with the Noril’sk basalts of the Siberian fluids. LIP

HFSE depletion. As is common in many large igneous In the (Hf/Sm)PM versus (Ta/La)PM diagram (Fig. provinces, the Siberian LIP lavas include both high and 9A) (LaFlèche et al., 1998), majority of the IAB–type low–Ti compositional types (Sharma et al., 1991; Fedor- gabbros and basalts (except basaltic andesites) plot in enko and Czamanske, 1997). The most important charac- the ﬂuid–related subduction metasomatism, but the OIB– teristics of the dominant low–Ti lavas are their IAB–type type gabbros display no subduction related metasoma- trace element patterns (Ivanov 2007), whereas those of tism and plot close to the OIB value. In addition, the the high–Ti lavas (such as meimechites) display OIB– metasomatism features of mantle source for the Chukotka type composition (Arndt et al., 1998). The high–Ti ba- gabbros and basalts may be further examined through the salts display chemical characteristics of deeper–derived plot of Sr/Nd versus Th/Yb (Woodhead et al., 1998). As melts controlled by garnet which has been attributed to shown in Figure 9B, the data points of the IAB–type gab- the direct involvement of a mantle plume (Lightfoot et bros and basalts (Except basaltic andesites) lie parallel al., 1993; Wooden et al., 1993). For the origin of low– to the Sr/Nd axis, demonstrating that the mantle source Ti basalts, several models have been proposed, such as was metasomatized by aqueous ﬂuids, whereas OIB–type partial melting of mantle lithosphere (Lightfoot et al., gabbros have low Th/Yb and Sr/Nd ratios that preclude 1990, 1993; Hawkesworth et al., 1995) or melting of dif- the important role of subduction derived components in ferent source regions within the upper mantle (Fedorenko their source. et al., 1996). The source of arc–like LIP magma is inter- In summary, we suggest that the parental magmas of preted to be an enriched mantle that sourced previous arc the IAB–type gabbros and basalts most probably originat- and back–arc magmatism (e.g., Puffer, 2001). Recently, ed in a supra–subduction zone setting and the mantle Ivanov (2007) pointed out that the IAB–like geochemical 264 M.T. Desta, A. Ishiwatari, S. Machi, S. Arai, A. Tamura, G.V. Ledneva, S.D. Sokolov, A.V. Moiseev and B.A. Bazylev

Figure 10. (A) (La/Nb)N versus (Th/Ta)N (after Neal et al., 2002), where N denotes normalized to primitive mantle (McDonough and Sun, 1995). (B) Nb/Yb versus Th/Yb diagram discriminating rocks enriched by input from subduction componenet, crustal contamination, and deep crustal recycling (modified after Pearce, 2008). SZE, subduction zone enrichment; CC, continental contamination; WPE, within–plate enrichment. The field of Siberian LIP is from Wooden et al. (1993) and Krivolutskaya et al. (2009); Bel’kov Island, Kuz’michev and Pease (2007); Central Atlantic magmatic province (CAMP), Callegaro et al. (2013); Ethiopian LIP, Beccaluva et al. (2009) and Desta et al. 2014; Etendeka, Gibson et al. (2000); Karoo LIP, Luttinen et al. (2010); Kamchatka arc, Churikova et al. (2001); Mariana arc, Tamura et al. (2014); East Pacific Rise, Turner et al. (2011); Hawaiian–Emperor Chain, Huang et al. (2005) are shown for comparison. Upper crust (UC) and lower crustal (LC) values are from Rudnick and Gao, (2003). NMORB, EMORB and OIB are from Sun and McDonough (1989). signature of the majority of the low–Ti laves is attributed northernmost Maymecha–Kotuy region). Recently, on to the wet magma source. Moreover, Ivanov et al. (2008) the basis of petrological and geochemical evidence Wang and Ivanov and Litasov (2014) discussed the similarity of et al. (2015) concluded that the generation of arc–like the calculated initial melt compositions of the southeast- LIPs (such as Karoo, Siberian, and Central Atlantic LIPs; ern Siberian flood basalt province (Angara–Taseevskaya see Fig. 10) ultimately related to fluid/melt derived from dolerite sills) with the initial melt compositions of the subducted slabs. modern arc of Eastern Kamchatka (Portnyagin et al., Trace element patterns for the IAB–type gabbros and 2007) in terms of trace elements, suggesting a similar basalts of Chukotka resemble to the low–Ti Nadezdinsky subduction–related processes are responsible for their ori- suite of Noril’sk region in terms of depletion in HFSE gin. Ivanov and Litasov (2014) concluded that the Sibe- (i.e., Nb, Ta, Zr, and P) (Fig. 6), which may suggest sim- rian LIP is formed in a far back–arc region of the Mon- ilar condition of formation. OIB–type gabbros have (La/ golia–Okhotsk subduction system and the influence of Nb)N between 0.7 and 1.0 and (Th/Ta)N between 0.5 and subduction in the trace element budget of flood basalts 0.8 (Fig. 10A) similar to typical plume related LIPs, e.g., reduces away from subduction zone (e.g., presence of Ethiopian LIP (Beccaluva et al., 2009; Desta et al., 2014) rocks with OIB signature such as meimechite from the and Hawaiian–Emperor Chain basalts (Huang et al., Siberian LIP 265

IV Figure 11. (A) and (B) Alz [percentage of tetrahedral sites occupied by Al or (Al × 100/2)] versus TiO2 (wt%) in clinopyroxene (after Loucks, 1990) from gabbroic and basaltic rocks, respectively.

2005), which can be interpreted as insignificant contributions in the magma are suggested by the occurrence – tion of continental material in their source. However, the of Ca rich clinopyroxene (Johannes, 1978). High PH2O OIB–type gabbros of Chukotka display relatively flat could also cause plagioclase to be Ca–rich (Johannes, HREE (Fig. 6) patterns compare with typical OIB, which 1978), which is consistent with the presence of An–rich may suggest shallow mantle melting/thin lithosphere or plagioclase crystals from IAB–type gabbros. In addition, presence of insignificant garnet in their source. Arculus and Wills (1980) described that hydrous basaltic Hydrous nature. The presence of hydrous minerals melts crystallize more Ca–rich (anorthitic) plagioclase is reported from several localities of Siberian LIP: amphi- than anhydrous melts. bole and biotite in Noril’sk I intrusion (Renne 1995), bio- In summary, the Triassic mafic rocks of Chukotka tite in meimechite from Maymecha–Kotuy region (Fedor- may represent the eastern margin of Siberian LIP based enko and Czamanske 1997); biotite in the olivine gabbro on the following points: (1) the presence of both OIB and from west Siberian basin (Reichow et al., 2002); biotite in IAB–like (i.e., depletion of HFSE) mafic rocks through- dolerite sills of Angara–Taseevskaya (Ivanov 2007); am- out Chukotka like that of Siberian LIP, (2) the hydrous phibole and biotite in the dolerite of New Siberian Islands nature of the magma to crystallize hydrous minerals, (3) (Kuz’michev and Pease 2007). However, it is important to platformal field occurrence of the studied rocks, and (4) note that amphibole is rarely a phenocryst in arc basalts the geochemical and petrological similarity between the (Davidson et al., 2007) and their scarcity can be attributed IAB–type gabbros of Chukotka and the Bel’kov Island to its instability at low pressures (e.g., Rutherford and dolerites. The last issue suggests that the eastern Chukot- Devine, 1988; Romick et al., 1992). Ivanov and Litasov ka was closer to the Siberian margin during Permo–Tri- (2014) have shown that the influence of the subduction assic time and later tectonically rifted and moved to the related deep water cycle on the Siberian LIP on the basis current position. of experimental data and numerical calculations. The ther- mochemical model by Wang et al. (2015) shows that slab– CONCLUSION triggered wet–upwelling produces large volumes of melt that may rise from the hydrous mantle transition zone. The Petrology and geochemistry of the Triassic Chukotka above lines of evidence support the hydrous nature of the basalts and gabbros from NE Russia were examined and Siberian LIP. IAB– and OIB–type rocks were identified. Calculated pa- The hydrous nature of the IAB–type gabbros and rental melts inferred from clinopyroxene analyses from basaltic rocks of Chukotka is evident from its high Ca IAB–type gabbros have trace element patterns similar to content of clinopyroxene and crystallization of abundant the basaltic rocks, suggesting that they are genetically re- – magmatic amphibole. High water pressure (PH2O) condi- lated and the basalts are the probable parental melt of 266 M.T. Desta, A. Ishiwatari, S. Machi, S. Arai, A. Tamura, G.V. Ledneva, S.D. Sokolov, A.V. Moiseev and B.A. Bazylev the IAB–type gabbros but not for the OIB–type gabbros. Journal of Petrology, 21, 743–799. Mineral and bulk rock chemical characteristics of the Arndt, N., Chauvel, C., Czamanske, G. and Fedorenko, V. (1998) – Two mantle sources, two plumbing systems: tholeiitic and al- IAB type gabbros and basalts are compatible with mag- kaline magmatism of the Maymecha River basin, Siberian mas from subduction zone geotectonic setting, whereas flood volcanic province. Contributions to Mineralogy and Pet- OIB–type gabbros show within plate magma characteris- rology, 133, 297–313. tics. The IAB–type gabbros and basalts were derived from Barsdell, M. and Berry, R.F. (1990) Origin and evolution of prim- fi itive island arc ankaramites from Western Epi, Vanuatu. Jour- a mantle source highly modi ed by subduction derived – fl nal of Petrology, 31, 747 777. uids. Basaltic andesite samples have derived from the Beccaluva, L., Bianchini, G., Natali, C. and Siena, F. (2009) Con- source mantle metasomatized by the fluids originated in tinental flood basalts and mantle plumes: a case study of the the subducted sediments. The geochemical similarity of northern Ethiopian Plateau. Journal of Petrology, 50, 1377– the studied IAB–type gabbroic and basaltic rocks of Chu- 1403. Bryan, S.E. and Ernst, R.E. (2008) Revised definition of Large kotka with Bel’kov dolerite and low–Ti Nadezhdinsky suit Igneous Provinces (LIPs). Earth–Science Reviews, 86, 175– ’ (Noril sk region) in terms of the presence of dominant hy- 202. drous minerals and negative HFSE anomaly, it is possible Callegaro, S., Marzoli, A., Bertrand, H., Chiaradia, M., Reisberg, that the eastern Chukotka was placed closer to the Siberian L., Meyzen, C., Bellieni, G., Weems, R.E. and Merle, R. LIP in Triassic, and the Triassic mafic magmatism in Chu- (2013) Upper and lower crust recycling in the source of CAMP basaltic dykes from southeastern North America. kotka may represents a marginal part of the Siberian LIP. Earth and Planetary Science Letters, 376, 186–199. Churikova, T., Dorendorf, F. and Worner, G. (2001) Sources and ACKNOWLEDGMENTS fluids in the mantle wedge below Kamchatka, evidence from across–arc geochemical variation. Journal of Petrology, 42, – This paper is based on the first author’s PhD course stud- 1567 1593. Coffin, M.F. and Eldholm, O. (2005) Large igneous provinces. In ies. We are greatly indebted to Tsuyoshi Miyamoto for the Encyclopedia of Geology (Selley, R.C., Cocks, L.R.M. and X–ray fluorescence analyses of bulk–rock composition Plimer, I.R. Eds.). Elsevier, Oxford, 315–323 and to staffs of Kanazawa University for their comments Davidson, J., Turner, S., Handley, H., Macpherson, C. and Dosse- and assistance with LA–ICP–MS trace element determina- to, A. (2007) Amphibole “sponge” in arc crust? Geology, 35, – tions. We would like to express our gratitude to Jun–Ichi 787 790. Degtyaryov, V.S. (1975) Petrochemical peculiarities of the Amgue- Kimura and an anonymous reviewer, whose thoughtful ma–Anyui diabase formation of the Chukotka fold region. In comments and valuable suggestions allowed us to im- Magmatism of North East Asia, part 2 (Shatalov, E.T. Ed.). prove the manuscript significantly. We also thank Prof. Magadan book publisher, Magadan, 160–175 (in Russian). Yildirim Dilek of Miami University (USA) for his com- Della–Pasqua, F. and Varne, R. (1997) Primitive ankaramitic magmas in volcanic arcs: a melt–inclusion approach. Canadian ments on our draft. The second author acknowledges the – – – fi – Mineralogist, 35,291 312. Grant in Aid for Scienti c research (C) 23540554 pro- Desta, M.T., Ayalew, D., Ishiwatari, A., Arai, S. and Tamura, A. vided by the Ministry of Education, Culture, Sports, Sci- (2014) Ferropicrite from the Lalibela area in the Ethiopian ence and Technology (MEXT), Japan. large igneous province. Journal of Mineralogical and Petro- logical Sciences, 109, 191–207. Elliott, T., Plank, T., Zindler, A., White, W. and Bourdon, B. REFERENCES (1997) Element transport from slab to volcanic front at the Mariana arc. Journal of Geophysical Research, 102, 14991– Adam, J. and Green, T. (2006) Trace element partitioning between 15019. mica– and amphibole–bearing garnet lherzolite and hydrous Fedorenko, V.A. and Czamanske, G.K. (1997) Results of new field basanitic melt: 1. Experimental results and the investigation and geochemical studies of the volcanic and intrusive rocks of of controls on partitioning behaviour. Contributions to Miner- the Maymecha–Kotuy area, Siberian flood–basalt province. alogy and Petrology, 152, 1–17. International Geological Review, 39, 479–531. Allen, M.B., Kheirkhah, M., Neill, I., Emami, M.H. and McLeod, Fedorenko, V.I., Lightfoot, P.C., Naldrett, A.J., Czamanske, G.K., C.L. (2013) Generation of arc and within–plate chemical sig- Hawkesworth, C.J., Wooden, J.L. and Ebel, D.S. (1996) Pet- natures in collision zone magmatism: Quaternary lavas from rogenesis of the flood–basalt sequence at Noril’sk, North Cen- Kurdistan Province, Iran. Journal of Petrology, 54, 887–911. tral Siberia. International Geology Review, 38, 99–135. Amato, J.M., Toro, J., Miller, E.L., Gehrels, G.E., Farmer, G.L., Figueroa, O., Deruelle, B. and Demaiffe, D. (2009) Genesis of Gottlieb, E.S. and Till, A.B. (2009) Late Proterozoic–Paleo- adakite–like lavas of Licancabur volcano (Chile–Bolivia, Cen- zoic evolution of the Arctic Alaska–Chukotka terrane based tral Andes). Comptes Rendus Geosci, 341, 310–318. on U–Pb igneous and detrital zircon ages: Implications for Gel’man, M.L. (1963) Triassic diabase association in the Anyui Neoproterozoic paleogeographic reconstructions. GSA Bulle- fold zone (Chukotka). Geologiya i Geofizika, 2, 127–134 (in tin, 121, 1219–1235. Russian). Arculus, R.J. and Wills, K.J.A. (1980) The petrology of plutonic Gibson, S.A., Thompson, R.N. and Dickin, A.P. (2000) Ferropi- blocks and inclusions from the Lesser Antilles island arc. crites: Geochemical evidence for Fe–rich streaks in upwelling Siberian LIP 267

mantle plumes. Earth and Planetary Science Letters, 174, 355– mafic–ultramafic complexes in the Central Ural Mountains. 374. Lithos, 95, 19–42. Green, D.H. (1976) Experimental testing of “equilibrium” partial Krivolutskaya, N.A. and Rudakova, A.V. (2009) Structure and melting of peridotite under water–saturated, high–pressure Geochemical characteristics of trap rocks from the Noril’sk conditions. Canadian Mineralogist, 14, 255–268. trough, northwestern Siberian craton. Geochemistry Interna- Green, T.H. and Pearson, N.J. (1985) Experimental determination tional, 47, 635–656. of REE partition coefficients between amphibole and basaltic Kuz’michev, A.B. and Pease, V.L. (2007) Siberian trap magmatism to andesitic liquids at high pressure. Geochimica et Cosmo- on the New Siberian Islands: constraints for Arctic Mesozoic chimica Acta, 49, 1465–1468. plate tectonic reconstructions. Journal of Geological Society, Hanson, G.N. and Langmuir, C.H. (1978) Modeling of major ele- 164, 959–968. ments in mantle–melts systems using trace element ap- LaFlèche, M.R., Camire, G. and Jenner, G.A. (1998) Geochemistry proaches. Geochimica et Cosmochimica Acta, 42, 725–742. of post–Acadian, Carboniferous continental intraplate basalts Hart, S.R. and Dunn, T. (1993) Experimental cpx/melt partitioning from the Maritimes Basin, Magdalen islands, Quebec, Cana- of 24 trace elements. Contributions to Mineralogy and Petrol- da. Chemical Geology, 148, 115–136. ogy, 113, 1–8. Lawver, L.A. and Scotese, C.R. (1990) A review of tectonic mod- Hawkesworth, C.J., Lightfoot, P.C., Fedorenko, V.A., Blake, S., els for the evolution of the Canada Basin. In The Geology of Naldrett, A.J., Doherty, W. and Gorbachev, N.S. (1995) Mag- North America, vol. L (Grantz, A., Johnson, L. and Sweeney, ma differentiation and mineralisation in the Siberian continen- J.F. Eds.). Geological Society of America, Boulder, Colorado, tal flood basalts. Lithos, 34, 61–88. The Arctic Ocean Region, 593–618. Himmelberg, R.G. and Loney, R.A. (1995) Characteristics and Lawver, L.A., Grantz, A. and Gahagan, L.M. (2002) Plate kine- perogenesis of Alaskan–type ultramafic–ultramafic intrusions, matic evolution of the present Arctic region since the Ordo- southeastern Alaska. US Geological Survey Professional Pa- vician. In Tectonic Evolution of the Bering Shelf–Chukchi per, 1564, 1–47. Sea–Arctic Margin and Adjacent Landmasses (Miller, E.L., Huang, S., Regelous, M., Thordarson, T. and Frey, F.A. (2005) Grantz, A. and Klemperer, S.L. Eds.). Geological Society of Petrogenesis of lavas from Detroit Seamount: Geochemical America, Special Papers, 360, 333–358. differences between Emperor Chain and Hawaiian volcanoes. Leake, B.E., Wooley, A.R., Arps, C.E.S., Birch, W.D., Gilbert, Geochemistry, Geophysics and Geosystems, 6, Q01L06, doi: M.C., Grice, J.D., Hawthorne, F.C., Kato, A., Kisch, H.J., 10.1029/2004GC000756. Krivovichev, V.G., Linthout, K., Laird, J., Mandarino, J.A´., Ishida, H., Morishita, T., Arai, S. and Shirasaka, M. (2004) Simul- Maresch, W.V., Nickel, E.H., Rock, N.M.S., Schumacher, taneous in–situ multi–element analysis of minerals on thin sec- J.C., Smith, D.C., Stephenson, N.C.N., Ungaretti, L., Whit- tion using LA–ICP–MS. Science Report Kanazawa Universi- taker, E.J.W. and Youzhi, G. (1997) Nomenclature of amphib- ty, 48, 31–42. oles: report of the Subcommittee on Amphiboles of the Inter- Ivanov, A.I. and Litasov, K.D. (2014) The deep water cycle and national Mineralogical Association, Commission on New flood basalt volcanism. International Geology Review, 56, 1– Minerals and Mineral Names. American Mineralogist, 82, 14. 1019–1037. Ivanov, A.V. (2007) Evaluation of different models for the origin Ledneva, G.V., Sokolov, S.D. and Pease, V.L. (2011) Permo–Tri- of the Siberian traps. In The Origin of Melting Anomalies: assic hypabyssal mafic intrusions and associated tholeiitic ba- Plates, Plumes and Planetary Processes (Foulger, G.R. and salts of the Kolyuchinskaya Bay, Chukotka (NE Russia): Jurdy, D.M. Eds.). Geological Society of America Special Pa- Links to the Siberian LIP. Journal of Asian Earth Sciences, per, 430, 669–691. 40, 737–745. Ivanov, A.V., Demonterova, E.I., Rasskazov, S.V. and Yasnygina, Ledneva, G.V., Bazylev, B.A., Layer, P.W., Ishiwatari, A., Sokolov, T.A. (2008) Low–Ti melts from the Southeastern Siberian S.D., Kononkova, N.N., Tikhomirov, P.L. and Novikova, Traps large igneous province: Evidence for a water–rich man- M.S. (2014) Intra–plate gabbroic rocks of Permo–Triassic to tle source? Journal of Earth System Science, 117, 1–21. Early–Middle Triassic dike–and sill province of Chukotka Ivanov, O.N. and Milov, A.P. (1975) The diabase association in the (Russia). International Conference on Arctic Margins, VI, Chukchi fold system and its relation to the basic magmatism 115–156. in the northern portion of the Pacific Mobile Belt. In Magma- Lee, C.–T.A., Luffi, P., Plank, T., Dalton, H. and Leeman, W.P. tism of North East Asia, part 2 (Shatalov, E.T. Eds.). Magadan (2009) Constraints on the depths and temperatures of basaltic Publishing House, Magadan, Russia, 155–159 (in Russian). magma generation on Earth and other terrestrial planets using Johannes, W. (1978) Melting of plagioclase in the system Ab–An– new thermobarometers for mafic magmas. Earth and Planetary H2O– and Qz–Ab–An–H2OatPH2O = 5 kbars an equilibrium Science Letters, 279, 20–33. problem. Contributions to Mineralogy and Petrology, 66, 295– Lightfoot, P.C., Naldrett, A.J., Gorbachev, N.S., Doherty, W. and 303. Fedorenko, V.A. (1990) Geochemistry of the Siberian Trap of Jolly, W.T., Lidiak, E.G., Dickin, A.P. and Way, T.–W. (2002) Re- the Noril’sk area, USSR, with implications for the relative cycling in the Puerto Rican mantle wedge, Greater Antilles contrubutions of crust and mantle to flood basalt magmatism. island arc. The Island Arc, 11:10–24. Contributions to Mineralogy and Petrology, 104, 631–644. Kimura, J.–I. and Yoshida, T. (2006) Contributions of slab fluid, Lightfoot, P.C., Hawkesworth, C.J., Hergt, J., Naldrett, A.J., Gor- wedge mantle, and crust to the origin of Quaternary lavas in bachev, N.S., Fedorenko, V.A. and Doherty, W. (1993) Remo- the NE Japan arc. Journal of Petrology, 47, 2185–2232, doi: bilisation of the continental lithosphere by a mantle plume: 10.1093/petrology/egl041. major–, trace–element, and Sr–,Nd–, and Pb–isotope evidence Krause, J., Brugmann, G.E. and Pushkarev, E.V. (2007) Accessory from picritic and tholeiitic lavas of the Noril’sk district, Sibe- and rock forming minerals monitoring the evolution of zoned rian Trap, Russia. Contributions to Mineralogy and Petrology, 268 M.T. Desta, A. Ishiwatari, S. Machi, S. Arai, A. Tamura, G.V. Ledneva, S.D. Sokolov, A.V. Moiseev and B.A. Bazylev

114, 171–188. naya, S. (2007) Constraints on mantle melting and composi- Loucks, R.R. (1990) Discrimination of ophiolitic from nonophio- tion and nature of slab components in volcanic arcs from vol- litic ultramafic–mafic allochthons in orogenic belts by the Al/ atiles (H2O, S, Cl, F) and trace elements in melt inclusions Ti ratio in clinopyroxene. Geology, 18, 346–349. from the Kamchatka Arc. Earth and Planetary Science Letters, Luttinen, A.V., Leat, P.T. and Furnes, H. (2010). Björnnutane and 255, 53–69. Sembberget basalt lavas and the geochemical provinciality of Puffer, J.H. (2001) Contrasting high filed strength element content Karoo magmatism in the western Dronning Maud Land, Ant- of continental flood basalts from plume versus reactivated–arc arctica. Journal of Volcanology and Geothermal Research, sources. Geology, 29, 675–678. 198, 1–18. P’yankov, A.Y. (1981) Gosudarstvennaya geologicheskaya karta McDonough, W.F. (1990) Constraints on the composition of the SSSR (National Geological map of USSR), 1:200,000 scale, continental lithospheric mantle. Earth and planetary Science Anyui–Chaun series, Sheet R–60 – XXI, XXII, MinGeo Rus- Letters, 1, 1–18. sia, Magadan. McDonough, W.F. and Sun, S.–s. (1995) The composition of the Reichow, M.K., Saunders, A.D., White, R.V., Pringle, M.S., Earth. Chemical Geology, 120, 223–253. Al’mukhamedov, A.I., Medvedev, A.I. and Kirda, N.P. McKenzie, D. and O’Nions, R.K. (1995) The source regions of (2002) 40Ar/39Ar dates from the West Siberian Basin: Siberian oceanic island basalts. Journal of Petrology, 36, 133–159. flood basalt province doubled. Science, 296, 1846–1849. Miller, E.L., Toro, J., Gehrels, G., Amato, J.M., Prokopiev, A., Reichow, M.K., Saunders, A.D., White, R.V., Al’Mukhamedov, Tuchkova, M.I., Akinin, V.V., Dumitru, T.A., Moore, T.E. A.I. and Medvedev, A.Y. (2005) Geochemistry and petrogen- and Cecile, M.P. (2006) New insights into Arctic paleogeog- esis of basalts from the Western Siberian Basin: an extension raphy and tectonics from U–Pb detrital zircon geochronology. of Permo–Triassic Siberian traps, Russia. Lithos, 79, 425–452. Tectonics, 25, 1–19. Renne, P.R. (1995) Excess 40Ar in biotite and hornblende from the Miyashiro, A. (1974) Volcanic rock series in island–arcs and active Norilsk 1 intrusion, Siberia: implication for the age of Sibe- continental margins. American Journal of Science, 274, 321– rian Traps. Earth and Planetary Science Letters, 131, 165–176. 355. Richards, J.P. and Kerrich, R. (2007) Special paper: adakite–like Morimoto, N., Fabries, J., Ferguson, A.K., Ginzburg, I.V., Ross, rocks: their diverse origins and questionable role in metallo- M., Seifert, F.A., Zussman, J., Aoki, K. and Gottardi, D. genesis. Economic Geology, 102, 537–576. (1988) Nomenclature of pyroxenes. American Mineralogist, Ridolfi, F., Renzulli, A. and Puerini, M. (2010) Stability and chem- 62, 53–62. ical equilibrium of amphibole in calcalkaline magmas: an Natal’in, B.A., Amato, J.M., Toro, J. and Wright, J.E. (1999) Pa- overview, new thermobarometric formulations and application leozoic rocks of northern Chukotka Peninsula, Russian Far to subduction–related volcanoes. Contributions to Mineralogy East: implication for the tectonics of the Arctic region. Tec- and Petrology, 160, 45–66. tonics, 18, 977–1003. Ridolfi, F. and Renzulli, A. (2012) Calcic amphiboles in calc–alka- Neal, C.R., Mahoney, J.J. and Chazey III, W.J. (2002) Petrogenesis line and alkaline magmas: thermobarometric and chemometric of Kerguelen Plateau and Broken Ridge LIP: Results from empirical equations valid up to 1,130 °C and 2.2 GPa. Con- ODP Leg 183. Journal of Petrology, 43, 1177–1205. tributions to Mineralogy and Petrology, 163, 877–895. Nokleberg, W.J., Parfenov, L.M., Monger, J.W.H., Norton, I.O., Romick, J.D., Kay, S.M. and Kay, R.W. (1992) The influence of Khanchuk, A.I., Stone, D.B., Scholl, D.W. and Fujita, K. amphibole fractionation on the evolution of calc–alkaline an- (1998) Phanerozoic Tectonic Evolution the Circum–North Pa- desite and dacite tephra from the Central Aleutians, Alaska. cific. U.S. Geological Survey, Open–File Report, 98–754. Contributions to Mineralogy and Petrology, 112, 101–118 Parfenov, L.M. (1984) Continental Margins and Island Arcs in the Rudnick, R.L. (1995) Making continental crust. Nature, 378, 571– Mesozoides of Northeastern Asia, Novosibirsk, Nauka (in 578. Russian). Rudnick, R.L. and Gao, S. (2003) Composition of the continental Parfenov, L.M., Natapov, L.M., Sokolov, S.D. and Tsukanov, N.V. crust. In The Crust (Rudnick, R.L. Ed.). Amsterdam, Elsevier, (1993) Terranes analysis and accretion in northeast Asia. The 1–70. Islands Arc, 2, 35–54. Rutherford, M.J. and Devine, J. (1988) The May 18, 1980, erup- Pearce, J.A. (1996) A user’s guide to basalt discrimination dia- tion of Mount St. Helens 3. Stability and chemistry of amphi- grams. In Trace Element Geochemistry of Volcanic Rocks: bole in the magma chamber. Journal of Geophysical Research, Applications for Massive Sulphide Exploration. Short Course 93, 11949–11959. Notes, vol. 12. Geological Association Canadian, 79–113. Seslavinsky, K.B. (1979) The South Anyui Suture (Western Chu- Pearce, J.A. (2008) Geochemical fingerprinting of oceanic basalts kotka). Doklady Akademii Nauk SSSR, 249, 1181–1185 (in with applications to ophiolite classification and the search for Russian). Archaean oceanic crust. Lithos, 100, 14–48. Sharma, M., Basu, A.R. and Nesterenko, G.V. (1991) Temporal Sr, Pearce, J.A. and Peate, D.W. (1995) Tectonic implications of the Nd–, and Pb–isotopic variations in the Siberian flood basalts: composition of volcanic arc lavas. Annual Review of Earth Implications for the plume–source characteristics. Earth and and Planetary Sciences, 23, 251–285. Planetary Science Letters, 113, 365–381. Plank, T. (2005) Constraints from thorium/lanthanum on sediment Sisson, T.W. (1991) Pyroxene–high silica rhyolite trace element recycling at subduction zones and the evolution of the conti- partition coefficients measured by ion microbrobe. Geo- nents. Journal of Petrology, 46, 921–944. chimica et Cosmochimica Acta, 55, 1575–1585. Plank, T. and Langmuir, C.H. (1992) Sediments melt and basaltic Smith, D.J. (2014) Clinopyroxene precursors to amphibole crust dehydrates at subduction zones. EOS, Transactions, sponge in arc crust. Nature Communications, doi:10.1038/ American Geophysical Union, 73, 637. ncomms5329. Portnyagin, M., Hoernle, K., Plechov, P., Mironov, N. and Khubu- Sobolev, A.V., Sobolev, S.V., Kuzmin, D.V., Malitch, K.N. and Siberian LIP 269

Petrunin, A.G. (2009) Siberian meimechites: origin and rela- evolution of the Chukotka geosynclines in the Early Triassic. tion to flood basalts and kimberlites. Russian Geology and Doklady Akademii Nauk SSSR, 130, 834–837 (in Russian). Geophysics, 50, 999–1033. Tribuzio, R., Tiepolo, M., Vannucci, R. and Bottazzi, P. (1999) Sokolov, S.D., Bondarenko, G.Ye., Morozov, O.L., Shekhovtsov, Trace element distribution within olivine–bearing gabbros V.A., Glotov, S.P., Ganelin, A.V. and Kravchenko–Berezhnoy, from the Northern Apennine ophiolites (Italy): evidence for I.R. (2002) The South Anyui Suture, NE Arctic Russia: facts post–cumulus crystallization in MOR–type gabbroic rocks. and problems to solve. In, Tectonic Evolution of the Bering Contributions to Mineralogy and Petrology, 134, 123–133. Shelf–Chukchi Sea–Arctic Margin and Adjacent Landmasses Tuchkova, M.I., Sokolov, S. and Kravchenko–Berezhnoy, I.R. (Miller, E.L., Grantz, A. and Klemperer, S.L. Eds.). Geolog- (2009) Provenance analysis and tectonic setting of the Triassic ical Society of America, Special Paper, 360, 209–224. clastic deposits in Western Chukotka, Northeast Russia. Ste- Sokolov, S.D., Ledneva, G.V. and Pease, V.L. (2009) New data on phan Mueller Special Publication Series, 4, 177–200. the age and genesis of igneous rocks in the Kolyuchinskaya Turner, S., Hawkesworth, C., Rogers, N., Bartlett, J., Worthington, Guba (Eastern Chukotka). Transactions (Doklady) of the Rus- T., Hergt, J., Pearce, J. and Smith, I. (1997) 238U–230Th dis- sian Academy of Sciences/Earth Science Section, 425A, 384– equilibrium, magma petrogenesis, and flux rates beneath the 388. depleted Tonga–Kermadec island arc. Geochimica et Cosmo- Stern, R.J. (2002) Subduction zones. Reviews of Geophysics, 40, chimica Acta, 61, 4855–4884. 3-1–3-38. Turner, S., Beier, C., Niu, Y. and Cook, C. (2011) U–Th–Ra dis- Sun, S.–s. and McDonough, W.F. (1989) Chemical and isotopic equilibria and the extent of off–axis volcanism across the East systematics of oceanic basalts: implications for mantle com- Pacific Rise at 9°30′N, 10°30′N, and 11°20′N. Geochemistry position and processes. In Magmatism in the Ocean Basins Geophysics Geosystems, 12, Q0AC12, doi:10.1029/ (Saunders, A.D. and Norry, M.J. Eds.). Geological Society 2010GC003403. Special Publication, 42, 313–345. Vatrushkina, E.V. and Tuchkova, M.I. (2014) Peculiarities of lith- Tamura, Y., Ishizuka, O., Stern, R.J., Nichols, A.R.L., Kawabata, ology and geochemistry of the Raucha Formation (upper Ju- H., Hirahara, Y., Chang, Q., Miyazaki, T., Kimura, J.–I., Emb- rassic, western Chukotka). Bulletin MOIP. Geology, Branch, ley, R.W. and Tatsumi, Y. (2014) Mission immiscible: distinct 89, 59–74 (in Russian). subduction components generate two primary magmas at Pa- Wang, X.–C., Wilde, S.A., Li, Q.–L. and Yang, Y.–N. (2015) gan volcano, Mariana arc. Journal of Petrology, 55, 63–101. Continental flood basalts derived from the hydrous mantle Tatsumi, Y. (1989) Migration of fluid phases and genesis of basalt transition zone. Nature Communications, doi:10.1038/ magmas in subduction zones. Journal of Geophysical Re- ncomms8700. search, 94, 4697–4707. Winchester, J.A. and Floyd, P.A. (1977) Geochemical discrimina- Taylor, S.R. and McLennan, S.M. (1995) The geochemical evolution of different magma series and their differentiation prod- tion of the continental crust. Review of Geophysics, 33, 241– ucts using immobile elements. Chemical Geology, 20, 325– 265. 343. Thirlwall, M.F., Smith, T.E., Graham, A.M., Theodorou, N., Hol- Wooden, J.L., Czamanske, G.K., Bouse, R.M., King, B.–S.W., lings, P., Davidson, J.P. and Arculus, R.J. (1994) High field Knight, R.J. and Siems, D.F. (1993) Isotopic and trace–ele- strength anomalies in arc lavas: source or process? Journal of ment constraints on mantle and crustal contributions to Sibe- Petrology, 35, 819–838. rian continental flood basalts, Noril’sk area, Siberia. Geo- Thompson, R.N. and Morrison, M.A. (1988) Asthenospheric and chimica et Cosmochimica Acta, 57, 3677–3704. lower–lithospheric mantle contributions to continental exten- Woodhead, J.D., Eggins, S.M. and Johnson, R.W. (1998) Magma sional magmatism: an example from the British Tertiary Prov- genesis in the New Britain Island Arc: further insights into ince. Chemical Geology, 68, 1–15. melting and mass transfer processes. Journal of Petrology, 39, Tiepolo, M. (1999) Determinazione sperimentale dei coefficient di 1641–1668. distribuzione solido/liquido in anfiboli di mantello: ruolo del Woodhead, J.D., Hergt, J.M., Davidson, J.P. and Eggins, S.M. controllo cristallochimico. Ph.D. Thesis, Università di Pavia. (2001) Hafnium isotope evidence for ‘conservative’ element Tiepolo, M., Oberti, R., Zanetti, A., Vannucci, R. and Foley, S. mobility during subduction zone processes. Earth and Plane- (2007) Tracee element partitioning between Amphibole and tary Science Letters, 192, 331–346. Silicate Melt. Reviews in Mineralogy and Geochemistry, 67, Zhao, J.H. and Zhou, M.F. (2007) Geochemistry of Neoproterozoic 417–452. mafic intrusions in the Panzhihua district (Sichuan Province, Tikhomirov, P.L., Kalinina, E.A., Moriguti, T., Makishima, A., Ko- SW China): Implications for subduction related metasoma- bayashi, K., Cherepanova, I.Yu. and Nakamura, E. (2012) The tism in the upper mantle. Precambrian Research, 152, 27–47. Cretaceous Okhotsk–Chukotka Volcanic Belt (NE Russia): Geology, geochronology, magma output rates, and implica- Manuscript received May 4, 2015 tions on the genesis of silicic LIPs. Journal of Volcanology Manuscript accepted October 4, 2015 and Geothermal Research, 221, 14–32. Til’man, S.M. and Sosunov, G.M. (1960) Some peculiarities of Manuscript handled by Jun–Ichi Kimura 270 M.T. Desta, A. Ishiwatari, S. Machi, S. Arai, A. Tamura, G.V. Ledneva, S.D. Sokolov, A.V. Moiseev and B.A. Bazylev

SUPPLEMENTARY TABLES: ANALYTICAL RESULTS OF MINERAL COMPOSITIONS

Table S1. Average EPMA analyses of clinopyroxenes (wt%) from basaltic rocks of Chukotka, NE Russia

* Total Fe expressed as FeO. Mg# = 100 × Mg (Mg + Fe2+). Abbreviations: Wo, wollastonite; En, enstatite; Fs, ferrosilite; n, number of analyses used for average. Siberian LIP 271

Table S2. Trace element analyses (ppm) of clinopyroxenes from basaltic rocks of Chukotka, NE Russia

Abbreviations: Cpx, clinopyroxene; n.d., not detected. 272 M.T. Desta, A. Ishiwatari, S. Machi, S. Arai, A. Tamura, G.V. Ledneva, S.D. Sokolov, A.V. Moiseev and B.A. Bazylev

Table S3. Average EPMA analyses clinopyroxenes (wt%) from gabbroic rocks of Chukotka, NE Russia

* Total Fe as FeO. Mg# = 100 × Mg (Mg + Fe2+). Abbreviations: Wo, wollastonite; En, enstatite; Fs, ferrosilite; n, number of analyses used for average. Siberian LIP 273

Table S4. Trace element analyses (ppm) of clinopyroxenes from gabbroic rocks of Chukotka, NE Russia

Abbreviations: Cpx, clinopyroxene; n.d., not detected. 274 M.T. Desta, A. Ishiwatari, S. Machi, S. Arai, A. Tamura, G.V. Ledneva, S.D. Sokolov, A.V. Moiseev and B.A. Bazylev

Table S5. Average EPMA analyses of amphiboles (wt%) from gabbroic and basaltic rocks of Chukotka, NE Russia

* Total Fe as FeO. Mg# = 100 × Mg (Mg + Fe2+). Abbreviations: n, number of analyses used for average. Siberian LIP 275

Table S6. Trace element compositions of amphiboles from IAB–type gabbros and basaltic rocks of Chukotka, NE Russia