Yaron Katzir

Post-collisional tectonomagmatic evolution in the northern Arabian–Nubian Shield: time constraints from ion-probe U–Pb dating of zircon

Abstract: Ion-probe U–Pb dating of plutonic rocks from the northern Arabian–Nubian Shield in Sinai and southern Israel constrains the timing of late East African batholithic post-collisional calc-alkaline (CA2) magmatism and within-plate alkaline to peralkaline (AL) magmatism to c. 635–590 Ma and c. 608–580 Ma, respectively. The earliest dated CA2 rocks are slightly deformed to undeformed, indicating that penetrative deformation ceased by c. 630 Ma. Within the CA2 suite a change from mafic to felsic magmatism is manifested in most of the region, peaking in a voluminous pulse of granodiorite to granite intrusion at 610–600 Ma. The AL magmatism started contemporaneously with the peak in CA2 felsic activity at c. 608 Ma and lasted until 580 Ma. It includes mostly alkaline and peralkaline granites, probably representing variable degrees of differentiation of similar parental magmas. Thus CA2 and AL granites do not represent different tectonic settings, but coeval derivation from variable sources during crustal extension. The majority of rocks dated in this study show minor to non-existent zircon inheritance and thus indicate very minor interaction with previously formed felsic crust. The rare zircon xenocrysts span a typical East African age range (900–607 Ma) and confirm the absence of older crustal components in the juvenile Arabian–Nubian Shield. Supplementary material: Geochemical and geochronological methods, sample descriptions and data are available at http://www.geolsoc.org.uk/SUP18327.

The geodynamic evolution of the Alpine orogen in the Cyclades (Aegean Sea, Greece): insights from diverse origins and modes of emplacement of ultramafic rocks

Abstract The Alpine orogen in the Cyclades, wherein both high-pressure metamorphic rocks and ultramafic rocks co-occur, is a key area in studying the emplacement of mantle rocks into the crust. Within the Cyclades three distinct ultramafic associations occur: (1) HP–LT ophiolitic mélanges of the Cycladic Blueschist Unit (CBU) on Evia and Syros; (2) meta-peridotites associated with migmatized leucogneisses on Naxos, which represent the deepest exposed levels of the CBU; (3) a greenschist-facies metamorphosed dismembered ophiolite juxtaposed on top of the CBU by an extensional detachment on Tinos. Most of the Cycladic ultramafic rocks were serpentinized prior to Alpine metamorphism, suggesting denudation prior to reburial. The Naxos meta-peridotites preserve, however, relict mantle assemblage and mantle-like oxygen isotope ratios, and thus indicate direct emplacement from the mantle into an underthrust continent during collision and HP metamorphism (M1). Thus conditions for M1 in the Naxos leucogneiss core are constrained by ultramafic assemblages to 550–650 °C and ≥14 kbar. Mafic blocks of the ophiolitic mélanges in the NW Cyclades span a wide range of chemical compositions indicating derivation from variable oceanic settings and sequential events of alteration and metasomatism. Given the comparable geochemical heterogeneity in the Syros and Evian mélange intervals, the garnet-bearing meta-basites of the Syros mélange record higher M1 temperatures (450–500 °C) than the garnet-free epidote blueschists of the Evian mélanges (400–430 °C). It follows that going southeastwards from Evia progressively deeper (i.e. hotter) levels of the subducted plate are exposed. Correspondingly, temperatures of the M2 overprint also increase from pumpellyite-bearing assemblages on southern Evia, through greenschists on Syros to upper-amphibolite, sillimanite-bearing gneisses on Naxos. The diverse P–T paths of the CBU form an array wherein the deeper a rock sequence is buried, the ‘hotter’ is its exhumation path. Such a pattern is predicted by thermal modelling of tectonically thickened crust unroofed by either erosion or uniform extension.

Oceanic and orogenic fluid–rock interaction in 18O/16O-enriched metagabbros of an ophiolite (Tinos, Cyclades)

Abstract Two spatial scales of fluid–rock interaction in an ophiolite suite are revealed by oxygen isotope and hydrogen isotope studies of metagabbros on the island of Tinos (Cyclades, Greece). Sequentially formed mineral generations in the metagabbros include relict igneous augite, hornblende of sub-seafloor hydrothermal origin, and actinolite and albite formed by regional greenschist-facies metamorphism during orogenesis. With the exception of augite (δ 18 O=4.4–5.6‰), the metagabbros are characterized by unusually high δ 18 O values: hornblende (5.8–7.4‰), actinolite (6.5–10.2‰), feldspar (14.6–14.9‰) and whole rocks (7.0–10.5‰). Hornblende δ D values range from −57 to −66‰. The high δ 18 O values and the δ D range of the hornblendes are compatible with interaction of oceanic gabbro with seawater that had previously been enriched in 18 O/ 16 O (δ 18 O=6.5–8‰) by isotopic exchange at moderate to high temperatures. The high degree of oceanic alteration in the layered gabbros, mass balance calculations of isotopic exchange, and field evidence for early oceanic thrusting suggest that seawater could have penetrated deeply into the ocean crust, becoming 18 O/ 16 O-enriched through isotopic exchange with gabbros at progressively increasing temperature. Upward, down-temperature flow of the high-δ 18 O water would be very effective in elevating the δ 18 O values of gabbros. The regional greenschist metamorphic overprint of the ophiolite, possibly the result of continued thrusting and piling up of nappes during obduction, is characterized by localized fluid–rock exchange. Actinolite in massive gabbroic layers has δ 18 O values (6.5–7.2‰) close to those of the hornblende, whereas in deformed meter-sized gabbroic blocks the amphiboles have significantly higher values (8.4–10.2‰). Likewise, albite in the gabbroic blocks has high δ 18 O values of ca. 15‰ that are ascribed to meter-scale exchange with 18 O-rich fluids derived from dehydration reactions in low-temperature hydrothermally altered basaltic host rock enclosing the blocks. Deformation-enhanced permeability facilitated fluid infiltration in gabbroic blocks, whereas the relatively undeformed, and therefore less permeable, massive gabbros experienced minor interaction with fluids. The orogenic fluid–rock interaction thus represents local-scale redistribution of hydrous mineral components introduced during seafloor hydrothermal exchange.

The origin, cooling and alteration of A-type granites in southern Israel (northernmost Arabian-Nubian shield): a multi-mineral oxygen isotope study

A multi-mineral oxygen isotope study sheds light on the origin, cooling and alteration of Late Neoproterozoic A-type granites in the Arabian–Nubian shield of southern Israel. The oxygen isotope ratio of zircon of the Timna monzodiorite, quartz syenite and alkaline granite are within the range of mantle zircon (δ 18 O(Zrn) = 5.3 ± 0.6‰, 2σ), supporting the co-genetic mantle-derived origin previously suggested based on geochemical data and similar ɛNd(T) values and U–Pb ages (610 Ma). Likewise, olivine norite xenoliths within the monzodiorite (δ 18 O(Ol) = 5.41 ± 0.07‰) may have formed as cumulate in a parent mantle-derived magma. Within the Timna igneous complex, the latest and most evolved intrusion, an alkaline granite, has the least contaminated isotope ratio (δ 18 O(Zrn) = 5.50 ± 0.02‰), whereas its inferred parental monzodiorite magma has slightly higher and more variable δ 18 O(Zrn) values (5.60 to 5.93‰). The small isotope variation may be accounted for either by small differences in the temperature of zircon crystallization or by minor contamination of the parent magma followed by shallow emplacement and intrusion by the Timna alkaline granite. The Timna alkaline granite evolved, however, from a non-contaminated batch of mantle-derived magma. The formation of Yehoshafat granite (605 Ma; δ 18 O(Zrn) = 6.63 ± 0.10‰), exposed ~30 km to the south of the mineralogically comparable Timna alkaline granite, involved significant contribution from supracrustal rocks. A-type granites in southern Israel thus formed by differentiation of mantle-derived magma and upper crustal melting coevally. Fast grain boundary diffusion modelling and measured quartz-zircon fractionations demonstrate that the Timna and Yehoshafat alkaline granites cooled very rapidly below 600 °C in accordance with being epizonal. One to three orders of magnitude slower cooling is calculated for 30 Ma older calc-alkaline granites of the host batholith, indicating a transition from thick orogenic to extended crust. Significant elevation of the δ 18 O of feldspars occurred through water–rock interaction at moderate temperatures (100–250 °C), most probably during a thermal event in Early Carboniferous times.

Muscovite dehydration melting in Si-rich metapelites: microstructural evidence from trondhjemitic migmatites, Roded, Southern Israel

Making a distinction between partial melting and subsolidus segregation in amphibolite facies migmatites is difficult. The only significant melting reactions at lowpressures, either vapour saturated or muscovite dehydration melting, do not produce melanocratic peritectic phases. If protoliths are Si-rich and K-poor, then peritectic sillimanite and K-feldspar will form in scarce amounts, and may be lost by retrograde rehydration. The Roded migmatites of southern Israel (northernmost Arabian Nubian Shield) formed at P = 4.5 ± 1 kbar and T ≤ 700 °C and include Si-rich, K-poor paragneissic paleosome and trondhjemitic leucosomes. The lack of K-feldspar in leucosomes was taken as evidence for the non-anatectic origin of the Roded migmatites (Gutkin and Eyal, Isr J Earth Sci 47:117, 1998). It is shown here that although the Roded migmatites experienced significant post-peak deformation and recrystallization, microstructural evidence for partial melting is retained. Based on these microstructures, coupled with pseudosection modelling, indicators of anatexis in retrograded migmatites are established. Phase diagram modelling of neosomes shows the onset of muscovite dehydration melting at 4.5 kbar and 660 °C, forming peritectic sillimanite and K-feldspar. Adjacent non-melted paleosomes lack muscovite and would thus not melt by this reaction. Vapour saturation was not attained, as it would have formed cordierite that does not exist. Furthermore, vapour saturation would not allow peritectic K-feldspar to form, however K-feldspar is ubiquitous in melanosomes. Direct petrographic evidence for anatexis is rare and includes euhedral plagioclase phenocrysts in leucosomes and quartz-filled embayments in corroded plagioclase at leucosome-melanosome interfaces. In deformed and recrystallized rocks muscovite dehydration melting is inferred by: (1) lenticular K-feldspar enclosed by biotite in melanosomes, (2) abundant myrmekite in leucosomes, (3) muscovite–quartz symplectites after sillimanite in melanosomes and associated with myrmekite in leucosomes. While peritectic K-feldspar formed in melanosomes by muscovite dehydration melting reaction, K-feldspar crystallizing from granitic melt in adjacent leucosome was myrmekitized. Excess potassium was used in rehydration of sillimanite to muscovite.

From ocean depths to mountain tops: Uplift of the Troodos ophiolite (Cyprus) constrained by low-temperature thermochronology and geomorphic analysis

The timing and mode of uplift of the Troodos ophiolite are constrained by low-temperature thermochronology combined with geomorphic analysis. Zircon (U-Th)/He and apatite fission track cooling ages in the Troodos plutonic sequence are all Cretaceous (83–106 Ma) and within error of published zircon U-Pb crystallization ages. This indicates early cooling of the oceanic crust and termination of spreading axis magmatism at ~90 Ma. Apatite (U-Th)/He ages decrease with reconstructed crustal depths from ~40 Ma near the top of the sheeted-dike complex to ~4 Ma within the mantle sequence. A prominent inflection point in the age versus depth curve defines the bottom of the exhumed helium partial retention zone and records the onset of rapid exhumation of the main Troodos massif at 6 ± 2 Ma. Inverse thermal modeling supports this conclusion, indicating that the timing of uplift is earlier than previously estimated. The boundaries of the mantle sequence exposed in the core of the Troodos structure closely overlap the boundaries of a concentric zone delineated by high local relief and higher channel steepness indices, indicating differential exhumation and uplift of this area relative to its surroundings. This zone also overlaps with a prominent negative Bouguer gravity anomaly. The timing and pattern of the Troodos ophiolite uplift suggest that it is driven by serpentinite diapirism, possibly triggered by Miocene reactivation of subduction along the Cyprean Arc. The worldwide ubiquity of suprasubduction zone ophiolites may thus reflect the importance of extensive serpentinization at the overthrusting mantle wedge in obduction processes.

Rare earth element evolution and migration in plagiogranites: a record preserved in epidote and allanite of the Troodos ophiolite

Plagiogranites from the Troodos ophiolite in Cyprus are occasionally epidotised, either partially or completely. Epidotisation phenomena include replacement of pre-existing minerals and filling of miarolitic cavities. In addition to epidote, miarolites in one plagiogranite body (located near the village of Spilia) contain coexisting ferriallanite-(Ce) and allanite-(Y). Textural and geochemical evidence indicates that late-stage REE-enriched granitic melt facilitated crystallisation of magmatic ferriallanite-(Ce). High REE contents persisted after fluid exsolution, causing crystallisation of allanite-(Y) from hydrothermal fluids in the miarolites. The REE pattern of the hydrothermal allanite-(Y) is characterised by LREE and Eu depletion, similar to the parent plagiogranitic magma. As allanite had sequestered most of the REE in the fluid, epidote took over as the principle hydrothermal mineral. Epidote in Troodos plagiogranites records a fluid evolutionary trend beginning with REE-rich–Eu-depleted similar to allanite-(Y) and gradually transforming into the REE-depleted–Eu-enriched pattern prevalent throughout ‘conventional’ sub-seafloor fluids. A comparison of allanite-bearing and allanite-absent plagiogranites from the same locality suggests that REE-bearing fluids migrated from the plagiogranites. Similar fluid evolution trends observed in diabase-hosted epidote, located adjacent to a large plagiogranite body, suggest influx of plagiogranite-derived REE-bearing fluids. Epidotisation in oceanic settings is usually considered to be the result of alteration by high fluxes of seawater-derived hydrothermal fluids. Although epidotisation by magmatic fluids has been suggested to occur in plagiogranites, our study shows that this autometasomatic process is the dominant mechanism by which epidosites form in plagiogranites. Furthermore, epidotisation of diabase has been attributed solely to seawater-derived fluids, but we show that it is possible for diabase-hosted epidosites to form by migration of plagiogranite-derived fluids.

Early Carboniferous anorogenic magmatism in the Levant: implications for rifting in northern Gondwana

ABSTRACT The Variscan orogenesis in Europe peaked during the Late Devonian–Early Carboniferous times when Gondwanan terranes collided with Laurasia. Hitherto it has been thought that Carboniferous tectonics in northern Arabia and the adjacent parts of NE Africa were broad swells (‘arches’) and depressions (‘basins’) that formed as a far-field contractional effect of the Variscan compression. The discovery of a 351 ± 3 Ma (U–Pb in zircon) within-plate felsic volcanism in the Helez borehole, southern coastal Israel, suggests that the Levant Arch is, instead, extensional in origin. Felsic volcanism was associated with gabbro underplating of the crust, an extreme (~50°C/km) crustal thermal gradient, major uplift, and truncation of the ≥2.5 km section. Taken together with the recent discovery of the ~340 Ma oceanic crust in the Eastern Mediterranean, the Levant Arch is interpreted as an uplifted shoulder of a rift, preceding ocean spreading.

The origin of celestine–quartz–calcite geodes associated with a basaltic dyke, Makhtesh Ramon, Israel

Spectacular celestine geodes occur in a Jurassic peri-evaporitic sequence (Ardon Formation) exposed in Makhtesh Ramon, southern Israel. The geodes are found only in one specific location: adjacent to an intrusive contact with a Lower Cretaceous basaltic dyke. Celestine, well known in sedimentary associations worldwide and considered as a low temperature mineral, may therefore be associated with magmatic-induced hydrothermal activity. Abundant fluid inclusions in celestine provide valuable information on its origin: gas-rich inclusions in celestine interiors homogenized at T>=200°C whereas smaller liquid-rich inclusions record the growth of celestine rims at T<=200°C. Near 0°C melting temperatures of some fluid inclusions and the occurrence of hydrous Ca-sulphate solid crystals in other inclusions indicate that celestine precipitated from variably concentrated Ca-sulphate aqueous solutions of meteoric origin. Celestine crystallized from meteoric water heated by the cooling basaltic dyke at shallow levels (c. 160 m) during a Lower Cretaceous thermal perturbation recorded by regional uplift and magmatism. The 87Sr/86Sr ratio of geode celestine, 0.7074, is similar to that measured in the dolostones of the host Jurassic sequence, but differs markedly from the non-radiogenic ratio of the dyke. Strontium in celestine was derived from dolostones preserving the 87Sr/86Sr of Lower Jurassic seawater, while sulphur (δ34S = 19.9‰) was provided by in situ dissolution of precursor marine gypsum (δ34S = 16.8‰) indicated by relict anhydrite inclusions in celestine. Low-temperature meteoric fluid flow during the Campanian caused alteration of the dyke into secondary clays and alteration of geodal celestine into quartz, calcite and iron oxides.